Home

R32C/100 Series C Compiler Package V.1.02 C Compiler User`s

1. Description 1 By using the above format to declare interrupt processing functions written in C NC100 generates the code for performing the following interrupt processing at the entry and exit points of the function In entry processing all registers of the Micro Procesor are saved to the stack In exit processing the saved registers are restored and control is returned to the calling function by the REIT instruction 2 You may specify either B or E of F in this declaration e B Instead of saving the registers to the stack when calling the function you can switch to the alternate registers This allows for faster interrupt processing e E Multiple interrupts are enabled immediately after entering the interrupt This improves interrupt response e F Return to th calling function by the FREIT instruction in exit processing e R Does not output the code that changes floating point rounding mode of FLG register to the nearest value e IN Only generates a vector table for interrupt functions and does not change generated code Use this switch primarily for fixed vectors 3 Interrupt vector numbers can be specified in a function declaration A variable vector table can be automatically generated by setting interrupt vector numbers before compiling the sources To use the assembly language startup program without specifying vector numbers refer to paragraph e Sett
2. Figure 2 6 Startup Program List 1 ncrt0 a30 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 21 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler if USTACKSIZE _ STACKSIZE equ 300h else for HEW STACKSIZE equ __USTACKSIZE _ endif if__ISTACKSIZE_ ISTACKSIZE equ 00h else for HEW ISTACKSIZE equ _ ISTACKSIZE__ endif list OFF include sect100 inc 5 list ON defines the user stack size defines the interrupt stack size defines the start address of interrupt vector table Includes sect100 inc Figure 2 7 Startup Program List 2 ncrt0 a30 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 22 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler SB AREA DEFINITION glb _ SB _ SB equ data_SB8_top INITIALIZE MACRO DEFINITION BZERO macro TOP_ SECT_ mov b 00H ROL mov TOP_ A1 mov sizeof SECT_ R7R5 sstr b endm BCOPY macro FROM_ TO_ SECT_ mov l FROM_ AO mov TO_Al mov sizeof SECT_ R7R5 smovf b endm Figure 2 8 Startup Program List 3 ncrt0 a30 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 23 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler INTERRUPT SECTION i insf start S 0 glb start section interrupt code align start 6 after reset this program will start Idc istack_top ISP istack p
3. REJ10J2009 0200 Rev 2 00 2010 04 01 221 2tENESAS R32C 100 Series C Compiler Package V 1 02 E Standard Library 9 Character String Value Convert Functions The following lists character string value convert functions Table E 10 Character String Value Convert Functions Function Contents Reentrant atof Converts a character string into a double type floatingpoint number O atoi Converts a character string into an int O atol Converts a character string into a long O strtod Converts a character string into a double O strtol Converts a character string into a long O strtou Converts a character string into an unsigned long type integer O 10 Multi byte Character and Multi byte Character String Manipulate Functions The following lists Multibyte Character and Multibyte Character string Manipulate Functions Table E 11 Multibyte Character and Multibyte Character String Manipulate Functions Function Contents Reentrant mblen Calculates the length of a multibyte character string O mbstowcs Converts a multibyte character string into a wide character string O mbtowc Converts a multibyte character into a wide character O westombs Converts a wide character string into a multibyte character string O wctomb Converts a wide character into a multibyte character O 11 Localization Functions The following lists localization functions Table E 12 Localization Function
4. Not enough core Insufficient swap area Increase the swap area Permission denied The processing system will not run Check access permission to the processing systems Or if access permission is OK check that the directory of the processing system is correctly set in the environment variable Yel can t open command file Can not open the command file specified by Specify the correct input file eij This error message is not normally displayed It is an internal error gt Compile options cannot be specified exceeding 99 characters too many options Result too large e This error message is not normally displayed It is an internal error Contact Renesas Solutions Corp This error message is not normally displayed It is an internal error Contact Renesas Solutions Corp Too many open files REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 303 R32C 100 Series C Compiler Package V 1 02 F Error Messages F 3 cpp100 Error Messages Table F3 to Table F5 list the error messages output by the cpp100 preprocessor and their countermeasures Table F 3 cpp100 Error Messages 1 3 NO Error message Description and countermeasure 1 illegal command option e Input filename specified twice Specify the input filename once only e The same name was specified for both input and output files Specify different names for input and output files Output filename
5. section fvector romdata org OFFFFFFDCH UDI word dummy_int OVER_FLOW word dummy_int BRKI word dummy_int word OFFFFFFFFH lword OFFFFFFFFH WDT word dummy_int word dummy_int NMI word dummy_int RESET word start IDcode DEFINITION l ID code check function id CodeChk IRI KIKI KIKI RIK IK II IK III III III IIIA IK IK III III IAI IIR kk IIIA IKI a 2 ey 3 End of R32C 100 start up 3 RII II KIKI KIA IK IK IIH IK IIIA IK IKI IAI IIR IK III IIA III IIR IAI kkk kk kkk a J Figure 2 32 Listing of sect100 inc in Single Chip Moce 4 e Setting Interrupt Vector Table In a program that uses interrupt processing set up an interrupt vector table by 1 Setting up the interrupt vector table of the vector section in sect100 inc The contents of interrupt vectors differ with each microcomputer type Make sure the interrupt vectors you ve set suit the microcomputer type you use For details refer to the user s manual of your microcomputer 1 When setting up the interrupt vector table in sect100 inc For programs that use interrupt processing change the interrupt vector table for the vector section in sect100 inc Figure 2 33 shows an example interrupt vector table REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 47 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler VARIABLE VECTOR SECTION f section vector romdata org VECTOR_ADR if _M I_ word dummy_
6. R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 SECTION rom_FAR ROMDATA ALIGN TO byte 66H byte 6cH ae byte 61H a byte 67H g byte 20H oo byte 3dH Set byte 20H 7 byte 25H byte 64H o byte OaH byte 00H END Compile End Time XX XXX XX XX XX XX XXXX Figure 1 11 Example Assembly Language Source File 2 sample a30 Let s look at the contents of the assembly language source files Items 1 to 2 correspond to 1 to 2 in Figure 1 10 1 Shows status of optimization option and information on the initial settings of the near and far attribute for ROM and RAM 2 When the nc100 command line option dsource dS is specified shows the contents of the C source file s as comments REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 12 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler Chapter 2 Basic Method for Using the Compiler This chapter describes how to start the compile driver nc100 and the command line options 2 1 Starting Up the Compiler 2 1 1 nc100 Command Format The nc100 compile driver starts the compiler commands cpp100 and ccom100 the assemble command as100 and the link command In100 to create a absolute module file The following information input parameters is needed in order to start nc100 1 Csource file s 2 Assembly language source file s 3 Relocatable object file s 4 Command line options opti
7. char port void func void port 0 Uses a variable before specifying pragma ADDRESS pragma ADDRESS port 100H Figure B 72 Cases where the specification of pragma ADDRESS has no effect The numeric representation in C language is used to write the absolute address in this declaration form REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 145 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma DMAC Specifies the DMAC register of a external variable Function The DMAC register inside CPU is assigned to the specified external variable Syntax pragma DMACAvariable nameADMAC Register Name Rules 1 You have to declare the variable specified to be pragma DMAC before description of pragma DMAC 2 It can be specified as pragma DMAC pragma DMAC register name and the type of a variable are as follows Register Name DMDO DCTO DCRO DDAO DDRO DSAO DSRO DMD1 DCT1 DCR1 DDA1 DDR1 DSA1 DSR1 DMD2 DCT2 DCR2 DDA2 DDR2 DSA2 DSR2 DMD3 DCT3 DCR3 DDA3 DDR3 DSA3 DSR3 Variable Type unsigned long To arbitrary models far pointer However the pointer to a function cannot be used 8 Two or more pragma DMAC cannot be declared to the same register 4 The amp address operator 0 function call operator subscript operator and gt ndirection operator cannot be specified to the variable specified by pragma DMAC 5 The variable specified by prag
8. lt A gt lt lt lt gt gt sizeof operator sizeof I REJ10J2009 0200 Rev 2 00 2010 04 01 7tENESAS 168 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications f Punctuators NC100 interprets the followings as punctuators e e e e o g Comment Comments are enclosed between and They cannot be nested Comments are enclosed between and the end of line C 2 2 Type a Data Type NC100 supports the following data type character type integral type structure union enumerator type void floating type b Qualified Type NC100 interprets the following as qualified type const volatile restrict near far REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 169 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications c Data Type and Size Table C 9 shows the size corresponding to data type Table C 9 Data Type and Bit Size Type Existence of sign Bit size Range of values _Bool No 8 0 1 char No 8 between 0 and 255 unsigned char signed char Yes 8 between 128 and 127 int Yes 16 between 32768 and 32767 short signed int signed short unsigned int No 16 between 0 and 65535 unsigned short int Yes 32 between 2147483648 and 2147483647 long signed int signed long unsigned int No 32 between 0 and 4294967295 unsigned long long long Yes
9. U predefined macro Undefines the specified predefined macro b Options Specifying Output Files Tabel 2 2 shows the command line option that specifies the name of the output machine language data file Tabel 2 2 Options for Specifying Output Files Option Function dir directory name Specifies the destination directory of the file s absolute module file map file etc generated by In100 ofile name Specifies the name s of the file s absolute module file map file etc generated by In100 This option can also be used to specify the destination directory Do not specify the filename extension c Version and command line Information Display Option Tabel 2 3 shows the command line options that display the cross tool version data and the command line informations Tabel 2 3 Options for Displaying Version Data and Command line informations Option Function v Displays the name of the command program and the command line during execution V Displays the startup messages of the compiler programs then finishes processing without compiling 1 Tf you do not specify command line options c E P or S nc100 finishes at In100 and output files up to the absolute load module file extension x30 are created REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 16 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler d Options for Debugging Tabel 2 4 show
10. gt Write correctly invalid member used e The member reference contains an error gt Write correctly invalid redefined type name of e The same identifier is defined more than once in typedef identifier Write the identifier correctly invalid return type e The type of return value of the function is incorrect gt Write it correctly invalid sign specifier e Specifiers signed unsigned are described twice or more Write the type specifier correctly invalid storage class for data e The storage class is erroneously specified gt Write it correctly invalid struct or union type e Structure or union members are referenced for the enumerated type of data Write it correctly ey invalid truth expression The void struct or union type is used in the first expression of a condition expression Use scalar type to write this expression ey invalid type specifier The same type specifier is described twice or more as in int int 1 or an incompatible type specifier is described as in float int 1 Write the type specifier correctly invalid type s bitfield A bit field of an invalid type is declared Use the integer type for bit fields invalid type specifier long long long Specifiers long are described thrice or more Check the type invalid unary operands Use of the unary operator is illegal Check the right side expression of the operator invalid unary operands Use
11. include Takes in the specified file line Specifies file s line numbers pragma Instructs processing for NC100 s extended function undef Undefines macros C 3 2 Preprocess Commands Reference The NC100 preprocess commands are described in more detail below assert Function Issues a warning if a constant expression results in zero 0 Format assertAconstant expression Description Issues a warning if a constant expression results in zero 0 Compile is continued however Warning cpp100 x c line xxJassertion warning REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 178 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications define Function Defines macros Format 1 defineAindentifierAlexical string opt 2 defineAidentifier identifier list opt Alexical string opt Description 3 Defines an identifier as macro 4 Defines an identifier as macro In this format do not insert any space or tab between the first identifier and the left parenthesis The identifier in the following code is replaced by blanks define SYMBOL When a macro is used to define a function you can insert a backslash so that the code can span two or more lines The following four identifiers are reserved words for the compiler RLE aaa Name of source file UNE Current source file line No _ DATE Date compiled mm dd yyyy
12. R32C 100 Series C Compiler Package V 1 02 3 Programming Technique b Returning Return Values from Assembler Functions When returning values from an assembler function to a C language program registers can be used through which to return the values for the integer pointer and floating point types Table 3 2 lists the rules on calls regarding return values Figure3 16 shows an example of how to write an assembler function to return a value Table 3 2 Return Value related Calling Rules Type of return value Rules char type ROL register _Bool type int type 16 bits RO register short int int type 82 bits R2RO0 register float type long type pointer type AO register long long type A1A0 register 32 high order and 32 low order bits stored in Al and AO registers double type respectively long double type structure type Immediately before the function call save the far address for the area for storing union type the return value to the stack Before execution returns from the called function that function writes the return value to the area indicated by the far address saved to the stack section program glo _asm_func _asm_func omitted mov 01A000H R2RO rts END Figure3 16 Example of Coding Assembler Function to Return long type Return Value c Referencing C Variables Because assembler functions are written in different files from the C program only the C gl
13. REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 101 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference near area Figure B 6 Memory Location of Pointer type Variable When near and far is explicitly specified determine the size of the address at which to store the variable and function that is written on the right side A declaration of pointer type variables that handle addresses is shown in Figure B 7 Example 1 short far ptr1 Example 1 short far ptr2 Figure B 7 Example of Declarning a Pointer Type Variable 1 As explained earlier unless near and far is specified the compiler handles the variable location as near and the variable type as far Therefore Examples 1 and 2 respectively are interpreted as shown in Figure B 8 Example 1 short far near ptr1 Example 2 short far far ptr2 Figure B 8 Example of Declaring a Pointer Type Variable 2 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 102 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference In Example 1 the variable ptr1 is a 4 byte variable that indicates the short type variable located in the far area The variable itself is located in the near area In Example 2 the variable ptr2 is a 4 byte variable that indicates the short type variable located in the far area The variable itself is located in the far area Memory mappings for Examples 1 and 2 are
14. section rom_EXT romdata align rom_EXT_top i section program _EXT code align Ei cn E S E E l FAR ROM SECTIONS section rom_FAR romdata org OFFE00000H rom_FAR_top Figure 2 30 Listing of sect100 inc in Single Chip Mode 2 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 45 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler section data_NEAR_INIT romdata org OFFFFOOOOH data_NEAR_INIT_top section data _MON1_INIT romdata align data_MON1_INIT_top section data MON2_INIT romdata align data_MON2_INIT_top section data _MON3_INIT romdata align data_MON3_INIT_top section data_MON4_INIT romdata align data_MON4_INIT_top section data _SB8_INIT romdata align data_SB8_INIT_top section data_SB16_INIT romdata align data_SB16_INIT_top section data _FAR_INIT romdata align data_FAR_INIT_top SWITCH TABLE SECTIONS section switch_table romdata align CODE SECTIONS section program code align section interrupt code align NEAR ROM SECTIONS section rom_NEAR romdata org OFFFF8000H srom_NEAR_top VARIABLE VECTOR SECTION f section vector i romdata org VECTOR_ADR vif _ MVT_ omitted endif l Figure 2 31 Listing of sect100 inc in Single Chip Moce 3 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 46 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler FIXED VECTOR SECTION
15. 04 option is specified Optimization is performed when all of the following conditions are met 1 Variables not including bit fields and unions 2 Variables for which the const qualifier is specified but are not specified to be volatile 3 Variables that are subject to initialization in the same C language source file 4 Variables that are initialized by constant or const qualified variables Ofile_inline inline expansion file OFI inline expansion file Function Supplement Inline expansion All inline functions are expanded in line Code generation for unreferenced static functions is suppressed Ifan inline expansion file is specified inline expansion is performed on global functions extending across a file boundary The inline expansion of global functions extending across a file boundary is performed on only those functions whose expanded size from to Y is within 150 lines including a comment line Note that the size can be changed using the option QOinline_line Although it normally is necessary that an inline function be declared before its entity can be defined use of this option permits the entity of an inline function to be defined before the inline function is declared The following shows an example of a program fragment where a function is inline expanded for the option Ofile_inline specified in it extern int i inline int func void The functio
16. Function Outputs one line to the stream Format include lt stdio h gt int fputs str stream Method function Argument const char _far str Pointer of the character string to be output FILE _far stream Pointer of the stream ReturnValue Returns 0 if output normally Returns any value other than 0 EOF if an error occurs Description Outputs one line to the stream REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 236 R32C 100 Series C Compiler Package V 1 02 E Standard Library fread Input Output Functions Function Reads fixed length data from the stream Format include lt stdio h gt size_t fread buffer size count stream Method function Argument void _far buffer Pointer of the location to be stored in size_t size Number of bytes in one data item size_t count Maximum number of data items FILE _far stream Pointer of stream ReturnValue Returns the number of data items input Description Reads data of the size specified in size from the stream and stores it in the buffer This is repeated by the number of times specified in count Ifthe end of the stream is encountered before the data specified in count has been input this function returns the number of data items read up to the end of the stream Interprets code 0x1A as the end code and ignores any subsequent data free Function Frees the allocated memory area Format include lt stdlib h gt void free cp Method
17. R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 11 ccom100Error message 5 13 Error message Description and countermeasure inline function s address used e An attempt is made to reference the address of an inline function Do not use the address of an inline function inline function s body is not declared previously Ye The body of an inline function is not defined Using an inline function define the function body prior to the function call inline function functionname is e The recursive call of an in line function cannot be carried recursion out Using an inline function No recursive interrupt function called e The function specified by pragma INTERRUPT is called Be careful not to call an interrupt handling function invalid environment variable e The variable name specified in the environment variable environment variable name NCKIN NCKOUT is specified by other than SJIS and EUC Check the environment variables used invalid function default argument Yoel The default argument to the function is incorrect This error occurs when the prototype declaration of the function with default arguments and those in the function definition section do not match Make sure they match invalid push An attempt is made to push void type in function argument etc The type void cannot be pushed invalid operand The operation contains an error Ch
18. _ TIME Time compiled hh mm ss The following are predefined macros in NC100 R32C100 NC100 __INT_16__ When compilation option fl16 fint_16 is used it is defined CHAR_SIGNED__ When compilation option fSC fsigned_char is used it is defined You can use the token string operator and token concatenated operator with tokens as shown below define debug s t printf x s Yd x t d x sx t When parameters are specified for this macro debug s t as debug 1 2 they are interpreted as follows define debug s t printf x1 d x2 d x1 x2 Macro definitions can be nested to a maximum of 20 levels as shown below define XYZ1 100 define XYZ2 XYZ1 abbreviated define XYZ20 XYZ19 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 179 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications error Function Suspends compilation and outputs the message to the standard output device Format errorAcharacter string Description Suspends compilation lexical string is found this command outputs that character string to the standard output device if elif else endif Function Performs conditional compilation Examines the expression true or false Format ifAconstant expression elifAconstant expression else endif Description If the value of the constant is true
19. section bss_MONS data align bss_MON3_top section data_MON4 data align data_MON4_ top section bss_MON4 data align bss_MON4_top STACK SECTION section stack data align blkb STACKSIZE align stack_top blkb ISTACKSIZE align istack_top Figure 2 29 Listing of sect100 inc in Single Chip Mode 1 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 44 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler HEAP SECTION section heap data align heap_top blkb HEAPSIZE ere ere ee ee ee 4 3 SB RELATIVE RAM SECTIONS Decent ee aia SERRE section data SB8 data 3 org 00008000H data_SB8_top l l section bss SB8 data align i bss_SB8_top E section data_SB16 data align data_SB16_top lhe section bss _SB16 data align bss_SB16_top E SE ee eR eS E S Le ee i FAR RAM SECTIONS You can remove this part because it is NETE S eee ot S T ne unnecessary E data_FAR_top Seeion cata BAR data saion In this case you need to remove the eae section bss_FAR data align nera T in the far area of bss _FAR_top ae pays a eee EXTENDED RAM SECTIONS E section data_EXT data org 00800000H data_EXT_top i section bss EXT data align bss_EXT_top l lre EXTENDED ROM SECTIONS Weenie ce neetn ones earn ar a nn oc l section data _EXT _ INIT romdata org OFFOO0000H data EXT_INIT_top
20. sin Calculates sine sinh Calculates hyperbolic sine sqrt Calculates the square root of a numeric value tan Calculates tangent tanh Calculates hyperbolic tangent REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 211 R32C 100 Series C Compiler Package V 1 02 E Standard Library Taa Function Defines the structures used in branch functions Function Contents longjmp Performs a global jump setjmp Sets a stack environment for a global jump signal h Function Defines declares necessary for processing asynchronous interrupts REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 212 R32C 100 Series C Compiler Package V 1 02 E Standard Library ee Function Defines declares the functions which have a variable number of real arguments stddef h Function Defines the macro names which are shared among standard include files stdio h Function Defines the FILE structure stream name and declares I O function prototypes Prototype declarations are made for the following functions Type Function Function Initialize init Initializes R32C 100 family input outputs clearerr Initializes clears error status specifiers Input fgetc Inputs one character from the stream getc Inputs one character from the stream getchar Inputs one character from stdin fgets Inputs one line from the stream
21. 1 Fatal error message This error message is not normally output Please contact nearest Renesas office with details of the message if displayed REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 301 R32C 100 Series C Compiler Package V 1 02 F Error Messages F 2 nc100 Error Messages Table F1 and Table F2 list the nc100 compile driver error messages and their countermeasures Table F 1 nc100 Error Messages 1 2 Error message Description and countermeasure Arg list too long e The command line for starting the respective processingsystem is longer than the character string defined bythe system Specify a NC100 option to ensure that the number ofcharacters defined by the system is not exceeded Usethe v option to check the command line used for eachprocessing block Cannot analyze error This error message is not normally displayed It is aninternal error Contact Renesas Solutions Corp command file line characters exceed 2048 There are more than 2048 characters on one or more lines in the command file Reduce the number of characters per line in the commandfile to 2048 max Core dump command_name The processing system indicated in parentheses caused a core dump The processing system is not running correctly Checkthe environment variables and the directory containingthe processing system If the processing system stilldoes not run correctly Please contact Renesas SolutionsCorp Exec format
22. CENESAS 7 D me T lt D D R32C 100 Series C Compiler Package V 1 02 C Compiler User s Manual Renesas FIEC ironics Rev 2 00 2010 04 www renesas co 10 11 12 Notice All information included in this document is current as of the date this document is issued Such information however is subject to change without any prior notice Before purchasing or using any Renesas Electronics products listed herein please confirm the latest product information with a Renesas Electronics sales office Also please pay regular and careful attention to additional and different information to be disclosed by Renesas Electronics such as that disclosed through our website Renesas Electronics does not assume any liability for infringement of patents copyrights or other intellectual property rights of third parties by or arising from the use of Renesas Electronics products or technical information described in this document No license express implied or otherwise is granted hereby under any patents copyrights or other intellectual property rights of Renesas Electronics or others You should not alter modify copy or otherwise misappropriate any Renesas Electronics product whether in whole or in part Descriptions of circuits software and other related information in this document are provided only to illustrate the operation of semiconductor products and application examples You are fully responsible for the inco
23. Depends on amount of available memory function Maximum number of parameters or macro call parameters 31max per macro Maximum number of characters in character string literals Depends on amount of available memory after concatenation Maximum size in bytes of object Depends on amount of available memory Maximum number of members per structure union Depends on amount of available memory Maximum number of enumerator constants per numerator Depends on amount of available memory Maximum number of levels of nesting of structures or Depends on amount of available memory unions per struct declaration list Maximum number of characters per character string Depends on operating system Maximum number of lines per file Depends on amount of available memory REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 165 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications C 2 Standard Language Specifications The chapter discusses the NC100 language specifications with the standard language specifications C 2 1 Syntax This section describes the syntactical token elements In NC100 the following are processed as tokens Key words Identifiers Constants Character literals Operators Punctuators Comment a KeyWords NC100 interprets the followings as key words Table C 5 Key Words List _asm _far _ near asm auto _ Bool _ inline
24. G2 OQutlineof ensi ietan niani airn ai a aaien aeaaee E anaE aN ERNE EESE SRi oi 330 G 2 1 Processing Outline of Zensni c cccccssssssssssssssssssssscsessssssssssssssessssessssssescsessssesessssescseessseseesssesnsaesssnsesseaeseeeees 330 G3 gt GAINS cens TAN saen an anA AAA ON hdgesesedios sues seeeesdsvavanveassevous ads ieeseodestecenienee 332 G 3 1 Input format iiss sccisadesisdscsatsisaui ad atiniiti danidaitnariiadniadini darad aud uidarhaniiacdiaaien 332 G 3 2 Option References nasrane enni a eai a a i aa ani ao eia aan iiia 333 GA Error Messagesof Ceri Sint cc d iceiecissscencccedss ccossd Siesaesaccdensecesctedsaleccessaieccusd ANA NAE NENEA AA aE 334 G 4 1 Error Messaees stvisitasaiditeaatinasasinilaiiniddianadainasadiniddinaaiiiadaiaidaiawadinagd ad 334 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 Chapter 1 Introduction to NC100 This chapter introduces the processing of compiling performed by NC100 and provides an example of program development using NC100 1 1 NC100 Components NC100 consists of the following five executable files 1 nc100 C Compile driver 2 igen100 C Inline generator 3 cpp100 C Preprocessor 4 ccom100 C Compiler 5 aopt100 Assembler optimizer 6 Call Walker amp gensni Stack analysis tool amp Stack information analysis utility 7 MapViewer Map Viewer REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 1 R32C 10
25. R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma INTCALL Declare a function called the INT instruction Example int asm_func unsigned long unsigned short Prototype declaration for the pragma INTCALL 25 asm_func R2R0 R1 assembler function void main void int i long l i 0x7FFD 0x007F asm_func i Calling the assembler function Figure B 74 Example of pragma INTCALL Declaration asm function 1 int c_func unsigned int unsigned int lt Prototype declaration for the C function pragma INTCALL 25 c_func You may NOT specify registers void main void int i j i 0x7FFD j 0x007F c_func i j Calling the C function Figure B 75 Example of pragma INTCALL Declaration C language function 2 Supplement To use the startup file included with the product alter the content of the vector section before use For details on how to alter it refer to Preparing the Startup Program REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 148 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma INTERRUPT Function Declares an interrupt handler Syntax 1 pragma INTERRUPTA B E F R V Ainterrupt handler name 2 pragmaINTERRUPTA B E F R Ainterrupt vector numberAinterrupt handler name 3 pragmaINTERRUPTA B E F R Ainterrupt handler name vect interrupt vector number
26. REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 141 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma STRUCT Control structure mapping Function 1 Inhibits packing of structures 2 Arranges structure members Syntax 1 pragma STRUCTAstructure_tagAunpack 2 pragma STRUCTAstructure_tagAarrange Description In NC100 structures are packed For example the members of the structure in Figure B 67 are arranged in the order declared without any padding struct s Member Type Size Mapped short i name location char C offset short j i short 16 bits 0 c char 8 bits 2 j shortt 16 bits 3 Figure B 67 Example Mapping of Structure Members 1 Rules 1 Inhibiting packing This NC100 extended function allows you to control the mapping of structure members Figure B 68 is an example of mapping the members of the structure in Figure B 67 using pragma STRUCT to inhibit packing struct s Member Mapped short i name location char c offset Figure B 68 Example Mapping of Structure Members 2 As shown Figure B 68 if the total size of the structure members is an odd number of bytes pragma STRUCT adds 1 byte as packing after the last member Therefore if you use pragma STRUCT to inhibit padding all structures have an even byte size REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 142 R32C 100 Series C Compiler Packa
27. Storage class Specifier NC100 supports the following storage class specifier o extern o static 2 Type Declarators NC100 supports the following type declarators Bool char int short Jong long long float double unsigned signed struct union e enum 3 Declaration Specifier Use the format of declaration specifier shown in Figure C 5 in NC100 Declarator Pointer opt declarator2 Declarator2 identifier list of dummy argument opt declarator declarator constant expressiono opi declarator list of dummy argument opt Only the first array can be omitted from constant expressions showing the number of arrays opt indicates optional items The list of dummy arguments is replaced by a list of type declarators in a prototype declaration Figure C 5 Format of Declaration Specifier REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 174 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications 4 Body of the Program Use the format of body of the program shown in Figure C 6 List of Variable Declaratoropt Compound Statement There is no body of the program in a prototype declaration which ends with a semicolon opt indicates optional items Figure C 6 Format of Body of the Program C 2 5 Statement NC100 supports the following Labelled Statement Compound Statement Expression Null Statement Selectio
28. The following types are supported for variable names d Variable name Array name integer e Struct name member name not including bit field members void func void short idata short a3 struct TAG short i short k asm MOVW PO FB idata asm MOVW RO FB a 2 asm MOVW PO FBI s i Remainder omitted asm MOVW FB FBI si a2 Figure B 18 Description example for specifying Figure B 19 shows an example for referencing an auto variable and its compile result REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 108 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference C source file void func void short idata 1 asm MOV W FB RO idata asm CMPW 00001H RO remainder omitted Assembly language source file compile result gt HE FUNCTION func HHH FRAME AUTO idata size 2 offset 4 HH FRAME AUTO __PAD2 size 1 offset 1 HH H FRAME AUTO __PAD1 size 1 offset 2 HH H ARG Size 4 Auto Size 4 Context Size 8 abbreviated HHEC SRC asm MOV W FB RO idata HHH ASM START MOV W _ 4 FB RO _line 5 HHC_SRC asm CMP W 00001H RO CMPW 00001H RO HHE ASM END remainder omitted Figure B 19 Example for Referencing an auto Variables You can also use the format show in Figure B 20 so that auto variables in an asm function use a 1 bit field Can not operate bit fie
29. invalid pragma SECTION unknown section base name The section name in pragma SECTION contains an error The section names that can be specified are data bss program rom interrupt and bas This line will be ignored Write the description correctly invalid pragma operand ignored An operand of pragma contains an error This line will be ignored Write the description correctly invalid function argument The function argument is not correctly written Write the function argument correctly invalid return type eje Y The expression of the return statement does not match the type of the function Make sure that the return value is matched to the type of the function or that the type of the function is matched to the return value invalid storage class for function change to extern An invalid storage class is used in function declaration It will be handled as extern when processed Change the storage class to extern Kanji in pragma ADDRESS e The line of pragma ADDRESS contains kanji code This line will be ignored Do not use kanji code in this declaration Kanji in pragma BITADDRESS e The line of pragma BITADDRESS contains kanji code This line will be ignored Do not use kanji code in this declaration keyword keyword are reserved for future A reversed keyword is used Change it to a different name large type was implicitly cast to small type eu ey y The up
30. malloc in the program Set the heap to 0 if you do not use these memory management functions Make sure that the heap section does not exceed the physical RAM area HEEP SIZE definition if__HEAP__ for HEW HEAPSIZE equ Oh else jf HEAPSIZE HEAPSIZE equ 300h else for HEW HEAPSIZE equ __HEAPSIZE _ endif endif Figure 2 21 Example of Setting Heap Section Size ncrt0 a30 f Setting the interrupt vector table Set the top address of the interrupt vector table to the part of Figure 2 22 in ncrt0 a30 The INTB Register is initialized by the top address of the interrupt vector table VECTOR_ADR equ OFFFFFBDCH Figure 2 22 Example of Setting Top Address of Interrupt Vector Table ncrt0 a30 The sample startup program has had values set for the tables listed below OFFFFFBDCH OFFFFFFDBH Interrupt vector table OFFFFFFDCH OFFFFFFFFH Fixed vector table Normally these set values do not need to be modified REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 36 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler g Setting the Processor Mode Register Set the processor operating mode to match the target system at address 04H Processor mode register in the part of nert0 a30 shown in Figure 2 23 i mov b 00H 04H processor mode Figure 2 23 Example Setting of Processor Mode Register ncrt0 a30 See the User s Manual of microcomputer you are usin
31. 0200 Rev 2 00 2010 04 01 171 7tENESAS R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications Table C 11 Types of Expressions and Their Elements 2 2 Type of expression Elements of expression Assign expression monadic expression expression monadic expression expression monadic expression expression monadic expression expression monadic expression expression monadic expression lt lt expression monadic expression gt gt expression monadic expression amp expression monadic expression expression monadic expression expression assignment expression Comma operator expression monadic expression C 2 4 Declaration There are following two types of declaration o Variable Declaration Function Declaration a Variable Declaration Use the format shown in Figure C 1 to declare variables storage class specifier Atype declarator Adeclaration specifier Ainitialization_expression Figure C 1 Declaration Format of Variable 1 Storage class Specifiers NC100 supports the following storage class specifiers extern auto static register typedef 2 Type Declarator NC100 supports the type declarators Bool char int short long long long float double unsigned signed struct union enum 3 Declaration Specifier Use the format of declaration specif
32. 1 n100 Command POwimarts sciisscasscesciseiscaccsasdedbacasssessacaatecascscacasecsesebelsscdeacacasobsadasdaancssosasceceorse TEAS EENE sai 13 2 1 2 Command TIE uari a oE E E NEE E AEN TENE EE EE ERN 14 2 1 3 Notes on NC100 Command Line OptionS sesssesseesseeseeesseesseesseesseesseesseesseesseesseenseessressersseesseessresserssrersseessee 15 2 1 4 nc100 Command Line Options ccccsesssssssessssssscssssssessesesescsescsescsescscscsssescsesesssssesesesesesesescaeseseseessesceneceseees 16 2 2 Preparing the Startup Progyam c cccccssscsesssssssescssscsescssscssescsesssessscsssescscscscscsssescecssseessessseseseseseacseseesaeecseeceneseaeees 21 2 2 1 Sample of Startup Program cccccccssssssssssessscscsescsescsescscscscsessscsesssssccsssesssesesescsesesescasscsescseecseeseaeacsescecscseseseees 21 2 2 2 Customizing the Startup Program ccccssssssssessssssssscscsssssssscssssssssessscsssescscscsescscscssscscscscsesecssssesesceeseesseseees 33 2 2 3 Customizing for NC100 Memory Mapping cccscssssssssssssssesesessssesesessesesnssssesesessesesesssseseeesesesnsseseeseaeaes 37 Chapter 3 Programming Technique c cccccssscssscsescsssesescssssssscssscssscssesssesescsescssscsescsesssusssnssesssesessacsescecscsescsescaescsescaneceesesesesees 50 3 1 Notessa maen ines dads is dewshehinn A A E aged boast oan Giese Oona Rie 50 3 1 1 Notes about Version up Of compiler ccccescssssssssescsescsesesescsescsescssscssssesesesesesesssesc
33. 290 E 3 Modifying Standard Library ccccccssssssessssssssessssssesessssssesessesesesesscseseesssesesessesesessesesesesseseseeseseeseseseeseseaeseeeeseaeens 291 E 3 1 piruictuire of VO Munctions arrani enen nni rn a RAN R N R NR RN 291 E 3 2 Sequence of Modifying I O Functions c c ccccssssssssesssescsssssssessssssescsesesescsescsescssscsescseessssesssesseeseseasseseeeseesees 292 Appendix F Error Messages i s sccsetieiseiecsceicveessvescacesusntesuessvciotieacbesteesisadeboveseceeaseieadobcnsnbese Aa e TEKE NRE TEDN TENi 301 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS R32C 100 Series C Compiler Package V 1 02 Contents Fl Message Format sninsmenonsissnmooiimonnina adenta inact n E E riaa 301 F2 TIC TOO Error Mossas sarine eei aa e a E e a N a E EE 302 E3 CPP LOO Error Messa SCS 5 so5c55ciescsidesccseessesscesessceccesasacecessavaccesdsssacevacde tse sasdeeveseddeseses0sveesasacsesesedssoess sesidoscesuseatscetiaveds 304 F4 CpplOO Warning Messavesicisscciesciedcasdscscicdacssscuisdancacaiades tiadanea cei dastaasaaarad NTARE 307 F5 CCOMITOO Error MESSAGES riiin iee ean E E Ea ERa oe E E N E RE 308 F6 ccecom100 Warning Messages e sseseesseesessesseescessesseneesssesesnesseeseeneosstssentoostssenseosteseossesstneesseostsssenetostesesseosseseeseesse 321 Appendix G Using gensni or the stack information File Creation Tool for Call Walker c ccccssccesseeseseseseessesees 330 GI Starting Call Walker escossa r EAEEREN NEER 330
34. 4 5 6 7 8 9 10 In this case the array array is located as ROM area By specifying this option you can locate the array in the RAM area You do not normally need to use this option fdouble_32 fD32 Modify generated code Function This option specifies that the double type be handled in 32 bit data length as is the float type Supplement 1 When specifying this option always make sure the prototype of the function is declared If no prototype declarations exist invalid code may be generated 2 When this option is selected the debug information for type double is handled as type float In the C watch window or global window etc of the emulator debugger or simulator debugger said information is displayed as type float REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 83 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference fenable_register fER ee storage class Function Allocates variables with a specified register storage class to registers Supplement When optimizing register assignments of auto variables it may not always be possible to obtain the optimum solution This option is provided as a means of increasing the efficiency of optimization by instructing register assignments in the program under the above situation When this option is specified the following register specified variables are forcibly assigned to registers Integral type variable Floating point variab
35. 64 between 9223372036854775808 and signed long long 9223372036854775807 unsigned long long No 64 18446744073709551615 float Yes 32 between 1 17549435e 38F and 3 40282347e 38F double Yes 64 between 2 2250738585072014e 308 and long double 1 7976931348623157e 308 far pointer No 32 between 0 and OxFFFFFFFF J The _Bool type can not specify to sign J If a char type is specified with no sign it is processed as an unsigned char type Ifan int or short type is specified with no sign it is processed as a signed int or signed short type Ifa long type is specified with no sign it is processed as a sign long type J Ifa long long type is specified with no sign it is processed as a sign long long type J If the bit field members of a structure are specified with no sign they are processed as unsigned o Can not specifies bit fields of long long type Type int is handled in 32 bits However if the compile option fI16 fint_16 is specified int is handled in 16 bits REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 170 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications C 2 3 Expressions Table C 10 and Table C 11 show the relationship between types of expressions and their elements Table C 10 Types of Expressions and Their Elements 1 2 Type of expression Elements of expression Primary expression identifier constant character literal ex
36. Declaration of Function NC100 allows you to use implicit arguments by assigning parameter default values when declaring a function s prototype By using this facility you can save the time and labor that would otherwise be required for writing frequently used values when calling a function B 4 2 Format of Default Argument Declaration of Function Figure B 46 shows the format used to declare the default arguments of a function Storage class specifier A Type declarator ADeclarator Dummy argument Default value or variable Figure B 46 Format for declaring the default arguments of a function Figure B 47 shows an example of declaration of a function and Figure B 48 shows a result of compiling of sample program which shows at Figure B 47 short func short i 1 short j 2 Declares the default values of parameters in the arguments to the function func as first argument 1 and second argument 2 void main void func lt The actual argument consists of the first argument 1 and the second argument 2 func 3 The actual argument consists of the first argument 3 and the second argument 2 func 3 5 The actual argument consists of the first argument 3 and the second argument 5 Figure B 47 Example for declaring the default arguments of a function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 122 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference HHHEC
37. Format nci00 osample as100 I In100 ms ncrt0 a30 sample c lt RET gt lt RET gt Return key Always specify the startup program first when linking Figure A 2 Example nc100 Command Line REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 63 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 nc100 Command Line Options A 2 1 Options for Controlling Compile Driver Table A 1 shows the command line options for controlling the compile driver Table A 1 Options for Controlling Compile Driver Option Function c Creates a relocatable file extension r30 and ends processing 1 Didentifier Defines an identifier Same function as define dsource Generates an assembly language source file extension a30 with a C Short form dS language source list output as a comment Not deleted even after assembling dsource_in_list In addition to the dsource dS function generates an assembly Short form dSL language list file Ist E Invokes only preprocess commands and outputs result to standard output I directory Specifies the directory containing the file s specified in include You can specify up to 256 directories P Invokes only preprocess commands and creates a file extension i S Creates an assembly language source file extension a80 and ends processing silent Suppresses the copyright message display at startup Upredefine
38. I E EEEE E E lt nczevassestettateasteatecsasaveanstavatee 3 1 3 NOTES nna oiea anA A A N EEEN EE ANERE E A NR EA E EES A ia 4 1 3 1 Notes about Version up of compiler c ecssssessssesessssescseescsesesessesesesscsesesessesesesseseseesseseseseseseseseseeeseaeeneseseans 4 1 3 2 Notes about the R32C s Type Dependent Part ccccccsssssssssescsessssssscssssssscesssescsescsescsescsescssseseecseessneeeseeeseees 4 1 4 Example Program Development cccssssssssssssssssssssssssssssnsssssssssssesssusssssaessssesssussssusuesssesscucseseaesasacsusseseaeeeseeaceneeees 5 1 5 NC 100 Output Files id eseinniaadiiesiinnialnsiatanielwadtincusiewiiesviiaadiaa sania laminas 7 1 5 1 Introduction to Output Files ccccccccsscssscssscscsescsescsesescscsssescscsesssescsssscsesesesescsescseacsescsescssscscesesesesesesssescseseeeaees 7 1 5 2 Preprocessed C Source Files ccccssssssssssssssscscscsssssscscssssescssssesescssssssessecsesceeesesesesseseseeseeseseeseseeseseseseeseaessesenes 8 1 5 3 Assembly Language Source Files ccccscsssssssssssssssscsssescsescsescssscsescssscsescssssesssesssesesesesescsesesescssacseensseseeeseneaces 10 Chapter 2 Basic Method for Using the Compiler c cccsssssssssssesssssssscssssssessssssescscesssesssescseseeseescsecacsceseseasaseeaeseesesessanenes 13 2 1 Starting Up the Compiler ccccesssssssssssssssesssescsesesescssscssscssssssssesssssscsessscsssessscsscescsescssscesessscessesescsesesceeseeseeesees 13 2 1
39. Mapping Sections to Memory Because of the effect on the memory attributes RAM and ROM of Micro Processor memory some sections can only be mapped to specific areas Apply the following rules when mapping sections to memory 1 Sections mapped to RAM stack section heap section data NEAR section bss NEAR section data _FAR section bss FAR section data_EXT section bss EXT section data _SB8 section bss SB8 section data _SB16 section bss SB16 section data MONI1 section bss MONI1 section data MONZ2 section bss MONZ2 section data_MONs section bss_MONS section data_MON4 section bss_MON4 section 2 Sections mapped to ROM program section program_EXT section interrupt section fvector section switch_table section rom_NEAR section rom_EXT section rom FAR section data NEAR INIT section data_MON1_INIT section data_FAR_INIT section data MON2_ INIT section data_EXT_INIT section data MON3_INIT section data_SB8_INIT section data MON4 INIT section data_SB16_INIT section Note also that some sections can only be mapped to specific memory areas in the Micro Processor memory space 1 Sections mapped only to OH O7FFFH OFFFF8000H OFFFFFFFFH near area data_NEAR section rom_NEAR section bss_NEAR section 2 Sections mapped only to OH O7FFFFFH OFF800000H OFFFFFFFFH farr area program section interrupt secti
40. R32C 100 Series C Compiler Package V 1 02 E Standard Library memicmp Memory Handling Functions Function Compares memory areas with alphabets handled as upper case letters Format include lt string h gt int memicmp s1 s2 n Method function Argument char _far s1 Pointer to the first memory area to be compared char _far s2 Pointer to the second memory area to be compared size tn Number of bytes to be compared ReturnValue Return Value 0 The two memory areas are equal Return Value gt 0 The first memory area s1 is greater than the other Return Value lt 0 The second memory area s2 is greater than the other Description Compares memory areas with alphabets handled as upper case letters When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization memmove Memory Handling Functions Function Moves the area of a character string Format include lt string h gt void _far memmove sl s2 n Method function Argument void _far s1 Pointer to be moved to const void _far s2 Pointer to be moved from size_t n Number of bytes to be moved ReturnValue Returns a pointer to the destination of movement Description When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 257 R32C 100 S
41. RAM monitor area are not subject to size limitations 4 The location address of the section allocated by pragma MONITORIn and a process to set the initial value for the external variable should be written in the startup program 5 If pragma MONITORInI is declared a number times for one and the same external variable the pragma MONITORIn declared first is effective 6 The external variables declared by pragma MONITORIn are not affected by pragma SECTION 7 The declaration of pragma MONITORIn has no effect if n in it is other than 1 4 If the compile option Wunknown_pragmal WUP or Wall is specified a warning is output 8 External variables with ROM attribute cannot be handled by pragma MONITORIn However if the compile option fconst_not_ROMI fCNR is specified these variables can be handled by pragma MONITORIn I pramga MONITORI i const int i lt Has no effect 9 Even when variable locations are changed by this function the addressing mode of generated code is not changed If the locations of variables that are to be stored in a near attribute RAM area addresses 00000000H to 00007FFFH are changed the section that contains those variables must be located in a near area except for the ROMDATA attribute sections of initial values REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 135 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma MONITORn Directive to specify
42. UARTO can be selected using the conditional compile commands as follows J To use UARTO ccceceeseseseeteees define UARTO 1 Specify these commands at the beginning of device c or specify following option when compiling To use UART0 ce ecccceseeeeteenees DUARTO To use both UARTs modify the file as follows 1 Delete the conditional compiling commands from the beginning of the device c file 2 Change the UARTO special register name defined in pragma EQU to a variable other than UART1 3 Reproduce the level 3 functions _sget and _sput for UARTO and change them to different variable names such as _sgetO and _sput0 4 Also reproduce the speed function for UARTO and change the function name to something like speedo This completes modification of device c Next modify the init function init c which makes the initial I O function settings then change the stream settings see below REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 293 R32C 100 Series C Compiler Package V 1 02 E Standard Library b Stream Settings The NC100 standard library has five items of stream data stdin stdout stderr stdaux and stdprn as external structures These external structures are defined in the standard header file stdio h and control the mode information of each stream flag indicating whether input or output stream and status information flag indicating error or EOF Table E 15 Stream Information St
43. Write it correctly pragma pragma name not function e A name is written in the pragma pragma name that is ignored not a function gt _ Write it with a function name pragma pragma name s function must e A function specified in the pragma pragma name is not be predeclared ignored declared For functions specified in a pragma pragmaname write prototype declaration in advance pragma pragma name s function must e A function specified in the pragma pragma name is not be prototyped ignored prototype declared For functions specified in a pragma pragmaname write prototype declaration in advance pragma pragma name s function e The type of return value for a function specified in the return type invalid ignored pragma pragma name is invalid gt Make sure the type of return value is any type other than struct union or double pragma pragma name unknown e The switch specified in the pragma pragma name is switch ignored invalid gt Write it correctly pragma pragma name variable e The variable specified in pragma pragma name is initialized initialization ignored initialized The specification of pragma pragma name will be nullified Delete either pragma pragma name or the initialize expression pragma ASM line too long then cut e The line in which pragma ASM is written exceeds the allowable number of characters 1 024 bytes gt _ Write it within 1 024 bytes REJ10J2009 0200 Rev 2 00 2010 04 01 7t
44. _far stream Pointer of stream const char _far format Pointer of the format specifying character string va_list ap Pointer of argument list ReturnValue Returns the number of characters output Description Output toa stream with format When writing pointers in variable length variables make sure they are a far type pointer vprintf Input Output Functions Function Output to stdout with format Format include lt stdarg h gt include lt stdio h gt int vprintf format ap Method function Argument const char _far format Pointer of the format specifying character string va_list ap Pointer of argument list ReturnValue Returns the number of characters output Description Output to stdout with format When writing pointers in variable length variables make sure they are a far type pointer REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 287 R32C 100 Series C Compiler Package V 1 02 E Standard Library vsprintf Input Output Functions Function Output to a buffer with format Format include lt stdarg h gt include lt stdio h gt int vfprintf s format ap Method function Argument char _far s Pointer of the location to be store const char _far format Pointer of the format specifying character string va_list ap Pointer of argument list ReturnValue Returns the number of characters output Description When writing pointers in variable length variables make sure they
45. a a ae Fint fune no proto hjk l 13 int i 1 int j l int k l return k j i void main void 4 1 i int sum sum func_proto 1 2 3 5 l sum func_no_proto 1 2 3 ee l ae E E E 1 This is the prototype declaration of function func_proto 2 This is the body of function func_proto Prototype declaration is entered so this is a new format 3 This is the body of function func_no_proto This is a description in K amp R format that is an old format 4 This is the body of function main 5 This calls function func_proto 6 This calls function func_no_proto Figure D 10 Sample Program for Calling a Function sample c The compile result of the above sample program is shown in the next page Figure D 11 shows the compile result of program part 2 that defines function func_proto Figure D 12 shows the compile result of program part 3 that defines function func_no_proto Figure D 13 shows the compile result of program part 4 that calls function func_proto and function func_no_proto REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 195 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules HH FUNCTION func_proto REGISTERARG i size 4 REGISTER R2RO 9 REGISTERARG j size 4 REGISTER R3R1 lt 8 REGISTERARG k size 4 REGISTER R6R4 lt 07 HHH ARG Size 0 Auto Size 0 Context Size 4 SECTION program CODE ALI
46. are a far type pointer REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 288 R32C 100 Series C Compiler Package V 1 02 E Standard Library wcstombs Multi byte Character Multitoyte Character String Manipulate Functions Function Converts a wide character string into a multibyte character string Format include lt stdlib h gt size_t _far wcstombs s wcs n Method function Argument char _far s Pointer to an area for storing conversion multibyte character string const wchar t _far wes Pointer to a wide character string size_t n Number of wide characters stored ReturnValue Returns the number of stored multibyte characters if the character string was converted correctly Returns 1 if the character string was not converted correctly wctomb Multi byte Character Multitoyte Character String Manipulate Functions Function Converts a wide character into a multibyte character Format include lt stdlib h gt int wctomb s wchar Method function Argument char _far s Pointer to an area for storing conversion multibyte character string wchar_t wchar wide character ReturnValue Returns the number of bytes contained in the multibyte characters Returns 1 if there is no corresponding multibyte character Returns 0 if the wide character is 0 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 289 R32C 100 Series C Compiler Package V 1 02 E Standard Library E 2 4 Using the Standard Librar
47. area for storing a conversion result character string const char _far s2 Pointer to the character string to be converted size tn Number of bytes converted ReturnValue Returns the number of characters converted Description When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 283 R32C 100 Series C Compiler Package V 1 02 E Standard Library tan Mathematical Functions Function Calculates tangent Format include lt math h gt double tan x Method function Argument double x arbitrary real number OOD ReturnValue Returns the tangent of given real number x handled in units of radian tanh Function Calculates hyperbolic tangent Format include lt math h gt double tanh x Method function Argument double x arbitrary real number ReturnValue Returns the hyperbolic tangent of given real number x tolower Function Converts the character from an upper case to a lower case Format include lt ctype h gt int tolower c Method macro Argument int c Character to be converted ReturnValue Returns the lower case letter if the argument is an upper case letter Otherwise returns the passed argument as is Description Converts the character from an upper case to a lower case REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 284 R32C 100 Series C Compil
48. as register variables rather than being moved to the auto area In this case when parameters are specified in an asm function the assembly language is output using the register names instead of the variable s FB offset 3 When referencing arguments in the asm function The compiler analyzes a program flow with respect to its interval in which variables including arguments and auto variables remain effective as it processes the program If arguments or auto variables are referenced in an asm function as shown in Figure B 36 the compiler will fail to keep track of the effective interval and cannot generate correct code Therefore if arguments or auto variables need to be referenced in an asm function you write always be sure to use the b or feature of the asm function for that reference void func void short i j asm mov w 2 FB 4 FB jst Figure B 36 Example cannot be referred to correctly In the above case because the compiler determines that i and j are not used within the function func it does not output codes necessary to construct the frame in which to reference the arguments For this reason the arguments cannot be referenced correctly 4 About branching within the asm function The compiler analyzes program flow in the intervals in which registers and variables respectively are effective as it processes the program Do not write statements for branching including conditional b
49. avoid errors in the case of functions that include code generated using the fJSRW option and that cause errors during linking Rules This preprocessing directive has no effect when the fJSRW option not specified Example extern void func int i pragma JSRA func void main void func 1 Figure B 81 Example of pragma JSRA oragma JSRW Calls a function with JSR W Function Calls a function using the JSR W instruction Syntax pragma JSRWAfunction name Description By default the JSR A instruction is used when calling a function that in the same file has no body definition However the pragma JSRW declared function are always called using JSR W This directive helps reduce ROM size Rules 1 You may NOT specify pragma JSRW for static functions 2 When function call with the JSR W instruction does not reach pragma JSRW declared function an error occurs at link time In this case you may not use pragma JSRW Example pragma JSRW func void main void func 1 Figure B 82 Example of pragma JSRW Supplement The pragma JSRW is valid only when directly calling a function It has no effect when calling indirectly REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 154 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma PAGE Function Declares the position to be changed for a new page in a list file that is output by an assemb
50. can be defined only once Change the name for either of the two functions redefinition tag of enum tag name An enumeration is defined twice Make sure that enumeration is defined only once redefinition tag of struct tag name Astructure is defined twice Make sure that a structure is defined only once redefinition tag of union tag name A union is defined twice Make sure that a union is defined only once reinitialized of variable name iu eiy 2 yy An initialize expression is specified twice for the same variable Specify the initializer only once restrict is duplicate A restrict is defined twice Make sure that a restrict is defined only once size of incomplete array type ey ey An attempt is made to find sizeof of an array of unknown size This is an invalid size Specify the size of the array size of incomplete type An undefined structure or union is used in the operand of the sizeof operator Define the structure or union first The number of elements of an array defined as an operand of the sizeof operator is unknown Define the structure or union first REJ10J2009 0200 Rev 2 00 2010 04 01 317 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 17 ccom100Error message 11 13 Error message Description and countermeasure size of void An attempt is made to find the size of void This is an invalid size The s
51. carry value O cos Calculates cosine O cosh Calculates hyperbolic cosine O exp Calculates exponential function O fabs Calculates the absolute value of a double precision floating point O number floor Calculates an integer borrow value O fmod Calculates the remainder O frexp Divides floating point number into mantissa and exponent parts O labs Calculates the absolute value of a long type integer O Idexp Calculates the power of a floating point number O log Calculates natural logarithm 0 log10 Calculates common logarithm O modf Calculates the division of a real number into the mantissa and exponent O parts pow Calculates the power of a number O sin Calculates sine O sinh Calculates hyperbolic sine O sqrt Calculates the square root of a numeric value O tan Calculates tangent O tanh Calculates hyperbolic tangent O 8 Integer Arithmetic Functions The following lists integer arithmetic functions Table E 9 Integer Arithmetic Functions Function Contents Reentrant abs Calculates the absolute value of an integer O bsearch Performs binary search in an array O div Divides an int type integer and calculates the remainder O labs Calculates the absolute value of a long type integer O Idiv Divides a long type integer and calculates the remainder O qsort Sorts elements in an array O rand Generates a pseudo random number O srand Imparts seed to a pseudo random number generating routine O
52. character 0 The target parameter must be a pointer to a character array of sufficient size to store the character string including the null character 0 If input stops when the maximum field width is reached the character string stored consists of the characters to that point plus the ending null character c Stores a character Space characters are not skipped If you specify 2 or more for the maximum field width multiple characters are stored However the null character 0 is not included The target parameter must be a pointer to a character array of sufficient size to store the character string e p The pointer of the argument is output e Stores the input characters while the one or more characters between land are input Storing stops when a character other than those between and is input If you specify the circumflex after only character other than those between the circumflex and are legal input characters Storing stops when one of the specified characters is input The target parameter must be a pointer to a character array of sufficient size to store the character string including the null character 0 which is automatically added J n Stores the number of characters already read in format conversion The target parameter must be a pointer to an integer J e E f g G Convert to floating point format If you specify modifier I the target parameter must be a pointer to a double type The default
53. ei ey oyu REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 304 R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 4 cpp100 Error Messages 2 3 No Error message Description and countermeasure 17 input filename buffer overflow e The input filename buffer has overflowed Note that the filename includes the path Reduce the length of the filename and path use the I option to specify the standard directory 18 not enough memory for macro e Insufficient memory for macro name and contents of include file not found macro Increase the swap area 21 include file not found e The include file could not be opened The include files are in the current directory and that specified in the I option and environment variable Check these directories 22 illegal file name error e Illegal filename Specify a correct filename 23 include file nesting over e Nesting of include files exceeds the limit 8 Reduce nesting of include files to a maximum of 8 levels 25 illegal identifier e Error in define Code the source file correctly 26 illegal operation e Error in preprocess commands if elseif assert operation expression Rewrite operation expression correctly 27 macro argument error e Error in number of macro parameters when expanding macro gt Check macro definition and reference and correct as necessary 28 input buffer over f
54. erroneous Therefore not all errors can be alarmed Wccom_max_warnings Warning Count WCMW Warning Count Warning Options Function This option allows you to specify an upper limit for the number of warnings output by ccom100 Supplement By default there is no upper limit to warning outputs Use this option to adjust the screen as it scrolls for many warnings that are output Notes For the upper limit count of warning outputs specify a number equal to or greater than 0 Specification of this count cannot be omitted When you specify 0 warning outputs are completely suppressed inhibited Werror_file lt file name gt Function Outputs error messages to the specified file Syntax nc100A Werror _file lt output error message file name gt Notes The format in which error messages are output to a file differs from one in which error messages are displayed on the screen When error messages are output to a file they are output in the format suitable for the tag jump function that some editors have REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 90 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Wignore_near_pointer WINP Warning Options Function Inhibits a warning when the near pointer is handled as a far pointer Supplement In the compiler the pointer attribute is fixed to far 82 bits The compiler by default ignores near qualifiers for the pointer after generating a warning If this o
55. example program development using NC100 The program is described below Items 1 to 4 correspond to the same numbers in Figure 1 3 1 The C source program AA c is compiled using nc100 then assembled using as100 to create the re locatable object file AA r30 2 The startup program ncrt0 a30 and the include file sect100 inc which contains information on the sections are matched to the system by altering the section mapping section size and interrupt vector table settings 3 The modified startup program is assembled to create the relocatable object file ncrt0 a30 4 The two relocatable object files AA r30 and ncrt0 a30 are linked by the linkage editor In100 which is run from nc100 to create the absolute module file AA x380 Y sect100 inc e nert0 r30 Figure 1 3 Program Development Flow Figure 1 3 is an example make file containing the series of operations shown in Figure 1 4 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 5 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 AA x30 nert0 r30 AA r30 nc100 oAA nert0 r30 AA r30 nert0 r30 ncrt0 a30 as100 ncrt0 a30 AA r30 AA c nc100 c AA c Figure 1 4 Example make File Figure 1 5 shows the command line required for nc100 to perform the same operations as in the make file shown in Figure 1 4 nc1i00 oAA nert0 a30 AAc lt RET gt Indicates the prompt lt RETs gt Indicates the Return key Specify nc
56. extensions a30 relocatable object files extension 130 absolute module files extension x30 or other files output by ccom100 as100 or In100 are generated 2 The file extension i generated by this option does not include the line command generated by the preprocessor To get a result that includes line try again with the E option S Compile driver control Function Creates assembly language source files extension a30 and ext and stops processing Notes When this option is specified no relocatable object files extension r30 absolute module files extension x30 or other files output by as100 or In100 are generated silent Compile driver control Function Suppresses the display of copyright notices at startup Upredefined macro Compile driver control Function Undefines predefined macro constants Syntax nc100A U predefined macro lt C source file gt Notes The maximum number of macros that can be undefined may be limited by the maximum number of characters that can be specified on the command line of the operating system of the host machine STDC_ _LINE_ FILE_ DATE_ and_TIME_ cannot be undefined REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 66 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 2 Options Specifying Output Files Table A 2 shows the command line option that specifies the name of the output absolute module file Table A 2 Options for Specifying Outp
57. f amp func Sorry compilation terminated because of these errors in main Figure B 57 Example of inappropriate code of inline function 3 6 Declaration of static data Tf static data is declared in an inline function the body of the declared static data is allocated in units of files For this reason if an inline function consists of two or more files this results in accessing different areas Therefore if there is static data you want to be used in an inline function declare it outside the function If a static declaration is found in an inline function NC100 generates a warning Renesas does not recommend entering static declarations in an inline function Figure B 58 inline int func intj static int i 0 i return i j Warning Message Warning ccom smp c line 3 static valuable in inline function gt Static int i 0 Figure B 58 Example of inappropriate code of inline function 4 7 Regarding debug information NC100 does not output C language level debug information for inline functions Therefore you need to debug inline functions at the assembly language level REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 128 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 6 Extension of Comments NC100 allows comments enclosed between and as well as C like comments starting with B 6 1 Overview of Comments In C comments must be
58. from the data_EXT_INIT section to the data_EXT section 6 pragma MONITORn Extended Functions Transfers initial values from the data MON1_INIT data_MON2_INIT data_MONs3_INIT and data MON4_INIT sections to data MON1 data_MON2 data_MON8 and data MON4 sections respectively Initializes the heap area Initializes the standard I O function library Initializes FB register Calls the main function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 33 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler b Modifying the Startup Program Figure 2 18 summarizes the steps required to modify the startup programs to match the target system d Set the size of stack sections e Set the size of heap sections oge opou f Set the interrupt base register g Set the processor operating mode 2 2 3 Customizing Memory Allocations n D 2 2 a Figure 2 18 Example Sequence for Modifying Startup Programs c Examples of startup modifications that require caution 1 Settings When Not Using Standard I O Functions The init function initializes the R32C 100 Series I O It is called before main in ncrt0 a30 Figure 2 19 shows the part where the init function is called If your application program does not use standard I O comment out the init function call from nert0 a30 if STANDARD IO _ glb __ init Call _init G jsra __init endif Figure 2 19 Part of ncrt
59. gets Inputs one line from stdin fread Inputs the specified items of data from the stream scanf Inputs characters with format from stdin fscanf Inputs characters with format from the stream sscanf Inputs data with format from a character string Output fputc Outputs one character to the stream putc Outputs one character to the stream putchar Outputs one character to stdout fputs Outputs one line to the stream puts Outputs one line to stdout fwrite Outputs the specified items of data to the stream perror Outputs an error message to stdout printf Outputs characters with format to stdout fflush Flushes the stream of an output buffer Fprintf Outputs characters with format to the stream sprintf Writes text with format to a character string vfprintf Output to a stream with format vprintf Output to stdout with format vsprintf Output to a buffer with format Return ungetc Sends one character back to the input stream Deter ferror Checks input output errors mination feof Checks EOF End of File REJ10J2009 0200 Rev 2 00 2010 04 01 7tENESAS 213 R32C 100 Series C Compiler Package V 1 02 E Standard Library stdlib h E aa Function Declares the prototypes of memory management and terminate functions Function Contents abort Terminates the execution of the program abs Calculates the absolute value of an integer atof Converts a character string into a double t
60. hex representation is included after Reduce the length of the hex character identifier member name is duplicated this declare ignored The member name is defined twice or more This declaration will be ignored Make sure that member names are declared only once identifier variable name is duplicated The variable name is defined twice or more This declaration will be ignored Make sure that variable names are declared only once identifier variable name is shadowed The auto variable which is the same as the name declared as an argument is used Use any name not in use for arguments illegal storage class for argument extern ignore ey An invalid storage class is used in the argument list of function definition Specify the correct storage class incomplete array access An incomplete multidimensional array has been accessed Specify the size of the multidimensional array incompatible pointer types The object type pointed to by the pointer is incorrect Check the pointer type incomplete return type eu eU ey An attempt is made to reference an return variable of incomplete type Check return variable incomplete struct member ey An attempt is made to reference an struct member of incomplete Define complete structs or unions first init elements overflow ignored The initialization expression exceeded the size of the variable to be initialized Make su
61. ine 7 rts Figure B 50 Compile Result of smp2 c smp2 a30 B 4 3 Restrictions on Default Argument Declaration of Function The default argument declaration of a function is subject to some restrictions as listed below These restrictions must be observed a When specifying a default value for multiple arguments When specifying a default value in a function that has multiple arguments always be sure to write values beginning with the last argument Figure B 51 shows examples of incorrect description void func inti int j 1 int k 2 correct void funcaiint i int j int k 2 correct void func3 int i 0 int j int k incorrect void funcA int i 0 intj int k 1 incorrect Figure B 51 Examples of Prototype Declaration b When specifying a variable for a default value When specifying a variable for a default value write the prototype declaration of a function after declaring the variable you specify If a variable is specified for the default value of an argument that is not declared before the prototype declaration of a function it is processes as an error REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 124 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 5 inline Function Declaration NC100 allows you to specify the inline storage class in the similar manner as in C By specifying the inline storage class for a function you can expand the function
62. inline This chapter describes specifications of the inline storage class B 5 1 Overview of inline Storage Class The inline storage class specifier declares that the specified function is a function to be expanded inline The inline storage class specifier indicates to a function that the function declared with it is to be expanded in line The functions specified as inline storage class have codes embedded directly in them at the assembly level B 5 2 Declaration Format of inline Storage Class The inline storage class specifier must be written in a syntactically similar format to that of the static and extern type storage class specifiers when declaring the inline storage class Figure B 52 shows the format used to declare the inline storage class inline Atype specifier A function Figure B 52 Declaration Format of inline Storage Class An example function declaration and its compile result are shown in Figure B 53 and Figure B 54 respectively inline short func short i Inline function declaration and definition return i void main void short S s func s Inline function call Figure B 53 Sample program of inline function sample c REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 125 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference SECTION program CODE ALIGN _file sample c align line 7 H HC_SRC glb main _main enter 04H EP
63. int val REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 99 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 1 Near and far Modifiers For the R32C 100 series microcomputers the addressing modes used for referencing and locating data vary around the boundary address 00007FFFH and OFFFF8000H NC100 allows you to control addressing mode switching by near and far qualifiers B 1 1 Overview of near and far Modifiers The near and far qualifiers select an addressing mode used for variables or functions near modifier Area of 00000000H 00007FFFH and OFFFF8000H OFFFFFFFFH far modifier 04 Area of 00000000H 007FFFFFH and OFF800000H 0OFFFFFFFFH The near and far modifiers are added to a type specifier when declaring a variable or function If you do not specify the near or far modifiers when declaring variables and functions NC100 interprets their attributes as follows J Variables 000 near attribute J const qualified constants far attribute J Functions far attribute Furthermore NC100 allows you to modify these default attributes by using the startup options of compile driver nc100 B 1 2 Format of Variable Declaration The near and far modifiers are included in declarations using the same syntactical format as the const and volatile type modifiers Figure B 1 is a format of variable declaration type specifier near or far variable Figure B 1 Format
64. is a pointer to a float type REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 268 R32C 100 Series C Compiler Package V 1 02 E Standard Library scant Input Output Functions Description 2 prevents data storage Specifying the asterisk prevents the storage of converted data in the parameter 3 Maximum field width e Specify the maximum number of input characters as a positive decimal integer In any one format conversion the number of characters read will not exceed this number J If before the specified number of characters has been read a space character a character that is true in function isspace or a character other than in the specified format is input reading stops at that character 4 I Lorh e I The results of d i 0 u and x conversion are stored as long int and unsigned long int The results of e E f g and G conversion are stored as double e h The results of d i o u and x conversion are stored as short int and unsigned short int J If I or h are specified in other than d i 0 u or x conversion they are ignored e L The results of e E f g and G conversion are stored as float REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 269 R32C 100 Series C Compiler Package V 1 02 E Standard Library setimp Execution Control Functions Function Saves the environment before a function call Format include lt setjmp h gt int set jmp env Method function Argument
65. is truncated to form an integer which is returned as a double type fmod Mathematical Functions Function Calculates the remainder Format include lt math h gt double fmod x y Method function Argument double x dividend double y divisor ReturnValue Returns a remainder that derives when dividend x is divided by divisor y fprintf Input Output Functions Function Outputs characters with format to the stream Format include lt stdio h gt int fprintf stream format argument Method function Argument FILE _far stream Pointer of stream const char _far format Pointer of the format specifying character string ReturnValue Returns the number of characters output Returns EOF if a hardware error occurs Description Argument is converted to a character string according to format and output to the stream Format is specified in the same way as in printf REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 235 R32C 100 Series C Compiler Package V 1 02 E Standard Library putc Input Output Functions Function Outputs one character to the stream Format include lt stdio h gt int fputc c stream Method function Argument int c Character to be output FILE _far stream Pointer of the stream ReturnValue Returns the output character if output normally Returns EOF if an error occurs Description Outputs one character to the stream fputs Input Output Functions
66. islower Checks whether the character is a lower case letter isprint Checks whether the character is printable including a blank ispunct Checks whether the character is a punctuation character isspace Checks whether the character is a blank tab or new line isupper Checks whether the character is an upper case letter isxdigit Checks whether the character is a hexadecimal character tolower Converts the character from an upper case to a lower case toupper Converts the character from a lower case to an upper case errno h Function Defines error number REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 207 R32C 100 Series C Compiler Package V 1 02 E Standard Library float h Function Defines the limits of internal representation of floating point values The following lists the macros that define the limits of floating point values In NC100 long double types are processed as double types Therefore the limits applying to double types also apply to long double types Macro name Contents Defined value DBL_DIG Maximum number of digits of double type 15 decimal precision DBL_EPSILON Minimum positive value where 2 2204460492503131e 16 1 0 DBL_EPSILON is found not to be 1 0 DBL MANT DIG Maximum number of digits in the mantissa part when a doubletype floating point value is matched to the radix in its representation 53 DBL MAX Maximum value that a double type variable can take on as value 1 79
67. it as the unsigned char type Since the range of values re presentable by the unsigned char type is 0 to 255 the variable c will never take on the value 1 Accordingly if there is any statement which logically has no effect like this example the compiler does not generate assembler code 3 Programs not executed No assembler codes are generated for programs which logically are not executed void func int i func2 i return i 10 lt Fragment not executed Figure3 5 Program Not Executed 4 Operation between constants Operation between constants is performed when compiling int func void int i 1 2 Operation on this part is performed when compiling return i Figure3 6 Program Not Executed REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 52 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique 5 Selection of optimum instructions Selection of optimum instructions as when outputting shift instructions for division multiplications is always performed regardless of whether optimization options are specified or not b About the volatile qualifier Use of the volatile qualifier helps to prevent the referencing of variables the order in which they are referenced the number of times they are referenced etc from being affected by optimization However avoid writing statements like those shown below which will be interpreted ambiguously int a int
68. location of the RAM monitor area Function Declares that the specified external variable be located in a section used exclusively for the RAM monitor area Syntax pragma MONITOR n Aexternal variable name n 1 4 Rules 1 Only external variables and external static variables can be specified 2 The area for the external variable declared by pragma MONITORIn is allocated to one of the sections listed below data MONI External variables that have initial values are located here bss MONIn External variables that do not have initial values are located here data MONI n _INIT Initial values of external variables that have initial values are located here 8 The declaration of pragma MONITORIn must be made before the external variable is defined 4 The external variable declared by pragma MONITOR n cannot be used in combination with other extended pragma directives However if pragma SBDATA and pragma MONITORIn are specified at the same time pragma SBDATA has priority At this time no warnings are output Note 1 pragma MONITORIn does not affect the op codes generated by the compiler Please pay attention to the near far attributes of variables 2 Even if external variables with different near far attributes coexist in a section used exclusively for the RAM monitor area no errors and warnings are assumed Please pay attention to the near far attributes of variables 3 The sections used exclusively for the
69. lt file name gt from the directory specified by environment variable INC100 ifit s not found 2 Takes in file name from the current directory Searches file name from the following directory in sequence if it s not found 1 The directory specified by nc100 s startup option I 2 The directory specified by environment variable INC100 3 If the macro expanded identifier is lt file name gt or file name this command takes in that file from the directory according to rules of search 1 or 2 The maximum number of levels of nesting is 40 An include error results if the specified file does not exist line Function Changes the line number in the file Format lineAintegerA file name Description Specify the line number in the file and the filename You can change the name of the source file and the line No REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 182 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications dem Function Instructs the system to process NC100 s extended functions Format 1 pragma ROMAvariable name 2 pragma SBDATAAvariable name 8 pragma SB16DATAAvariable name 4 pragma SECTIONApredetermined section nameAaltered section name 5 pragma STRUCTAtag name of structureAunpack 6 pragma STRUCTAtag name of structureAarrange 7 pragma ADDRESSAvariable nameAabsolute address 8 pragma DMACAvariable nameADMAC register name 9 pragma I
70. mainly o Maps each section in sequence e Sets the starting addresses of the sections e Defines the size of the stack and heap sections Sets the interrupt vector table e Sets the fixed vector table c Modifying the sect100 inc Figure 2 27 summarizes the steps required to modify the startup programs to match the target system 2 2 2 Customizing the Startup Program v d Map order each section and set starting addresses oge opou our 00 H9S e Set the special page vector table Figure 2 27 Example Sequence for Modifying Startup Programs d Mapping and Order Sections and Specifying Starting Address Map and order the sections to memory and specify their starting addresses mapping programs and data to ROM and RAM in the sect100 inc include file of the startup program The sections are mapped to memory in the order they are defined in sect100 inc Use the as100 pseudo instruction ORG to specify their starting addresses Figure 2 28 is an example of these settings section program code align org OFFFF0000H Specifies the starting address of the program section Figure 2 28 Example Setting of Section Starting Address If no starting address is specified for a section that section is mapped immediately after the previously defined section REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 41 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 1 Rules for
71. max Maximum number of parameters that can be specified in nc100 command line option as100 Depends on amount of available memory Maximum number of parameters that can be specified in nc100 command line option n100 Depends on amount of available memory Maximum nesting levels of compound statements iteration control structures and selection control structures Depends on amount of available memory Maximum nesting levels in conditional compiling Depends on amount of available memory Number of pointers modifying declared basic types arrays and function declarators Depends on amount of available memory Number of function definitions Depends on amount of available memory Number of identifiers with block scope in one block Depends on amount of available memory Maximum number of macro identifiers that can be simultaneously defined in one source file Depends on amount of available memory Maximum number of macro name replacements Number of logical source lines in input program Depends on amount of available memory Depends on amount of available memory Maximum number of levels of nesting include files 40max Maximum number of case names in one switch statement with no nesting of switch statement Depends on amount of available memory Total number of operators and operands that can be defined in if and elif Depends on amount of available memory Size of sta
72. not 0 the commands if and elif process the program that follows elifis used in a pair with if ifdef or ifndef else is used in a pair with if Do not specify any tokens between else and the line feed You can however insert a comment endif indicates the end of the range controlled by if Always be sure to enter endif when using command if Combinations of if elif else endif can be nested There is no set limit to the number of levels of nesting but it depends on the amount of available memory Cannot use the sizeof operator cast operator or variables in a constant expression REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 180 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications ifdef elif else endif Function Performs conditional compilation Examines the macro defined or not Format ifdefAidentifier HelifAconstant expression else endif Description Jf an identifier is defined ifdef processes the program that follows You can also describe the following if defined Aidentifier if defined identifier else is used in a pair with ifdef Do not specify any tokens between else and the line feed You can however insert a comment elifis used in a pair with if ifdef or ifndef endif indicates the end of the range controlled by ifdef Always be sure to enter endif when using comm
73. not too large most functions can be called with the JSR W instruction Tn this case ROM size will be reduced by doing as follows First Compile with the fJSRW option and check functions which are indicated as errors at link time Then change declarations for the error functions only into declarations using pragma JSRA function name When you use the OGJ option the JMP instruction at the time of a link is chosen REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 54 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique d Other methods Tn addition to the above the ROM capacity can be compressed by changing program description s as shown below 1 Change a relatively small function that is called only once to an inline function 2 Replace an if else statement with a switch statement This is effective unless the variable concerned is a simple variable such as an array pointer or structure 3 For a function which returns a value in only the range of char type declare its return value type with char 4 For variables used overlapping a function call do not use a register variable 3 2 2 Speeding Up Startup Processing The ncrt0 a30 startup program includes routines for clearing the bss area This routine ensures that variables that are not initialized have an initial value of 0 as per the C language specifications For example the code shown in Figure3 9 does not initialize the variable which must therefore be
74. of Variable added near far modifier Figure B 2 is an example of variable declaration Figure B 3 is a memory map for that variable short near in_data short far if data void func void remainder omitted Figure B 2 Example of Variable Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 100 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference _in_data near area far area na a _if_ data Figure B 3 Memory Location of Variable B 1 3 Format of Pointer type Variable The pointer type variables are always a far type 4 bytes variable If type near is specified in the declaration of a pointer type variable the compiler outputs a warning message Near pointer not supported near qualifier ignored and ignores the near qualifier An example declaration of a pointer type variable is shown in Figure B 4 Example short ptr Figure B 4 Example of Declaring a Pointer Type Variable 1 Because the variables are located near and take on the pointer variable type far the description in Figure B 4 is interpreted as in Figure B 5 Example short far near ptr Figure B 5 Example of Declaring a Pointer Type Variable 2 The variable ptr is a 4 byte variable that indicates the short type variable located in the far area The ptr itself is located in the near area Memory mapping for the above example is shown in Figure B 6
75. of warnings output by ccom100 Werror_file lt fle name gt WEF Outputs error messages to the specified file Wignore_near_pointer WINP Inhibits a warning when the near pointer is handled as a far pointer Wlarge_to_small WLTS Outputs a warning about the tacit transfer of variables in descending sequence of size Wmake _tagfile WMT If an error or warning occurred a tag file is output for each file Wnesting_comment WNC Outputs a warning for a comment including Wno_stop WNS Prevents the compiler stopping when an error occurs Wno_used_argument WNUA Outputs a warning for unused argument of functions Wno_used_function WNUF Displays unused global functions when linking Wno_used_static_function WNUSF For one of the following reasons a static function name is output that does not require code generation Wno_warning_stdlib WNWS Specifying this option while Wnon_prototype or Wall is specified inhibits Alarm for standard libraries which do not have prototype declaration Wnon_prototype WNP Outputs warning messages for functions without prototype declarations Wstdout None Outputs error messages to the host machine s standard output stdout Wstop_at_link WSAL Stops linking the source files if a warning occurs during linking to suppress generation of absolute module files Also a return value 10 is returned to the host OS Wstop_at_warning WSAW Stops compiling the source files if a warning occurs du
76. on as value ULLONG_MAX Maximum value that an unsigned long 18446744073709551615 long inttype variable can take on as value 7tENESAS R32C 100 Series C Compiler Package V 1 02 E Standard Library locale h Function Defines declares macros and functions that manipulate program localization The following lists locale functions Function Contents localeconv Initializes struct Iconv setlocale Sets and searches the locale information of a program math h eee Function Declares prototype of mathematical function The following lists mathematical functions Function Contents acos Calculates arc cosine asin Calculates arc sine atan Calculates arc tangent atan2 Calculates arc tangent ceil Calculates an integer carry value cos Calculates cosine cosh Calculates hyperbolic cosine exp Calculates exponential function fabs Calculates the absolute value of a double precision floating point number floor Calculates an integer borrow value fmod Calculates the remainder frexp Divides floating point number into mantissa and exponent parts labs Calculates the absolute value of a long type integer Idexp Calculates the power of a floating point number log Calculates natural logarithm log10 Calculates common logarithm modf Calculates the division of a real number into the mantissa and exponent parts pow Calculates the power of a number
77. or greater than 1 eu ey The function or variable declaration without const qualification is const qualified on the entity definition side Make sure the function or variable declaration and the const qualification on the entity definition side are matched no const in previous declaretion y There is a declaration that is not referenced Delete the declaration xxx was declared but never referenced REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 329 R32C 100 Series C Compiler Package V 1 02 G Using gensni or the Stack information File Creation Tool for Call Walker Appendix G Using gensni or the stack information File Creation Tool for Call Walker Before Call Walker or the stack analysis tool of the High performance Embedded Workshop can be used you must have stack information files extension sni as the input files for it You use gensni or the stack information file creation tool for Call Walker to create these stack information files from the absolute module file G 1 Starting Call Walker To start Call Walker select Call Walker that is registered to the High performance Embedded Workshop or select the tool from the Tools menu of the High performance Embedded Workshop After starting Call Walker choose Import Stack File from the File menu and select a stack information file as the input file for Call Walker G 2 Outline of gensni G 2 1 Processing Outline of gensni gensn1 is the tool to create
78. performance a Test Environment Table C 1 shows the standard PC environment Table C 1 Standard PC Environment Item Type of PC OS Version PC environment IBM PC AT or compatible Windows XP Windows Me Windows 98 Windows 2000 Windows NT 4 0 b C Source File Coding Specifications Table C 2 shows the specifications for coding NC100 C source files Note that estimates are provided for items for which actual measurements could not be achieved Table C 2 Specifications for Coding C Source Files Item Specification Number of characters per line of source file 512 bytes characters including the new line code Number of lines in source file 65535 max REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 163 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications c NC100 Specifications Table C 3 to Table C 4 lists the NC100 specifications Note that estimates are provided for items for which actual measurements could not be achieved Table C 3 NC100 Specifications 1 2 Item Specification Maximum number of files that can be specified in nc100 Depends on amount of available memory Maximum length of filename Depends on operating system Maximum number of macros that can be specified in nc100 command line option D Depends on amount of available memory Maximum number of directories that can be specified in nc100 command line option I 256
79. performed collectively rather than for each function call made Example In the example shown below the stack pointer is corrected each time func1 and then func2 is called so that the stack pointer is corrected twice If this option is specified the stack pointer is corrected only once char func char char char char func2 char char char void main void char i 1 char j 2 char k 3 char I m incih i m func2 i j K Passien nede Notes Use of the option Osp_adjust helps to reduce the ROM capacity and at the same time to speed up the processing However the amount of stack used may increase REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 80 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Ostatic_to_inline OSTI A static function is treated as an inline function Function Supplement Notes REJ10J2009 0200 Rev 2 00 2010 04 01 81 A static function is treated as an inline function and the assembling code which carried out inline deployment is generated When the following conditions are fulfilled a static function is treated as an inline function and the assembling code which carried out inline deployment is generated 1 Substance is described before the function call It is aimed at a static function A function call and the body of that function must be written in the same source file When you specify Ofile_inline
80. positive field width 3 Precision Specify a positive integer after If you specify only with no value it is interpreted as zero The function and default value differs according to the conversion type Floating point type data is output with a precision of 6 by default However no decimal places are output if you specify a precision of 0 J d i 0 u x and X conversion 1 If the number of columns in the result of conversion is less than the specified number the beginning is padded with Zeros 2 Tf the specified number of columns exceeds the minimum field width the specified number of columns takes precedence REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 261 R32C 100 Series C Compiler Package V 1 02 E Standard Library printf Description 3 If the number of columns in the specified precision is less than the minimum field width the field width is processed after the minimum number of columns have bee processed 4 The default is 1 5 Nothing is output if zero with converted by zero minimum columns e s conversion 1 Represents the maximum number of characters 2 Tf the result of conversion exceeds the specified number of characters the remainder is discarded 3 There is no limit to the number of characters in the default 4 If you specify an asterisk for the precision the integer of the parameter specifies the precision 5 Ifthe parameter is a negative value specification of the precision is inv
81. s2 Pointer to the character string to be concatenated from size_t n Number of characters to be concatenated ReturnValue Returns a pointer to the concatenated character string area Description Concatenates character strings s1 and n characters from character string s2 The concatenated string ends with NULL When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 277 R32C 100 Series C Compiler Package V 1 02 E Standard Library strncmp String Handling Function Function Compares character strings n characters Format include lt string h gt int strncmp s1 s2 n Method function Argument const char _far s1 Pointer to the first character string to be compared const char _far s2 Pointer to the second character string to be compared size tn Number of characters to be compared ReturnValue ReturnValue 0 The two character strings are equal ReturnValue gt 0 The first character string s1 is greater than the other ReturnValue lt 0 The second character string s2 is greater than the other Description Compares each byte of n characters of two character strings ending with NULL When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization strncpy String Handling Function Function C
82. said warning from being output D 1 5 Array Types Array types are mapped contiguously to an area equal to the product of the size of the elements in bytes and the number of elements They are mapped to memory in the order in which the elements appear Figure D 3 is an example of mapping REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 188 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules Example char c 5 0 1 2 3 4 address higher Figure D 3 Example of Placement of Array D 1 6 Structure types Structure types are mapped contiguously in the order of their member data Figure D 4 is an example of mapping Example struct TAG char C 3 bytes s address higher Figure D 4 Example of Placement of Structure 1 Normally there is no word alignment with structures The members of structures aremapped contiguously To use word alignment use the pragma STRUCT extended function pragma STRUCT adds a byte of padding if the total size of the members is odd Figure D 5 is an example of mapping Example pragma STRUCT TAG unpac struct TAG char C short i s address higher Figure D 5 Example of Placement of Structure 2 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 189 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules D 1 7 Unions Unions occupy an area equal to the maximum data size of their me
83. section rom NEAR romdata org OFFFF8000H om N EAR_top Figure 2 14 Startup Program List 9 Sect100 inc REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 29 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler VARIABLE VECTOR SECTION section vector romdata org VECTOR_ADR Jif _MT_ word dummy_int BRK software int 0 word dummy_int reservation area software int 1 word dummy_int uart5 trance NACK software int 2 word dummy_int uart5 receive ACK software int 3 word dummy_int uart6 trance NACK software int 4 word dummy_int uart6 receive ACK software int 5 word dummy_int uart5 uart6 bus collision software int 6 word dummy_int reservation area software int 7 word dummy_int DMAO software int 8 word dummy_int DMA1 software int 9 word dummy_int DMA2 software int 10 word dummy_int DMA3 software int 11 word dummy_int TIMER AO software int 12 word dummy_int TIMERA1 software int 13 word dummy_int TIMER A2 software int 14 word dummy_int TIMER A3 software int 15 word dummy_int TIMER A4 software int 16 word dummy_int uartO trance NACK software int 17 word dummy_int uart0 receive ACK software int 18 word dummy_int uart1 trance NACK software int 19 word dummy_int uart1 receive ACK software int 20 word dummy_int TIMER BO software int 21 word dummy_int TIMER B1 software int 22 word d
84. shown in Figure B 9 short far ptr1 short far ptr2 4bytes _ptr2 4bytes near area far area sn Figure B 9 Memory Location of Pointer type Variable B 1 4 Declaration of function A function s near and far allocation attributes are always far If you specify the near attribute in function declaration the system outputs a warning message function must be far with your near declaration ignored B 1 5 near and far Control by nc100 Command Line Options NC100 handles functions as belonging to the far attribute and variables data as belonging to the near attribute if you do not specify the near and far attributes NC100 s command line options allow you to modify the default attributes of functions and variables data These are listed in the table below Table B 3 Command Line Options Command Line Options Function fnear_ROM fNROM Change the default attribute of ROM data to near ffar RAMCfFRAM Change the default attribute of RAM data to far REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 103 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 1 6 Function of Type conversion from near to far The program in Figure B 10 performs a type conversion from near to far int func int far int far f ptr int near n_var void main void f ptr amp n_var assigns the near address to the far pointer abbreviated func amp n_var prototyp
85. specified twice Specify the output filename once only The command line ends with the o option Specify the name of the output file after the o option The I option specifying the include file path exceeds the limit Specify the I option 8 times or less The command line ends with the I option Specify the name of an include file after the I option The string following the D option is not of a character type letter or underscore that can be used in a macro name Illegal macro name definition Specify the macro name correctly and define the macro correctly The command line ends with the D option Specify a macro filename after the D option The string following the U option is not of a character type letter or underscore that can be used in a macro name Define the macro correctly You specified an illegal option on the cpp100 command line Specify only legal options eu eU ey eu ey ey 11 cannot open input file Input file not found Specify the correct input file name 12 cannot close input file Input file cannot be closed Check the input file name 14 cannot open output file Cannot open output file Specify the correct output file name 15 cannot close output file Cannot close output file Check the available space on disk 16 cannot write output file Error writing to output file Check the available space on disk Yew ey
86. stdin X fscanf Inputs characters with format from the stream X sscanf Inputs data with format from a character string X Output fputc Outputs one character to the stream X putc Outputs one character to the stream X putchar Outputs one character to stdout X fputs Outputs one line to the stream X puts Outputs one line to stdout X fwrite Outputs the specified items of data to the stream X perror Outputs an error message to stdout X printf Outputs characters with format to stdout X fflush Flushes the stream of an output buffer X fprintf Outputs characters with format to the stream X sprintf Writes text with format to a character string X vfprintf Output to a stream with format X vprintf Output to stdout with format X vsprintf Output to a buffer with format X Return ungetc Sends one character back to the input stream X Determination ferror Checks input output errors X feof Checks EOF End of File X 4 Memory Management Functions The following lists memory management functions Table E 5 Memory Management Functions Function Contents Reentrant calloc Allocates a memory area and initializes it to zero 0 X free Frees the allocated memory area X malloc Allocates a memory area X realloc Changes the size of an allocated memory area X REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 219 R32C 100 Series C Compiler Package V 1 02 E Standard Library 5 Memory Handling Functions Th
87. t get bitfield address by amp operator The bit field type is written with the amp operator Do not use the amp operator to write the bit field type can t get inline function s address by amp operator The amp operator is written in an inline function Do not use the amp operator in an inline function can t get size of bitfield The bit field type is written with the sizeof operator Do not use the sizeof operator to write the bitfield type can t get void value ey ey ey ely An attempt is made to get void type data as in cases where the right side of an assignment expression is the void type Check the data type can t output to filename The file cannot be wrote Check the rest of disk capacity or permission of the file can t open file name The file cannot be opened Check the permission of the file can t set argument eu ew ey The type of an actual argument does not match prototype declaration The argument cannot be set in a register argument Correct mismatch of the type cannot refer to the range outside of the stack frame A location outside the stack frame area is referenced Reference the correct location case value is duplicated Yew ely The value of case is used more than one time Make sure that the value of case that you used once is not used again within one switch statement REJ10J2009 0200 Rev 2 00 2010 04 01 309 7tENESAS R32
88. than that freed the area is secured from the high address of the contiguously empty area created by free toward the low address Heap area Low Sas rea ae es High Unused area REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 252 R32C 100 Series C Compiler Package V 1 02 E Standard Librar malloc Memory Management Functions Rule Ifthe amount of memory to be secured is larger than that freed the area is secured from the lowest address of the unused memory toward the high address Heap area Low Pree S rete unsa E ussa High 2 Ifno memory area has been freed with free Ifthere is any unused area that can be secured the area is secured from the lowest address of the unused memory toward the high address Heap area Low eae gt Unused area pass High Ifthere is no unused area that can be secured malloc returns NULL without any memory being secured Note No garbage collection is performed Therefore even if there are lots of small unused portions of memory no memory is secured and malloc returns NULL unless there is an unused portion of memory that is larger than the specified size REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 253 R32C 100 Series C Compiler Package V 1 02 E Standard Library mblen Multi byte Character Multitoyte Character String Manipulate Functions Function Calculates the length of a multibyte character stri
89. the limitations on the number of command line characters imposed by Microsoft Windows TM etc a Command file input format nc100A command line option A lt file name gt command line option Prompt lt gt Mandatory item Optional item A Space Figure 2 3 Command File Command Line Format nc100 c test cmd g lt RET gt lt RET gt Return key Always specify the startup program first when linking Figure 2 4 Example Command File Command Line Command files are written in the manner described below ncrt0 a30 lt CR gt Command File description sample1 c sample2 r30 lt CR gt g as100 kCR gt lt CR gt Denotes carriage return o lt CR gt sample lt CR gt Figure 2 5 Example Command File description REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 14 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler b Rules on command file description The following rules apply for command file description Only one command file can be specified at a time You cannot specify multiple command files simultaneously No command file can be specified in another command file Multiple command lines can be written in a command file New line characters in a command file are replaced with space characters The maximum number of characters that can be written in one line of a command file is 2 048 An error results when this limit is exceeded c P
90. the location of the RAM monitor area Example pragma MONITORI i pragma MONITORI c short i short j 0x0100 Figure B 62 Example Use of pragma MONITORn Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 136 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma ROM Map to rom section Function Syntax Description Rules Example Maps specified data variable to rom section pragma ROMAvariable name This extended function is valid only for variables that satisfy one or other of the following conditions Non extern variables defined outside a function Variables for which an area is secured Variables declared as static within the function 1 Ifyou specify other than a variable it will be ignored 2 No error occurs if you specify pragma ROM more than once 3 The data is mapped to a rom section with initial value 0 if you do not include aninitialization expression C language source program pragma ROM i unsigned short i Variable i which satisfies condition 1 void func void static short i 20 Variable i which satisfies condition 2 remainder omitted Assembly language source program SECTION rom_FAR ROMDATA ALIGN glb i i Variable i which satisfies condition 1 byte 00H byte 00H ___S0_i HEC s name is i Variable i which satisfies condition 2 word 0014H Figure B 63 Example Use
91. the variables in C e Specifying the symbol name of storage class extern and static variables in the C program using the names of the variables in C The following shows precautions to be observed when using the asm function e Do not destroy register contents in the asm function The compiler does not check the inside of the asm function o If registers are going to be destroyed write push and pop instructions using the asm function to save and restore the registers 1 For the purpose of expression in this user s manual the subroutines written in the assembly language are referred to as assembler functions Those written with asm in a C language program are referred to as asm functions or inline assemble description REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 107 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 2 2 Specifying FB Offset Value of auto Variable The storage class auto and register variables including arguments written in the C language are referenced and located as being offset from the Frame Base Register FB They may be mapped to registers as a result of optimization The auto variables which are mapped to the stack can be used in the asm function by writing the program as shown in Figure B 17 below asm op code R1 FB variable name Figure B 17 Description Format for Specifying FB Offset Only two variable name can be specified by using this description format
92. to make the prototype declaration a warning is output and pragma PARAMETER is ignored 2 Follow the following rules in the prototype declaration a Note also that the number of parameters specified in the prototype declaration must match that in the pragma PARAMETER declaration b The following types cannot be declared as parameters for an assembler function in a pragma PARAMETER declaration o structure type and union type c The assembler functions shown below cannot be declared e Functions returning structure or union type Example short asm_func short short Prototype declaration for the assembler function pragma PARAMETER asm_func RO R1 void main void short i j i Ox7FFD j Ox007F asm_func i j Calling the assembler function Figure B 78 Example of pragma PARAMETER Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 151 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 7 4 Use of the other extension function NC100 includes the following extended function for embedding assembler description inline pragma__ ASMMACRO Function Declares defined a function by assembler macro Syntax pragma __ASMMACRO function name register name Rules 1 Always put the prototype declaration before the pragma _ASMMACRO declaration Assembler macro function be sure to declare static 2 Can t declare the function of no parameter Parameter is passed
93. used in asm Q ey An undefined variable name is used in the asm statement Define the variable unknown variable variable name An undefined variable name is used Define the variable unknown variable variable name used An undefined variable name is used Define the variable void array is invalid type int array assumed ee ey An array cannot be declared as void Processing will be continued assuming it has type int Write the type specifier correctly void value can t return The value converted to void by cast is used as the return from a function Write correctly while struct union statement struct or union is used in the expression of a while statement Use scalar type while void statement void is used in the expression of a while statement Use scalar type REJ10J2009 0200 Rev 2 00 2010 04 01 319 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 19 ccom100Error message 13 13 Error message Description and countermeasure multiple pragma EXT4MPTR spointer e A pointer variable decleared by pragma EXT4MPTR is ignored NC30 only duplecate Declare the variable only one time zero size array member e the array which size is zero Declare the array size e The structure members include an array whose size is zero Arrays whose size is zero cannot be members of a structure function name is resursion
94. via register Please specify the register matching the parameter type 8 Please append the underscore _ to the head of the definition assembler macro name 4 The following is a return value related calling rules You can t declare structure and union type as the return value char and _Bool types ROL float types R2RO int fint_16 use RO double types A1A0 short types int fint_16 does nt use R2RO long long type A1A0 long types pointer types AO 5 If you change the register s data save the register to the stack in entry processing of assembler macro function and the saved register restore in exit processing Example static short max_w short short Be sure to declare static pragma___ ASMMACRO max_w RO R2 pragma ASM _max_w macro max w R2 RO0 The return value is set to RO register endm pragma ENDASM short S void test_func void s max_w 2 3 Figure B 79 Example of pragma _ AMMACRO REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 152 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference ud et ma alias D et ma ENDASM Inline assembling Function Syntax Description Rules Example Supplement Specifies assembly code in C pragma ASM assembly statements pragma ENDASM The line s between pragma ASM and pragma ENDASM are output without modifying anything t
95. word fsigned_char SC Modify generated code Function Handles type char without sign specification as type signed char fswitch_other_section SOS Modify generated code Function This option outputs a ROM table for a switch statement to some other section than a program section Supplement Section name is switch_table Notes This option does not normally need to be used fuse _FPU UF Modify generated code Function Outputs FPU instruction Supplement When using this option you need to link nc100fpu lib instead of nc100lib lib as the standard library If you executed a range of operations from compile to link after specifying this option from the compiler driver the libraries to be linked are automatically changed Note that if this option is used in combination with the compile option fint_16 f116 you need to link nc100116fpu lib instead of nc100hb lib REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 87 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 7 Library Specifying Option Table A 8 lists the startup options you can use to specify a library file Table A 8 Library Specifying Option Option Function Libraryfilename Specifies a library file that is used by In100 when linking files llibrary file name Function Syntax Notes Specifies a library file that is used by In100 when linking files The file extension can be omi
96. written between and In C anything following B 6 2 Comment Format When you include on a line anything after the is treated as a comment Figure B 59 shows comment format comments Figure B 59 Comment Format Figure B 60 shows example comments void func void inti This is commentes intj This is commentes omitted Figure B 60 Example Comments B 6 3 Priority of and The priority of and is such that the one that appears first has priority Therefore a written between a to the new line code does not have an effect as signifying the beginning of a comment Also a written between and does not have an effect as signifying the beginning of a comment REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 129 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 7 pragma Extended Functions B 7 1 Index of pragma Extended Functions Following index tables show contents and formation for pragma extended functions a Using Memory Mapping Extended Functions Table B 4 Memory Mapping Extended Functions 1 2 Extented function Description pragma EXTMEM Declares that data or function be located in an area that cannot be accessed in address 0 relative addressing Syntax pragma EXTMEMAvariable name pragma EXTMEMA function name Example pragma EXTMEM val pragma EXTMEM func
97. 0 i exitd E2 11 This passes all arguments via a stack 12 This is the start address of function func_no_proto Figure D 12 Compile Result of Sample Program sample c 2 In the compile result 2 3 of the sample program sample c listed in Figure D 10 all arguments are passed via a stack since function func_no_proto is written in K amp R format Furthermore since the arguments of the function are not passed via register the symbol name of the function s start address is derived from func_no_proto described in the C language source file by prefixing it with _ underbar hence _func_no_proto REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 197 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules HH H FUNCTION main HH H FRAME AUTO sum size 4 offset 4 HHH ARG Size 4 Auto Size 4 Context Size 8 align ine 17 C_SRC glb _main _main enter 04H ine 20 C SRO sum fune_proto 1 2 3 2 1 mov l 00000003H R6R4 13 movl 00000002H R3R1 l l movl 00000001H R2RO l jsr func_proto L movl___R2RO 4 FB E Eee ee ae ine 21 HHHC SRO L Sum fune no_proto 1 23 LL push l 00000003H 714 push 00000002H i push 00000001H jsr _func_no_proto add 0cH SP L Mov __ R2RO 4FB ____ oona ine 22 HHEC SRC exitd E3 END Figure D 13 Compile Result of Sample Program sample c 3 Figure D 13 part 13 calls func_proto and pa
98. 0 Series C Compiler Package V 1 02 1 Introduction to NC100 1 2 NC100 Processing Flow Figure 1 1 illustrates the NC100 processing flow ae C language source file Compile driver igen 100 Inline generator cpp 100 aopt100 Assembly language source file Preprocessor Compiler Assembler optimizer Stack information analysis utility Map Viewer ensni Map viewer g p Stack analysis tool Call Walker Assembler Relocatable file Motorola S format file or Intel HEX format file Imc100 Absolute module file Software in NC100 package a File processed by NC100 Figure 1 1 NC100 Processing Flow REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 2 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 1 2 1 nc100 nc100 is the executable file of the compile driver By specifying options nc100 can perform the series of operations from compiling to linking You can also specify for the as100 relocatable macro assembler and four for the In100 linkage editor by including the as100 and In100 command line options when you start nc100 1 2 2 igeni00 igen 100 is the executable file of the inline generator igen100 calls cpp100 1 2 3 cpp100 cpp100 is the executable file for the preprocessor cpp100 processes macros starting with define include etc and performs conditional compiling Gif telse endif etc 1 24
99. 0 a30 Where init Function is Called If you are using only sprintf and sscanf the init function does not need to be called 1 The init function also initializes the microcomputer hardware for standard in put output functions By default the R32C 100 is assumed to be the microcomputer that it initializes When using standard input output functions the init function etc may need to be modified depending on the system in which the microcomputer is to be used REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 34 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 2 Settings When Not Using Memory Management Functions To use the memory management functions calloc and malloc etc not only is an area allocated in the heap section but the following settings are also made in ncrt0 a30 1 Initialization of external variable char _mnext Initializes the heap_top label which is the starting address of the heap section 2 Initialization of external variable unsigned_msize Initializes the HEAPSIZE expression which sets at 2 2 2 e heap section size Figure 2 20 shows the initialization performed in ncrt0 a30 initialize heap manager jf HEAP 1 glb __mnext glb __msize mov l heap_top _mnext mov l HEAPSIZE _msize endif Figure 2 20 Initialization When Using Memory Management Functions ncrt0 a30 If you are not using the memory management functions comment out the whole initializa
100. 01H s_val Assembly language source file compile result _func ine 5 H C_SRC asm mov w 01H ext_val AHHH ASM START mov w 01H _ext_val Move to_ext_val ine 6 C_SRC asm mov w 01H s_val mov w 01H _S0_s val Move to _S0_s val HHH ASM END ine 7 C_ SRC rts E1 glo _ext val SECTION bss_NEAR DATA ALIGN 0 s val HHH C s name is s_val blkb 2 END Figure B 27 Example of Referencing extern and static Variables You can use the format shown in Figure B 26 to use 1 bit bit fields of extern and static variables in asm functions Can not operate bit fields og greater than 2 bits asm op code b FB bit field name Figure B 28 Format for Specifying Symbol Names REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 113 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference You can specify one variable name using this format See Figure B 29 for an example struct TAG char bitO 1 char bit1 1 char bit2 1 char bit3 1 s void func void asm bset b s bit1 Figure B 29 Example of Specifying Symbol Bit Position Figure B 30 shows the results of compiling the C source file shown in Figure B 29 HHH FUNCTION func HHH ARG Size 4 Auto Size 0 Context Size 4 SECTION program CODE ALIGN _file bitfield c align _ line 8 HHEC SRC glb _func _func _line 9 HHC SRC asm bset b s bit1 AHHH AS
101. 1 Pointer to the first memory area to be compared const void _far s2 Pointer to the second memory area to be compared size tn Number of bytes to be compared ReturnValue Return Value 0 The two memory areas are equal Return Value gt 0 The first memory area s1 is greater than the other Return Value lt 0 The second memory area s2 is greater than the other Description Compares each of n bytes of two memory areas When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization memcpy Memory Handling Functions Function Copies n bytes of memory Format include lt string h gt void _far memcpy s1 s2 n Method macro default or function Argument void _far s1 Pointer to the memory area to be copied to const void _far s2 Pointer to the memory area to be copied from size tn Number of bytes to be copied ReturnValue Returns the pointer to the memory area to which the characters have been copied Description Usually the program code described by macro is used for this function In using the function in a library please describe it as undef memcpy after description of include lt string h gt Copies n bytes from memory S2 to memory S1 When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 256
102. 1 RENESAS 92 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Wno_used_function WNUF Warning Options Function Displays unused global functions when linking Notes When selecting this option be sure to specify the finfo option at the same time Wno_used_static_function WNUSF Warning Options Function For one of the following reasons a static function name is output that does not require code generation The static function is not referenced from anywhere in the file static functions are made inline by use of the Ostatic_to_inline OSTD option Notes If a function name is written for the initializer of an array as shown below the compiler handles the function as referenced even though it may not actually be referenced during program operation In the example given below although the functions f4 and f5 are not referenced the compiler handles them as referenced Example void a 5 void f1 f2 f8 f4 15 for i 0 i lt 3 i afi Wno_warning_stdlib WNWS Function Specifying this option while Wnon_prototype or Wall is specified inhibits Alarm for standard libraries which do not have prototype declarations Wnon_prototype WNP Warning Options Function Outputs warning messages for functions without prototype declarations or if the prototype declaration is not performed for any function Supplement Function arguments can be p
103. 12 2tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 12 ccom100Error message 6 13 Error message Description and countermeasure invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid operands The operator contains an error Check the expressions on the left and right sides of the operator invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid lt lt operands The lt lt operation contains an error Check the expressions on the left and right sides of the operator invalid lt lt operands The lt lt operation contains an error Check the expressions on the left and right sides of the operator invalid lt operands The lt operation contains an error Check the expressions on the left and right sides of the operator invalid operand The operation contains an error Check the expressions on the left and right sides of the operator invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid gt operands The gt operation contains an error Check
104. 17549435e 38F FLT_MIN_10_EXP Minimum value of the power of 10 that can 37 be represented as a float type floating point numeric value FLT_MIN_EXP Maximum value of the power of the radix 125 that can be represented as a float type floating point numeric value REJ10J2009 0200 Rev 2 00 2010 04 01 208 7tENESAS R32C 100 Series C Compiler Package V 1 02 E Standard Library Macro name Contents Defined value FLT RADIX Radix of exponent in floating point 2 representation FLT_ROUNDS Method of rounding off a floating point number 1 Rounded to the nearest whole number REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 209 R32C 100 Series C Compiler Package V 1 02 E Standard Library limits h Function REJ10J2009 0200 Rev 2 00 2010 04 01 210 Defines the limitations applying to the internal processing of the compiler The following lists the macros that define these limits Macro name Contents Defined value MB LEN MAX Maximum value of the number of 1 multibyte character type bytes CHAR BIT Number of char type bits 8 CHAR_MAX Maximum value that a char type variable 255 or 127 When a compile can take on as value option fSC fsigned_char is specified CHAR_MIN Minimum value that a char type variable 0 or 128 When a compile can take on as value option fSC fsigned_char is specified SCHAR_MAX Maximu
105. 2 B Extended Functions Reference B 3 4 Sing Japanese Characters as Character Constants Figure B 43 shows the format for using Japanese characters as character constants Li Figure B 43 Format of Kanji code Description in Character Strings As with character strings precede the character constant with L and process it as a wchar_t type If as 3C in you use two or more characters as the character constant only the first character 3 lt becomes the character constant Figure B 44 shows examples of how to write Japanese character constants include lt stdlib h gt void func void wehar t JC 5 JC O L X JC L F JCIZ L FE JC 3 L 3 remainder omitted Figure B 44 Format of Kanji Character Constant Description Figure B 45 is a memory map of the array to which the character constant in Figure B 44 has been assigned JC 0 JC 1 JC 2 10 bytes JCs JC 4 address higher Figure B 45 Memory Location of wchar_t Type Character Constant Assigned Array REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 121 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 4 Default Argument Declaration of Function NC100 allows you to define default values for the arguments of functions in the same way as with the C facility This chapter describes NC100 s facility to declare the default arguments of functions B 4 1 Overview of Default Argument
106. 200 Rev 2 00 2010 04 01 RENESAS 238 R32C 100 Series C Compiler Package V 1 02 E Standard Library fwrite Input Output Functions Function Outputs the specified items of data to the stream Format include lt stdio h gt size_t fwrite buffer size count stream Method function Argument const void _far buffer Pointer of the output data size_t size Number of bytes in one data item size_t count Maximum number of data items FILE _far stream Pointer of the stream ReturnValue Returns the number of data items output Description Outputs data with the size specified in size to the stream Data is output by the number of times specified in count If an error occurs before the amount of data specified in count has been input this function returns the number of data items output to that point REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 239 R32C 100 Series C Compiler Package V 1 02 E Standard Library getc Function Reads one character from the stream Format include lt stdio h gt int getc stream Method macro Argument FILE _far stream Pointer of stream ReturnValue Returns the one input character Returns EOF if an error or the end of the stream is encountered Description Reads one character from the stream Interprets code 0x1A as the end code and ignores any subsequent data getchar Input Output Functions Function Reads one character from stdin Format include
107. 4 01 RENESAS 118 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 3 Description of Japanese Characters NC100 allows you to include Japanese characters in your C source programs This chapter describes how to do so B 3 1 Overview of Japanese Characters In contrast to the letters in the alphabet and other characters represented using one byte Japanese characters require two bytes NC100 allows such 2 byte characters to be used in character strings character constants and comments The following character types can be included o kanji e hiragana full size katakana half size katakana Only the following kanji code systems can be used for Japanese characters in NC100 J EUC excluding user defined characters made up of 3 byte code Shift JIS SJIS B 3 2 Settings Required for Using Japanese Characters The following environment variables must be set in order to use kanji codes default specifies Environment variable specifying input code system ________ NCKIN J Environment variable specifying output code system NCKOUT Figure B 39 is an example of setting the environment variables Include the following in your autoexec bat file set NCKIN SJIS set NCKOUT SJIS Figure B 39 Example Setting of Environment Variables NCKIN and NCKOUT In NC100 the input kanji codes are processed by the cpp100 preprocessor cpp100 changes the codes to EUC codes In the last stage of token anal
108. 5 Facility to control the array of structures 6 Facility to control the array of structures 7 Facility to specify absolute addresses for input output variables 8 Facility to specify the DMAC register of a external variable 9 Facility to declare functions using software interrupts 10 Facility to declare functions using software interrupts 11 Facility to write interrupt functions 12 Facility to declare assembler functions passed via register 13 Facility to describe inline assembler 14 Facility to describe inline assembler 15 Facility to declare functions calling with JSR A instruction 16 Facility to declare functions calling with JSR W instruction 17 Facility to output PAGE 18 Facility to declare Assembler macro function 19 Facility to declare interface functions of service call of realtime OS for R32C series 20 Facility to declare interface functions of service call of realtime OS for R32C series 21 Facility to declare alarm handler functions of realtime OS for R32C series 22 Facility to declare cyclic handler functions of realtime OS for R32C series 23 Facility to declare kernel interrupt handler functions of realtime OS for R32C series 24 Facility to declare task start functions of realtime OS for R32C series 25 Declares exclusion of address 0 relative addressing 26 Declares exclusion of address 0 relative addressing You can only specify the above 25 processing functions with pragma If you specif
109. 76931348623157e 308 DBL MAX _10_EXP Maximum value of the power of 10 that can be represented as a double type floating point numeric value 308 DBL_MAX EXP Maximum value of the power of the radix that can be represented as a doubletype floating point numeric value 1024 DBL_MIN Minimum value that a double type variable can take on as value 2 2250738585072014e 308 DBL_MIN_10_EXP Minimum value of the power of 10 that can be represented as a double type floating point numeric value 307 DBL_MIN_EXP Minimum value of the power of the radix that can be represented as a doubletype floating point numeric value 1021 FLT_DIG Maximum number of digits of float type decimal precision 6 FLT_EPSILON Minimum positive value where 1 0 FLT_EPSILON is found not to be 1 0 1 19209290e 07F FLT_MANT_DIG Maximum number of digits in the mantissa part when a _ float type floating point value is matched to the radix in its representation 24 FLT_MAX Maximum value that a float type variable can take on as value 8 40282347e 38F FLT_MAX_10_EXP Maximum value of the power of 10 that can be represented as a float type floating point numeric value 38 FLT_MAX_EXP Maximum value of the power of the radix that can be represented as a float type floating point numeric value 128 FLT_MIN Minimum value that a float type variable can take on as value 1
110. 9 0200 Rev 2 00 2010 04 01 322 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 22 ccom100 Warning Messages 3 9 Warning message Description and countermeasure assignment from volatile pointer to non volatile pointer The volatile property is lost by assignment from volatile pointer to non volatile pointer Check the statement description If the description is correct ignore this warning assignment in comparison statement You put an assignment expression in a comparison statement You may confuse with Check on it block level extern variable initialize forbid ignored An initializer is written in extern variable declaration in a function Delete the initializer or change the storage class can t get address from register storage class variable The amp operator is written for a variable of the storage class register Do not use the amp operator to describe a variable of the storage class register can t get size of bitfield The bit field is used for the operand of the sizeof operator Write the operand correctly can t get size of function A function name is used for the operand of the sizeof operator Write the operand correctly can t get size of function unit size 1 assumed The pointer to the function is incremented 4 or decremented Processing will be continued by assuming the increment or decrement value is 1
111. A1 A2 A3 Save registers 01H ROL R2R0 R3R1 R6R4 R7R5 A0 A1 A2 A3 lt Restore registers Return to C program Figure3 18 Example Coding of Interrupt Processing Assembler Function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 60 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique e Notes on Calling C Functions from Assembler Functions Note the following when calling a function written in C from an assembly language program 1 Cal the C function using a label preceded by the underscore _ or the dollar 2 Make sure the registers used in the assembler functions are saved before calling any C language function and that they are restored after returning from the C language function 3 3 3 Notes on Coding Assembler Functions Note the following when writing assembly language functions subroutines that are called from a C program a Notes on Handling B and U flags When returning from an assembler function to a C language program always make sure that the B and U flags are in the same condition as they were when the function was called b Notes on Handling FB Register If you modified the FB frame base register in an assembler function you may not be able to return normally to the C language program from which the function was called c Notes on Handling General purpose and Address Registers The general purpose registers R2RO R3R1 R6R4 and R7R5 and address registers AO A1 A2 and A3 ca
112. AS 201 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules R2R0 Argument x i R3R1 RORO Argument y j sed R3R1 Argument z k Variable j R6R4 Variable k lt spP Return address SP lt FB lt FB OLD FB Stack usage state for Stack usage state of 2 Stack usage state of 3 start of 1 main When ENTER instruction When JSR _ instruction completed completed OLD FB ESP Figure D 17 Process at entry to the function and process to call the function func REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 202 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules R2RO Argument x i R3R1 R7R5 aren R6R4 AO E FB of FB of Function Function main main R2RO Return value of func Variable sum Return Return address address lt SpP lt sp S n lt FB lt FB OLD FB OLD FB Stack uage state of 4 Stack uage state of 5 Stack uage state of 6 Stack uage state of 7 When ENTER instruction When EXITED instruction completed completed Figure D 18 Process to build the stack frame used in the function func and process to return from the function func to the function main REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 203 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules D 4 Securing auto Variable Area Variables of storage class auto are placed in the stack o
113. C 100 Series C Compiler Package V 1 02 F Error Messages Table F 9 ccom100 Error Messages 3 13 Error message Description and countermeasure conflict declare of variable name The variable is defined twice with different storage classes each time Use the same storage class to declare a variable twice conflict function argument type of variable name The argument list contains the same variable name Change the variable name declared register parameter function s body declared The function body for the function declared with pragma PARAMETER is defined in C Do not define in C the body for such function default function argument conflict The default value of an argument is declared more than once in prototype declaration Make sure that the default value of an argument is declared only once default is duplicated The default value is used more than one time Use only one default within one switch statement do while struct union statement The struct or union type is used in the expression of the do while statement Use the scalar type for an expression in the dowhile statement do while void statement y The void type is used in the expression of the dowhile statement Use the scalar type for an expression in the dowhile statement duplicate frame position defind Auto variable is described more than twice variable name Write the type specifier corre
114. C 100 series NC100 converts programs written in C into assembly language source files for the R32C 100 series You can also specify compiler options for assembling and linking to generate hexadecimal files that can be written to the microcomputer Please be sure to read the precautions written in this manual before using NC100 Microsoft MS DOS Windows and Windows NT are either registered trademarks or trademarks or Microsoft Corporation in the United States and other countries HP UX is a registered trademark of Hewlett Packard Company IBM and AT are registered trademarks of International Business Machines Corporation Intel and Pentium are registered trademarks of Intel Corporation Adobe and Acrobat are registered trademarks of Adobe Systems Incorporated Netscape and Netscape Navigator are registered trademarks of Netscape Communications Corporation in the U S and other countries All other brand and product names are trademarks registered trademarks or service marks of their respective holders Terminology The following terms are used in the NC100 User Manuals Term Meaning NC100 Compiler package for R32C 100 series nc100 Compile driver and its executable file AS100 Assembler package included in Compiler package for R32C 100 series as100 Relocatable macro assembler and its executable file een toa Integrated development environment of attachment Workshop Description of Symbols The following symbo
115. Compiler Package V 1 02 B Extended Functions Reference inline void func int i void main void func 1 Error Message Error ccom sample c line 5 inline function s body is not declared previously gt func 1 Sorry compilation terminated because of these errors in main Figure B 55 Example of inappropriate code of inline function 1 Furthermore if any function is defined as an inline function after being used as an ordinary function the specification of inline has no effect and all of such a definition is handled as static functions Figure B 56 int func int i void main void func 1 inline int func int i return i Warning Message Warning ccom smp c line 10 inline function is called as normal function before change to static function Figure B 56 Example of inappropriate code of inline function 2 5 Regarding the address of an inline function The inline function itself does not have an address Therefore if the amp operator is used for an inline function the software assumes an error Figure B 57 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 127 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference inline int func int i void main void return i int f int f amp func Error Message Error ccom sample c line 10 can t get inline function s address by amp operator gt
116. Do not increment or decrement the pointer to a function char array initialized by wchar_t string The array of type char is initialized with type wchar_t Make sure that the types of initializer are matched case value is out of range The value of case exceeds the switch parameter range Specify correctly character buffer overflow The size of the string exceeded 512 characters Do not use more than 511 characters for a string character constant too long Olt ee ope There are too many characters in a character constant characters enclosed with single quotes Write it correctly constant variable assignment In this assign statement substitution is made for a variable specified by the const qualifier Check the declaration part to be substituted for cyclic or alarm handler function has argument The function specified by pragma CYCHANDLER or ALMHANDLER is using an argument The function cannot use an argument Delete the argument enumerator value overflow size of unsigned char The enumerator value exceeded 255 Do not use more than 255 for the enumerator otherwise do not specify the startup function fchar_enumerator enumerator value overflow size of e The enumerator value exceeded 65535 unsigned int Do not use more than 65535 to describe the enumerator enum s bitfield e Anenumeration is used as a bit field member Use a different type of member ex
117. ENESAS 321 R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 21 ccom100 Warning Messages 2 9 Warning message Description and countermeasure pragma directive conflict e pragma of different functions is specified for one function gt Write it correctly pragma DMAC duplicate e The same pragma DMAC is defined twice gt Write it correctly e pragma DMAC variable must be far pointer to object for variable name ignored The pragma DMAC declared variable must be a far pointer to an object or incomplete type The DMAC declaration was ignored Define pragma DMAC correctly pragma DMAC variable must be unsigned long for variable name ignored ey The pragma DMAC declared variable must be of an unsigned long type The DMAC declaration was ignored Define pragma DMAC correctly pragma DMAC s variable must be pre declared ignored ey Variable declared by pragma DMAC needs a type declaration Write it correctly pragma DMAC register conflict Multiple variables are allocated to the same register Write it correctly pragma DMAC unknown register name used Unknown register is used in pragma DMAC declaration Write it correctly pragma JSRA illegal location ignored Do not put pragma JSRA inside function scope Write pragma JSRA outside a function pragma JSRW illegal location ignored Do not put pragma JSRW inside functi
118. F pragma AD porA 0x60 endif Figure C 14 Usage Example of Predefined Macros REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 185 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules Appendix D C Language Specification Rules This appendix describes the internal structure and mapping of data processed by NC100 the extended rules for signs in operations etc and the rules for calling functions and the values returned by functions D 1 Internal Representation of Data D 1 1 Integral Type Table D 1 shows the number of bytes used by integral type data Table D 1 Data Size of Integral Type Type Existence of sign Bit size Range of values _Bool No 8 0 1 char No 8 between 0 and 255 unsigned char signed char Yes 8 between 128 and 127 int Yes 16 between 32768 and 32767 short signed int signed short unsigned int No 16 between 0 and 65535 unsigned short int Yes 32 between 2147483648 and 2147483647 long signed int signed long unsigned int No 32 between 0 and 4294967295 unsigned long long long Yes 64 between 9223372036854775808 and signed long long 9223372036854775807 unsigned long long No 64 18446744073709551615 float Yes 32 between 1 17549435e 38F and 3 40282347e 38F double Yes 64 between 2 2250738585072014e 308 and long double 1 7976931348623157e 308 near pointer No 16 between 0 and OxFFFF far pointer No 32 between 0 and OxFFFFFFFF J The _B
119. FILE_iobl define stdin amp _iob 0 Fundamental input define stdout amp _iob 1 Fundamental output define stdaux amp _iob 2 Fundamental auxialiary input output define stdprn amp _iob 3 Fundamental printer output define stderr stdout NC no support a aaa SALA ERER ED ERASE ORAL E LEER CLA ERO TSEME SERRE RE TROND OR ONE define _IOREAD 1 Read only flag define _IOWRT 2 Write only flag define _IOEOF 4 End of file flag define _IOERR 8 Error flag define _IORW 16 Read and write flag define _NFILE 4 Stream number define _TEXT 1 Text mode flag define _BIN 2 Binary mode flag remainder omitted Figure E 4 Stream Definition in stdio h Let s look at the elements of the file structures shown in Figure E 4 Items 1 to 6 correspond to 1 to 6 in Figure E 4 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 296 R32C 100 Series C Compiler Package V 1 02 E Standard Library 1 char _buff Functions scanf and fscanf read one character ahead during input If the character is no use function ungetc is called and the character is stored in this variable If data exists in this variable the input function uses this data as the input data 2 int_cnt Stores the _buff data count 0 or 1 8 int_flag Stores the read only flag IOREAD the write only flag JOWRT the read write flag IORW the end of file flag IOEOF and the
120. Functions to execute and terminate programs and for jumping from the currently executing function to another function 7 Mathematical Functions These functions require time e Therefore pay attention to the use of the watchdog timer 8 Integer Arithmetic Functions Functions for performing calculations on integer values 9 Character String Value Convert Functions Functions for converting character strings to numerical values 10 Multi byte Character and Multi byte Character String Manipulate Functions Functions for processing multi byte characters and multi byte character strings 11 Locale Functions Locale related functions REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 216 R32C 100 Series C Compiler Package V 1 02 E Standard Library E 2 2 List of Standard Library Functions by Function 1 String Handling Functions The following lists String Handling Functions Table E 2 String Handling Functions Type Function Contents Reentrant Copy strepy Copies a character string O strncpy Copies a character string n characters O Concatenate strcat Concatenates character strings O strncat Concatenates character strings n characters O Compare stremp Compares character strings O streoll Compares character strings using locale information O stricmp Compares character strings All alphabets are handled as O upper case letters strncmp Compares character
121. GN _file test c align ine 4 H C_ SRC glb func_proto func_proto 10 ine 5 C_SRC return k j i add R3R1 R6R4 gt j add R6R4 R2R0 rts E1 7 This passes the third argument k via stack 8 This passes the first argument i via register 9 This passes the second argument j via register 10 This is the start address of function func_proto Figure D 11 Compile Result of Sample Program sample c 1 In the compilation result 1 of the sample program sample c in Figure D 10 the first second and third arguments are passed via registers because the function func_proto has its prototype declared Furthermore since the arguments to the function are passed via registers the symbol name for the beginning address of the function is taken after func_proto written in the C language source file by prefixing it with the dollar mark namely func_proto REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 196 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules HHH FUNCTION func_no_proto H FRAME ARG i sze 4 SC offset8 o0 i 11 FRAME ARG j size 4 offset12 l FRAME ARG ___k size 4 offset16 ____ _____ HH H ARG Size 12 Auto Size 0 Context Size 8 align line 12 HHHC SRC glo _func_no_proto _func_no_proto 12 enter 00H ine 13 H C_SRC return k j i mov 16 FB R2RO k add l 12 FB R2RO gt j add l 8 FB R2R
122. HEAPSIZE _ msize endif initialize standard I O i 11 if STANDARD IO _ 1 glb init Call _init G jsra _ init endif invoke main function 12 idc 0H FB for DEBUGGER glb _main jsra _main 10 Initializes the heap area Comment out this line if no memory management function is used 11 Calls the init function which initializes standard I O Comment out this line if no I O function is used 12 Calls the main function Interrupt is not enable when calls main function Therefore permits interrupt by FSET command when uses interrupt function Figure 2 10 Startup Program List 5 ncrt0 a30 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 25 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler exit function 13 glb _exit glib exit _ exit End of execution exit jmp _exit einsf dummy interrupt function 14 glb dummy_int dummy_int reit end IRI II KIKI KI KIER IK IIH IIIA IK IAI III IIIS IK III III IIIA IK IAI kk kkk kkk a End of R32C 100 start up 3 RIKKI AIK II KI KIER IK IIH III III IK IKI III IK IK IIIA kk dekk kkk k IAI III III a 13 exit function 14 Dummy interrupt processing function Figure 2 11 Startup Program List 6 ncrt0 a30 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 26 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler II KIKI AK III KI
123. KIER IK III KIKI IERIE I III IIIA III IKI IK III IARI C COMPILER for R32C 100 Copyright C XXXX Renesas Electronics Corp and Renesas Solutions Corp All rights reserved This program is applicable when using the basic I O library ncrt0 a30 startup program ld sect100 inc v X X XXXX XX XX XX XX XX XXX Exp Arrangement of section NEAR RAM SECTIONS section data_SB8 data org 00000400H data_SB8_top section bss_SB8 data align bss_SB8_top section data NEAR data align data _ NEAR _top section bss _NEAR data align bss_NEAR_top section data MONI data align data_MON1_top section bss_MON1 data align bss_MON1_top section data_MON2 data align data_MON2_top section bss_MON2 data align bss_MON2_top section data_MON3 data align data_MON3_top section bss_MONS data align bss_MON3_top section data_MON4 data align data_MON4_ top section bss_MON4 data align bss_MON4_top STACK SECTION section stack data align blkb STACKSIZE align stack_top blkb ISTACKSIZE align istack_top Figure 2 12 Startup Program List 7 sect100 inc REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 27 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler HEAP SECTION section heap data align heap_top blkb HEAPSIZE SB RELATIVE RAM SECTIONS section dat _SB8 data i org 00008000H data_SB8_top section bss _SB8 data align bss_
124. M START bset 1 S Reference to bitfield bitO of structure s HHH ASM END _line 10 HHHEC SRC rts E1 SECTION bss NEAR DATA ALIGN glb _s S blkb 1 END Figure B 30 Example of Referencing Bit Field of Symbol REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 114 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 2 5 Specification Not Dependent on Storage Class The variables written in C language can be used in the asm function without relying on the storage class of that variable auto register extern or static variable Consequently any variable written in C language can be used in the asm function by writing it in the format shown in Figure B 312 asm op code RO variable name Figure B 31 Description Format Not Dependent on Variable s Storage Class You can only specify one variable name using this format Figure B 32 shows examples of referencing register variables and the results of compiling C source file extern int e_val void func void int f_val register int r_val static int s_ val asm mov w 1 e_val Reference to external variable asm mov w 2 f_vall lt Reference to auto variable asm mov w 3 r_val lt Reference to register variable asm mov w 4 s_ val lt Reference to static variable asm mov w f_val r_val Assembly language source file compile result glb _ fu
125. NTCALLAint NoAassembler function name register name register name 10 pragma INTCALLAint NoAC language function name 11 pragma INTERRUPTA B E F R V Ainterrupt handling vector numberAinterrupt handling function name 12 pragma PARAMETERAassembler function name register name register name 13 pragma ASM 14 pragma ENDASM 15 pragma JSRAAfunction name 16 pragma JSRWAfunction name 17 pragma PAGE 18 pragma __ASMMACROAfunction name register name 19 pragmaMRCALLAS stack sizeAINT numberA function code service call name type of argument 20 pragma MRPARAMETERAservice call name type of quotation 21 pragma ALMHANDLERAalarm handler function name 22 pragma CYCHANDLERAcyclic handler function name 23 pragma INTHANDLERA E R Ainterrupt handler function name 24 pragma TASKAtask start function name 25 pragma EXTMEM variable name 26 pragma EXTMEM function name REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 183 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications _ Description 1 Facility to arrange in the rom section 2 Facility to describe variables using SB relative addressing 3 Facility to describe variables using SB relative 16 bit displacement addressing 4 Facility to alter the section base name
126. Reads the format of integral constant if the base of value is zero ReturnValue ReturnValue 0L Does not constitute a number ReturnValue OL Returns the configured number in long type Description When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 282 R32C 100 Series C Compiler Package V 1 02 E Standard Library strtoul Character String Value Convert Function Function Converts a character string into an unsigned long type integer Format include lt string h gt unsigned long strtoul s endptr base Method function Argument const char _far s Pointer to the converted character string char _far _far endptr Pointer to the remaining character strings that have not been converted int base Base of values to be read in 0 to 36 Reads the format of integral constant if the base of value is zero ReturnValue ReturnValue 0L Does not constitute a number ReturnValue OL Returns the configured number in long type Description When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization strxfrm Character String Value Convert Functions Function Converts a character string using locale information Format include lt string h gt size_t strxfrm sl s2 n Method function Argument char _far s1 Pointer to an
127. SAS R32C 100 Series C Compiler Package V 1 02 C Compiler User s Manual Publication Date Apr 1 2010 Rev 2 00 Renesas Electronics Corporation Published by 1753 Shimonumabe Nakahara ku Kawasaki shi Kanagawa 211 8668 Japan Edited by Renesas Solutions Corp 2010 Renesas Electronics Corporation All rights reserved Printed in Japan R32C 100 Series C Compiler Package V 1 02 C Compiler User s Manual 2rCENESAS Renesas Electronics Corporation 1753 Shimonumabe Nakahara ku Kawasaki shi Kanagawa 211 8668 Japan REJ10J2009 0200
128. SB8_top i section data _SB16 data align data_SB16_top section bss _SB16 data align bss_SB16_top FAR RAM SECTIONS section data _FAR data T data_FAR_top section bss FAR data align bss_FAR_top EXTENDED RAM SECTIONS section data_EXT data org 00800000H data_EXT_top section bss EXT data align bss_EXT_top EXT ENDED ROM SECTIONS section data _EXT_ INIT romdata org OFF000000H data_EXT_INIT_top section rom_EXT romdata align rom_EXT_top section _ program_EXT code align FAR ROM SECTIONS l section rom_FAR romdata org OFFE00000H rom_FAR_top Figure 2 13 Startup Program List 8 sect100 inc REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 28 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler section data_NEAR_INIT romdata org OFFFFOOOOH data_NEAR_INIT_top section data MON1_INIT romdata align data_MON1_INIT_top section data_MON2_INIT romdata align data_MON2_INIT_top section data MONS INIT romdata align data_MON3_INIT_top section data MON4_INIT romdata align data_MON4_INIT_top section data_SB8_INIT romdata align data_SB8_INIT_top section data_SB16_INIT romdata align data_SB16_INIT_top section data_FAR_INIT romdata align data_FAR_INIT_top SWITCH TABLE SECTIONS section _ switch_table romdata align CODE SECTIONS oes section program code align section interrupt code align NEAR ROM SECTIONS nn
129. SEN E EE E i HHC_ SRC s func s l movw 4 FB RO s i _line 2 HHC_ SRC movw RO 2 FB i l l _line 3 HHC_SRC return i Inline storage class have codes mov w RO R1 embedded directly add w 0001H RO l l ine 10 l C_SRC s func s mov w R1 4 FB s l ine SSS H HC_SRC exitd E1 END Figure B 54 Compile Result of sample program smp a30 B 5 3 Restrictions on inline Storage Class When specifying the inline storage class pay attention to the following 1 Regarding the parameter of inline functions The parameter of an in line function cannot be used by structure and union It becomes a compile error 2 Regarding the indirect call of inline functions The indirect call of an in line function cannot be carried out It becomes a compile error when a indirect call is described 3 Regarding the recursive call of inline functions The recursive call of an in line function cannot be carried out It becomes a compile error when a recursive call is described 4 Regarding the definition of an inline function When specifying inline storage class for a function be sure to define the body of the function in addition to declaring it Make sure that this body definition is written in the same file as the function is written The description in Figure B 55 is processed as an error in NC100 REJ10J2009 0200 Rev 2 00 2010 04 01 126 2tENESAS R32C 100 Series C
130. SRC glb _main _line HHHEC SRC mov w mov w jsr ine H C_SRC mov w mov w jsr _line HHHC SRC mov w mov w jsr _line HEC SRC rts omitted _main 5 func 0002H R1 0001H RO func 6 func 3 0002H R1 0003H RO func 7 func 3 5 0005H R1 0003H RO func 8 second argument 2 first argument 1 second argument 2 first argument 3 second argument 5 first argument 3 Note In NC100 arguments are stacked in revere order beginning with the argument that is declared last in the function In this example arguments are passed via registers as they are processed Figure B 48 Compiling Result of smp1 c smp1 a30 A variable can be written for the argument of a function Figure B 49 shows an example where default arguments are specified with variables Figure B 50 shows a compile result of the sample program shown in Figure B 49 short near sym short func short i sym Default argument is specified with a variable void main void func Function is called using variable sym as argument omitted Figure B 49 Example for specifying default argument with a variable smp2 c REJ10J2009 0200 Rev 2 00 2010 04 01 123 2tENESAS R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference _main _line6 mov w _sym 16 RO0 Function is called using variable sym as argument jsr func
131. The two character strings are equal ReturnValue gt 0 The first character string s1 is greater than the other ReturnValue lt 0 The second character string s2 is greater than the other Description Compares each byte of n characters of two character strings ending with NULL However all letters are treated as uppercase letters When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization strobrk String Handling Functions Function Searches the specified character in a character string from the other character string Format include lt string h gt char _far strpbrk s1 s2 Method function Argument const char _far s1 Pointer to the character string to be searched in const char _far s2 Pointer to the character string of the character to be searched for ReturnValue Returns the position pointer where the specified character is found first Returns NULL if the specified character cannot be found Description Searches the specified character s2 from the other character string in s1 area You cannot search for 0 When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 279 R32C 100 Series C Compiler Package V 1 02 E Standard Library strrchr String Handling Functions Function Searches t
132. a Ug This is a syntax error Do not this syntax because t is reserved for extended use in the future Write the description correctly this function used before with non default argument ey A function once used is declared as a function hat has a default argument Declare the default argument before using a unction this interrupt function is called as normal function before Yoel A function once used is declared in pragma NTERRUPT An interrupt function cannot be called Check the ontent of pragma too big octal character The character constant or the octal constant in he character string exceeded the limit value 255 n decimal Do not use a value greater than 255 to describe he constant too few parameters Arguments are insufficient compared to the number f arguments declared in prototype declaration Check the number of arguments too many parameters Arguments are excessive compared to the number f arguments declared in prototype declaration Check the number of arguments unknown pragma STRUCT xxx ey pragma STRUCTxxx cannot be processed his line will be ignored Write correctly Unknown debug option dx The option dx cannot be specified Specify the option correctly Unknown function option Wxxx The option Wxxx cannot be specified Specify the option correctly Unknown function option fx The option fx cannot be specified Specify the option co
133. a command Figure B 84 Description Example of Assembler Macro Function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 156 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 8 3 Commands that Can be Written by Assembler Macro Function The following shows the assembler commands that can be written using assembler macro functions and their functionality and format as assembler macro functions ABS Function Returns the absolute value of val Syntax include lt asmmacro h gt When calculated in 8 bits static signed char abs_b signed char val When calculated in 16 bits static short int abs_w short int val When calculated in 32 bits static long int abs_1 long int val MAX Function Returns the value val1 or val2 whichever is found larger by comparison Syntax include lt asmmacro h gt When calculated in 8 bits static signed char max_b signed char vall signed char val2 When calculated in 16 bits static short int max_w short int vall short int val2 When calculated in 32 bits static long int max_l long int vall long int val2 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 157 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference MIN Function Returns the value val1 or val2 whichever is found smaller by comparison Syntax include lt asmmacro
134. aas ra Type onein iene ests AA O A R EN A E a E AEA 186 D 1 2 Floating Type ssenssiinie oi nn NEN AONANE AREAS ONNEA 187 D 1 3 En merat r PY PC ve ends ss scezcecsesesasocledstsosnsesecstsesesleseecdescvscectesclosctesivacessvsdeasedesededeesvsedotee eececsinesisesuestsissdeicsbestee 188 D 1 4 POumiber DY Peis s islsadacechtssdasdesachcaceancshsecdusaseseutadeceesaehielastasadoceassesde0daencoceaesdadeesaaddedestagecesshiasanaachcaatangdacecandbseeteehes 188 D 1 5 Array yP enen aAA NAAN N R dh AA N EA 188 D 1 6 Structure types aneneen e a a a a E R e N 189 D 1 7 LOSE a lE AAE EE A E EE E 190 D 1 8 Bitheld Types iti zstes di sesicitvaseitesibed a aN EEE T T E E N i ha 190 D2 Sen Erenson RULES earnais a a ELEN AAAA ONEONE 191 D3 Tunto Call Rules esenieee n aa E A E E E NE 192 D 3 1 Rules of Return Value c cccccssssssesssesssescssscsescssscsescssscsescssesssssssesssesssessseseseseseacaeacseacsescacscacenseeseneacaeacseacaeecsees 192 D 3 2 Rules on Argument Transfer cccccccscscssscssscssscssscscscsescscsssssesessscsssscssscscseseseacsescsescssscsescseesseseeseaesesesescassesees 192 D 3 3 Rules for Converting Functions into Assembly Language Symbols ccccssssssesssessesssesssseseessseaeeeeeees 194 D 3 4 Interface between FUunctions c cccccsssssssssssssssssssssssssssssssssssssssssssssssssesssnssssssesessesssussssesssasacsesacsessesseaeeeseees 199 DE S curing auto Variable Area E E A AEE A 204 D 5 Rules of Escaping of the Re
135. alid J e E and f conversion n where n is the precision numerals are output after the decimal point e g and G conversion Valid characters in excess of n where n is the precision are not output 4 TLiorh J I d i 0 u x X and n conversion is performed on long int and unsigned long int parameters o h d i o u x and X conversion is performed on short int and unsigned short int parameters 1 Tf I or h are specified in other than d i o u x X or n conversion they are ignored J L e E f g and G conversion is performed on double parameters 1 In the standard C specifications variables e E f and g conversions are performed in the case of Lon long double parameters In NC100 long double types are processed as double types Threfore if you specify L the parameters are processed as double types REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 262 R32C 100 Series C Compiler Package V 1 02 E Standard Library putc Input Output Functions Function Outputs one character to the stream Format include lt stdio h gt int putc c stream Method macro Argument int c Character to be output FILE _far stream Pointer of the stream ReturnValue Returns the output character if output normally Returns EOF if an error occurs Description Outputs one character to the stream putchar Function Outputs one character to stdout Format include lt stdio h gt int putchar c Met
136. alizes the stream Also calls speed and init_prn in the function to make the initial settings of the UART and Centronics output device init is normally used by calling it from the startup program isalnum Character Handling Functions Function Checks whether the character is an alphabet or numeral A Z a z 0 9 Format include lt ctype h gt int isalnum c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if an alphabet or numeral Returns 0 if not an alphabet nor numeral Description Determines the type of character in the parameter REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 242 R32C 100 Series C Compiler Package V 1 02 E Standard Library isaloha Character Handling Functions Function Checks whether the character is an alphabet A Z a z Format include lt ctype h gt int isalpha c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if an alphabet Returns 0 if not an alphabet Description Determines the type of character in the parameter iscntrl Character Handling Functions Function Checks whether the character is a control character 0x00 0x1f 0x7 Format include lt ctype h gt int iscntrl c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if a numeral Returns 0 if not a control character D
137. alues are indicated by preceding the value with REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 167 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications Table C 7 Extended Notation List Notation Escape sequence Notation Trigraph sequence single quote constant octal y quotation mark x constant hexadecimal backslash 72 express character question mark express character a bell express character b backspace 2 express character f form feed lt express character n line feed 29 express character r return 22 gt express character t horizontal tab T express character v vertical tab express character d Character Literals Character literals are written in double quote marks as in character string The extended notation shown in Table C 7 for character constants can also be used for character literals e Operators NC100 can interpret the operators shown in Table C 8 Table C 8 Operators List monadic operator logical operator amp amp binary operator conditional operator comma operator T address operator amp pointer operator k bitwise operator lt lt assignment operators gt gt amp r amp relational operators gt
138. am is a possibility of different actions If you do not want the source line information to be modified use the One_break_source_debug CONBSD option to suppress optimization dp Optimization Function Although the ROM size may somewhat increase optimization is performed to obtain the fastest speed possible This option can be specified along with the g and O options OR_MAX ORM Optimization Function Optimizes ROM size in preference to speed When this option is used the effect is same with O5 O50A OGJ OR fD32 NA fUF options OS_MAX OSM Optimization Function Although the ROM size may somewhat increase optimization is performed to obtain the fastest speed possible When this option is used the effect is same with 04 OGJ OGTT OS OSA OSTT OLU 10 fD32 fUF options Ocompare_byte_to_word OCBTW Optimization Function Compares consecutive bytes of data at contiguous addresses in words Notes This option is only valid if you specify option O to 5 OR OR_MAX ORM OS or OS_MAX OSM REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 73 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Oconst OC Optimization Function Supplement Optimizes code generation by replacing reference to variables to declared by the const qualifier with constants This is effective even when other than the
139. and ifdef Combinations of ifdef else endif can be nested There is no set limit to the number of levels of nesting but it depends on the amount of available memory ifndef elif else endif Function Performs conditional compilation Examines the macro defined or not Format ifndefAidentifier HelifAconstant expression else endif Description Ifan identifier isn t defined ifndef processes the program that follows You can also describe the followings if A defined A identifier ifA defined A identifier else is used in a pair with ifndef Do not specify any tokens between else and the line feed You can however insert a comment elifis used in a pair with if ifdef or ifndef endif indicates the end of the range controlled by ifndef Always be sure to enter endif when using command ifndef Combinations of ifndef else endif can be nested There is no set limit to the number of levels of nesting but it depends on the amount of available memory REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 181 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications include Function Takes in the specified file Format G includeA lt file name gt 2 includeA file name 8 includeAidentifier Description 1 Takes in lt file name gt from the directory specified by nc100 s command line option I Searches
140. aracter that cannot be used in octal notation Use numerals 0 to 7 to describe octal constants octal_character is out of range The octal constant contains some character that cannot be used in octal notation Use numerals 0 to 7 to describe octal constants overflow in floating value converting to integer A very large floating point number that cannot be stored in integer type is being assigned to the integer type Reexamine the assignment expression REJ10J2009 0200 Rev 2 00 2010 04 01 326 2tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 26 com100 Warning Messages 7 9 Warning message Description and countermeasure old style function declaration The function definition is written in format prior to ANSI ISO C Write the function definition in ANSI ISO format prototype function is defined as non prototype function before ey The non prototyped function is redefine prototype declaration Write identifier correctly Unite ways to declare function type redefined type e Redwfine typedef Check typedef redefined type name of qualify e The same identifier is defined twice or more in typedef gt e register parameter function used before as stack parameter function The function for register argument is used as a function for stack argument before Write a prototype declaration before using the function RESTRICT qualif
141. assed via a register by writing a prototype declaration Increased speed and reduced code size can be expected by passing arguments via a register Also the prototype declaration causes the compiler to check function arguments Increased program reliability can be expected from this Therefore Renesas recommends using this option whenever possible REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 93 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Wsitdout Warning Options Function Outputs error messages to the host machine s standard output stdout Supplement Use this option to save error output etc to a file by using Redirect in the Microsoft Windows TM Notes In this Compiler for Microsoft Windows TM errors from as100 and In100 invoked by the compile driver are output to the standard output regardless of this option Wstop_at_link WSAL Warning Options Function Stops linking the source files if a warning occurs during linking to suppress generation of absolute module files Also a return value 10 is returned to the host OS Wstop_at_warning WSAW Warning Options Function Stops compiling the source files if a warning occurs during compiling and returns the compiler end code 10 Supplement If a warning occurs when compiling the compilation by default is terminated with the end code 0 terminated normally Use this option when you are using the make utility etc and want to sto
142. at the variables and functions declared with pragma EXTMEM too can have their location addresses handled by an ordinary far pointer Rules 1 Declaration with pragma EXTMEM is ignored unless it declares a variable name or function name 2 pragma EXTMEM is not applied to the static variables declared within a function 3 This extended feature has priority over the near and far qualifiers declared 4 This extended feature cannot be used in combination with other extended features of pragma for one variable or function at a time Example pragma EXTMEM extfunc pragma EXTMEM extvar short extfunc void short far extvar short far p void func void extvar extfunc p amp extvar Figure B 61 Example Use of pragma EXTMEM Declaration Supplement When this extended feature is specified the compiler generates the following sections Name Attribute Content data_EXT DATA Data with initial values declared with pragma EXTMEM bss_EXT DATA Data without initial values declared with pragma EXTMEM rom_EXT ROMDATA const qualified data declared with pragma EXTMEM data_EXT_ INIT ROMDATA Initial value of data_EXT section program_EXT CODE Function code declared with pragma EXTMEM REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 134 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma MONITORn Directive to specify the
143. ation Table A 4 Options for Debugging Option Option g Outputs debug information to an assembly language source file extension a30 This makes C level debugging of programs possible genter Always outputs an enter instruction when calling a function Be sure to specify this option when using the debugger s stack trace function i Outputting debugging information Function Outputs debugging information to an assembler source file extension a30 Notes When debugging your program at the C language level always specify this option Specification of this option does not affect the code generated by the compiler genter Outputting enter instruction Function Always output an enter instruction when calling a function Notes 1 When using the debugger s stack trace function always specify this option Without this option you cannot obtain the correct result 2 When this option is specified the compiler generates code to reconstruct the stack frame using the enter command at entry of the function regardless of whether or not it is necessary Consequently the ROM size and the amount of stack used may increase REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 69 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 5 Optimization Options Table A 5 shows the command line options for optimizing program execution speed and ROM capacity Table A 5 Optimiza
144. bits static void smovf_w const void _far pl void _far p2 unsigned long int count When calculated in 32 bits static void smovf_l1 const voiid _far pl void _far p2 unsigned long int count SMOVU Function Strings are transferred from the source address indicated by p1 to the destination address indicated by p2 in the address incrementing direction until zero is detected There is no return value Syntax include lt asmmacro h gt When calcul ated in 8 bits static void When calcul smovu_b const void _far pl void _far p2 ated in 16 bits static void smovu_w const void _far pl void _far p2 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 160 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference SOUT Function Strings are transferred in the address incrementing direction from the source address indicated by p1 to the destination address indicated by p2 as many times as indicated by count There is no return value Syntax include lt asmmacro h gt When calculated in 8 bits staticvoid sout_b const void_far pl volatile void_far p2 unsigned long int count When calculated in 16 bits staticvoid sout_w const void_far pl volatilevoid_far p2 unsigned long int count When calculated in 32 bits staticvoid sout_l const void_far pl volatile void_far p2 unsigned long int count SSTR Function Strings are s
145. ble declared in a function the pragmaSB16DATA declaration is ignored as invalid 5 The variable declared to be pragma SB16DATA is placed in a SBI6DATA attribute section when allocating memory for it 6 If pragma SBI6DATA is declared for ROM data declaration of pragma SB16DATA becomes invalid Example pragma SBI6DATA sym data int far sym_data void func void sym_data 1 Figure B 64 Example Use of oragma SB16DATA Declaration Supplement NC100 is premised on an assumption that the SB register will be initialized after reset and will thereafter be used as a fixed quantity 1 Do not write a pragma SB16DATA declaration for ROM data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 138 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma SBDATA SB Relative Addressing Using of 8bit displacement Variable Description Function Function Declares that the data uses SB relative addressing of 8bit displacement Syntax pragma SBDATAAvaluable name Description The R32C 100 series allows you to choose instructions that can be executed efficiently by using SB relative addressing pragma SBDATA declares that SB relative addressing can be used for the variable when referencing data This facility helps to generate ROMefficient code Rules 1 The variable declared to be pragma SBDATA is declared by the assembler s pseudo instruction SBSYM 2 If pragma SBDATA is sp
146. break case char const continue default do double else enum extern far float for goto if inline int long near register restrict return short signed sizeof static struct switch typedef union unsigned void volatile while b Identifiers Identifiers consist of the following elements e The 1st character is a letter or the underscore A to Z a to z or __ J The 2nd and subsequent characters are alphanumerics or the underscore A to Z a to z 0 to 9 or_ Identifiers can consist of up to 200 characters However you cannot specify Japanese characters in identifiers c Constants Constants consists of the followings J Integer constants e Floating point constants o Character constants REJ10J2009 0200 Rev 2 00 2010 04 01 166 2tENESAS R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications 1 Integer constants Tn addition to decimals you can also specify octal and hexadecimal integer constants Table C 6 shows the format of each base decimal octal and hexadecimal Table C 6 Specifying Integer Constants Base Notation Structure Example Decimal None 0123456789 15 Octal Start with 0 zero 01234567 017 Hexadecimal Start with OX or Ox 0123456789ABCDEF OXF or Oxf 0123456789abcdef Determine the type of the integer constant in the following order according to the value Octal and hexadecimal signed int unsigned int signed long unsigned long signed long long unsi
147. bss rom SB8 SB16 FAR far attribute EXT SB8 SBDATA attribute data bss MON1 MON2_ SB16 SB16DATA attribute data bss MON3 MON4 EXT EXTMEM attribute data bss rom MON1 MONITOR attribute data bss MON2 MONITOR2 attribute data bss MON3 MONITOR3 attribute data bss MON4 MONITOR4 attribute data bss Tabel 2 12 shows the contents of sections other than those based on the naming rules described above Tabel 2 12 Section Names Section name Contents fvector This section stores the contents of the Micro Processor s fixed vector heap This memory area is dynamically allocated during program execution by memory management functions e g malloc This section can be allocated at any desired location of the Micro Processor RAM area program Stores programs stack This section is used as a stack This section can be allocated to any desired location of the RAM areas in the microcomputer switch_table The section to which the branch table for switch statements is allocated This section is generated only with the fSOS option vector This section stores the contents of the Micro Processor s interrupt vector table The interrupt vector table can be allocated at any desired location of the Micro Processor s entire memory space by intb register relative addressing For more information refer to the Micro Processor Hardware Manual These sections are mapped to memory according to the settings in t
148. c that is first encountered in character string g Returns NULL when character string s does not contain character c Description Searches for character c starting from the beginning of area s You can also search for 0 When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization 1 There must be adequate space to accommodate s1 plus s2 REJ10J2009 0200 Rev 2 00 2010 04 01 273 2tENESAS R32C 100 Series C Compiler Package V 1 02 E Standard Library strcmp String Handling Functions Function Compares character strings Format include lt string h gt int strcmp sl s2 Method Macro function Argument const char _far s1 Pointer to the first character string to be compared const char _far s2 Pointer to the second character string to be compared ReturnValue ReturnValue 0 The two character strings are equal ReturnValue gt 0 The first character string s1 is greater than the other ReturnValue lt 0 The second character string s2 is greater than the other Description Usually the program code described by macro is used for this function In using the function in a library please describe it as undef strcmp after description of include lt string h gt Compares each byte of two character strings ending with NULL When you specify options O 3 5 OR or OS the system may selects another func
149. ccomi00 ccom100 is the executable file of the compiler itself C source programs processed by cpp100 are converted to assembly language source programs that can be processed by as100 1 2 5 aopt100 aopt100 is the assembler optimizer It optimizes the assembler codes output by ccom100 1 26 Call Walker amp gensni CallWalker is the utility to graphically display the relationship between stack sizes and function calls that is needed for program operation Similarly gensni is the utility to analyze the necessary information CallWalker loads a stack information file x30 that is output by gensni to display the amount of stacks used The amount of stacks used by an assembly program that cannot be output to a stack information file can be added or edited by using the editing facility making it possible to find the total amount of stacks used in the entire system The edited information for the amount of stacks used can be saved or loaded as a call information file cal Before CallWalker amp gensni can be used the compile driver s startup option finfo must be specified during compilation so that inspector information will be added to the absolute module file x30 1 2 7 MapViewer MapViewer is the execution file for the map viewer By processing the absolute module file x30 MapViewer graphically shows a post link memory mapping To use MapViewer specify the compile driver startup option finfo when compiling so that the absolu
150. cified when linking become valid and functions with the same names in the library file are excluded Figure E 7 shows method 1 Figure E 8 shows method 2 nc1i00 c g osample ncert0 a30 device r30 init r30 sample c lt RET gt This example shows the command line when device c and init c are modified Figure E 7 Method of Directly Linking Modified Source Programs make lt RET gt Figure E 8 Method of Updating Library Using Modified Source Programs REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 300 R32C 100 Series C Compiler Package V 1 02 F Error Messages Appendix F Error Messages This appendix describes the error messages and warning messages output by NC100 and their countermeasures F 1 Message Format If during processing NC100 detects an error it displays an error message on the screen and stops the compiling process Figure F 1 to Figure F3 shows the format of error messages and warning messages nc100 error message Figure F 1 Format of Error Messages from the nc100 Compile Driver Error cpp100 error No filename line No error message Error ccom filename line No error message Fatal ccom filename line No error message 4 Figure F 2 Format of Command Error Messages Warning cpp1 00 warning No filename line No warning message Warning ccom filename line No warning message Figure F 3 Format of Command Warning Messages
151. cimal o o Converts the integer in the parameter to an unsigned octal e x Converts the integer in the parameter to an unsigned hexadecimal Lowercase abcdef are equivalent to OAH to OFH e X Converts the integer in the parameter to an unsigned hexadecimal Uppercase ABCDEF are equivalent to OAH to OFH o c Outputs the parameter as an ASCII character o s Converts the parameter after the string far pointer char and up to a null character 0 or the precision to a character string Note that wchar_t type character strings cannot be processed e p Outputs the parameter pointer all types in the format 24 bits address J n Stores the number of characters output in the integer pointer of the parameter The parameter is not converted 1 In the standard library included with your product the character string pointer is a far pointer All printf functions handle s with a far pointer Note that scanf functions use a near pointer by default REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 260 R32C 100 Series C Compiler Package V 1 02 E Standard Library printf Input Output Functions Description e Converts a double type parameter to the exponent format The format is d dddddde dd e E Same as e except that E is used in place of e for the exponent f Converts double parameters to d dddddd format g Converts double parameters to the format specified in e or f Normally f conversion but c
152. ck frame that can be secured per function in bytes 64K max Number of variables that can be defined in pragma ADDRESS Depends on amount of available memory Maximum number of levels of nesting parentheses Depends on amount of available memory Number of initial values that can be defined when defining variables with initialization expressions Depends on amount of available memory Maximum number of levels of nesting modifier declarators Depends on stack size of YACC Maximum number of levels of nesting declarator parentheses Depends on stack size of YACC Maximum number of levels of nesting operator parentheses Maximum number of valid characters per internal identifier or macro name Depends on stack size of YACC Depends en smountefevatablememery 200 max Maximum number of valid characters per external identifier Dependsorrameuntofavailablememery 200 max Maximum number of external identifiers per source file Depends on amount of available memory Maximum number of identifiers with block scope per block Depends on amount of available memory REJ10J2009 0200 Rev 2 00 2010 04 01 164 2tENESAS R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications Table C 4 NC100 Specifications 2 2 Item Specification Maximum number of macros per source file Depends on amount of available memory Maximum number of parameters per function call and per
153. cks whether the character is an upper case letter A Z Format include lt ctype h gt int isupper c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if an upper case letter Returns 0 if not an upper case letter Description Determines the type of character in the parameter isxdigit Character Handling Functions Function Checks whether the character is a hexadecimal character 0 9 A Fa f Format include lt ctype h gt int isxdigit c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if a hexadecimal character Returns 0 if not a hexadecimal character Description Determines the type of character in the parameter REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 247 R32C 100 Series C Compiler Package V 1 02 E Standard Library labs Function Calculates the absolute value of a long type integer Format include lt stdlib h gt long labs n Method function Argument long n Long integer ReturnValue Returns the absolute value of a long type integer distance from 0 Idexp Function Calculates the power of a floating point number Format include lt math h gt double ldexp x exp Method function Argument double x Float point number int exp Power of number ReturnValue Returns x exp power of 2 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 248 R32C 100 Seri
154. cro names reserved by cpp100 FILE_ LINE DATE _ _ TIME Use a different macro name 82 assertion warning The result of an assert operation expression was 0 Check the operation expression eu ey 83 garbage argument Characters other than a comment exist after a preprocess command Specify characters as a comment string after the preprocess command y 84 escape sequence out of range for e An escape sequence in a character constant or character character string constant exceeded 255 characters Reduce the escape sequence to within 255 characters 85 redefined e A previously defined macro was redefined with different contents Check the contents against those in the previous definition 87 within comment e Acomment includes gt Do not nest comments 88 Environment variable NCKIN e Environment variable NCKIN is not valid must be SJIS or EUC gt Set SJIS or EUC to NCKIN 90 Macro name in if is not e An undefined macro name in if is used defined so it s tereated as 0 Check the macro definition REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 307 R32C 100 Series C Compiler Package V 1 02 F Error Messages F 5 ccom100 Error Messages Table E7 to Table F19 list the ccom100 compiler error messages and their countermeasures Table F 7 ccom100 Error Messages 1 13 Error message Description and countermeasure pragma PRAGMA na
155. ctly Empty declare Only storage class and type specifiers are found Write a declarator float and double not have sign y ely s Specifiers signed unsigned are described in float or double Write the type specifier correctly floating point value overflow ey The floating point immediate value exceeds the representable range Make sure the value is within the range floating type s bitfield A bit field of an invalid type is declared Use the integer type to declare a bitfield for struct union statement ee The struct or union type is used in the second expression of the for statement Use the scalar type to describe the second expression of the for statement for void statement y e The 2nd expression of the for statement has void Use the scalar type as the 2nd expression of the for statement function initialized An initialize expression is described for function declaration Delete the initialize expression function member declared Amember of struct or union is function type Write the members correctly function returning a function declared eu ey y The type of the return value in function declaration is function type Change the type to pointer to function etc REJ10J2009 0200 Rev 2 00 2010 04 01 310 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 10 ccom100Error message 4 13 Error messag
156. d line options that display the cross tool version data Table A 3 Options for Displaying Version Data Option Function v Displays the name of the command program and the command line during execution V Displays the startup messages of the compiler programs then finishes processing without compiling I Display command program name Function Compiles the files while displaying the name of the command program that is being executed Notes Use lowercase v for this option V DEJEVAERIOI EIEI Function Displays version data for the command programs executed by the compiler then finishes processing Supplement Use this option to check that the compiler has been installed correctly The R32C 100 Series C Compiler package Release Notes list the correct version numbers of the commands executed internally by the compiler If the version numbers in the Release Notes do not match those displayed using this option the package may not have been installed correctly See the R32C 100 Series C Compiler package Release Notes for details of how to install the NC100 package Notes 1 Use uppercase V for this option 2 Ifyou specify this option all other options are ignored REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 68 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 4 Options for Debugging Table A 4 lists the debugging startup options that output C language level debug inform
157. d macro Undefines the specified predefined macro I Compile driver control Function Creates a relocatable object file extension r30 and finishes processing Notes Tf this option is specified no absolute module file extension x30 or other file output by In100 is created Didentifier Compile driver control Function The function is the same as the preprocess command define Delimit multiple identifiers with spaces Syntax nc100A Didentifier constant A lt C source file gt constand is optional Notes The number of identifiers that can be defined may be limited by the maximum number of characters that can be specified on the command line of the operating system of the host machine 1 Tf you do not specify command line options c E P or S nc100 finishes at In100 and output files up to the absolute load module file extension x30 are created REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 64 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference dsource dS Comment option Function Generates an assembly language source file extension a30 with a C language source list output as a comment Not deleted even after assembling Supplement 1 When the S option is used the option dsouce dS is automatically enabled The generated files a30 and r30 are not deleted 2 Use this option when you want to output C language source lists to the assembly list file ds
158. diately after a reset It performs the following process mainly Sets the top address _SB_ of the SBDATA area it is accessing area to used the SB relative addressing mode Sets the processor s operating mode Initializes the stack pointer ISP Register and USP Register Initializes SB register Initializes INTB register Initializes the data near area a Default bss_NEAR sections are cleared to 0 2 When far qualified variables are used bss_FAR sections are cleared to 0 8 pragma SBDATA Extended Functions bss _SB8 sections are cleared to 0 4 pragma SB16DATA Extended Functions bss_SB16 sections are cleared to 0 5 pragma EXTMEM Extended Functions bss_EXT sections are cleared to 0 6 pragma MONITORn Extended Functions bss_MONI1 bss MON2 bss MON3 bss_MON4 sections are cleared to 0 e Transfers initial values from the ROM section in which they are stored to a data area that has initial values 1 Default Transfers initial values from the data_NEAR_INIT section to the data NEAR section 2 When far qualified variables are used Transfers initial values from the data_FAR_INIT section to the data_FAR section 8 pragma SBDATA Extended Functions Transfers initial values from the data_SB8_INIT section to the data_SB8 section 4 pragma SB16DATA Extended Functions Transfers initial values from the data_SB16_INIT section to the data_SB16 section 5 pragma EXTMEM Extended Functions Transfers initial values
159. dification Options Option Short form Function fansi None Makes fnot_reserve_far_and_near fnot_reserve_asm and fextend_to_int valid fconst_not_ROM CNR Does not handle the types specified by const as ROM data fdouble_32 fD32 This option specifies that the double type be handled in 32 bit data length as is the float type fenable_register fER Make register storage class available fextend_to_int ETI Performs operation after extending char type or short type data to the int type data Extended according to ANSI standards ffar_RAM FFRAM Changes the default attribute of RAM data to far finfo None Outputs the information required for the Inspector Call Walker and Map Viewer to the absolute module file x30 fint_16 fI16 Does handle int type at the 16 bit width JSRW None Changes the default instruction for calling functions to JSR W fnear ROM NROM Changes the default attribute of ROM data to near fno_align NA Does not align the start address of the function fno_switch_table NST When this option is specified the code which branches since it compares is generated to a switch statement fnot_address_volatile ANAV Does not regard the variables specified by pragma ADDRESS pragma EQU as those specified by volatile fnot_reserve_asm FNRA Exclude asm from reserved words Only _asm is valid fnot_reserv
160. e Description and countermeasure function returning an array The type of the return value in function declaration is an array type Change the type to pointer to function ete handler function called The function specified by pragma HANDLER is called Be careful not to call a handler identifier variable name is duplicated The variable is defined more than one time Specify variable definition correctly if struct union statement ey eU ey The struct or union type is used in the expression of the if statement The expression must have scalar type if void statement The void type is used in the expression of the if statement The expression must have scalar type illegal storage class for argument inline ignored eu ey An inline function is declared in declaration statement within a function Declare it outside a function illegal storage class for argument nterrupt ignored ey An interrupt function is declared in declaration statement within a function Declare it outside a function incomplete array access An attempt is made to reference an array of incomplete Define size of array incomplete return type eu ey An attempt is made to reference an return variable of incomplete type Check return variable incomplete struct get by An attempt is made to reference or initialize an array of incomplete structs or unions that do not ha
161. e c 1 5 2 Preprocessed C Source Files The cpp100 processes preprocess commands starting with Such operations include header file contents macro expansion and judgments on conditional compiling The C source files output by the preprocessor include the results of cpp100 processing of the C source files Therefore do not contain preprocess lines other than pragma and line You can refer to these files to check the contents of programs processed by the compiler The file extension is 1 Figure 1 8 and Figure 1 9 are examples of file output l typedef struct iobuf 1 l char _buff l int _cnt l int _flag l i int _mod i int _furnc_in void int _func_out int l FILE omitted i l typedef long fpos_t I typedef unsigned long size_t l extern FILE_iobj Figure 1 8 Example Preprocessed C Source File 1 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 8 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 i typedef char _far va_list 1 extern int getc FILE _far extern int getchar void extern int putc int FILE _ far extern int putchar int extern int feof FILE _ far externint ferror FILE _ far extern int fgetc FILE _ far extern char _far fgets char _far int FILE _ far extern int fputc int FILE _ far extern int fputs const char _far FILE _ far omitted extern int printf con
162. e declaration for function with far pointer to parameter specifies near address parameter at the function call Figure B 10 Type conversion from near to far When converted to type far the pointer is sign extended with the most significant bit of the near address 16 bit quantity B 1 7 Declaration of function In NC100 functions are always located in the far area Therefore do not write a near declaration for functions If a function is declared to take on a near attribute NC100 outputs a warning and continues processing by assuming the attribute of that function is far Figure B 11 shows a display example where a function is declared to be near nc100 S smp c R32C 100 Series C Compiler V X XX Release XX Copyright C XXXX XXXX XXXX Renesas Electronics Corp and Renesas Solutions Corp All rights reserved smp c Warning ccom smp c line 3 function must be far gt func Figure B 11 Example Declaration of Function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 104 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 1 8 Function for Specifying near and far in Multiple Declarations As shown in Figure B 12 if there are multiple declarations of the same variable the type information for the variable is interpreted as indicating a combined type extern int far idata int idata int idata 10 void func void remainder omitted This Declarati
163. e following lists memory handling functions Table E 6 Memory Handling Functions Type Function Contents Reentrant Initialize bzero Initializes a memory area by clearing it to zero O Copy bcopy Copies characters from a memory area to another O memcpy Copies characters n bytes from a memory area to another O memset Set a memory area by filling with characters O Compare memcmp Compares memory areas n bytes O memicmp Compares memory areas with alphabets handled as O upper case letters Move memmove Moves the area of a character string O Search memchr Searches a character from a memory area O 6 Execution Control Functions The following lists execution control functions Table E 7 Execution Control Functions Function Contents Reentrant abort Terminates the execution of the program O longjmp Performs a global jump O setjmp Sets a stack environment for a global jump O REJ10J2009 0200 Rev 2 00 2010 04 01 220 2tENESAS R32C 100 Series C Compiler Package V 1 02 E Standard Library 7 Mathematical Functions The following lists mathematical functions Table E 8 Mathematical Functions Function Contents Reentrant acos Calculates arc cosine O asin Calculates arc sine O atan Calculates arc tangent O atan2 Calculates arc tangent O ceil Calculates an integer
164. e of gensni Function Shows the startup message of gensni and terminates processing without performing anything No stack information files are generated REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 333 R32C 100 Series C Compiler Package V 1 02 G Using gensni or the Stack information File Creation Tool for Call Walker G 4 Error Messages of gensni G 4 1 Error Messages Table G 2 lists the error messages output by gensni along with the contents of errors and the corrective actions to be taken Table G 2 List of Error Messages of gensni Error Messages Content of Error and Corrective Action usage gensni V o out_file in_file The input format is incorrect Specify a correct input format Can t open file XXX The absolute module file cannot be opened Check whether the file exists and the file attribute Can t create file XXX Ueu eju The stack information file cannot be created Check the file and folder attributes Check the free disk space available Ilegal file format XXX The content of the absolute module file is incorrect No stack information file can be created Check whether the absolute module file is the one that you created with NC100 Also check whether finfo and g are specified in the compiler options Not enough memory Memory could not be allocated for gensni Check the available memory size of your PC REJ10J2009 0200 Rev 2 00 2010 04 01 334 7tENE
165. e_far_and_near fNRFAN_ Exclude far and near from reserved words Only _far and _near are valid fnot_reserve_inline FNRI Exclude far and near from reserved words Only _inline is made a reserved word fsigned_char fSC Handles type char without sign specification as type signed char fswitch_other_section fSOS This option outputs a ROM table for a switch statement to some other section than a program section fuse _FPU FUF Outputs FPU instruction 1 unsigned char type signed char type short type and unsigned short type data evaluated under ANSI rules is always extended to the int type data This is because operations on char types c1 c2 2 c3 for example would otherwise result in an overflow and failure to obtain the intended result REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 18 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler g Library Specifying Option Tabel 2 7 lists the startup options you can use to specify a library file Tabel 2 7 Library Specifying Option Option Function Libraryfilename Specifies a library file that is used by In100 when linking files h Warning Options Tabel 2 8 shows the command line options for outputting warning messages for contraventions of nc100 language specifications Tabel 2 8 Warning Options Option Short form Function Wall None Displays mes
166. ecautions on Transporting between NC Series Compilers NC100 basically is compatible with Renesas C compilers NCxx at the language specification level including extended functions However there are some differences between the compiler this manual and other NC series compilers as described below a Difference in default near far The default near far in the NC series are shown in Table 3 3 Therefore when transporting the compiler this manual to other NC series compilers the near far specification needs to be adjusted Table 3 3 Default near far in the NC Series Compiler RAM data ROM data Program NC100 near far far Fixed However pointer type is far Fixed NC308 near far far Fixed However pointer type is far NC30 near far far Fixed REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 62 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Appendix A Command Option Reference This appendix describes how to start the compile driver nc100 and the command line options The description of the command line options includes those for the as100 assembler and 1n100 linkage editor which can be started from nc100 A 1 nc100 Command Format nc100A commana line option A lt assembly language source file name A relocatable object file name A C source file name gt Prompt lt gt Mandatory item Optional item A Space Figure A 1 nc100 Command Line
167. ecified for anything other than a variable it is ignored as invalid 8 If the specified variable is a static variable declared in a function the pragma SBDATA declaration is ignored as invalid 4 The variable declared to be pragma SBDATA is placed in a SBDATA attribute section when allocating memory for it 5 As opposed to the same variable pragma SBDATA pragma SB16DATA cannot be specified simultaneously 6 If pragma SBDATA is declared for ROM data the data is not placed in a SBDATA attribute section Example pragma SBDATA sym data struct sym_data char bitO 1 char bit1 1 char bit2 1 char bit3 1 char bit4 1 char bit5 1 char bit6 1 char bit7 1 sym_data void func void sym_data bit1 0 omitted Figure B 65 Example Use of pragma SBDATA Declaration Supplement NC100 is premised on an assumption that the SB register will be initialized after reset and will thereafter be used as a fixed quantity REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 139 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma SECTION Function Syntax Description Changes the names of sections generated by NC100 pragma SECTIONAsection nameAnew section name Specifying the program section data section and rom section in a pragma SECTION declaration changes the section names of all subsequent functions Specifying a bss section in a pragma SECTION declaration change
168. eck each expression Also note that the expressions on the left and right sides of must be of the same type invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid amp amp operands y e The amp amp operation contains an error Check the expressions on the left and right sides of the operator invalid amp operands The amp operation contains an error Check the expression on the right side of the operator invalid amp operands Uey The amp operation contains an error Check the expressions on the left and right sides of the operator invalid 0 operand y e The expression on the left side of is not a function Write a function or a pointer to the function in the left side expression of invalid operands Tf multiplication the operation contains an error If is the pointer operator the right side expressionis not pointer type For a multiplication check the expressions on the left and right sides of the operator For a pointer check the type of the right side expression invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid operands The operation contains an error Check the expressions on the left and right sides of the operator REJ10J2009 0200 Rev 2 00 2010 04 01 3
169. edence and only the assembly language source files will be generated J c Stop after creating relocatable files J S Stop after creating assembly language source files REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 97 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference Appendix B Extended Functions Reference To facilitate its use in systems using the R32C 100 series NC100 has a number of additional extended functions This appendix B describes how to use these extended functions excluding those related to language specifications which are only described in outline Table B 1 Extended Functions 1 2 Extended feature Description near far qualifiers Specifies the addressing mode to access data 007FFFFFH and OFF800000H OFFFFFFFFH All functions take on far attributes asm function 1 Assembly language can be directly included in C programs It can also be included outside functions Example asm MOV W 0 RO 2 You can specify variable names within functions only Example 1 asm MOV W RO SS FB f Example 2 asm MOV W RO SS s Example 3 asm MOV W RO S f 8 You can include dummy asm functions as a means of partially suppressing optimization within functions only Example asm Japanese characters 4 Permits you to use Japanese characters in character strings Example LRF 5 Permits you to use Japanese character
170. els cass cdd ected ccd cutavn dossccduersdobot si ated ccesvadedodud c dbset aisead aia eii a iais 70 A 2 6 Generated Code Modification Options cccccscsssssssscsssesssescsescsescsssescscscssssssssseesssesssesesesesescsesesescsssnsceneceseees 82 A 2 7 library Specifying Option sis ciiscasiiacsiicaniianws ansignmiaainrdanidanindanuisaumaiaanidanmae 88 A 2 8 Wearriinta Options Manae e a Ne e a a E aA edsssessschee 89 A 2 9 Assemble and Link Options isss csscesccseescsthscdaseducesdedatsssascuesdbacsctesdedasteticedacsceaseuetscusdaeeedadedasesdedeecadesescasdasstease s 96 A 3 Notes on Command Line Options cccccessssssssscssscssssssesssesssescscscsesescseesssssesssesesesessscssseseseacscscacsescesssesesesesescseaeas 97 A 3 1 Coding Command Line Options cccscsssssssssssssscssssssssssssssessssssssesscseseesssesssucscseseesesescscscsesessesesseseasseeseaeess 97 A 3 2 Priority of Options for Controlling cccceccscsssescsessscsescssscssscsssscscscsessscscsesescscscsescscsssesecssecessseecaeessesceeeeeneees 97 Appendix B Extended Functions Reference c ccssssssssssssesessssssescsesesesnssesesessssesessesesesessesesesussesesesussessssesesesseseseeeeseseseeneaeaes 98 Ba Neararnd far Modiiters t c3 si00 ci ditieiatseae leiden dheadgtiiie sien aa bal iendaawinas 100 B 1 1 Overview of near and far Modifiers scassaccaisiinaaidainauisacuiiaaiaaiadaimiamauiaas 100 B 1 2 Format of Variable Declaration ccccssssssssssssssssses
171. ements are written in assembly language pragma ENDASM Syntax pragmaAASM pragma AENDASM Example pragma ASM mov w RO R1 add w R1 02H pragma ENDASM pragma JSRA Calls functions using JSR A as the JSR instruction Syntax pragma JSRAAfunction name Example pragma JSRA func pragma JSRW Calls functions using JSR W as the JSR instruction Syntax pragma JSRWAfunction name Example pragma JSRW func pragma PAGE Indicates a new page point in the assembler listing file Syntax pragmaAPAGE Example pragma PAGE REJ10J2009 0200 Rev 2 00 2010 04 01 2tENESAS 133 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 7 2 Using Memory Mapping Extended Functions NC100 includes the following memory mapping extended functions pragma EXTMEM Declares exclusion of address 0 relative addressing Function Declares a variable or function to be located in an area that cannot be accessed in address 0 relative addressing Syntax pragma EXTMEMAvariable name pragma EXTMEMA function name Description In address 0 relative addressing that is normally used to access an external variable directly or call a function directly no addresses in the range O0O800000H OF F7FFFFFH can be accessed pragma EXTMEM declares that a variable or function be located in such an inaccessible area The variables and functions declared with pragma EXTMEM are accessed in address register relative addressing etc Note th
172. enced and its contents changed between interrupt handling and non interrupt handling routines or between tasks under realtime OS always be sure to use exclusive control such as volatile specification Also use exclusive control for bit field structures which have different member names but are mapped into the same RAM 3 1 2 Notes about the R32C s Type Dependent Part When writing to or reading a register in the SFR area it may sometimes be necessary to use a specific instruction Because this specific instruction varies with each type of MCU consult the user s manual of your MCU for details In this case write the instruction directly in the program using the ASM function Tn this compiler the instructions which cannot be used may be generated for writing and read out to the register of SFR area When accessing registers in the SFR area in C language make sure that the same correct instructions are generated as done by using asm functions regardless of the compiler s version and of whether optimizing options are used or not When you describe like the following examples as C language description to a SFR area in this compiler may generate the assembler code which carries out operation which is not assumed since the interrupt request bit is not normal REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 50 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique pragma ADDRESS TAOIC 006Ch R32C 100 MCU s Timer AO inte
173. ended Functions Reference B 2 3 Specifying Register Name of register Variable The storage class auto and register variables including arguments may be mapped to registers by the compiler The variables mapped to registers can be used in the asm function by writing the program as shown in Figure B 23 below asm op code Variable name Figure B 23 Description Format for Register Variables You can only specify two variable name using this format Figure B 24 shows examples of referencing register variables and the results of compiling C Source file void func void register short i 1 asm mov w R1 i Assembly language source file compile result HHH FUNCTION func HH H ARG Size 4 Auto Size 0 Context Size 4 SECTION program CODE ALIGN _file reg c align _line 2 H C_SRC glb _func _func _line 3 H C_ SRC register short i 1 mov w 0001H RO i _line 4 AHHC SRC asm mov w R1 i HHH ASM START mov w RO R1 RO register is transferred to RO register HHHH ASM END Figure B 24 Example for Referencing a Register Variable In NC100 register variables used within functions are managed dynamically At anyone position the register used for a register variable is not necessarily always the same one Therefore if a register is specified directly in an asm function it may after compiling operate differently We therefore strongly suggest using this function to check the reg
174. ently NC100 allows you to accomplish an efficient function call by declaring the prototype of a function This means that unless a function is declared of its prototype in NC100 arguments of that function are placed on the stack following the rules listed in Table 3 1 when calling the function Table 3 1 Rules for Using Stack for Parameters Data type s Rules for pushing onto stack char type Expanded into the int type when stacked short type float type Expanded into the double type when stacked otherwise type Not expanded when stacked For this reason NC100 may require redundant type expansion unless you declare the prototype of a function Prototype declaration of functions helps to suppress such redundant type expansion and also makes it possible to assign arguments to registers All this allows you to accomplish an efficient function call b Using SB Register Efficiently Using the SB register based addressing mode you can reduce the size of your application program ROM size NC100 allows you to declare variables that use the SB register based addressing mode by writing the description shown in Figure3 8 pragma SBDATA val int val Figure3 8 Example of variable declaration using SB based addressing mode c Compressing ROM Size Using Compile Option fUSRW When calling a function defined outside the file in NC100 the function is called with the JSR A instruction However if the program is
175. er Table D 7 Rules on Argument Transfer via Register NC30 Argument Type of argument Registers used First argument _Bool RiL register char int R1 register near pointer Second argument int R2 register near pointer REJ10J2009 0200 Rev 2 00 2010 04 01 193 7tENESAS R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules 2 Passing arguments via stack All arguments that do not satisfy the register transfer requirements are passed via stack The Table D 8 Table D 9 and Table D 10 summarize the methods used to pass arguments Table D 8 Rules on Passing Arguments to Function NC100 Type of argument First argument Second Third Fourth fifth and following argument argument argument arguments _Bool type char ROL register RiL register Stack Stack Stack int 16 bits RO register R1 register R4 register R5 register Stack short int 82 bits R2RO register R3R1 register R6R4 register R7R5 register Stack float long pointer AO register A2 register R6R4 register R7R5 register Stack long long A1A0 register A3A2 register Stack Stack Stack double long double Table D 9 Rules on Passing Arguments to Function NC308 l Third and following Type of argument First argument Second argument arguments _Bool ROL register Stack Stack char int RO register Stack Stack near pointer Table D 10 Rules on Passing Arg
176. er Package V 1 02 E Standard Library toupper Character Handling Functions Function Converts the character from a lower case to an upper case Format include lt ctype h gt int toupper c Method macro Argument int c Character to be converted ReturnValue Returns the upper case letter if the argument is a lower case letter Otherwise returns the passed argument as is Description Converts the character from a lower case to an upper case REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 285 R32C 100 Series C Compiler Package V 1 02 E Standard Library ungetc Input Output Functions Function Returns one character to the stream Format include lt stdio h gt int ungetc c stream Method macro Argument int c Character to be returned FILE _far stream Pointer of stream ReturnValue Returns the returned one character if done normally Returns EOF ifthe stream is in write mode an error or EOF is encountered or the character to be sent back is EOF Description Returns one character to the stream Interprets code 0x1A as the end code and ignores any subsequent data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 286 R32C 100 Series C Compiler Package V 1 02 E Standard Library Viprintf Input Output Functions Function Output to a stream with format Format include lt stdarg h gt include lt stdio h gt int vfprintf stream format ap Method function Argument FILE
177. eries C Compiler Package V 1 02 E Standard Library memset Memory Handling Functions Function Set a memory area Format include lt string h gt void _far memset s c n Method macro or function Argument void _far s Pointer to the memory area to be set at int c Data to be set size_t n Number of bytes to be set ReturnValue Returns the pointer to the memory area which has been set Description Usually the program code described by macro is used for this function In using the function in a library please describe it as undef memset after description of include lt string h gt Sets n bytes of data c in memory s When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization modf Function Calculates the division of a real number into the mantissa and exponent parts Format include lt math h gt double modf val pd Method function Argument double val arbitrary real number double _far pd Pointer to an area for storing an integer ReturnValue Returns the decimal part of a real number REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 258 R32C 100 Series C Compiler Package V 1 02 E Standard Library perror Function Outputs an error message to stderr Format include lt stdio h gt void perror s Method function Argument const char _far s Pointer to a character string attached before a message ReturnVal
178. error Corrupted processing system executable file Reinstall the processing system Ignore option You specified an illegal option for NC100 Specify the correct option illegal option y e You specified options greater than 100 characters for as100 or ln100 Reduce the options to 99 characters or less Invalid argument ey This error message is not normally displayed It is aninternal error Contact Renesas Solutions Corp Invalid option eu eyy The required parameter was not specified in option 2 Specify the required parameter after You specified a space between the option and its parameter Delete the space between the option and its parameter Invalid option o Y ey No output filename was specified after the o option Specify the name of the output file Do not specify the filename extension Invalid suffix xxx You specified a filename extension not recognized by NC100 other than c i a30 r30 x30 Specify the filename with the correct extension REJ10J2009 0200 Rev 2 00 2010 04 01 302 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 2 nc100 Error Messages 2 2 Error message Description and countermeasure No such file or directory e The processing system will not run gt Check that the directory of the processing system is correctly set in the environment variable
179. error flag IOERR _JOREAD IOWRT _IORW These flags specify the stream operating mode They are set during stream initialization _JOEOF IOERR These flags are set according to whether an EOF is encountered or error occurs in the T O function 4 imt_mod Stores the flags indicating the text mode TEXT and binary mode _BIN e Text mode Echo back of I O data and conversion of characters See the source programs fgetc c and fputc c of the fgetc and fputc functions for details of echo back and character conversion J Binary mode No conversion of I O data These flags are set in the initialization block of the stream 5 int func in 0 When the stream is in read only mode IOREAD or read write mode IORW stores the level 3 input function pointer Stores a NULL pointer in other cases This information is used for indirect calling of level 3 input functions by level 2 input functions int func out When the stream is in write mode IOWRT stores the level 3 output function pointer If the stream can be input JOREAD or _IORW and is in text mode it stores the level 3 output function pointer for echo back Stores a NULL pointer in other cases This information is used for indirect calling of level 3 output functions by level 2 output functions REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 297 R32C 100 Series C Compiler Package V 1 02 E Standard Library Set values for all elements other than char_buff in the stream initia
180. es are located in the near area by default Use this option when you want the RAM data to be located in other areas than the near area 64 Kbytes area finfo Modify generated code Function Outputs the information required for the Call Walker and Map Viewer Supplement When using Call Walker and Map Viewer the absolute module file x30 output by this option is needed fint 16 f116 Modify generated code Function Does handle int type at the 16 bit width Supplement When using this option you need to link nc100i16 lib instead of nc100lib lib as the standard library If you executed a range of operations from compile to link after specifying this option from the compiler driver the libraries to be linked are automatically changed The default size of type int when this option is not specified is 32 bits Note that if this option is used in combination with the compile option fuse_FPUCfIUF you need to link nc100i16fpu lib instead of nc100lib lib USRW Modify generated code Function Changes the default instruction for calling functions to JSR W Supplement When calling a function that has been defined external to the source file the JSR A command is used by default This option allows it to be changed to the JSR W command Change to the JSR W command helps to compress the generated code size This option is useful when the program is relatively small not exceeding 32 Kbytes in size or ROM compression
181. es C Compiler Package V 1 02 E Standard Library Idiv Function Divides a long type integer and calculates the remainder Format include lt stdlib h gt ldiv_t ldiv number denom Method function Argument long number Dividend long denom Divisor ReturnValue Returns the quotient derived by dividing number by denom and the remainder of the division Description Returns the quotient derived by dividing number by denom and the remainder of the division in the structure ldiv_t Idiv_t is defined in stdlib h This structure consists of members long quot and longrem localeconv Function Initializes struct Iconv Format include lt locale h gt struct lconv _far localeconv void Method function Argument No argument used ReturnValue Returns a pointer to the initialized struct Iconv REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 249 R32C 100 Series C Compiler Package V 1 02 E Standard Library log Mathematical Functions Function Calculates natural logarithm Format include lt math h gt double log x Method function Argument double x arbitrary real number ReturnValue Returns the natural logarithm of given real number x Description This is the reverse function of exp log10 Mathematical Functions Function Calculates common logarithm Format include lt math h gt double 1lo og10 x Method function Argument double x arbitrary real number ReturnValue Ret
182. escription Determines the type of character in the parameter REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 243 R32C 100 Series C Compiler Package V 1 02 E Standard Library isdigit Character Handling Functions Function Checks whether the character is a numeral 0 9 Format include lt ctype h gt int isdigit c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if a numeral Returns 0 ifnot a numeral Description Determines the type of character in the parameter isgraoh Character Handling Functions Function Checks whether the character is printable except a blank Ox21 Ox7e Format include lt ctype h gt int isgraph c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if printable Returns 0 if not printable Description Determines the type of character in the parameter REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 244 R32C 100 Series C Compiler Package V 1 02 E Standard Library islower Function Checks whether the character is a lower case letter a z Format include lt ctype h gt int islower c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if a lower case letter Returns 0 if not a lower case letter Description Determines the type of character in the parameter isprint Character Handling Func
183. eturns 0 if the value of given real number x is outside therange of 1 0 to 1 0 Otherwise returns a value in the range from 0 to n radian asin Mathematical Functions Function Calculates arc sine Format include lt math h gt double asin x Method Function Variable double x arbitrary real number ReturnValue Assumes an error and returns 0 if the value of given real number x is outside the range of 1 0 to 1 0 Otherwise returns a value in the range from n 2 to m 2 radian atan Function Calculates arc tangent Format include lt math h gt double atan x Method function Variable double x arbitrary real number ReturnValue Returns a value in the range from n 2 to n 2 radian REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 224 R32C 100 Series C Compiler Package V 1 02 E Standard Library atan2 Mathematical Functions Function Calculates arc tangent Format include lt math h gt double atan2 x y Method function Variable double x arbitrary real number double y arbitrary real number ReturnValue Returns a value in the range from n to n radian Function Converts a character string into a double type floating point number Format include lt stdlib h gt double atof s Method function Variable const char _far s Pointer to the converted character string ReturnValue Returns the value derived by converting a character string into a double precision floating po
184. f the micro processor For a C language source file like the one shown in Figure D 19 if the areas where variables of storage class auto are valid do not overlap each other the system allocates only one area which is then shared between multiple variables void func void int i j K for i 0 i lt 0 i process scope of i abbreviated for j OXFF j lt 0 j process scope of j abbreviated for K 0 k lt 0 k f process scope of k Figure D 19 Example of C Program In this example the effective ranges of three auto variables i j and k do not overlap so that a two byte area offset 1 from FB is shared Figure D 20 shows an assembly language source file generated by compiling the program in Figure D 19 4 FUNCTION func HR FRAME AUTO k size 4 offset 4 1 HAR FRAME AUTO j size 4 offset 4 2 HHH FRAME AUTO i size 4 offset 4 3 section program _file auto c _line 3 glib _func _ func enter 04H remainder omitted As shown by 1 2 and 3 the three auto variables share the FB offset 2 area Figure D 20 Example of Assembly Language Source Program REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 204 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules D 5 Rules of Escaping of the Register The rules of Escaping of the register when call C function as follows a The rules of Escaping of the re
185. file Check whether the file actually exists can t open C source filename for error message The source file in error cannot be opened Check whether the file exists argument type given both places eU ey y Argument declaration in function definition overlaps an argument list separately given Choose the argument list or argument declaration for this argument declaration array of functions declared The array type in array declaration is defined as function Specify scalar type struct union for the array type array size is not constant integer eij The number of elements in array declaration is not a constant Use a constant to describe the number of elements asm0 s string must have only 1 b b is described more than twice in asm statement Make sure that b is described only once asm s string must not have more than 3 or or is described more than thrice in asm statement Make sure that is described only twice auto variable s size is zero eu ey ey An array with 0 elements or no elements was declared in the auto area Correct the coding bitfield width exceeded The bit field width exceeds the bit width of the data type Make sure that the data type bit width declared in the bit field is not exceeded bitfield width is not constant integer The bit width of the bit field is not a constant Use a constant to write the bit width can
186. float data Fixed point location s fixed point part sign 1bit e characteristic part 8bits m fixed point part 23bits Figure D 1 Single precision floating point data format REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 187 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules 2 Double precision floating point data format Figure D 2 shows the format for binary floating point double and long double data 63 52 48 40 32 24 16 8 0 E ee eee Fixed point location s fixed point part sign bit e characteristic part 11bits m fixed point part 52bits Figure D 2 Double precision floating point data format D 1 3 Enumerator Type Enumerator types have the same internal representation as unsigned int types Unless otherwise specified integers 0 1 2 are applied in the order in which the members appear D 1 4 Pointer Type Table D 3 shows the number of bytes used by pointer type data Table D 3 Data Size of Pointer Types Type Existence of sign Bit Size Range pointers No 32 between 0 and OxFFFFFFFF All pointers are handled as the far pointer Therefore the compiler outputs a warning Near pointer not supported near qualifier ignored to the effect that the pointer variables declared as a near pointer will be handled as a far pointer Note however that if the compile option WINP Wignore_near_pointer is specified the compiler inhibits
187. floating point numbers Ono_logical_or_combine ONLOC Suppresses the optimization that puts consecutive OR together Ono_asmopt ONA Inhibits starting the assembler optimizer aopt100 Osp_adjust OSA Optimizes removal of stack correction code This allows the necessary ROM capacity to be reduced However this may result in an increased amount of stack being used Ostatic_to_inline OSTI A static function is treated as an inline function The effects of main optimization options are shown in Table A 6 Table A 6 Effect of each Optimization Options Option O OR OS OSA SPEED faster lower faster faster ROM size decrease decrease increase decrease usage of stack decrease same same increase REJ10J2009 0200 Rev 2 00 2010 04 01 7tENESAS 70 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference 1 a Optimization Function Optimizes speed and ROM size to the maximum This option can be specified with g options O3 is assumed if you specify no numeric no level O1 Makes O3 Ono_bit Ono_break_source_debug and Ono_float_const_fold valid O2 Makes no diffrence with O1 O3 Optimizes speed and ROM size to the maximum O4 O3 and Oconst valid O5 Effect the best possible optimization in common sub expressions if the option OR is concurrently specified effects the best possible optimization in transfer and comparis
188. function Argument void _far cp Pointer to the memory area to be freed ReturnValue No value is returned Description Frees memory areas previously allocated with malloc or calloc Noprocessing is performed if you specify NULL in the parameter REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 237 R32C 100 Series C Compiler Package V 1 02 E Standard Library frexo Mathematical Functions Function Divides floating point number into mantissa and exponent parts Format include lt math h gt double frexp x prexp Method function Argument double x float point number int _far prexp Pointer to an area for storing a 2 based exponent ReturnValue Returns the floating point number x mantissa part fscant Function Reads characters with format from the stream Format include lt stdio h gt int fscanf stream format argument Method function Argument FILE _far stream Pointer of stream const char _far format Pointer of the input character string ReturnValue Returns the number of data entries stored in each argument Returns EOF if EOF is input from the stream as data Description Converts the characters input from the stream as specified in format and stores them in the variables shown in the arguments Argument must be a pointer to the respective variable Interprets code 0x1A as the end code and ignores any subsequent data Format is specified in the same way as in scanf REJ10J2009 0
189. g for details of the Processor Mode Register 2 2 3 Customizing for NC100 Memory Mapping a Structure of Sections In the case of a native environment compiler the executable files generated by the compiler are mapped to memory by the operating system such as UNIX However with cross environment compilers such as NC100 the user must determine the memory mapping With NC100 storage class variables variables with initial values variables without initial values character string data interrupt processing programs and interrupt vector address tables etc are mapped to Micoro Processor series memory as independent sections according to their function The names of sections consist of a base name and attribute as shown below Section Base Name Figure 2 24 Section Names Tabel 2 10 shows Section Base Name and Tabel 2 11 shows Attributes Tabel 2 10 Section Base Names Section base name Content data Stores data with initial values bss Stores data without initial values rom Stores character strings and data specified in pragma ROM or with the const modifier REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 37 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler Tabel 2 11 Section Naming Rules Attribute Meaning Target section base name INIT Section containing initial values of data data NEAR FAR NEAR near attribute data
190. ge V 1 02 B Extended Functions Reference pragma STRUCT Control structure mapping Rules 2 Arranging members This NC100 extended function allows you to map the all odd sized structure members first followed by even sized members Figure B 69 shows the offsets when the structure shown in Figure B 68 is arranged using pragma STRUCT Example REJ10J2009 0200 Rev 2 00 2010 04 01 143 struct s Member Mapped short i name location char c offset short j k Figure B 69 Example Mapping of Structure Members 3 You must declare pragma STRUCT for inhibiting packing and arranging the structure members before defining the structure members pragma STRUCT TAG unpack struct TAG int i char C s1 Figure B 70 Example of pragma STRUCT Declaration 7tENESAS R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 7 3 Using Extended Functions for Target Devices NC100 includes the following extended functions for target devices pragma ADDRESS Specify absolute address of I O variable Function Specifies the absolute address of a variable For near variables the specified address is within the bank Syntax pragma ADDRESSAvariable nameAabsolute address Description The absolute address specified in this declaration is expanded as a character string in an assembler file and defined in pseudo instruction EQU The format for writing the numerica
191. gister cccsscssssssssssesssesssescsescsescsssescscssscssscesssesssesesescsesessscscsescecscessseseseacsesesescseecsees 205 Appendix E Standard Lry isa ciscastins hsteachaniaeuia datiacanadniaainaaataaual aaa ciaanabinacnaaniaaana 206 E 1 Standard Header Files 0 ccccesssssssssssesssssssscecsescsescsessscscsescscscsescessssesseasssssssssesesssssesessaesescecseesesseseseseeeseseseeeseneaes 206 E 1 1 Contents of Standard Header Files cccsesssssssssscsessssssesssesscsesssessescssssesesessssesessesesesesessesesessssesesseseeeeseseees 206 E 1 2 Standard Header Files Reference ccccssscssssescsssssescssssssecsesssescsescsescsescscscsssesesessesessscsssescsescsesseesceeeeesees 207 E 2 Standard Function Reference ccccsessssessssssesesessssesessssesesescsesesssesesesscsesesseseseseeseseseeseseseseesssesseseseeseseseseeseseseees 216 E 2 1 Overview of Standard Library c cccccesssssssescssscssssssessscssscsescssscssssssscsescsssssseseseseseseseseseseseseseaeeeeeeseseseneseaeaes 216 E 2 2 List of Standard Library Functions by Function cccccsssessssssssssseescsescsescsescsescssscssscssessnscstssenssesesescsesees 217 E 2 3 Standard Function Reference ccccccscssscsssesssssescsesssescssscssescsesssesssescscscsescscscssscscscssssesscessseseseecaeseseeceeeeesees 223 E 2 4 Using the Standard Library ccccscccsscsssesssescsescscscscsescsescscscscsnscssscesssssesesesssesssessscscsesescscscsesseseaeaeseseseasscsees
192. gister when call C function as follows e Register which use in called C function b Register which should escaping in the entrance procedure of the called function e None REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 205 R32C 100 Series C Compiler Package V 1 02 E Standard Library Appendix E Standard Library E 1 Standard Header Files When using the NC100 standard library you must include the header file that defines that function This appendix details the functions and specifications of the standard NC100 header files E 1 1 Contents of Standard Header Files NC100 includes the 15 standard header files shown in Table E 1 Table E 1 List of Standard Header Files Header File Name Contents assert h Outputs the program s diagnostic information ctype h Declares character determination function as macro errno h Defines an error number float h Defines various limit values concerning the internal representation of floating points limits h Defines various limit values concerning the internal processing of compiler locale h Defines declares macros and functions that manipulate program localization math h Declares arithmetic logic functions for internal processing setjmp h Defines the structures used in branch functions signal h Defines declares necessary for processing asynchronous interrupts stdarg h Defines declares the functions which have a variable number of real argument
193. gned long long o Decimal signed int signed long signed long long Adding the suffix U or u or Lor 1 or LL or I results in the integer constant being processed as follows 1 Unsigned constants Specify unsigned constants by appending the letter U or u after the value The type is determined from the value in the following order o unsigned int unsigned long unsigned long long 2 long type constants Specify long type constants by appending the letter L or 1 The type is determined from the value in the following order Octal and hexadecimal signed long unsigned long signed long long unsigned long long o Decimal signed long long unsigned long long 8 long type constants Specify long long type constants by appending the letter LL or Il The type is determined from the value in the following order Octal and hexadecimal signed long long unsigned long long e Decimal signed long long 2 Floating point constants If nothing is appended to the value floating point constants are handled as double types To have them processed as float types append the letter F or f after the value If you append Lor 1 they are treated as long double types 3 Character constants Character constants are normally written in single quote marks as in character You can also include the following extended notation escape sequences and trigraph sequences Hexadecimal values are indicated by preceding the value with x Octal v
194. h gt When calculated in 8 bits static signed char min_b signed char vall signed char val2 When calculated in 16 bits static short int min_w short int vall short int val2 When calculated in 32 bits static long int min_l long int vall long int val2 RMPA Function Initial value init Number of times count The result is returned after performing a sum of products operation assuming p1 and P2 as the start addresses where multipliers are stored Syntax include lt asmmacro h gt When calculated in 8 bits static long long rmpa_b long long init unsigned long int count const signed char _far pl const signed char _far p2 When calculated in 16 bits static long long rmpa_w long long init unsigned long int count const short int _far pl const short int _far p2 When calculated in 32 bits static long long rmpa_l long long init unsigned long int count const long int _far pl const long int _far p2 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 158 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference SIN _ SSS Function Strings are transferred from a fixed source address that is indicated by p1 to the destination address indicated by p2 as many times as indicated by count in the address incrementing direction There is no return value Syntax include lt asmmacro h gt When calculated in 8 bits static void sin_b volatileconst vo
195. har type ROL register _Bool type int type 16 bits RO register short int type int type 82 bits R2R0 register float type long type pointer type AO register long long type A1A0 register 82 high order and 32 low order bits stored in Al and AO registers double type respectively long double type struct type Immediately before the function call save the far address for the area for storing union type the return value to the stack Before execution returns from the called function that function writes the return value to the area indicated by the far address saved to the stack D 3 2 Rules on Argument Transfer NC100 uses registers or stack to pass arguments to a function 1 Passing arguments via register When the conditions below are met the system uses the corresponding Registers Used listed in Table D 5 Table D 6 and Table D 6 to pass arguments e Function is prototype declared and the type of argument is known when calling the function Variable argument is not used in prototype declaration o For the type of the argument of a function the Argument and Type of Argument in Table D 5 Table D 6 and Table D 7 are matched 2 NC100 uses a via register transfer only when entering prototype declaration e when writing a new format Consequently all arguments are passed via stack when description of K amp R format is entered description of old format Note also that if a description format where pr
196. he specified character from the end of a character string Format include lt string h gt char far strrchr s c Method function Argument const char _far s Pointer to the character string to be searched in int c Character to be searched for ReturnValue Returns the position of character c that is last encountered in character string s Returns NULL when character string s does not contain character c Description Searches for the character specified in c from the end of area s You can search for 0 When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization strspn String Handling Functions Function Calculates the length number of specified characters that are found in the character string Format include lt string h gt size_t strspn sl s2 Method function Argument const char _far s1 Pointer to the character string to be searched in const char _far s2 Pointer to the character string of the character to be searched for ReturnValue Returns the length number of specified characters Description Calculates the size of the first character string consisting of characters in s2 from area s1 and searches the characters from the beginning of s1 You cannot search for 0 When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimi
197. he startup program include file sect100 inc You can modify the include file to change the mapping Figure 2 25 shows the how the sections are mapped according to the sample startup program s include file sect100 inc 1 near and far are the qualifiers specific to NC100 Use of these qualifiers makes it possible to specify addressing modes explicitly near The accessible addresses range from 00000000H to 00007FFFH and from OFFFF8000H to OFFFFFFFFH far The accessible addresses range from 00000000H to 007FFFFFH and from OF F800000H to OFFFFFFFFH REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 38 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 00000000H 00000400H data_SB8 section bss MONT section bss MON1 section dat MON2 section bss MON2 section C da FAR socion dat FAR section 00008000H bss FAR section 00800000H data EXT section bss _ EXT section L eaw a o gae iowa puxa D Pale Wyy ewayu _ tel eae Y4S OFFOO0000H data_EXT_INIT section rom_EXT section program_EXT section OFFEO0000H OFFFFOO00H rom FAR section data_NEAR_INIT section data_MON1_ INIT section E oo mh E OFFFFFBDCH vector section data_MON2_INIT section OFFFFFDCH OFFFFFFFH section fvector Pare WOY fewelu ee S pore AIOWWOW jeux Figure 2 25 Example Section Mapping 1 Also Figure 2 26 shows the how the sections are mapped according to the sa
198. hod macro Argument int c Character to be output ReturnValue Returns the output character if output normally Returns EOF if an error occurs Description Outputs one character to stdout REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 263 R32C 100 Series C Compiler Package V 1 02 E Standard Library puts SEA Functions Function Outputs one line to stdout Format include lt stdio h gt int puts str Method macro Argument char _far str Pointer of the character string to be output ReturnValue Returns 0 if output normally Returns 1 EOF if an error occurs Description Outputs one line to stdout The null character 0 at the end of the character string is replaced with the new line character n REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 264 R32C 100 Series C Compiler Package V 1 02 E Standard Library qsort Integer Arithmetic Functions Function Sorts elements in an array Format include lt stdlib h gt void qsort base nelen size cmp el e2 Method function Argument void _far base Start address of array size_t nelen Element number size_t size Element size int emp Compare function ReturnValue No value is returned Description Sorts elements in an array REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 265 R32C 100 Series C Compiler Package V 1 02 E Standard Library rand Function Generates a pseudo random number Format incl
199. id_far pl void_far p2 unsigned long int count When calculated in 16 bits static void sin_w volatileconst void_far pl void_far p2 unsigned long int count When calculated in 32 bits static void sin_l volatileconst void_far pl void_far p2 unsigned long int count SMOVB eee Function Strings are transferred from the source address indicated by p1 to the destination address indicated by p2 as many times as indicated by count in the address decrementing direction There is no return value Syntax include lt asmmacro h gt When calculated in 8 bits static void smovb_b const void _far pl void _far p2 unsigned long int count When calculated in 16 bits static void smovb_w const void _far pl void _far p2 unsigned long int count When calculated in 32 bits static void smovb_l1 const voiid _far pl void _far p2 unsigned long int count REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 159 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference SMOVF Function Strings are transferred from the source address indicated by p1 to the destination address indicated by p2 as many times as indicated by count in the address incrementing direction There is no return value Syntax include lt asmmacro h gt When calculated in 8 bits static void smovf_b const void_far pl void _far p2 unsigned long int count When calculated in 16
200. ide expression invalid operands y The unary operation or postfix operation contains an error For the unary operator check the right side expression For the postfix operator check the leftside expression invalid gt used The left side expression of gt is not struct or union The left side expression of gt must have struct or union invalid s condition The ternary operator is erroneously written Check the ternary operator invalid array type Ye y ey Incomplete arrays cannot be declared Specify the number of elements in the multidimensional array invalid operation for pointer to incomplete type Invalid calculation for the pointer to an incomplete type Define members of a structure or define complete structs Invalid pragma OS Extended function interrupt number The INT No in pragma OS Extended function is invalid Specify correctly Invalid pragma INTCALL interrupt number The INT No in pragma INTCALL is invalid Specify correctly Invalid pragma SPECIAL special page number NC30 NC308 only Oe s The number or format specification written with pragma SPECIAL is incorrect Specify the number or format correctly invalid pragma INTERRUPT vector number The number or format specification written with pragma INTERRUPT is incorrect Specify the number or format correctly invalid CAST operand The cast operation contains an e
201. ier can set only pointer type The RESTRICT qualifier is declared outside a pointer Declare it in only a pointer section name interrupt no more used p eu ey y The section name specified by pragma SECTION uses imterrupt A section name interrupt cannot be used Change it to another size of incomplete type An undefined structure or union is used in the operand of the size of operator Define the structure or union first size of incomplete array type The number of elements of an array defined as an operand of the size of operator is unknown Define the structure or union first An attempt is made to find size of of an array of unknown size This is an invalid size Specify the size of the array size of void An attempt is made to find the size of void This is an invalid size The size of void cannot be found standard library function name need include file name This standard library function is used without its header file included Be sure to include the header file static variable in inline function static data is declared within a function that is declared in storage class inline Do not declare static data in an inline function string size bigger than array size The size of the initialize expression is greater than that of the variable to be initialized Make sure that the size of the initialize expression is equal to or smaller than the
202. ier shown in Figure C 2 in NC100 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 172 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications Declarator Pointer opt declarator2 Declarator2 identifier declarator declarator2 constant expression opt declarator2 list of dummy arguments opt Only the first array can be omitted from constant expressions showing the number of arrays opt indicates optional items Figure C 2 Format of Declaration Specifier 4 Initialization expressions NC100 allows the initial values shown in Figure C 3 in initialization expressions integral types constant integral types array constant constant character types constant character types array character literal constant pointer types character literal pointer array character literal character literal Figure C 3 Initial Values Specifiable in Initialization Expressions REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 173 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications b Function Declaration Use the format shown in Figure C 4 to declare functions function declaration definition storage class specifier Atype declarator Adeclaration specifier Amain program function declaration prototype declaration storage class specifier Atype declarator Adeclaration specifier Figure C 4 Declaration Format of Function 1
203. igent VO 8 software int 52 word dummy_int intelligent YO 9 software int 53 word dummy_int intelligent YO 10 software int 54 Figure 2 15 Startup Program List 10 Sect100 inc REJ10J2009 0200 Rev 2 00 2010 04 01 30 7tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword lword dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int omitted dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int omitted dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int dummy_int intelligent I O 11 reservation area reservation area reservation area CAN1WU reservation area reservation area reservation area reservation area reservation a
204. igit The hex constant contains some character that cannot be used in hex notation Use numerals 0 to 9 and alphabets A to F and a to f to describe hex constants No initialized of variable name It is probable that the register variables are used without being initialized Make sure the register variables are assigned the appropriate value No storage class amp data type in declare global storage class amp int type assumed The variable is declared without storage class and type specifiers It will be handled as int when processed Write the storage class and type specifiers non initialized variable variable name is used It is probable that uninitialized variables are being referenced Check the statement description This warning can occur in the last line of the function In such a case check the description of the auto variables etc in the function If the description is correct ignore this warning non prototyped function used A function is called that is not declared of the prototype This message is output only when you specified the Wnon_prototype option Write prototype declaration Or delete the option Wnon_prototype non prototyped function declared A prototype declaration for the defined function cannot be found Displayed only when the WNP option is specified Write a prototype declaration octal constant is out of range The octal constant contains some ch
205. igure 1 10 and Figure 1 11 are examples of the output files When the nc100 command line option dsource dS is specified the assembly language source files contain the contents of the C source file as comments REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 10 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 _ LANG CX XX XX XXX REV X C Compiler OUTPUT ccom100 Version X XX XX XXX Copyright C XXXX Renesas Electronics Corp 3 and Renesas Solutions Corp All Rights Reserved Compile Start Time XXX XX XX XX XX XX XXXX 3 COMMAND LINE ccom100 dS o sample a30 sample i aeea ea a EPRI Normal Optimize OFF G i ROM size Optimize OFF l Speed Optimize OFF Default ROM is far Default RAM is near GLB __SB SB __ SB FB 0 HHH FUNCTION main HHH ARG Size 4 Auto Size 0 Context Size 4 SECTION program CODE ALIGN _file sample c align ine 6 C_ SRC glb _main _main ine 9 H C_SRC flag CLR mov 00000000H R2RO_ flag line 11 H C_SRC printf flag d n flag 2 push R2RO flag push __ TO jsr _ printf add l 08H SP ine 13 C_ SRC rts E1 omitted glo __iob glib getc glb _getchar glib putc glib putchar glib feof glb ferror glib fgetc glib fgets glib fputc omitted Figure 1 10 Example Assembly Language Source File 1 sample a30 REJ10J2009 0200 Rev 2 00 2010 04 01 11 2tENESAS
206. ile Format include lt stdio h gt int feof stream Method macro Argument FILE _far stream Pointer of stream ReturnValue Returns true other than 0 if the stream is EOF Otherwise returns NULL 0 Description Determines if the stream has been read to the EOF Interprets code 0x1A as the end code and ignores any subsequent data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 232 R32C 100 Series C Compiler Package V 1 02 E Standard Library ferror Function Checks input output errors Format include lt stdio h gt int ferror stream Method macro Argument FILE _far stream Pointer of stream ReturnValue Returns true other than 0 if the stream is in error Otherwise returns NULL 0 Description Determines errors in the stream Interprets code 0x1A as the end code and ignores any subsequent data fflush Function Flushes the stream of an output buffer Format include lt stdio h gt int fflush stream Method function Argument FILE _far stream Pointer of stream ReturnValue Always returns 0 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 233 R32C 100 Series C Compiler Package V 1 02 E Standard Library fgetc Input Output Functions Function Reads one character from the stream Format include lt stdio h gt int fgetc stream Method function Argument FILE _far stream Pointer of stream ReturnValue Returns the one input character Re
207. ine FNRI Exclude far and near from reserved words Only _inline is made a reserved word fsigned_char fSC Handles type char without sign specification as type signed char fswitch_other_section SOS This option outputs a ROM table for a switch statement to some other section than a program section fuse_FPU FUF Outputs FPU instruction 2 unsigned char type signed char type short type and unsigned short type data evaluated under ANSI rules is always extended to the int type data This is because operations on char types c1 c2 2 c3 for example would otherwise result in an overflow and failure to obtain the intended result REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 82 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference fansi Function Validates the following command line options fnot_reserve_asm Removes asm from reserved words fnot_reserve_far_and_near Removes far and near from reserved words fnot_reserve_inline Removes inline from reserved words fextend_to_int Extends char type data to int type data to perform operations Supplement When this option is specified the compiler generates code in conformity with ANSI standards fconst_not_ROM f CNR Modify generated code Function Does not handle the types specified by const as ROM data Supplement The const specified data by default is located in the ROM area Take a look at the example below int const array 10 1 2 3
208. ing Extended Functions for Target Devices Table B 6 Extended Functions for Use with Target Devices 1 2 Extended function Description pragma ADDRESS Specifies the absolute address of a variable For near variables this specifies the address within the bank Syntax pragma ADDRESSAvariable nameAabsolute address Example pragma ADDRESS port0O 2H pragma DMAC Specifies the DMAC register of a external variable Syntax pragma DMACAvariable nameADMAC register name Example pragma DMAC dsa0 DSAO pragma INTCALL Declares a function written in assembler called in a software interrupt Gnt instruction Syntax pragma INTCALLAINT No Aassembler function name register name Example pragma INTCALL 25 func RO R1 pragma INTCALLAINT No AC language function name Example pragma INTCALL 25 func Always be sure to declare the prototype of the function before entering this declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 131 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference Table B 7 Extended Functions for Use with Target Devices 2 2 Extended function Description pragma INTERRUPT Declares an interrupt handling function written in C language This declaration causes code to perform a procedure for the interrupt handling function to be generated at the entry or exit to and from the functio
209. ing an interrupt vector table in Section 2 2 2 Customizing the Startup Program REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 149 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma INTERRUPT Rules 1 A warning is output when compiling if you declare interrupt processing functions that take parameters 2 Awarning is output when compiling if you declare interrupt processing functions that return a value Be sure to declare that any return value of the function has the void type 8 Only functions for which the function is defined after a pragma INTERRUPT declaration are valid 4 No processing occurs if you specify other than a function name 5 No error occurs if you duplicate pragma INTERRUPT declarations 6 You cannot specify both switch E and switch B at the same time 7 If different interrupt vector numbers are written in the same interrupt handling function the vector number declared later is effective 8 _ V and other switches cannot be used at the same time pragma INTTERUPT _ intr vect 10 pragma INTTERUPT _ intr vect 20 The interrupt vector number 20 is effective Figure B 76 Example for writing different interrupt vector numbers Example extern int int_counter pragma INTERRUPT B i func void i_func void int_counter 1 Figure B 77 Example of pragma INTERRUPT Declaration Supplement 1 To use the start
210. initialized to 0 by clearing the bss area during the startup routine staticint i Figure3 9 Example Declaration of Variable Without Initial Value In some instances it is not necessary for a variable with no initial value to be cleared to 0 In such cases you can comment out the routine for clearing the bss area in the startup program to increase the speed of startup processing zero clear BSS BZERO bss_SB8 top bss _SB8 BZERO bss_SB16_top bss_SB16 BZERO bss_NEAR_top bss NEAR BZERO bss_FAR_top bss FAR BZERO bss_MON1_top bss_MON1 BZERO bss_MON2_top bss_MON2 BZERO bss_MON3_top bss_MON3 BZERO bss_MON4_top bss_MON4 Figure3 10 Commenting Out Routine to Clear bss Area 1 The external variables in RAM which do not have initial values are referred to as bss REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 55 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique 3 3 Linking Assembly Language Programs with C Programs 3 3 1 Calling Assembler Functions from C Programs a Calling Assembler Functions Assembler functions are called from C programs using the name of the assembler function in the same way that functions written in C would be The first label in an assembler function must be preceded by an underscore _ However when calling the assembly function from the C program the underscore is omitted The calling C program must include a prototype declaration f
211. int BRK software int 0 omitted word dummy int DMAO software int 8 word dummy_int DMA1 software int 9 word dummy_in DMA2 software int 10 omitted word dummy_int uart trance NACK software int 19 word dummy_int uari receive ACK software int 20 word dummy_int TIMER BO software int 21 omitted word dumny int INT5 software int 26 word dummy_int INT4 software int 27 omitted word dummy_int uart2 trance NACK software int 33 word dummy_int uart2 receive ACK software int 34 omitted word dummy_int software int 255 dummy_int is a dummy interrupt processing function Figure 2 33 Interrupt Vector Address Table REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 48 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler Follow the procedure described below to alter the interrupt vector table of the vector section in sect100 inc 1 Externally declare the interrupt processing function in the GLB as100 pseudo instruction 2 The labels of functions created by NC100 are preceded by the underscore _ Therefore the names of interrupt processing functions declared here should also be preceded by the underscore 3 Replace the names of the interrupt processing functions with the names of interrupt processing functions that use the dummy interrupt function name dummy_int corresponding to the appropriate interrupt table in the vector address table Figure 2 34 is an exa
212. int number atoi Function Converts a character string into an int type integer Format include lt stdlib h gt int atoi s Method function Variable const char _far s Pointer to the converted character string ReturnValue Returns the value derived by converting a character string into an int type integer REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 225 R32C 100 Series C Compiler Package V 1 02 E Standard Library atol Function Converts a character string into a long type integer Format include lt stdlib h gt long atol s Method function Variable const char _far s Pointer to the converted character string ReturnValue Returns the value derived by converting a character string into a long type integer REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 226 R32C 100 Series C Compiler Package V 1 02 E Standard Library bcopy Memory Handling Functions Function Copies characters from a memory area to another Format include lt string h gt void bcopy srce dtop size Method function Variable char _far src Start address of the memory area to be copied from char _far dtop Start address of the memory area to be copied to unsigned long size Number of bytes to be copied ReturnValue Copies the number of bytes specified in size from the beginning of the area specified in src to the area specified in dtop bsearch Integer Arithmetic Functions Function Performs binary search in an array Forma
213. ion bss NEAR section bss MONI section bss FAR section bss MONZ2 section bss EXT section bss_ MONS section bss SB8 section bss_MONG4 section bss SB16 section rom_NEAR section rom_FAR section rom_EXT section 2 Example Section Mapping in Single Chip Mode Figure 2 29 to Figure 2 32 are examples of the sect100 inc include file which is used for mapping sections to memory in single chip mode REJ10J2009 0200 Rev 2 00 2010 04 01 43 7tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler AIK IKK AK III KI RIK IK III IIR IEK IERIE I IIIS II III IKI III KIKI kkk kk kkk kkk Copyright C XXXX Renesas Electronics Corp and Renesas Solutions Corp All rights reserved C COMPILER for R32C 100 omitted ld sect 00 inc v X X XXXX XX KX XX XX XX XXXXX Exp A IK eeke ee keke dee RIK IK ekeke ede kekeke ek eee keee kek IKI eke kekek keke IK IIIA KIER k kkk kkk kkk kkk 3 Arrangement of section NEAR RAM SECTIONS section data _SB8 data org 00000400H data_SB8_top section bss_SB8 data align bss_SB8_top section data NEAR data align data_NEAR_top section bss_NEAR data align bss_NEAR_top section data MONI data align data_MON1_top section bss_MON1 data align bss_MON1_top section data_MON2 data align data_MON2_top section bss_MON2 data align bss_MON2_top section data_MON3 data align data_MON3_top
214. is a control character O isdigit Checks whether the character is a numeral O isgraph Checks whether the character is printable except a blank O islower Checks whether the character is a lower case letter O isprint Checks whether the character is printable including a blank O ispunct Checks whether the character is a punctuation character O isspace Checks whether the character is a blank tab or new line O isupper Checks whether the character is an upper case letter O isxdigit Checks whether the character is a hexadecimal character O tolower Converts the character from an upper case to a lowercase O toupper Converts the character from a lower case to an uppercase O REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 218 R32C 100 Series C Compiler Package V 1 02 E Standard Library 3 Input Output Functions The following lists Input Output functions Table E 4 Input Output Functions Type Function Contents Reentrant Initialize init Initializes R32C series s input outputs X clearerror Initializes clears error status specifiers X Initialize fgete Inputs one character from the stream X getc Inputs one character from the stream X getchar Inputs one character from stdin X fgets Inputs one line from the stream X gets Inputs one line from stdin X fread Inputs the specified items of data from the stream X scanf Inputs characters with format from
215. is desired Notes Conversely if a function is called that is located 32 Kbytes or more forward or backward from the calling position the JSR W command causes an error when linking This error can be avoided by a combined use with pragma JSRA REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 85 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference fnear_ROM fNROM Modify generated code Function Changes the default attribute of ROM data to near Supplement The ROM data const specified variables etc are located in the far area by default By specifying this option you can locate the ROM data in the near area fno_align NA Modify generated code Function Does not align the start address of the function fno_switch_table INST Modify generated code Function When this option is specified the code which branches since it compares is generated to a switch statement Supplement Only when code size becomes smaller when not specifying this option the code which used the jump table is generated Notes For such a large function whose code size is larger than 32 Kbytes if code which contains a jump table for a switch statement is generated the program may not be branched to an appropriate address In that case be sure to specify this option Please note that when a code which cannot be branched properly because of not specifying this option is generated the compiler assembler and linkage edito
216. ister variables 1 Tf the variables need to be forcibly mapped to registers using the register qualifier specify the option fenable_register fER when compiling REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 111 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 2 4 Specifying Symbol Name of extern and static Variable Extern and static storage class variables written in C are referenced as symbols You can use the format shown in Figure B 25 to use extern and static variables in asm functions asm op code R1 variable name Figure B 25 Description Format for Specifying Symbol Name Only two variable name can be specified by using this description format The following types are supported for variable names Variable name Array name integer o Struct name member name not including bit field members short idata short af 3 struct TAG short i short k s void func void asm MOVW PO idata asm MOVW RO al2 asm MOVW_ RO si remainder omitted Figure B 26 Example for Specifying Symbol Names See Figure B 27 for examples of referencing extern and static variables REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 112 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference C source file extern shot ext_val void func void static short s_val asm mov w _ 01H ext_val asm mov w
217. ize of void cannot be found Sorry stack frame memory exhaust max 128 bytes but now nnn bytes A maximum of 128 bytes of parameters can be secured on the stack frame Currently nnn bytes have been used Reduce the size or number of parameters Sorry compilation terminated because of these errors in function name An error occurred in some function indicated by function name Compilation is terminated Correct the errors detected before this message is output Sorry compilation terminated because of too many errors Errors in the source file exceeded the upper limit 50 errors Correct the errors detected before this message is output struct or enum s tag used for union The tag name for structure and enumerated type is used as a tag name for union Change the tag name struct or union s tag used for enum The tag name for structure and union is used as a tag name for enumerated type Change the tag name struct or union enum does not have long or sign Type specifiers long or signed are used for the struct union enum type specifiers Write the type specifier correctly switch s condition is floating The float type is used for the expression of a switch statement Use the integer type or enumerated type switch s condition is void The void type is used for the expression of a switch statement Use the integer type or enumerated type switch s condition must integer ey In
218. jmp_buf env Pointer to the area where environment is saved ReturnValue Returns the numeric value given by the argument of longjmp Description Saves the environment to the area specified in env setlocale Localization Functions Function Sets and searches the locale information of a program Format include lt locale h gt char _far setlocale category locale Method function Argument int category Locale information search section information const char _far locale Pointer to a locale information character string ReturnValue Returns a pointer to a locale information character string Returns NULL if information cannot be set or searched sin Function Calculates sine Format include lt math h gt double sin x Method function Argument double x arbitrary real number ReturnValue Returns the sine of given real number x handled in units of radian REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 270 R32C 100 Series C Compiler Package V 1 02 E Standard Library sinh Mathematical Functions Function Calculates hyperbolic sine Format include lt math h gt double sinh x Method function Argument double x arbitrary real number ReturnValue Returns the hyperbolic sine of given real number x sprintf Input Output Functions Function Writes text with format to a character string Format include lt stdio h gt int sprintf pointer format argument Me
219. l values therefore depends on the assembler as follows Append B or b to binary numbers Append O or o to octal numbers Write decimal integers only Append H or h to hexadecimal numbers If the number starts with letters A to F precede it with 0 Rules 1 All storage classes such as extern and static for variables specified in pragma ADDRESS are invalid 2 Variables specified in pragma ADDRESS are valid only for variables defined outside the function 3 pragma ADDRESS is valid for previously declared variables 4 pragma ADDRESS is invalid if you specify other than a variable 5 No error occurs if a pragma ADDRESS declaration is duplicated but the last declared address is valid 6 A warning occurs if you include an initialization expression and an initialization expression is invalid 7 Normally pragma ADDRESS operates on I O variables so that even though volatile may not actually be specified the compiler processes them assuming volatile is specified Example pragma ADDRESS port 24H int O void func void io 10 Figure B 71 pragma ADDRESS Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 144 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma ADDRESS Note Supplement Ifa variable is used prior to the specification of pragma ADDRESS as shown in Figure B 72 the specification of pragma ADDRESS has no effect
220. laring a variable twice typedef initialized An initialize expression is described in the variable declared with typedef Delete the initialize expression uncomplete array pointer operation An incomplete multidimensional array has been accessed to pointer Specify the size of the multidimensional array undefined label label used The jump address label for goto is not defined in the function Define the jump address label in the function union or enum s tag used for struct The tag name for union and enumerated types is used as a tag name for structure Change the tag name unknown function argument variable name An argument is specified that is not included in the argument list Check the argument unknown member member name used A member is referenced that is not registered as any structure or union members Check the member name unknown pointer to structure identifier variable name The left side expression of gt is not the structure or union type Use struct or union as the left side expression of gt unknown size of struct or union A structure or union is used which has had its size not determined Declare the structure or union before declaring a structure or union variable unknown structure identifier variable name The left side expression of dose not have struct or union Use the struct or union as it unknown variable variable name
221. lds og greater than 2 bits asm op code b FB bit field name Figure B 20 Format for Specifying FB Offset Bit Position You can only specify one variable name using this format Figure B 21 is an example REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 109 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference void func void struct TAG char bit0 1 char bit 1 char bit2 1 char bit3 1 s asm bset b FB s bit1 Figure B 21 Example for Specifying FB Offset Position Figure B 22 shows examples of referencing auto area bit fields and the result of compiling C source file void func void struct TAG char bit0 1 char bit1 1 char bit2 1 char bit3 1 s asm bset b FB s bit1 Assembly language source file compile result AHH FUNCTION func HHH FRAME AUTO _ PADS size 1 offset 1 HHH FRAME AUTO __PAD2 size 1 offset 2 HHH FRAME AUTO __PAD1 size 1 offset 3 HHH FRAME AUTO s size 1 offset 4 HH H ARG Size 4 Auto Size 4 Context Size 8 SECTION program CODE ALIGN _file bit c align ine 2 C_ SRC glb _func _func enter 04H ine 9 H C_SRC asm bset b FB s bit1 HHHH ASM START bset 1 4FB s HHH ASM END _line 10 H C_SRC exitd Figure B 22 Example of Referencing auto Area Bit Field REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 110 R32C 100 Series C Compiler Package V 1 02 B Ext
222. le Pointer variable Notes Because register specification in some cases has an adverse effect that the efficiency decreases be sure to verify the generated assembly language before using this specification fextend_to_int fETI Modify generated code Function Performs operation after extending chartype or short type data to the int type data Extended according to ANSI standards Supplement unsigned char type signed chartype short type and unsigned short type data evaluated under ANSI rules is always extended to the int type data This extension is provided to prevent a problem in chartype arithmetic operations e g cl c2 2 c38 that the char type overflows in the middle of operation and that the result takes on an unexpected value An example is shown below void main void char cl char c2 200 char C3 2 c1 c2 2 c3 In this case the char type overflows when calculating c2 2 so that the correct result may not be obtained Specification of this option helps to obtain the correct result The reason why extension into the int type is disabled by default is because it is conducive to increasing the ROM efficiency any further REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 84 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference ffar_ RAM fFRAM Modify generated code Function Change the default attribute of RAM data to far Supplement The RAM data variabl
223. ler Syntax pragma PAGE Description Putting the line pragma PAGE in C source code the PAGE pseudo instruction is output at the corresponding line in the compiler generated assembly source This instruction causes page ejection asesmbler output assembly list file Rules 1 You cannot specify the character string specified in the header of the assembler pseudo instruction PAGE 2 You cannot write a pragma PAGE in an auto variable declaration Example void func void int i j for i 0 i lt 10 i func2 pragma PAGE i Figure B 83 Example of pragma PAGE REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 155 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 8 assembler Macro Function B 8 1 Outline of Assembler Macro Function NC100 allows part of assembler commands to be written as C language functions Because specific assembler commands can be written directly in a C language program you can easily tune up the program B 8 2 Description Example of Assembler Macro Function Assembler macro functions can be written in a C language program in the same form as C language functions as shown in Figure B 84 When using the facility of any assembler macro function be sure to include asmmacro h include lt asmmacro h gt Includes the assembler macro function definition file pe on char b 20 void func void rmpa_b 0 19 a b asm Macro Function rmp
224. liary I O to match the system 3 Set the level 3 function pointer for standard auxiliary I O 4 Delete the conditional compile commands for the speed function and change to function speedO for UARTO These settings allow both UARTs to be used However functions using the standard I O stream cannot be used for standard auxiliary I O used by UARTO Therefore only use functions that take streams as parameters Figure E 6 shows how to change the init function void init void omitted stdaux gt _flag __IORW 2 set read write mode omitted stdaux gt _mod _ TEXT 2 set text mode omitted stdaux gt _func in _sget0 3 set UARTO level 3 input function omitted stdaux gt _func_out _sput0 3 set UARTO level 3 input function omitted speed _96 _B8 _PN _S2 4 set UARTO speed function init_prn 2 to 4 correspond to the items in the description of setting above Figure E 6 Modifying the init Function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 299 R32C 100 Series C Compiler Package V 1 02 E Standard Library c Incorporating the Modified Source Program There are two methods of incorporating the modified source program in the target system 1 Specify the object files of the modified function source files when linking 2 Use the makefile under MS Windows makefile dos supplied in the NC100 package to update the library file In method 1 the functions spe
225. lization block The standard library file supplied in the NC100 package initializes the stream in function init which is called from the ncrt0 a30 startup program Figure E 5 shows the source program for the init function include lt stdio h gt FILE _iob 4 Void init void void init void stdin gt _cnt stdout gt _cnt stdaux gt _cnt stdprn gt _cnt 0 stdin gt _flag _IOREAD stdout gt _flag _IOWRT stdaux gt _flag _IORW stdorn gt _flag _IOWRT stdin gt _mod _TEXT stdout gt _mod _ TEXT stdaux gt _mod _ BIN stdprn gt _mod _TEXT stdin gt _func_in _ sget stdout gt _func_in NULL stdaux gt _func_in _sget stdprn gt _func_in NULL stdin gt _func_out _ sput stdout gt _func_out _ sput stdaux gt _func_out _ sput stdprn gt _func_out _pput ifdef UARTO speed _96 _B8 _PN _S2 else UART1 default speed _96 B8 PN S2 endif init_prn Figure E 5 Source file of init function init c REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 298 R32C 100 Series C Compiler Package V 1 02 E Standard Library In systems using the two R32C 100 series UARTs modify the init function as shown below In the previous subsection we set the UARTO functions in the device c source file temporarily as _sget0 _sput0 and speedO 1 Use the standard auxiliary I O stdaux for the UARTO stream 2 Set the flag flag and mode _mod for standard auxi
226. low e Input line buffer overflow occurred when reading source file s Or buffer overflowed when converting macros Reduce each line in the source file to a maximum of 1023 characters If you anticipate macro conversion modify the code so that no line exceeds 1023 characters after conversion 29 EOF in comment e End of file encountered in a comment Correct the source file 31 EOF in preprocess command e End of file encountered in a preprocess command Correct the source file 32 unknown preprocess command e An unknown preprocess command has been specified Only the following preprocess commands can be used in CPP100 include define undef if ifdef ifndef Helse endif elseif line assert pragma error 33 new_line in string e A newline code was included in a character constant or character string constant Correct the program 34 string literal out of range 509 e Acharacter string exceeded 509 characters characters Reduce the character string to 509 characters max 35 macro replace nesting over e Macro nesting exceeded the limit 20 Reduce the nesting level to a maximum of 20 41 include file error e Error in include instruction gt Correct the include REJ10J2009 0200 Rev 2 00 2010 04 01 305 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 5 cpp100 Error Messages 3 3 No Error message Description and countermeasure 43 illegalid name e E
227. ls are used in the NC100 manuals Symbol Description A gt MS Windows TM prompt lt RET gt Return key lt gt Mandatory item Optional item A Space or tab code mandatory A Space or tab code optional omitted Indicates that part of file listing has been omitted Additional descriptions are provided where other symbols are used REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS R32C 100 Series C Compiler Package V 1 02 Contents Chapter 1 Trntroduction to NG TOO sisscisitscsccasicscaceisscuesanisaes guid sescdaaits eugenics chs snaea guia senddaaand cheassicaedias TTEA 1 1 1 NC100 Components saeriicntendernei anaiita eead iia aii e edat aan 1 1 2 NC100 Processing FIOW ccsscssscsssesssescssscsescsescssscsscssssesesssesssesssescsesesescssscsessseessussesesessssseseacacscsesescsesceeseeesscesseeseees 2 1 2 1 VIG LOO a ns O vs svedeas3 EO os sa dagnayas tue anaes sbvatsaed sue patseanian a sans aaa eaiaa sid dea ETR 3 1 2 2 TOTAL OO EA E teste E EE A EE A TEE EEE A E 3 1 2 3 e01 SAKO 0 OE E E E A E copbenasediohgasedacesscaescansaviscazbes 3 1 2 4 CCOMD LOO E E E E veel E A eden dah ete Gites eeeees 3 1 2 5 F210 0 RUC O eupen e eere tere tert E ier ererr ise ere re vee rere creer ere tere ereee ree N aE 3 1 2 6 Call Walker amp Sensi ccccccccscssscsscesssesssescssscsssessssssssssssessscsesesesescsesesescsssessessseseessesssesssescsescsesesescassessecasesesesess 3 1 2 7 IMA VIC WER 5 ER EE AE
228. lt stdio h gt int getchar void Method macro Argument No argument used ReturnValue Returns the one input character Returns EOF if an error or the end of the file is encountered Description Reads one character from stream stdin Interprets code 0x1A as the end code and ignores any subsequent data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 240 R32C 100 Series C Compiler Package V 1 02 E Standard Library gets Input Output Functions Function Reads one line from stdin Format include lt stdio h gt char _far gets buffer Method function Argument char _far buffer Pointer of the location to be stored in ReturnValue Returns the pointer of the location to be stored the same pointer as given by the argument if normally input Returns the NULL pointer if an error or the end of the file is encountered Description Reads character string from stdin and stores it in the buffer The new line character n at the end of the line is replaced with the null character 0 Interprets code 0x1A as the end code and ignores any subsequent data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 241 R32C 100 Series C Compiler Package V 1 02 E Standard Library init Input Output Functions Function Initializes the stream Format include lt stdio h gt void init void Method function Argument No argument used ReturnValue No value is returned Description Initi
229. lues of assembler functions J Functions that return structures or unions 3 You can use the following registers for parameters when calling double types long types 64 bit registers R3R1R2R0 R7RS5R6R4 A1A0 A8A2 float types long types int types far far pointer 32 bit registers R2R0 R3R1 R6R4 R7R5 AO Al A2 A3 short types int types fint_16 option use 16 bit registers RO R1 R2 R3 R4 R5 R6 R7 char types _Bool types 8 bit registers ROL ROH R1L R1H R2L R2H R3L R3H e There is no differentiation between uppercase and lowercase letters in register names 4 You can only use decimals for the INT Numbers Declaring functions of which the body is written in C 1 Before a pragma INTCALL declaration be sure to include a prototype declaration If there is no prototype declaration a warning is output and the pragma INTCALL declaration is ignored 2 You cannot specify register names in the parameters of functions that include the pragma INTCALL declaration 3 Observe the following in the prototype declaration 1 In the prototype declaration you can only declare functions in which all parameters are passed via registers as in the function calling rules 2 You cannot declare the following functions as the return values of functions e Functions that return structures or unions 4 You can only use decimals for the INT Numbers REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 147
230. ly resulting in the assembled program performing improper operations pragma ADDRESS taQic_addr O06CH Timer AO interrupt control register struct char ilvl 3 charir 1 An interrupt request bit chardmy 4 ta0ic void wait_until_IR_is ON void while taOic ir 0 Waits for taOic ir to become 1 ta0ic ir 0 Returns 0 to taOic ir when it becomes 1 Please compile after taking the following measures if the manipulation instructions is generated to bit operation of SFR area Make sure that no BTSTC and BTSTS instructions are generated after these side steppings Optimization options other than O05 are used When you use the optimization option of O5 please use together with O5A Aninstruction is directly described in a program using an ASM function O50A Optimization Function Inhibits code generation based on bit manipulating instructions when the optimization option 05 is selected REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 72 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference VU Optimization Function Optimizes ROM size in preference to speed This option can be specified with g and Q options Notes When this option is used the source line information may partly be modified in the course of optimization Therefore if this options is specified when your program is running on the debugger your progr
231. ly the functions whose expanded size from to Y is within 150 lines including a comment line Oglb_jmp OGJ Optimization Function Global jump is optimized REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 76 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Oglobal_to_inline OGTI Handles global functions as inline declared Function Handles global functions as inline declared functions Gnline functions to generate inline expanded assemble code Supplement When the following conditions are met the compiler handles global functions as inline functions to generate inline expanded assemble code 1 Operation is performed on the global function whose body is written prior to a function call A function call and the body of that function must be written in the same source file If the Ofile_inline option is selected this condition should be ignored 2 Ifno addresses are acquired in the program for the target global function 3 Ifthe target global function is recursively called The following shows an example of how a global function to be inline expanded will be written extern int i The function func is inline expanded in pds te se Pe eee ah ee ee AP ee ts respective places of the function main in which int func void E i i return i r void main void int s func s func Notes 1 Assembler code for the body of the global functi
232. m Function fansi None Makes fnot_reserve_far_and_near fnot_reserve_asm and fextend_to_int valid fconst_not_ROM fCNR Does not handle the types specified by const as ROM data fdouble_32 fD32 This option specifies that the double type be handled in 32 bit data length as is the float type fenable_register FER Make register storage class available fextend_to_int FETI Performs operation after extending char type or short type data to the int type data Extended according to ANSI standards ffar_ RAM FFRAM Changes the default attribute of RAM data to far finfo None Outputs the information required for the Inspector Call Walker and Map Viewer to the absolute module file x30 fint_16 f116 Does handle int type at the 16 bit width JSRW None Changes the default instruction for calling functions to JSR W fnear ROM NROM Changes the default attribute of ROM data to near fno_align FNA Does not align the start address of the function fno_switch_table NST When this option is specified the code which branches since it compares is generated to a switch statement fnot_address_volatile ANAV Does not regard the variables specified by pragma ADDRESS pragma EQU as those specified by volatile fnot_reserve_asm FNRA Exclude asm from reserved words Only _asm is valid fnot_reserve_far_and_near fNRFAN_ Exclude far and near from reserved words Only _far and _near are valid fnot_reserve_inl
233. m value that a signed char type 127 variable can take on as value SCHAR_MIN Minimum value that a signed char type 128 variable can take on as value INT MAX Maximum value that a int type variable 32767 or 2147483647 When can take on as valueMaximum value that a compile option a int type variable can take on as value 116 fint_16 is specified INT MIN Minimum value that a int type variable 32768 or 2147483648 When can take on as value a compile option 116 fint_16 is specified SHRT_ MAX Maximum value that a short int type 32767 variable can take on as value SHRT_MIN Minimum value that a short imt type 32768 variable can take on as value LONG_MAX Maximum value that a long type variable 2147483647 can take on as value LONG_MIN Minimum value that a long type variable 2147483648 can take on as value LLONG_MAX Maximum value that a signed long 9223372036854775807 long type variable can take on as value LLONG_MIN Minimum value that a signed long 9223372036854775808 longtype variable can take on as value UCHAR_MAX Maximum value that an unsigned 255 char type variable can take on as value UINT_MAX Maximum value that an unsigned int type 65535 or 4294967295 When variable can take on as value a compile option 116 fint_16 is specified USHRT_MAX Maximum value that an unsigned short 65535 int type variable can take on as value ULONG_MAX Maximum value that an unsigned long 4294967295 int type variable can take
234. ma INTCALL are written in C the parameters are passed via the stack When the body of functions declared in pragma INTCALL are written in C specify the parameters are being passed via the stack pragma PARAMETER function s e A register which is specifed in the function decleared by register not allocated pragma PARAMETER can not be allocated Use the correct register const is duplicate e const is described more than twice Write the type qualifier correctly far amp near conflict e far near is described more than twice Write near far correctly far is duplicate e far is described more than twice Write far correctly near is duplicate e near is described more than twice Write near correctly static is illegal storage class for e An appropriate storage class is used in argument agument declaration Use the correct storage class volatile is duplicate volatile is described more than twice Write the type qualifier correctly REJ10J2009 0200 Rev 2 00 2010 04 01 308 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 8 ccom100 Error Messages 2 13 Error message Description and countermeasure can t read C source from filename line number for error message The source line is in error and cannot be displayed The file indicated by filename cannot be found or the line number does not exist in the
235. ma DMAC is processed as that to which volatile specification is carried out even if there is no volatile specification Example void _far dda0 pragma DMAC ddad DDAO void func void unsigned char buff 1 0 dda0 buff Figure B 73 pragma DMAC Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 146 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma INTCALL Declare a function called by the INT instruction Function Declares a function called by a software interrupt by the int instruction Syntax 1 pragmaINTCALLAINT No Aassembler function name register name registername 2 pragma INTCALLAINT No AC function name Description This extended function declares the assembler function called by a software interrupt with the INT number Rules Declaring assembler functions 1 Before a pragma INTCALL declaration be sure to include an assembler function prototype declaration If there is no prototype declaration a warning is output and the pragma INTCALL declaration is ignored 2 Observe the following in the prototype declaration 1 Make sure that the number of parameters in the prototype declaration matches those in the pragma INTCALL declaration 2 You cannot declare the following types in the parameters in the assembler function o structure types o union types 8 You cannot declare the following functions as the return va
236. ma ENDASM e pragma ASM does not have matching pragma ENDASM Write pragma ENDASM No declarator The declaration statement is incomplete Write a complete declaration statement Not enough memory The memory area is insufficient Increase the memory or the swap area not have long char Type specifiers long and char are simultaneously used Write the type specifier correctly not have long float Type specifiers long and float are simultaneously used Write the type specifier correctly not have long short Type specifiers long and short are simultaneously used Write the type specifier correctly not static initializer for variable name ee eu eU eb ey The initialize expression of static variable contains an error This is because the initialize expression is a function call for example Write the initialize expression correctly not struct or union type ey y The left side expression of gt is not the structure or union type Use the structure or union type to describe the left side expression of gt redeclare of variable name An variable name has been declared twice Change the name for either of the two variable name redeclare of enumerator An enumerator has been declared twice Change the name for either of the two enumerators redefine function function name Ueu eye The function indicated by function name is defined twice The function
237. mand input example REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 332 R32C 100 Series C Compiler Package V 1 02 G Using gensni or the Stack information File Creation Tool for Call Walker G 3 2 Option References The startup options of gensni are listed in Table G 1 Table G 1 gensni Command option Option short form function o file name None Specify a stack information file name If this option is not specified the stack information file is named after the input file by changing its file extension to sni If an extension is specified sni file name the specified extension is changed to sni If no extensions are specified the extension sni is assumed V None Shows the startup message of gensni and terminates processing without performing anything No stack information files are generated O Stack information file Specify a stack information file name Function If this option is not specified the stack information file is named after the input file by changing its file extension to sni Ifan extension is specified the stack information file name the specified extension is changed to sni If no extensions are specified the extension sni is assumed Description Use of this option permits you to change the stack information file name as necessary The extension can also be changed V Terminate processing after showing the startup messag
238. mbers Table D 6 is an example of mapping Example union TAG char C short i 4 bytes size of lo long lo s address higher Figure D 6 Example of Placement of Union D 1 8 Bitfield Types Bitfield types are mapped from the least significant bit Figure D 7 is an example of mapping Example f bit7 bitO struct BTAG J char b0 1 char b1 1 char b2 1 char b3 1 char b4 1 char b5 1 char b6 1 char b7 1 s Figure D 7 Example of Placement of Bitfield 1 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 190 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules If a bitfield member is of a different data type it is mapped to the next address Thus members of the same data type are mapped contiguously from the lowest address to which that data type is mapped Example bit7 bito struct BTAG 1 byte char bO 1 char b1 1 2 bytes char b2 1 char b3 1 char b4 1 short b56 2 address higher char b7 1 s Figure D 8 Example of Placement of Bitfield 2 o Note a Ifno sign is specified the default bitfield member type is unsigned b Can not specifies bit fields of long long type D 2 Sign Extension Rules Under the ANSI and other standard C language specifications char type data is sign extended to int type data for calculations etc This specification prevents the maximum value for char types being exceeded
239. me e The same function is defined twice in pragma name function name redefined gt Make sure that pragma name is declared only once pragma PRAGMA name function e The arguments used for the function specified with the argument is long long or double pragma program name function name are the long long type or the double type The long long type and double type cannot be used in the functions specified with the pragma program name function name Use other types pragma PRAGMA name amp function prototype mismatched The function specified by pragma PRAGMA name does not match the contents of argument in prototype declaration Make sure it is matched to the argument in prototype declaration pragma PRAGMA name s argument is struct or union function The struct or union type is specified in the prototype declaration for the function specified by pragma PRAGMA name Specify the int or short type 2 byte pointer type or enumeration type in the prototype declaration pragma PRAGMA name must be declared before use A function specified in the pragma PRAGMAname declaration is defined after call for that function Declare a function before calling it pragma BITADDRESS variable is not _Bool type The variable spcified by pragma BITADDRESS is not _Bool type Use the _Bool type to declare the variable pragma INTCALL function s argument on stack When the body of functions declared in prag
240. mitted v int i int far j Figure B 14 Example of Variable Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 106 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 2 asm Function NC100 allows you to include assembly language routines asm functions in your C source programs B 2 1 Overview of asm Function The asm function is used for including assembly language code in a C source program As shown in Figure B 15 the format of the asm function is asm where an assembly language instruction that conforms to the AS100 language specifications is included between the double quote marks pragma ADDRESS ta0_int 6CH char taO_ int void func void abbreviated ta0_ int 0x07 lt Permits timer AO interrupt asm FSET Set interrupt enable flag Figure B 15 Example of Description of asm Function 1 Compiler optimization based on the positional relationship of the statements can be partially suppressed using the code shown in Figure B 16 asm Figure B 16 Example of Coding asm Function 2 The asm function used in NC100 not only allows you to include assembly language code but also has the following extended functions J Specifying the FB offset of storage class auto variables in the C program using the names of the variables in C J Specifying the register name of storage class register variables in the C program using the names of
241. mple of registering the UART1 send interrupt processing function uarttrn word dummy_int uart0 receive for user glo _uarttrn Process 1 above word _uarttrn uart 1 trance for user Process 2 above remainder omitted Figure 2 34 Example Setting of Interrupt Vector Addresses REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 49 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique Chapter 3 Programming Technique This chapter describes precautions to be observed when programming with the C compiler NC100 3 1 Notes Renesas Electronics Corp are not designed or manufactured for use in a device or system that is used under circumstances in which human life is potentially at stake Please contact Renesas Electronics Corp Renesas Solutions Corp or an authorized Renesas Semiconductor product distributor when considering the use of a product contained herein for any specific purposes such as apparatus orsystems for transportation vehicular medical aerospace nuclear or undersea repeater use 3 1 1 Notes about Version up of compiler The machine language instructions assembly language generated by NC100 vary in contents depending on the startup options specified when compiling contents of version up etc Therefore when you have changed the startup options or upgraded the compiler version be sure to reevaluate the operation of your application program Furthermore when the same RAM data is refer
242. mple startup program s include file sect100 inc used pragma SB16DATA Extended Functions See the B 7 pragma Extended Functions and 2 2 1 f pragma SB16DATA for the pragma SB16DATA Extended Functions REJ10J2009 0200 Rev 2 00 2010 04 01 39 2tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 00000000H eae YHS 00000400H data NEAR section bss NEAR section dat MON1 section bss MON1 section dat MON2 section bss MON2 section Pale WyYy ewu 00008000H data SB8 section vo bss _SB8 section ay data SB16 section go bss_SB16 section Bo dat FAR section m bss_FAR section 3 D 3 D pl CE 00800000H data EXT section Oo bss EXT section 7 _ OFF000000H data _EXT_INIT section rom_EXT section program _EXT section rom_FAR section data_NEAR_INIT section data_MON1_ INIT section data_MON2_ INIT section data_MONS3_INIT section OFFE00000H OFFFF0000H OFFFF8000H rom NEAR section INTB Ooo OFFFFFBDCH vector section OFFFFFDCH OFFFFFFFH section fvector a Pale OY ewelu ee rove Aowew jewelxy Figure 2 26 Example Section Mapping 2 REJ10J2009 0200 Rev 2 00 2010 04 01 40 2tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler b Outline of memory mapping setup file 1 About sect100 inc This program is included from ncrt0 a30 It performs the following process
243. n Syntax pragma INTERRUPTA B E F R V interrupt handling function name pragma INTERRUPTA B E F R interrupt vector number interrupt handling function name pragma INTERRUPTA B E F R interrupt handling function name vect interrupt vector number Example pragma INTERRUPT int_func pragma INTERRUPT B int_func pragma INTERRUPT 10 int_func pragma INTERRUPT E 10 int_func pragma INTERRUPT int_func vect 10 pragma INTERRUPT R int_func pragma PARAMETER Declares that when calling an assembler function the parameters are passed via specified registers Syntax pragma PARAMETERA function name register name Example pragma PARAMETER asm_func RO R1 Always be sure to declare the prototype of the function before entering this declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 132 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference c The Other Extensions Table B 8 The Other Extensions Extended feature Description pragma _ ASMMACRO Declares defined a function by assembler macro Syntax pragma __ASMMACROA function name register name Example pragma __ASMMACRO max_w RO R2 pragma ASM Specifies an area in which stat
244. n Figure C 12 goto identifier continue break return expression opt opt indicates optional items Figure C 12 Format of Jump Statement REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 176 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications g Assembly Language Statement Use the format of assembly language shown in Figure C 13 asm Literals literals assembly language statement Figure C 13 Format of Assembly Language Statement REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 177 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications C 3 Preprocess Commands Preprocess commands start with the pound sign and are processed by the cpp100 preprocessor This chapter provides the specifications of the preprocess commands C 3 1 List of Preprocess Commands Available Table C 12 lists the preprocess commands available in NC100 Table C 12 List of Preprocess Commands Command Function Hassert Outputs a warning when a constant expression is false define Defines macros elif Performs conditional compilation else Performs conditional compilation endif Performs conditional compilation error Outputs messages to the standard output device and terminates processing Hif Performs conditional compilation ifdef Performs conditional compilation ifndef Performs conditional compilation
245. n Statement Iteration Statement Jump Statement Assembly Language Statement a Labelled Statement Use the format of labelled statement shown in Figure C 7 Identifier statement case constant statement default statement Figure C 7 Format of Labelled Statement b Compound Statement Use the format of compound statement shown in Figure C 8 list of declarationsoptlist of statementsopt opt opt indicates optional items Figure C 8 Format of Compound Statement REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 175 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications c Expression Null Statement Use the format of expression and null statement shown in Figure C 9 expression expression null statement Figure C 9 Format of Expression and Null Statement d Selection Statement Use the format of selection statement shown in Figure C 10 if expression statement if expression statement else statement switch expression statement Figure C 10 Format of Selection Statement e iteration Statement Use the format of iteration statement shown in Figure C 11 while expression statement do statement while expression for expression op jexpression opt jexpression opt statement opt indicates optional items Figure C 11 Format of Iteration Statement f Jump statement Use the format of jump statement shown i
246. n func is inline expanded within the respective places of the function main in which void main void it is called int S s func s func inline int func void return i Furthermore if an inline expansion file is specified the global functions extending across a file boundary can be inline expanded An example of a program fragment where a function is inline expanded for the option Ofile_inline inline expansion file specified in it In the example below inline expansion is performed on the source file main c for the option Ofile_inline sub c specified in it REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 74 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Ofile_inline inline expansion file OFI inline expansion file Inline expansion file main c The function func defined in sub c is inline expanded within the respective extern int func void places of the function main defined in extern int i main c where func is called void main void int S s func s func file sub c extern int func void extern int Notes 1 Declaration of an inline function and the definition of the entity of the inline function must be written in one and the same file 2 No structures or unions can be used for the arguments to an inline function If this precaution is neglected a compile error ma
247. n have their contents modified in assembler functions without a problem d Passing Parameters to an Assembler Function Use the pragma PARAMETER function if you need to pass parameters to a function written in assembly language The parameters are passed via registers Figure3 19 shows the format asm_func in the figure is the name of an assembler function short asm_func short short Prototype declaration of assembler function pragma PARAMETER asm_func RO R1 Figure3 19 Prototype declaration of assembler function pragma PARAMETER passes arguments to assembler functions via 32 bit general purpose registers R2RO R3R1 R6R4 and R7R5 16 bit general purpose registers RO R1 R2 R3 R4 R5 R6 and R7 Sbit general purpose registers ROL ROH R1L R1H R2L R2H R3L and R3H and address registers A0 A1 A2 and A3 In addition the 32 bit general purpose registers are combined to form 64 bit registers R3R1R2R0 R7R5R6R4 A1A0 and A3AQ2 for the parameters to be passed to the Note that an assembler function s prototype must always be declared before the pragma PARAMETER declaration However you cannot declare the struct and union types in a pragma PARAMETER declaration Also cannot declare the functions returning structure or union types as the function s return values REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 61 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique 3 4 Other 3 4 1 Pr
248. n the product s quality grade as indicated below You must check the quality grade of each Renesas Electronics product before using it in a particular application You may not use any Renesas Electronics product for any application categorized as Specific without the prior written consent of Renesas Electronics Further you may not use any Renesas Electronics product for any application for which it is not intended without the prior written consent of Renesas Electronics Renesas Electronics shall not be in any way liable for any damages or losses incurred by you or third parties arising from the use of any Renesas Electronics product for an application categorized as Specific or for which the product is not intended where you have failed to obtain the prior written consent of Renesas Electronics The quality grade of each Renesas Electronics product is Standard unless otherwise expressly specified in a Renesas Electronics data sheets or data books etc Standard Computers office equipment communications equipment test and measurement equipment audio and visual equipment home electronic appliances machine tools personal electronic equipment and industrial robots High Quality Transportation equipment automobiles trains ships etc traffic control systems anti disaster systems anti crime systems safety equipment and medical equipment not specifically designed for life support Specific Aircraft ae
249. nc _ func enter 04H _line 7 H C_SRC asm mov w 1 e_val HHH ASM START mov w 1 _e val 16 Reference to external variable _line 8 H C_SRC asm mov w 2 f_val mov w 2 4 FB Reference to auto variable line 9 H C_SRC asm mov w 3 r_val mov w 3 R2RO lt Reference to register variable line 10 H C_ SRC asm mov w 4 s_val mov w 4 S0_s val 16 lt Reference to static variable _line 11 H C_SRC asm mov w f_val r_vall mov w 4 FB R2RO HHH ASM END Figure B 32 Example for Referencing Variables of Each Storage Class 1 Tt does not restrict being assigned to a register even if it specifies a register qualified 2 Whether it is arranged at which storage class should actually compile and please check it REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 115 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 2 6 Selectively suppressing optimization Tn Figure B 38 the dummy asm function is used to selectively suppress a part of optimization pragma ADDRESS port 02H struct port char bitO 1 char bit1 1 char bit2 1 char bit3 1 char bit4 1 char bit5 1 char bit6 1 char bit7 1 port Optimization results in any steps to set void func void the two port bits separately being combined as one step ee Opimizaton P DHE pot port bit0 0x01 Optimization gt bset OOH _port port bit1 0x01 bse
250. ne Oinline_line OIL This option changes the size number of lines of the function to be inline expanded Oglb_jmp OGI Global jump is optimized Oglobal_to_inline OGTI Handles global functions as inline declared Oloop_unrolll oop coun OLU Unrolls code as many times as the loop count without revolving the loop statement The loop count can be omitted When omitted this option is applied to a loop count of up to 5 Ono_bit ONB Suppresses optimization based on grouping of bit manipulations Ono_break_source_debug ONBSD Suppresses optimization that affects source line data Ono_float_const_fold ONFCF Suppresses the constant folding processing of floating point numbers Ono_logical_or_combine ONLOC Suppresses the optimization that puts consecutive OR together Ono_asmopt ONA Inhibits starting the assembler optimizer aopt100 Osp_adjust OSA Optimizes removal of stack correction code This allows the necessary ROM capacity to be reduced However this may result in an increased amount of stack being used Ostatic_to_inline OSTI Astatic function is treated as an inline function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 17 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler f Generated Code Modification Options Tabel 2 6 shows the command line options for controlling nc100 generated assembly code Tabel 2 6 Generated Code Mo
251. ng Format include lt stdlib h gt int mblen s n Method function Argument const char _far s Pointer to a multibyte character string size_t n Number of searched byte ReturnValue Returns the number of bytes in the character string if s configures a correct multibyte character string Returns 1 if s does not configure a correct multibyte character string Description Returns 0 if s indicates a NULL character mbstowcs Multi byte Character Multitoyte Character String Manipulate Functions Function Converts a multibyte character string into a wide character string Format include lt stdlib h gt size_t mbstowcs wcs s n Method function Argument wchar_t _far wcs Pointer to an area for storing conversion wide character string const char _far s Pointer to a multibyte character string size_t n Number of wide characters stored ReturnValue Returns the number of characters in the converted multibyte character string Returns 1 if s does not configure a correct multibyte character string REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 254 R32C 100 Series C Compiler Package V 1 02 E Standard Library mbtowc Multi byte Character Multitoyte Character String Manipulate Functions Function Converts a multibyte character into a wide character Format include lt stdlib h gt int mbtowc wcs s n Method function Argument wchar_t _far wcs Pointer to an area for storing co
252. ns with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 281 R32C 100 Series C Compiler Package V 1 02 E Standard Library strtok Function Divides some character string from a character string into tokens Format include lt string h gt char _far strtok sl s2 Method function Argument char _far s1 Pointer to the character string to be divided up const char _far s2 Pointer to the punctuation character to be divided with ReturnValue Returns the pointer to the divided token when character is found Returns NULL when character cannot be found Description Jn the first call returns a pointer to the first character of the first token A NULL character is written after the returned character In subsequent calls when s1 is NULL this instruction returns each token as it is encountered NULL is returned when there are no more tokens in s1 When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization strtol Character String Value Convert Function Function Converts a character string into a long type integer Format include lt string h gt long strtol s endptr base Method function Argument const char _far s Pointer to the converted character string char _far _far endptr Pointer to the remaining character strings that have not been converted int base Base of values to be read in 0 to 36
253. nversion wide character string const char _far s Pointer to a multibyte character string size_t n Number of wide characters stored ReturnValue Returns the number of wide characters converted if s configure a correct multibyte character string Returns 1 if s does not configure a correct multibyte character string Returns 0 if s indicates a NULL character memchr Memory Handling Functions Function Searches a character from a memory area Format include lt string h gt void _far memchr s c n Method function Argument const void _far s Pointer to the memory area to be searched from int c Character to be searched size tn Size of the memory area to be searched ReturnValue Returns the position pointer of the specified character c where it is found Returns NULL if the character c could not be found in the memory area Description Searches for the characters shown in c in the amount of memory specified in n starting at the address specified in s When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 255 R32C 100 Series C Compiler Package V 1 02 E Standard Library memcmp Memory Handling Functions Function Compares memory areas n bytes Format include lt string h gt int memcmp s1 s2 n Method function Argument const void _far s
254. o a return value 10 is returned to the host OS Wstop_at_warning WSAW Stops compiling the source files if a warning occurs during compiling and returns the compiler end code 10 Wundefined_macro WUM Warns you that undefined macros are used in if Wuninitialize_variable WUV Outputs a warning about auto variables that have not been initialized Wunknown_pragma WUP Outputs warning messages for non supported pragma Wmultiple_tentative_definitions WMTD Outputs a warning when there are multiple tentative definitions for one and the same variable name Wignore_near_pointer WINP Inhibits a warning when the near pointer is handled as a far pointer REJ10J2009 0200 Rev 2 00 2010 04 01 19 2tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler i Assemble and Link Options Tabel 2 9 shows the command line options for specifying as100 and In100 options Tabel 2 9 Assemble and Link Options Option Function asl0O0A lt Option gt Specifies options for the as100 link command If you specify two or more options enclose them in double quotes In100A lt Option gt Specifies options for the In100 assemble command If you specify two or more options enclose them in double quotes REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 20 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 2 2 Preparing the Startup Program For C lang
255. o or reading a register in the SFR area it may sometimes be necessary to use a specific instruction Because this specific instruction varies with each type of MCU consult the user s manual of your MCU for details In this case write the instruction directly in the program using the ASM function In this compiler the instructions which cannot be used may be generated for writing and read out to the register of SFR area When accessing registers in the SFR area in C language make sure that the same correct instructions are generated as done by using asm functions regardless of the compiler s version and of whether optimizing options are used or not When you describe like the following examples as C language description to a SFR area in this compiler may generate the assembler code which carries out operation which is not assumed since the interrupt request bit is not normal pragma ADDRESS TAOIC 006Ch R32C 100 Timer AO interrupt control register struct char ILVL 3 char IR 1 An interrupt request bit char dmy 4 TAOIC void wait_until_IR_is ON void while TAOIC IR 0 Waits for TAOIC IR to become 1 TAOIC IR 0 Returns 0 to TAOIC IR when it becomes 1 Figure 1 2 C language description to SFR area REJ10J2009 0200 Rev 2 00 2010 04 01 2tENESAS 4 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 1 4 Example Program Development Figure 1 3 shows the flow for the
256. o the generated assembly source file Writing pragma ASM be sure to use it in combination with pragma ENDASM NC100 suspends processing if no pragma ENDASM is found the corresponding pragma ASM 1 In assembly language description do not write statements which will cause the register contents to be destroyed When writing such statements be sure to use the push and pop instructions to save and restore the register contents 2 Within the pragma ASM to pragma ENDASM section do not reference arguments and auto variables 3 Within the pragma ASM to pragma ENDASM section do not write a branch statement including conditional branch which may affect the program flow void func void int i j for i O i lt 10 i func2 This area is output directly to an assembly language file MOV W 0FFH RO omitted FSET pragma ENDASM Figure B 80 Example of pragma ASM ENDASM It is this assembly language program written between pragma ASM and pragma ENDASM that is processed by the C preprocessor REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 153 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma JSRA Calls a function with JSR A Function Calls a function using the JSR A instruction Syntax pragma JSRAAfunction name Description Calls all functions declared using pragma JSRA using the JSR A instruction pragma JSRA can be specified to
257. obal variables can be referenced When including the names of C variables in an assembler function precede them with an underscore _ Also in assembler language programs external variables must be declared using the assembler pseudo instruction GLB Figure3 17 is an example of referencing the C program s global variable counter from the assembler function asm_funce REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 59 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique C program unsigned short void main void omitted Assembler function glb _asm_func omitted mov w counter _ counter _ counter RO C program global variable External declaration of C program s global variable Reference Figure3 17 Referencing a C Global Variable d Notes on Coding Interrupt Handing in Assembler Function If you are writing a program function for interrupt processing the following processing must be performed at the entry and exit 1 2 3 Save the registers R2RO R3R1 R6R4 R7R5 AO Al A2 and A3 at the entry point Restore the registers R2RO R3R1 R6R4 R7R5 AO A1 A2 and A3 at the exit point Use the REIT instruction to return from the function Figure3 18 is an example of coding an assembler function for interrupt processing section glb _int_func pushm mov b omitted popm reit END program _ func R2R0 R3R1 R6R4 R7R5 A0
258. of pragma ROM Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 137 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference pragma SB16DATA SB Relative Addressing Using of 16bit displacement Variable Description Function Function Declares that the data uses SB relative addressing of 16bit displacement Syntax pragma SBDATAAvaluable name Description The R32C 100 series allows you to choose instructions that can be executed efficiently by using SB relative addressing Section accessed by SB relative addressing When it has arranged to the far area pragma SBI6DATA declares that SB relative addressing of 16bit displacement can be used for the variable when referencing data This facility helps to generate ROM efficient code Rules 1 Section accessed by SB relative addressing when using pragma SBI6DATA It is necessary to arrange to a far domain Therefore it is necessary to change specification of the section arrangement by the start up file For details of how to modify the startup file see Chapter 2 2 2 Customizing the Startup Program and Chapter 2 2 3 2 2 3 Customizing for NC100 Memory Mapping in the Operation part of the NC100 User s Manual 2 As opposed to the same variable pragma SBDATA pragma SB16DATA cannot be specified simultaneously 8 If pragma SB16DATA is specified for anything other than a variable it is ignored as invalid 4 If the specified variable is a static varia
259. of the unary operator is illegal Check the right side expression of the operator Use of the unary operator is illegal Check the right side expression of the operator invalid unary operands invalid unary operands Use of the unary operator is illegal Check the right side expression of the operator The void type specifier is used with long or singed Write the type specifier correctly invalid void type eu ei el ew ed ely ed ely invalid void type int assumed The void type variable cannot be declared Processing will be continued by assuming it to be the int type Write the type specifier correctly invalid size of bitfield Get the bitfield size Not write bitfield on this decraration invalid switch statement The switch statement is illegal Write it correctly label label redefine The same label is defined twice within one function Change the name for either of the two labels long long type s bitfield Specifies bitfield by long long type Can not specifies bit fields of long long type ew eu eu ey ey The argument type is not the type declared in prototype declaration Check the argument type mismatch prototyped parameter type y REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 316 R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 16 ccom100Error message 10 13 Error message Description and countermeasure No prag
260. ointer Idc 0080H FLG switch to usp 7 Idc stack_top SP stack pointer Idc data_SB8 top SB sb register fset b Idc data_SB8_top SB bsb register fclr b Idc VECTOR_ADR INTB vector address zero clear BSS 8 BZERO bss _SB8 top bss _SB8 i BZERO bss _SB16_top bss _SB16 BZERO bss_NEAR_top bss_NEAR BZERO bss FAR_top bss FAR BZERO bss_EXT_top bss_EXT BZERO bss_MON1_top bss MON1 BZERO bss_MON2_top bss MON2 BZERO bss_MONS3_top bss MONS BZERO bss_MON4_ top bss MON4 initialize DATA 9 BCOPY data_SB8_INIT top data_SB8 top data_SB8 BCOPY data_SB16_INIT_top data_SB16_top data_SB16 BCOPY data_NEAR_INIT_top data_NEAR_top data NEAR BCOPY data_FAR_INIT top data_FAR_top data_FAR BCOPY data_EXT_INIT_top data_EXT_top data_EXT BCOPY data_MON1_INIT_top data_MON1_top data_MON1 BCOPY data_MON2_INIT_top data_MON2_top data_MON2 BCOPY data_MON3_INIT_top data_MON3_top data_MON3 BCOPY data_MON4_INIT_top data_MON4_top data_MON4 6 After a reset execution starts from this label start 7 Sets IPL and each flags 8 Clears the bss section to zeros 9 Moves the initial values of the data section to RAM Figure 2 9 Startup Program List 4 ncrt0 a30 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 24 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler initialize heap manager 10 it HEAP 11 glb __mnext glb __msize mov heap_top _mnext mov
261. on switch_table section bss FAR section data FAR section bss MONI section data_MON1 section bss MON2 section data_MON2 section bss_MONS section data MON3 section bss MON4 section data MON4 section vector section rom FAR section 3 Sections mapped only to OFFFFFFDCH OFFFFFFFFH REJ10J2009 0200 Rev 2 00 2010 04 01 42 fvector section 2tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 4 Sections mapped to any area for the R32C 100 series stack section heap section data_EXT section bss EXT section rom_EXT section program EXT section data NEAR INIT section data _FAR INIT section data EXT_INIT section data SB8_INIT section data _SB16_INIT section data MONL1_INIT section data MON2_INIT section data MON3_INIT section data MON4 INIT section If any of the following data sections have a size of 0 they need not be defined program_EXT section switch_table section data NEAR section data NEAR INIT section data _FAR section data _FAR_INIT section data_EXT section data_EXT_INIT section data_MON1 section data_MON1_INIT section data_MON2 section data MON2_INIT section data_MONS section data _MON3_INIT section data_MON4 section data_MON4_ INIT section data_SB8 section data _SB8_ INIT section data_SB16 section data_SB16_INIT sect
262. on handled as an inline function is always generated 2 Ifa function needs to be forcibly handled as an inline function be sure to declare it as inline 3 Inline expansion is performed on only the functions whose expanded size from f to is within 150 lines including a comment line Note that the size can be changed using the option Oinline_line REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 77 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Oloop_unroll oop count OLU oop count Unrolls a loop Function Unrolls code as many times as the loop count without revolving the loop statement The loop count can be omitted When omitted this option is applied to a loop count of up to 5 Supplement Unrolled code is output for only the for statements where the number of times they are executed is known Specify the upper limit count for which times for is revolved in the target for statement to be unrolled By default this option is applied to the for statements where for is revolved up to five times Notes The ROM size increases for reasons that the for statement is revolved Ono_bit ONB Suppression of optimization Function Suppresses optimization based on grouping of bit manipulations Supplement When you specify O or OR or OS optimization is based on grouping manipulations that assign constants to a bit field mapped to the same memory area into one routine Because it is not
263. on is interpreted as the following externint far idata 10 void func void remainder omitted Figure B 12 Integrated Function of Variable Declaration As shown in this example if there are many declarations the type can be declared by specifying near or far in one of those declarations However an error occurs if there is any contention between near and far specifications in two or more of those declarations You can ensure consistency among source files by declaring near or far using a common header file common header file common h l externint far data C source file C source file a c b c include common h include common h l l void main void i l i int data 10 data 1 l l l a a a a a a se a a a ye ee a Figure B 13 Example of Common header file Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 105 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 1 9 Notes on near and far Attributes a Notes on near and far Attributes of Functions Functions always assume the far attribute Do not declare functions with near NC100 will output a warning when you declare the near attribute for a function b Notes on near and far Modifier Syntax Syntactically the near and far modifiers are identical to the const modifier The following code therefore results in an error int i far j This is not per
264. on of character strings Gf the option OS is concurrently specified However a normal code may be unable to be outputted when fulfilling the following conditions With a different variable points out the same memory position simultaneously within a single function and they point to an identical address When these variables are used in one and the same function Exsample int a 3 int p amp a void test void int b p 9 a 10 b p By applying optimization p will be transposed to 9 printf b d expect b 10 n b result b 9 expect 10 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 71 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference 1 a Optimization Notes When the O5 optimizing options is used the compiler generates in some cases BTSTC or BTSTS bit manipulation instructions In R32C 100 the BTSTC and BTSTS bit manipulation instructions are prohibited from rewriting the contents of the interrupt control registers However the compiler does not recognize the type of any register so should BTSTC or BTSTS instructions be generated for interrupt control registers the assembled program will be different from the one you intend to develop When the O5 optimizing options is used in the program shown below a BTSTC instruction is generated at compilation which prevents an interrupt request bit from being processed correct
265. on scope Write pragma JSRA outside a function pragma PARAMETER function s address used ew eu eU ey ey The address of the function specified by pragma PARAMETER is referenced Do not reference that address pragma control for function duplicate ignored ey Two or more of INTERRUPT TASK HANDLER CYCHANDLER or ALMHANDLER are specified for the same function in pragma Be sure to specify only one of INTERRUPT TASK HA NDLER CYCHANDLER orALMHANDLER pragma unknown switch ignored Invalid switch is specified to pragma pragma declaration is ignored Write switch correctly auto is illegal storage class An incorrect storage class is used Specify the correct storage class register is illegal storage class An incorrect storage class is used Specify the correct storage class argument is define by typedef typedef ignored eu ey ey Specifier typedef is used in argument declaration Specifier typedef will be ignored Delete typedef assign far pointer to near pointer bank value ignored ey The bank address will be nullified when substituting the far pointer for the near pointer Check the data types near or far assignment from const pointer to non const pointer The const property is lost by assignment from const pointer to non const pointer Check the statement description If the description is correct ignore this warning REJ10J200
266. onal These items are specified on the command line Figure 2 1 shows the command line format Figure 2 2 is an example In the example the following is performed 1 Startup program ncrt0 a30 is assembled 2 C source program sample c is compiled and assembled 3 Relocatable object files ncrt0 r30 and sample r30 are linked The absolute module file sample x30 is also created The following command line options are used option o option as100 1 option In100 ms o Specifies machine language data file sample x30 2222222222 Specifies output of list file extension lst at assembling Specifies output of map file extension map at linking nc100A command line option A assembly language source file name A relocatable object file name A lt C source file name gt Prompt lt gt Mandatory item Optional item A Space Figure 2 1 nc100 Command Line Format REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 13 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler nci00 osample as100 I n100 ms ncrt0 a30 sample c lt RET gt lt RET gt Return key Always specify the startup program first when linking Figure 2 2 Example nc100 Command Line 2 1 2 Command File The compile driver can compile a file which has multiple command options written in it i e a command file after loading it into the machine Use of a command file helps to overcome
267. onversion to e type when the exponent is 4 or less or the precision is less than the value of the exponent e G Same as g except that E is used in place of e for the exponent Left aligns the result of conversion in the minimum field width The default is right alignment J Adds or to the result of signed conversion By default only the is added to negative numbers e Blank By default a blank is added before the value if the result of signed conversion has no sign eo Adds 0 to the beginning of o conversion Adds Ox or 0X to the beginning when other than 0 in x or X conversion Always adds the decimal point in e E and f conversion Always adds the decimal point in g and G conversion and also outputs any Os in the decimal place 2 Minimum field width o Specifies the minimum field width of positive decimal integers e When the result of conversion has fewer characters than the specified field width the left of the field is padded The default padding character is the blank However 0 is the padding character if you specified the field with using an integer preceded by 0 o If you specified the flag the result of conversion is left aligned and padding characters always blanks inserted to the right Ifyou specified the asterisk for the minimum field width the integer in the parameter specifies the field width If the value of the parameter is negative the value after the flag is the
268. ool type can not specify to sign e If a char type is specified with no sign it is processed as an unsigned char type J If an int or short type is specified with no sign it is processed as a signed int or signed short type J If a long type is specified with no sign it is processed as a sign long type J Tf a long long type is specified with no sign it is processed as a sign long long type REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 186 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules J Tf the bit field members of a structure are specified with no sign they are processed as unsigned Can not specifies bit fields of long long type Type int is handled in 32 bits However if the compile option f116 fint_16 is specified int is handled in 16 bits D 1 2 Floating Type Table D 2 shows the number of bytes used by floating type data Table D 2 Data Size of Floating Type Type Existence of sign Bit Size Range of values float Yes 32 between 1 17549435e 38F and 3 40282347e 38F double Yes 64 between 2 2250738585072014e 308 and long double 1 7976931348623157e 308 NC100 s floating point format conforms to the format of IEEE Institute of Electrical and Electronics Engineers standards The following shows the single precision and double precision floating point formats 1 Single precision floating point data format Figure D 1 shows the format for binary floating point
269. opies a character string n characters Format include lt string h gt char _far strncpy s1 s2 n Method function Argument char _far s1 Pointer to the character string to be copied to const char _far s2 Pointer to the character string to be copied from size tn Number of characters to be copied ReturnValue Returns a pointer to the character string at the destination of copy Description Copies n characters from character string s2 to area s1 If character string s2 contains more characters than specified in n they are not copied and 0 is not appended Conversely if s2 contains fewer characters than specified in n 0 s are appended to the end of the copied character string to make up the number specified in n When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 278 R32C 100 Series C Compiler Package V 1 02 E Standard Library strnicmp String Handling Functions Function Compares character strings n characters All alphabets are handled as uppercase letters Format include lt string h gt int strnicmp sl s2 n Method function Argument char _far s1 Pointer to the first character string to be compared char _far s2 Pointer to the second character string to be compared size tn Number of characters to be compared ReturnValue ReturnValue 0
270. option ignore this condition 2 When address acquisition is omitted in the program to the static function 3 When the recursive call of the static function has not been carried out The following shows an example of how a static function to be inline expanded will be written extern int i Function func is a function a ee ne eS rg ae ee inline deployment is carried out in each place 1 static intfunc void currently called within main return i void main void int S s func s func 1 The assembler code to description of substance of the static function which became inline function treatment is always generated However it is not generated when using it together with the option Ofile_ inline 2 About a function it is compulsorily In treating as an inline function it is in a function Please make an inline declaration 8 Inline expansion is performed on only the functions whose expanded size from f to Yis within 150 lines including a comment line Note that the size can be changed using the option Oinline_line 7tENESAS R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 6 Generated Code Modification Options Table A 7 shows the command line options for controlling nc100 generated assembly code Table A 7 Generated Code Modification Options Option Short for
271. or aging degradation or any other appropriate measures Because the evaluation of microcomputer software alone is very difficult please evaluate the safety of the final products or system manufactured by you Please contact a Renesas Electronics sales office for details as to environmental matters such as the environmental compatibility of each Renesas Electronics product Please use Renesas Electronics products in compliance with all applicable laws and regulations that regulate the inclusion or use of controlled substances including without limitation the EU RoHS Directive Renesas Electronics assumes no liability for damages or losses occurring as a result of your noncompliance with applicable laws and regulations This document may not be reproduced or duplicated in any form in whole or in part without prior written consent of Renesas Electronics Please contact a Renesas Electronics sales office if you have any questions regarding the information contained in this document or Renesas Electronics products or if you have any other inquiries Note 1 Renesas Electronics as used in this document means Renesas Electronics Corporation and also includes its majority owned subsidiaries Note 2 Renesas Electronics product s means any product developed or manufactured by or for Renesas Electronics R32C 100 Series C Compiler Package V 1 02 Preface Preface NC100 is the C compiler for the Renesas 32 bit microcomputer R32
272. or the assembler function Figure3 11 is an example of calling assembler function asm_func extern void asm_func void lt Assembler function prototype declaration void main omitted i asm_func Calls assembler function Figure3 11 Example of Calling Assembler Function Without Parameters samplet c glb _main _main omitted jsr l _asm_func Calls assembler function preceded by _ rts Figure3 12 Compiled result of sample c sample1 a30 b When assigning arguments to assembler functions When passing arguments to assembler functions use the extended function pragma PARAMETER This pragma PARAMETER passes arguments to assembler functions via 32 bit general purpose registers R2RO R3R1 R6R4 R7R5 16 bit general purpose registers RO R1 R2 R3 R4 R5 R6 R7 or 8 bit general purpose registers ROL ROH R1L R1H R2L R2H R3L R3H and address registers AO A1 A2 A3 For 64 bit quantities the compiler uses the 32 bit general purpose register pair R3R1R2RO and R7R5R6R4 or address register pair A1A0 and A3AQ REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 56 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique The following shows the sequence of operations for calling an assembler function using pragma PARAMETER 1 Write a prototype declaration for the assembler function before the pragma PARAMETER declaration You must also declare the parameter ty
273. ototype declaration is entered for the function new format and a description of the K amp R format old format coexist in given statement the system may fail to pass arguments to the function correctly for reasons of language specifications of the C language Therefore we recommend using a prototype declaring description format as the standard format to write the C language source files for NC100 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 192 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules Table D 5 Rules on Argument Transfer via Register NC 100 Argument Type of argument Registers used First argument _Bool char ROL register int 16 bits short RO register int 82 bits R2RO register float long pointer AO register long long double A1A0 register long double Second argument _Bool char RiL register int 16 bits short R1 register int 32 bits R3R1 register float long pointer A2 register long long double A8A2 register long double Third argument int 16 bits short R4 register int 82 bits R6R4 register float long pointer R6R4 register Fourth argument int 16 bits short R5 register int 82 bits R7R5 register float long pointer R7R5 register Table D 6 Rules on Argument Transfer via Register NC308 Argument Type of argument Registers used First argument _Bool ROL register char int RO register near point
274. ource_in_ list dSL List File option Function In addition to the dsource dS function generates an assembly language list file filename extension lst E Compile driver control Function Invokes only preprocess commands and outputs results to standard output Notes When this option is specified no assembly source file extensions a30 re locatable object files extension r30 absolute module files extension x30 or other files output by ccom100 as100 or 1n100 are generated ldirectory Compile driver control Function Specifies the directory name in which to search for files to be referenced by the preprocess command include Max specified 256 directory Syntax ncl00A IdirectoryA lt C source file gt Supplement An example of specifying two directories dirl and dir2 for the I option is shown below nc100 Idir1 Idir2 sample c lt RET gt Indicates the prompt lt RET gt Indicates the Return key Notes The number of directories that can be defined may be limited by the maximum number of characters that can be specified on the command line of the operating system of the host machine REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 65 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference P Compile driver control Function Invokes only preprocess commands creates a file extension i and stops processing Notes 1 When this option is specified no assembly source file
275. own below Format maximum field width modifier I L or Wconversion specification character Example format 5ld ReturnValue Returns the number of data entries stored in each argument Returns EOF if EOF is input from stdin as data Description Converts the characters read from stdin as specified in format and stores them in the variables shown in the arguments Argument must be a far pointer to the respective variable The first space character is ignored except in c and conversion Interprets code 0x1A as the end code and ignores any subsequent data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 267 R32C 100 Series C Compiler Package V 1 02 E Standard Library scanf Input Output Functions Description 1 Conversion specification symbol e d Converts a signed decimal The target parameter must be a pointer to an integer e i Converts signed decimal octal and hexadecimal input Octals start with 0 Hexadecimals start with Ox or OX The target parameter must be a pointer to an integer Cd u Converts an unsigned decimal The target parameter must be a pointer to an unsigned integer o o Converts a signed octal The target parameter must be a pointer to an integer o x X Converts a signed hexadecimal Uppercase or lowercase can be used for OAH to OFH The leading Ox is not included The target parameter must be a pointer to an integer s Stores character strings ending with the null
276. p compile processing when a warning occurs Wundefined_macro WUM Warning Options Function Warns you that undefined macros are used in if REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 94 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Wuninitialize_variable WUV Warning Options Function Outputs a warning for uninitialized auto variables This option is effective even when Wall is specified Supplement If an auto variable is initialized in conditional jump by for example a if or a for statement in the user application the compiler assumes it is not initialized Therefore when this option is used the compiler outputs a warning for it Wunknown_pragma WUP Warning Options Function Outputs warning messages for non supported pragma Supplement By default no alarm is generated even when an unsupported unknown pragma is used When you are using only the NC series compilers use of this option helps to find misspellings in pragma Notes When you are using only the NC series compilers Renesas recommends that this option be always used when compiling REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 95 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 9 Assemble and Link Options Table A 10 shows the command line options for specifying as100 and 1n100 options Table A 10 Assemble and Link Options Option Function asl00A lt Option gt S
277. pe s 2 Declare the name of the register used by pragma PARAMETER in the assembler function s parameter list Figure3 13 is an example of using pragma PARAMETER when calling the assembler function asm_func extern short asm_func short short pragma PARAMETER asm_func RO R1 lt Parameters are passed via the RO and R1 registers to the assembler function void main void short i 0x02 short j 0x05 asm_funci i j Figure3 13 Example of Calling Assembler Function With Parameters sample2 c SECTION program CODE ALIGN _file smp2 c align ine 5 C_SRC glb _main _main ine 6 HHHC SRC short i 0x02 mov w 0002H RO i Parameters are passed via the RO and R1 ine 7 registers to the assembler function H C_SRC short j 0x05 mov w 0005H R1 j _line 9 H C_SRC asm_funci j jsr _asm_func Calls assembler function preceded by _ ine 10 H HC_SRC rts Figure3 14 Compiled result of sample2 c sample2 a30 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 57 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique c Limits on Parameters in pragma PARAMETER Declaration The following parameter types cannot be declared in a pragma PARAMETER declaration o structure types and union type parameters Furthermore return values of structure or union types cannot be defined as the return values of assembler functions 3 3 2 Writing Assemble
278. pecifies options for the as100 link command If you specify two or more options enclose them in double quotes In100A lt Option gt Specifies options for the In100 assemble command If you specify two or more options enclose them in double quotes as100 Option Assemble link option Function Specifies as100 assemble command options If you specify two or more options enclose them in double quotes Syntax ncl00A as100A option1A option A lt C source file gt Notes Do not specify the as100 options C O PSFP T or V In100 Option Assemble link option Function Specifies options for the In100 link command You can specify a maximum of four options If you specify two or more options enclose them in double quotes Syntax nc100A In100A option1A option A lt C source file name gt Notes Do not specify the In100 options G O ORDER L T V or file REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 96 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 3 Notes on Command Line Options A 3 1 Coding Command Line Options The NC100 command line options differ according to whether they are written in uppercase or lowercase letters Some options will not work if they are specified in the wrong case A 3 2 Priority of Options for Controlling If you specify both the following options in the NC100 command line the S option takes prec
279. per bytes word of the value may be lost by assignment from large type to a smaller type Check the type If the description is correct ignore this warning mismatch prototyped parameter type The argument type is not the type declared in prototype declaration Check the argument type REJ10J2009 0200 Rev 2 00 2010 04 01 325 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 25 ccom100 Warning Messages 6 9 Warning message Description and countermeasure meaningless statements deleted in optimize phase Meaningless statements were deleted during optimization Delete meaningless statements meaningless statement The tail of a statement is You may confuse with Check on it mismatch function pointer assignment eU ey The address of a function having a register argument is substituted for a pointer to a function that does not have a register argument i e a nonprototyped function Change the declaration of a pointer variable for function to a prototype declaration multi character character constant A character constant consisting of two characters or more is used Use a wide character L xx when two or more characters are required near far is conflict beyond over typedef The type defined by specifying near far is again defined by specifying near far when referencing it Write the type specifier correctly No hex d
280. plement Tf one of three options O8 or greater OR or OS is specified when compiling as in the example shown below the compiler optimizes code generation by combining logical OR Example if a amp 0x01 a amp 0x02 a amp 0x04 Optimized if a amp 0x07 Tn this case the variable a is referenced up to three times but after optimization it is referenced only once However if the variable a has any effect on I O references etc the program may become unable to operate correctly due to optimization In such a case specify this option to suppress the optimization to combine logical OR Note however that if the variable is declared with volatile logical OR are not combined for optimization Ono_asmopt ONA Inhibits starting the assembler optimizer Function Inhibits starting the assembler optimizer aopt100 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 79 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Osp_adjust OSA Removing stack correction code after calling a function Function Optimizes code generation by combining stack correction codes after function calls Please use this option together with O 1 5 Supplement Because the area for arguments to a function normally is deal located for each function call made processing is performed to correct the stack pointer If this option is specified processing to correct the stack pointer is
281. pragma MONITORn Declares that data be located in a special section for the RAM monitor Syntax pragma MONITOR1Avariable name Example pragma MONITORI val pragma ROM Maps the specified variable to rom Syntax pragma ROMAvariable name Example pragma ROM val pragma SB16DATA Declares that the data uses SB relative addressing of 16 bit displacement Syntax pragma SB16DATAAvariable name Example pragma SB16DATA val pragma SBDATA Declares that the data uses SB relative addressing of 8bit displacement Syntax pragma SBDATAAvariable name Example pragma SBDATA val pragma SECTION Changes the section name generated by NC100 Syntax pragma SECTIONAsection nameAnew sect ion nam Example pragma SECTION bss nonval data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 130 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference Table B 5 Memory Mapping Extended Functions 2 2 Extented function Description pragma STRUCT 1 Inhibits the packing of structures with the specified tag Syntax pragma STRUCTAstructure tagAunpack Example pragma STRUCT TAG unpack 2 Arranges members of structures with the specified tag and maps even sized members first Syntax pragma STRUCTAstructure tagAarrange Example pragma STRUCT TAG arrange b Us
282. pression primary expression Postpositional expression Postpositional expression expression Postpositional expression list of parameters Postpositional expression identifier Postpositional expression gt identifier Postpositional expression Postpositional expression Postpositional expression Monadic expression monadic expression monadic expression monadic operator cast expression sizeof monadic expression sizeof type name Monadic expression Cast expression type name cast expression cast expression Expression expression expression expression expression expression expression Additional and expression expression subtraction expressions expression expression Bitwise shift expression expression lt lt expression expression gt gt expression Relational expressions expression expression lt expression expression gt expression expression lt expression expression gt expression Equivalence expression expression expression expression expression Bitwise AND expression amp expression Bitwise XOR expression expression Bitwise OR expression expression Logical AND expression amp amp expression Logical OR expression expression Conditional expression expression expression expression REJ10J2009
283. ption Reference Wmultiple_tentative_definitions WMTD Function Outputs a warning when there are multiple tentative definitions for one and the same variable name Supplement If variables are declared outside a function by not using an initializer and without a storage class specifier or with storage class static such a declaration is referred to as tentative definition If this option is specified the compiler outputs a warning when such a declaration is encountered two or more times Wnesting_comment WNC Warning Options Function Generates a warning when comments include Supplement By using this option it is possible to detect nesting of comments Wno_stop WNS Warning Options Function Prevents the compiler stopping when an error occurs Supplement The compiler compiles the program one function at a time If an error occurs when compiling the compiler by default does not compile the next function Also another error may be induced by an error giving rise to multiple errors In such a case the compiler stops compiling When this option is specified the compiler continues compiling as far as possible Notes A system error may occur due to erroneous description in the program In such a case the compiler stops compiling even when this option is specified Wno_used_argument WNUA Warning Options Function Outputs a warning for unused arguments function REJ10J2009 0200 Rev 2 00 2010 04 0
284. ption is specified the compiler inhibits a warning that near qualifiers for the pointer are ignored Wlarge_to_small WLTS Warning Options Function Outputs a warning about the substitution of variables in descending sequence of size Supplement A warning may be output for negative boundary values of any type even when they fit in the type This is because negative values are considered under language conventions to be an integer combined with the unary operator For example the value 32768 fits in the signed int type but when broken into and 32768 the value 32768 does not fit in the signed int type and consequently becomes the signed long type Therefore the immediate value 32768 is the signed long type For this reason any statement like int i 32768 gives rise to a warning Notes Because this option outputs a large amount of warnings warning output is suppressed for the type conversions listed below Assignment from char type variables to char type variables Assignment of immediate values to char type variables Assignment of immediate values to float type variables Wmake_tagfile WMT Warning Options Function Outputs error messages to the tag file of source file by source file when an error or warning occurs Supplement This option with Werror_file WEF option can t specify REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 91 R32C 100 Series C Compiler Package V 1 02 A Command O
285. r Functions a Method for writing the called assembler functions The following shows a procedure for writing the entry processing of assembler functions 1 Specify section names using the assembler pseudo command SECTION 2 Global specify function name labels using the assembler pseudo command GLB 3 Add the underscore _ to the function name to write it as label 4 When modifying the B and U flags within the function save the flag register to the stack beforehand 1 The following shows a procedure for writing the exit processing of assembler functions 1 Ifyou modified the B and U flags within the function restore the flag register from the stack 2 Write the RTS instruction Do not change the contents of the SB and FB registers in the assembler function If the contents of the SB and FB registers are changed save them to the stack at the entry to the function then restore their values from the stack at the exit of the function Figure3 15 is an example of how to code an assembler function In this example the section name is program which is the same as the section name output by NC100 section program code align 1 glb _asm_func 2 _asm_func 3 pushc FLG 4 mov l SYM1 R3R1 popc FLG 5 rts 6 END Figure3 15 Example Coding of Assembler Function 1 Do not change the contents of B and U flags in the assembler function REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 58
286. r do not output any warning or error message fnot_address_volatile fNAV Modify generated code Function Does not handle the global variables specified by pragma ADDRESS or pragma EQU or the static variables declared outside a function as those that are specified by volatile Supplement If I O variables are optimized in the same way as for variables in RAM the compiler may not operate as expected This can be avoided by specifying volatile for the I O variables Normally pragma ADDRESS or pragma EQU operates on I O variables so that even though volatile may not actually be specified the compiler processes them assuming volatile is specified This option suppresses such processing Notes You do not normally need to use this option REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 86 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference fnot_reserve_asm ANRA Modify generated code Function Removes asm from the list of reserved words Supplement _asm that has the same function is handled as a reserved word fnot_reserve_far_and_near fNRFAN Modify generated code Function Removes far and near from list of reserved words Supplement _far and _near that has the same function is handled as a reserved word fnot_reserve_inline fNRI Modify generated code Function Does not handle inline as a reserved word Supplement inline that has the same function is handled as a reserved
287. r initialized iu ely The initial value of the enumerator is incorrectly specified by writing a variable name for example Write the initial value of the enumerator correctly invalid function argument y An argument which is not included in the argument list is declared in argument definition in function definition Declare arguments which are included in the argument list invalid function s argument declaration e The argument of the function is erroneously declared gt Write it correctly invalid function declare e The function definition contains an error Check the line in error or the immediately preceding function definition invalid initializer e The initialization expression contains an error This error includes excessive parentheses many initialize expressions a static variable in the function initialized by an auto variable or a variable initialized by another variable Write the initialization expression correctly invalid initializer of variable name The initialization expression contains an error This error includes a bit field initialize expression described with variables for example Write the initialization expression correctly invalid initializer on array The initialization expression contains an error Check to see if the number of initialize expressions in the parentheses matches the number of array elements and the number of structure members invalid initiali
288. ranching in the asm function that may affect the program flow b About Register e Do not destroy registers within the asm function If registers are going to be destroyed use push and pop instructions to save and restore the registers o NC100 is premised on condition that the SB register is used in fixed mode after being initialized by the startup program If you modified the SB register write a statement to restore it at the end of consecutive asm functions as shown in Figure B 37 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 117 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference asm SB 0 asm LDC 0H SB SB changed asm MOVW RO _port SB omitted asm SB __SB_ asm LDC _SB_ SB SB retumed to original state Figure B 37 Restoring Modified Static Base SB register Do not modified the FB register by the asm functions because which use for the stack flame pointer c Notes on Labels The assembler source files generated by NC100 include internal labels in the format shown in Figure B 38 Therefore you should avoid using labels in an asm function that might result in duplicate names Labels consisting of one uppercase letter and one or more numerals A1 C9830 Labels consisting of two or more characters preceded by the underscore _ _ LABEL START Figure B 38 Label Format Prohibited in asm Function REJ10J2009 0200 Rev 2 00 2010 0
289. re that the number of initialize expressions does not exceed the size of the variables to be initialized inline function is called as normal function before change to static function The function declared in storage class inline is called as an ordinary function Always be sure to define an inline function before using it REJ10J2009 0200 Rev 2 00 2010 04 01 324 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 24 ccom100 Warning Messages 5 9 Warning message Description and countermeasure integer constant is out of range The value of the integer constant exceeded the value that can be expressed by unsigned long Use a value that can be expressed by unsigned long to describe the constant interrupt function called The function specified by pragma INTERRUPT is called Be careful not to call an interrupt handling function interrupt function can t return value The interrupt handling function specified by pragma INTERRUPT is using a return value Return values cannot be used in an interrupt function Delete the return value interrupt function has argument The interrupt handling function specified by pragma INTERRUPT is using an argument Arguments cannot be used in an interrupt function Delete the argument invalid pragma EQU The description of pragma EQU contains an error This line will be ignored Write the description correctly
290. rea reservation area reservation area reservation area Audio interface 0 software int 55 software int 56 software int 57 software int 58 software int 59 software int 60 software int 61 software int 62 software int 63 software int 64 software int 65 software int 66 software int 67 software int 68 Sound field processor software int 69 reservation area reservation area reservation area reservation area reservation area reservation area reservation area reservation area INT8 INTZ INT6 CANO trance CANO receive CANO error CAN trance CAN receive CAN1 error reservation area reservation area reservation area reservation area reservation area reservation area reservation area reservation area uart7 trance uart7 receive uart8 trance uart8 receive software int 128 software int 129 software int 130 software int 131 software int 132 software int 70 software int 71 software int 72 software int 73 software int 89 software int 90 software int 91 software int 92 software int 93 software int 94 software int 95 software int 96 software int 97 software int 98 software int 99 software int 100 software int 101 software int 102 software int 103 software int 104 software int 105 software int 120 software int 121 software int 122 software in
291. ream information Name stdin Standard input stdout Standard output stderr Standard error output error is output to stdout stdaux Standard auxiliary I O stdprn Standard printer output The stream corresponding to the NC100 standard library functions shown shaded in Figure E 3 are fixed to standard input stdin and standard output stdout The stream cannot be changed for these functions The output direction of stderr is defined as stdout in define The stream can only be changed for functions that specify pointers to the stream as parameters such as fgetc and fpute REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 294 R32C 100 Series C Compiler Package V 1 02 E Standard Library Figure E 3 fRelationship of Functions and Streams Figure E 4 shows the stream definition in stdio h REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 295 R32C 100 Series C Compiler Package V 1 02 E Standard Library i ESPEN ERERD ERAS a REEEARERDARARERS AERA ER SERS SARSERIATEA BERBERS ERD ARE standard I O header file omitted typedef struct _iobuf char _ buff Store buffer for ungetc lt 1 int _cnt Strings number in _buff 1 or 0 lt 2 int _flag Flag 3 int _mod Mode lt 4 int _furnc_in void Pointer to one byte input function lt 5 int _func_out int Pointer to one byte output function lt 6 FILE define _ OBUF_DEF omitted extern
292. recautions to be observed when using a command file A directory path can be specified for command file names An error results if the file does not exist in the specified directory path Command files for In100 whose file name extension is cm are automatically generated in order for specifying files when linking Therefore existing files with the file name extension cm if any will be overwritten Do not use files which bear the file name extension cm along with this compiler You cannot specify two or more command files simultaneously If multiple files are specified the compiler displays an error message Too many command files 2 1 3 Notes onNC100 Command Line Options a Notes on Coding nc100 Command Line Options The nc100 command line options differ according to whether they are written in uppercase or lowercase letters Some options will not work if they are specified in the wrong case b Priority of Options for Controlling Compile driver There are the following priorities in the opinion about control of compile driver E P S C High Priority low gt Therefore if the following two options are specified at the same time for example e c Finish processing after creating a relocatable file extension r30 J S Finish processing after creating an assembly language source file extension a30 the S option has priority That is to say the compile driver does not perform any further proce
293. ring compiling and returns the compiler end code 10 Wundefined_macro WUM Warns you that undefined macros are used in if Wuninitialize_variable WUV Outputs a warning about auto variables that have not been initialized Wunknown_pragma WUP Outputs warning messages for non supported pragma Wmultiple_tentative_definitions WMTD Outputs a warning when there are multiple tentative definitions for one and the same variable name REJ10J2009 0200 Rev 2 00 2010 04 01 89 7tENESAS R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Wall Warning Options Function Indicates all detectable alarms Supplement 1 The alarms indicated here do not include those that may be generated when Wlarge_to_small WLTS and Wno_used_argument WNUA and Wno_used_static_function WNUSPF are used 2 The alarms indicated here are equivalent to those of the options Wnon_prototype WNP Wunknown_pragma WUP Wnesting_comment WNC and Wuninitialize_variable WUV 8 Alarms are indicated in the following cases too When the assignment operator is used in the if statement the for statement or a comparison statement with the amp amp or operator When is written to which should be specified When function is defined in old format Notes These alarms are detected within the scope that the compiler assumes on its judgment that description is
294. rom a memory area to another memcpy Copies characters n bytes from a memory area to another memset Set a memory area by filling with characters Compare memcmp Compares memory areas n bytes memicmp Compares memory areas with alphabets handled as uppercase letters Search memchr Searches a character from a memory area time h Function Declares the functions necessary to indicate the current calendar time and defines the type REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 215 R32C 100 Series C Compiler Package V 1 02 E Standard Library E 2 Standard Function Reference Describes the features and detailed specifications of the standard function library of the compiler E 2 1 Overview of Standard Library NC100 has 119 Standard Library items Each function can be classified into one of the following 11 categories according to its function 1 String Handling Functions Functions to copy and compare character strings etc 2 Character Handling Functions Functions to judge letters and decimal characters etc and to covert uppercase to lowercase and vice versa 8 TO Functions Functions to input and output characters and character strings These include functions for formatted I O and character string manipulation 4 Memory Management Functions Functions for dynamically securing and releasing memory areas 5 Memory Manipulation Functions Functions to copy set and compare memory areas 6 Execution Control Functions
295. rospace equipment submersible repeaters nuclear reactor control systems medical equipment or systems for life support e g artificial life support devices or systems surgical implantations or healthcare intervention e g excision etc and any other applications or purposes that pose a direct threat to human life You should use the Renesas Electronics products described in this document within the range specified by Renesas Electronics especially with respect to the maximum rating operating supply voltage range movement power voltage range heat radiation characteristics installation and other product characteristics Renesas Electronics shall have no liability for malfunctions or damages arising out of the use of Renesas Electronics products beyond such specified ranges Although Renesas Electronics endeavors to improve the quality and reliability of its products semiconductor products have specific characteristics such as the occurrence of failure at a certain rate and malfunctions under certain use conditions Further Renesas Electronics products are not subject to radiation resistance design Please be sure to implement safety measures to guard them against the possibility of physical injury and injury or damage caused by fire in the event of the failure of a Renesas Electronics product such as safety design for hardware and software including but not limited to redundancy fire control and malfunction prevention appropriate treatment f
296. rporation of these circuits software and information in the design of your equipment Renesas Electronics assumes no responsibility for any losses incurred by you or third parties arising from the use of these circuits software or information When exporting the products or technology described in this document you should comply with the applicable export control laws and regulations and follow the procedures required by such laws and regulations You should not use Renesas Electronics products or the technology described in this document for any purpose relating to military applications or use by the military including but not limited to the development of weapons of mass destruction Renesas Electronics products and technology may not be used for or incorporated into any products or systems whose manufacture use or sale is prohibited under any applicable domestic or foreign laws or regulations Renesas Electronics has used reasonable care in preparing the information included in this document but Renesas Electronics does not warrant that such information is error free Renesas Electronics assumes no liability whatsoever for any damages incurred by you resulting from errors in or omissions from the information included herein Renesas Electronics products are classified according to the following three quality grades Standard High Quality and Specific The recommended applications for each Renesas Electronics product depends o
297. rrectly Unknown function option gx The option gx cannot be specified Specify the option correctly Unknown optimize option mx The option mx cannot be specified Specify the option correctly Unknown optimize option Ox The option Ox cannot be specified Specify the option correctly Unknown option x Yew ey ew eye eu ey ey The option x cannot be specified Specify the option correctly REJ10J2009 0200 Rev 2 00 2010 04 01 328 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 28 ccom100 Warning Messages 9 9 Warning message Description and countermeasure unknown pragma pragma specification Unsupported pragma is written used Check the content of pragma This warning is displayed only when the Wunknown_pragma WUP option is specified wchar_t array initialized by char string e The initialize expression of the wchar_t type is nitialized by a character string of the char type gt Make sure that the types of the initialize expression re matched zero divide in constant folding e The divisor in the divide operator or remainder alculation operator is 0 Use any value other than 0 for the divisor ey zero divide ignored The divisor in the divide operator or remainder alculation operator is 0 Use any value other than 0 for the divisor zero width for bitfield The bit field width is 0 Write a bit field equal to
298. rror The void type cannot be cast to any other type it can neither be cast from the structure or union type nor can it be cast to the structure or union type Write the expression correctly invalid asm0 s argument ey The variables that can be used in asm statements are only the auto variable and argument Use the auto variable or argument for the statement invalid bitfield declare The bit field declaration contains an error Write the declaration correctly invalid break statements Ueu eju The break statement is put where it cannot be used Make sure that it is written in switch while dowhile and for invalid case statements The switch statement contains an error Write the switch statement correctly invalid case value The case value contains an error Write an integral type or enumerated type constant invalid cast operator Ueu eye Use of the cast operator is illegal Write the expression correctly REJ10J2009 0200 Rev 2 00 2010 04 01 314 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 14 ccom100Error message 8 13 Error message Description and countermeasure invalid continue statements The continue statement is put where it cannot be used Use it in a while do while and for block invalid default statements The switch statement contains an error Write the switch statement correctly invalid enumerato
299. rror in following macro name or argument in define command FILE_ LINE_ DATE _ __TIME Correct the source file 44 token buffer over flow Token character buffer of define overflowed Reduce the number of token characters Ervor in undef Correct the source file 45 illegal undef command usage eu eU ey 46 undefid not found The following macro names to be undefined in undef were not defined FILE_ LINE_ DATE TIME Check the macro name 52 illegal ifdef ifndef command usage Error in ifdef Correct the source file elseif or else were used without if ifdef ifndef Use elseif or else only after if ifdef ifndef 53 elseif else sequence erro 54 endif not exist No endif to match if ifdef ifndef Add endif to the source file endif was used without if ifdef ifndef Use endif only after if ifdef ifndef 55 endif sequence error Error in line Correct the source file 61 illegal line command usage Yew ey ew eU ey REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 306 R32C 100 Series C Compiler Package V 1 02 F Error Messages F 4 cpp100 Warning Messages Table E6 shows the warning messages output by cpp100 and their countermeasures Table F 6 cpp100 Warning Messages No Warning Messages Description and countermeasure 81 reserved id used e You attempted to define or undefine one of the following ma
300. rrupt control register struct char ILVL 3 char IR 1 An interrupt request bit char dmy 4 TAOIC void wait_until_IR_is ON void while TAOIC IR 0 Waits for TAOIC IR to become 1 TAOIC IR 0 Returns 0 to TAOIC IR when it becomes 1 Figure3 1 C language description to SFR area 3 1 3 About Optimization a Regular optimization The following are always optimized regardless of whether optimization options are specified or not 1 Meaningless variable access For example the variable port shown below does not use the readout results so that readout operations are deleted extern int port void func void port Figure3 2 Example of a Meaningless Variable Access Optimized Although the intended operation in this example is only to read out port the readout code actually is not optimized before being output To suppress optimization add the volatile qualifier as shown in Figure3 2 extern int volatile port void func void port Figure3 3 Example of a Meaningless Variable Access Optimization Suppressed REJ10J2009 0200 Rev 2 00 2010 04 01 51 2tENESAS R32C 100 Series C Compiler Package V 1 02 3 Programming Technique 2 Meaningless comparison int func char c int i if c 1 i 1 else i 0 return i Figure3 4 eaningless Comparison In the case of this example because the variable c is written as char the compiler treats
301. rt 14 calls func_no_proto REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 198 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules D 3 4 Interface between Functions Figure D 17 and Figure D 18 show the process for building and freeing the stack frame in the program shown in Figure D 14 Shown in Figure D 15 and Figure D 16 are the assembly language programs derived by compiling the program in Figure D 14 iint func int int int void main void int ans inti 0x1111 lt Argument to func int j 0x2222 Argument to func int k 0x3333 lt Argument to func ans func i j k int func int x int y int z int sum int s 0x4444 int t 0x5555 int u Ox6666 suM S t U X Y Z return sum Return value to main Figure D 14 Example of C Language Sample Program REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 199 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules HH H FUNCTION main HH H FRAME AUTO ans size 4 offset 4 HHH ARG Size 4 Auto Size 4 Context Size 8 SECTION program CODE ALIGN _file interface c align ine 4 C_ SRC glb _main _main lt 1 enter 04H 2 ine 6 C_SRC inti 0x1111 mov l 00001111H R2RO _line 7 H HC_SRC int j 0x2222 mov 00002222H R3R1 ine 8 HHHC SRC int k 0x3333 mov 00003333H R6R4 _line 9 C_ SRC ans func i j K jsr func lt 3 mo
302. rt0 a30 first when linking Figure 1 5 Example nc100 Command Line REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 6 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 1 5 NC100 Output Files This chapter introduces the preprocess result C source program output when the sample program sample c is compiled using NC100 and the assembly language source program 1 5 1 Introduction to Output Files With the specified command line options the nc100 compile driver outputs the files shown in Figure 1 6 Below we show the contents of the files output when the C source file smp c shown in Figure 1 7 is compiled assembled and linked See the AS100 User Manual for the relocatable object files extension r30 print files extension Ist and map files extension map output by as100 and In100 C language source file igen100 cop100 C source file from preprocesser ccom100 Assembly language source file Relocatable object file Ec output file of nc100 Absolute module file Figure 1 6 Relationship of nc100 Command Line Options and Output Files REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 7 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 include lt stdio h gt define CLR 0 define PRN 1 void main void int flag flag CLR ifdef PRN printf flag d n flag endif Figure 1 7 Example C Source File sampl
303. s Function Contents Reentrant localeconv Initializes struct Iconv O setlocale Sets and searches the locale information of a program O REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 222 R32C 100 Series C Compiler Package V 1 02 E Standard Library E 2 3 Standard Function Reference The following describes the detailed specifications of the standard functions provided in NC100 The functions are listed in alphabetical order Note that the standard header file extension h shown under Format must be included when that function is used abort Function Terminates the execution of the program abnormally Format include lt stdlib h gt void abort void Method function Variable No argument used ReturnValue No value is returned Description Terminates the execution of the program abnormally Note Actually the program loops in the abort function abs Function Calculates the absolute value of an integer Format include lt stdlib h gt int abs n Method function Variable int n Integer ReturnValue Returns the absolute value of integer n distance from 0 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 223 R32C 100 Series C Compiler Package V 1 02 E Standard Library acos Mathematical Functions Function Calculates arc cosine Format include lt math h gt double acos x Method function Variable double x arbitrary real number ReturnValue Assumes an error and r
304. s stddef h Defines the macro names which are shared among standard include files stdio h 1 Defines the FILE structure 2 Defines a stream name 3 Declares the prototype of input output functions stdlib h Declares the prototypes of memory management and terminate functions string h Declares the prototypes of character string and memory handling functions time h Declares the functions necessary to indicate the current calendar time and defines the type REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 206 R32C 100 Series C Compiler Package V 1 02 E Standard Library E 1 2 Standard Header Files Reference Following are detailed descriptions of the standard header files supplied with NC100 The header files are presented in alphabetical order The NC100 standard functions declared in the header files and the macros defining the limits of numerical expression of data types are described with the respective header files assert h Function Defines assert function ctype h Function Defines declares string handling function The following lists string handling functions Function Contents isalnum Checks whether the character is an alphabet or numeral isalpha Checks whether the character is an alphabet iscntrl Checks whether the character is a control character isdigit Checks whether the character is a numeral isgraph Checks whether the character is printable except a blank
305. s as specified by cout until the data that does not match val is encountered Syntax include lt asmmacro h gt When calculated in 8 bits static void _far swhile_b unsigned char val const void _far from unsigned long int count When calculated in 16 bits static void _far swhile_w unsigned short int val const void _far from unsigned long int count When calculated in 32 bits static void _far swhile_l1 unsigned long iont val const void _far from unsigned long int count REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 162 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications Appendix C Overview of C Language Specifications In addition to the standard versions of C available on the market C language specifications include extended functions for embedded system C 1 Performance Specifications C 1 1 Overview of Standard Specifications NC100 is a cross C compiler targeting the R32C 100 series In terms of language specifications it is virtually identical to the standard full set C language but also has specifications to the hardware in the R32C 100 series and extended functions for embedded system e Extended functions for embedded system near far modifiers and asm function etc o Floating point library and host machine dependent functions are contained in the standard library C 1 2 Introduction to NC100 Performance This section provides an overview of NC100
306. s for character constants Example Lk 6 Permits you to write Japanese characters in comments Example RF Shift JIS and EUC code are supported but can t use the half size character of Japanese KATA KANA Default argument declaration Default value can be defined for the argument of a function for function Example extern int func int 1 char 0 Example 2 extern int func int a char 0 When writing a variable as a default value be sure to declare the variable used as a default value before declaring the function Write default values sequentially beginning immediately after the argument REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 98 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference Table B 2 Extended Functions 2 2 Extended feature Description Inline storage class Functions can be inline developed by using the inline storage class specifier inline Example inline func int i Always be sure to define the body of an inline function before using the inline function Extension of Comments You can include C like comments Example This is a comment pragma Extended functions You can use extended functions for which the hardware of R32C 100 series in C language macro assebler function You can describe some assembler command as the function of C Example 1 signed char abs_b signed char val Example 2 long int abs_l long
307. s the command line options for outputting the symbol file for the C source file Tabel 2 4 Options for Debugging Option Function g Outputs debugging information to an assembler source file extension a30 Therefore you can perform C language level debugging genter Always outputs an enter instruction when calling a function Be sure to specify this option when using the debugger s stack trace function e Optimization Options Tabel 2 5 shows the command line options for optimizing program execution speed and ROM capacity Tabel 2 5 Optimization Option Option Short form Function O 1 5 None Optimizes the program to be efficient in both speed and ROM size at each level O50A None Inhibits code generation based on bit manipulating instructions when the optimization option O5 is selected OR None Optimizes the program as much as possible by placing priority on ROM size OS None Optimizes the program as much as possible by placing priority on speed OR_MAX ORM Maximum optimization of ROM size followed by speed OS_MAX OSM Maximum optimization of speed followed by ROM size Ocompare_byte_to_word OCBTW_ Compares consecutive bytes of data at contiguous addresses in words Oconst OC Performs optimization by replacing references to the const qualified external variables with constants Ofile_inline OFI All inline functions are expanded inli
308. s the names of all data sections defined in that file If you need to add or change section names after using this function to change section names change initialization etc in the startup program for the respective sections The program data rom and bss sections can have their names changed a number of times in one and the same file All other sections cannot have their names changed twice or more Example C source program pragma SECTION program proi void func void remainder omitted Assembly language source program FUNCTION func section pro1 CODE ALIGN _file smp c line 9 glb _func _func pragma SECTION data datai Change name of data section from data to data1 Changes name of program section to pro1 Maps to pro1 section int i Maps to datai_NE section void func void remainder omitted pragma SECTION data data2 int j Maps to data2 _NE section void sub void remainder omitted Figure B 66 Example Use of pragma SECTION Declaration REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 140 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma SECTION Supplement When modifying the name of a section note that the section s location attribute e g _NEor_NED is added after the section name Note String data and const data without initial values are output with the rom section name that is last declared
309. sage for all detectable warnings however not including alarms output by Wlarge_to_small and Wno_used_argument Wccom_max_warnings WCMW This option allows you to specify an upper limit for the Warning Count number of warnings output by ccom100 Werror_file lt file name gt WEF Outputs error messages to the specified file Wlarge_to_small WLTS Outputs a warning about the tacit transfer of variables in descending sequence of size Wmake _tagfile WMT Outputs error messages to the tag file of source file by source file Wnesting_comment WNC Outputs a warning for a comment including Wno_stop WNS Prevents the compiler stopping when an error occurs Wno_used_argument WNUA Outputs a warning for unused argument of functions Wno_used_function WNUF Displays unused global functions when linking Wno_used_static_function WNUSF For one of the following reasons a static function name is output that does not require code generation Wno_warning_stdlib WNWS Specifying this option while Wnon_prototype or Wall is specified inhibits Alarm for standard libraries which do not have prototype declaration Wnon_prototype WNP Outputs warning messages for functions without prototype declarations Wstdout None Outputs error messages to the host machine s standard output stdout Wstop_at_link WSAL Stops linking the source files if a warning occurs during linking to suppress generation of absolute module files Als
310. scanf and printf TO functions These functions are normally called high level I O functions These high level I O functions are combinations of hardware dependent lowlevel I O functions In R82C 100 series application programs the I O functions may need to be modified according to the target system s hardware This is accomplished by modifying the source file for the standard library This chapter describes how to modify the NC100 standard library to match the target system The entry vedrsion does not come with source files for the standard function library Therefore the standard function library cannot be customized for the entry version E 3 1 Structure of VO Functions As shown in Figure E 1 the I O functions work by calling lower level functions level 2 level 3 from the level 1 function For example fgets calls level 2 fgetc and fgetc calls a level 3 function Only the lowest level 3 functions are hardware dependent I O port dependent in the Micro Processor If your application program uses an I O function you may need to modify the source files for the level 3 functions to match the system Input function Figure E 1 Calling Relationship of I O Functions REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 291 R32C 100 Series C Compiler Package V 1 02 E Standard Library E 3 2 Sequence of Modifying I O Functions Figure E 2 outlines how to modify the I O functions to match the target system a Modify the le
311. sescscscseacscsnecseeseseseseseseneaes 50 3 1 2 Notes about the R32C s Type Dependent Part ccccsssssssssessssssesessescsesesessesesesseseseseesesesesesseseeeeseseeeseaes 50 3 1 3 About Optimi Ze GIO 1 anen cka desde AR OAN abby AAN 51 3 1 4 Precautions on Using register Variables cccccssssssssssesssscssssssssssescssesssescesesesssssscsescesesessssesesessseesseneseasaseees 53 3 1 5 About Startup Farid hii g s ocscctcescctiactetdaceatessccenttheiesdeostibcheed slensteispsesleestasssceuscederveseoasradonstssatdesesdecnsteivalocsesistbes 53 32 For Greater Code HMcleney ssscscisdscccscidinasndissisaiaennecasdieiedt stiansaisced RSA KANANEN sin sedtudaeeuesannnaan states 54 3 2 1 Programming Techniques for Greater Code Efficiency c ccccsssssscsesesessssscsessscsssccsssesssescsescecseessescseseeseees 54 3 2 2 Speeding Up Startup Processing cccscsssssssssssssssssesssssssssssesesssessssesessssesssessssesesesseseseseseseseeseeesseseseeeeaeaes 55 3 3 Linking Assembly Language Programs with C Programs c c ccssesssssssssssssssscsseesssescsescsescssscsescssscseecsnecsnsceneness 56 3 3 1 Calling Assembler Functions from C Programs cccscssssssssssssssescsescsescsesescsessscscsssescsescsessecesssessscsesescseecees 56 3 3 2 Writing Assembler Functions s iscsscsssssecicsastascrcrseed aii ania aanins damsaiaauiianmad 58 3 3 3 Notes on Coding Assembler FUunctions ccccsscssssssssssssesssescsescsescsssesescsesssescsesssesesesesesese
312. sesescscssscseecseesees 112 B 2 5 Specification Not Dependent on Storage Class ccsscsssccsssesssescssscsescssscsescssecsescscsssssesceseessseseseacsescseseesees 115 B 2 6 Selectively suppressing optimization ccccceccsescscssssssssessescsesesescsescsescsescseecsescsescesssesesesessscsssescecssscseecseesees 116 B 2 7 Notes on the asm Function c ccccssscssscsssssssssscssscscsesssescsescscsescsescssscsssscesssessecsesescaescseacssscscsesseeseeeessseeseeseees 116 B 3 Description of Japanese Characters c ccccsessssssescssssssscssssssssssessscsssescscscsesescsescsescscssscessscsesesesesesescseeeseenseeeseeseees 119 B 3 1 Overview of Japanese Characters ccccccsssssscsssscsssesssescsescsescssscsescsescsescsnssesssesesssesesesssescseseecacessnseessseeseeseees 119 B 3 2 Settings Required for Using Japanese Characters cccsssssssssscsssescsescsesssescsssssesessssssscssseesssesesescscsessseasens 119 B 3 3 Japanese Characters in Character Strings c cccccssscsssssssescsssssssccsssescsesesescssscsescssscseecseessesensseseseseacaseesees 120 B 3 4 Sing Japanese Characters as Character Constants cccccsssssscsssesssescsessscscsssescscssessescessesssescsescsescsescasens 121 B4 Default Argument Declaration of Function ccceessssssssesssessescssesssescssssessssecsseseesenessseescseseeeseessaeesesseseseseaeeneaes 122 B 4 1 Overview of Default Argument Declaration of Function ccccccesssessssscssscsssessesscscsessscscsescscsce
313. sni files for Call Walker gensni generates a stack information file extension sni by processing the absolute module file extension x30 Before gensni can be used there must be an absolute module file extension x30 available Specify the compile option finfo during compilation to generate that file The processing flow of NC100 is shown in Figure G 1 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 330 R32C 100 Series C Compiler Package V 1 02 G Using gensni or the Stack information File Creation Tool for Call Walker C language source file Compile driver nc100 igen100 ccom100 Compiler aopt100 Assembly language source file Assembler as100 Relocatabale file Linker In100 Inline generator ee cpp100 Preprocessor Assembler optimizer nc100 Command option Absolute module Stack information file Indicates the software included in the NC100 ae Files output by NC100 and gensni Figure G 1 Processing flow of NC100 REJ10J2009 0200 Rev 2 00 2010 04 01 331 ztENESAS R32C 100 Series C Compiler Package V 1 02 G Using gensni or the Stack information File Creation Tool for Call Walker G 3 Starting gensni If Call Walker is started from the High performance Embedded Workshop gensni is automatically executed However if Call Walker is started from other than the High performance Embedded Workshop gensni is not automatically executed In this case start gensni from the Windo
314. sscscsesescacsesesseseseeesensaes 163 C 1 2 Introduction to NC100 Performance c ccccscsssssssssescssscssscssesesescsesssescsescscscssscecscsescsescseessescensseseacaescaeacsees 163 C 2 Standard Language Specifications cccccscssssssscsssesssescsescsescsescscsessssscscssscsssccsssesssesesescsesesesesescseeseceesseeeseeesensees 166 C 2 1 POV TUCAK PIII AA E EA EE A N E EA A E 166 C 2 2 TE aa E dastccszsgeniets A steiata a etleatunnitien E EE OEE 169 C 2 3 BWxXPresslOnsss assiscitiadesiavnsinaasned a R NN AAEE NGINA 171 C 2 4 E E a O EAEAN EIN E EN ENN A ENEA N E E N E E O AA 172 C 2 5 SIIKI ATES a1 v EAE AAE E A EETA EEA A A 175 G3 Pr process Commands assinsmnannnnnnn NAR AAA NNN 178 C 3 1 List of Preprocess Commands Available c ccccsesssescssscssscsssssscsssesssescsescsescscsescscsssescecseseseseesssesesesescsesesees 178 C 3 2 Preprocess Commands Reference ccsssssssesssescssscsssescsessssssscsssessscseseseseseseacsescsescssscsescseeseasansseseacsescaescsees 178 C 3 3 Predenined MACOS annn A E 185 C 3 4 Usage of predefined Macros siscisseiasiisrissdilaaisindsnrcaieiiaawieaaisin anid ian NNA 185 Appendix D C Language Specification Rules 0 c ccccccssssssssssssesescssscsescsssssnscssscsssssesssesssescscscsessscsescsescsussesesesesescaeseseseasscaees 186 D 1 Internal Representation of Data ccccesssssssssssssssssssssssssssssssssssssssussssessssssssssussssesesessesesussesesesaeaesassessasesseaeeessees 186 D 1 1 M
315. ssescsesees 122 B 4 2 Format of Default Argument Declaration of Function cccccssssssescssssesseesssescsescsescssscseecseecseecseeseeeeeeeees 122 B 4 3 Restrictions on Default Argument Declaration of Function c cccscsesesesesesssescsesccssescsssesesescsesssescseaees 124 Bd inline Function Declaration as sisissecssceetsssescassasiencascsaastjisccetstisacdcassabicdcnignavca A N AS TARN 125 B 5 1 Overview of inline Storage Class c cccscssssssssssssssessssscssscscsescsescscscssscscsessssscscsssessecsesesesescseacseeeesesssesesensaes 125 B 5 2 Declaration Format of inline Storage Class cccsessssscsssssssssscsssesssescsesesescsescssscssscsescssecsescsnsceseesescsesesees 125 B 5 3 Restrictions on inline Storage Class ccccsssesssssssescssssssscssessscsssessscsssescscscssscscscsssescsessessseessseaescsescaeecsees 126 BO Extension of Comments sc i cs sieecetcteccteeedecentisisecencastescecetstencetetsenscvisacececessacenchelebsecsasiatschetsebsassondetssnecdivbestandbestce 129 B 6 1 Overview ofl Comin Ont E E E A A Gosek dleteseaneaertes 129 B 6 2 Comment ROM at ene a a EREN EEP ENEE NE O O EN 129 B 6 3 Priority Ob M and A ain S T ERS ENO 129 B 7 pragma Extended Functions ccccccssssssscscssscssssessssscscsescscscsescsescscscscscscscsescsnscsessssesesesesessseseseecsnsseeseeseseseseseneaes 130 B 7 1 Index of pragma Extended Functions c ccccssssssssssssssssssssssssssssssssssssssssesessssessseessseseeesseasses
316. ssing after assembling In this case it only generates an assembly language source file If you want to create a re locatable file simultaneously with an assembly language source file use the option dsource shortcut dS REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 15 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 2 1 4 nc100 Command Line Options a Options for Controlling Compile Driver Tabel 2 1 shows the command line options for controlling the compile driver Tabel 2 1 Options for Controlling Compile Driver Option Function c Creates a relocatable file extension r30 and ends processing Didentifier Defines an identifier Same function as define dsource Generates an assembly language source file extension a30 with a C Short form dS language source list output as a comment Not deleted even after assembling dsource_in_list Tn addition to the dsource function generates an assembly language Short form dSL list file Ist E Invokes only preprocess commands and outputs result to standard output I directory Specifies the directory containing the file s specified in include You can specify up to 256 directories P nvokes only preprocess commands and creates a file extension i S Creates an assembly language source file extension a80 and ends processing silent Suppresses the copyright message display at startup
317. ssscsescssssessssssssescssssssssseseaescesesessseecasseeseaesseeesseseseseaeesens 100 B 1 3 Format of Pointer type Variable a ccisissc scasts scosssccdeascdecsedsahasesss ecousddsieshosucesaneseecasedtaseceuts cedevadsacdesiecstvanense 101 B 1 4 Declaration Of HumChOM 5255 05icc20 sch icseraseidotaertsedvaiaioarsdanes eth seateised adiarsiaaibeaieuaniaataaia eurandadanwene 103 B 1 5 near and far Control by nc100 Command Line Option ccccescsssesesesesesescsessssscesseseeesesssesssessscscseaees 103 B 1 6 Function of Type conversion from near to far c c cccccessssssscsesssessssssssssscsssscssscsesescscscsescscsescseacssseecsessessensaes 104 B 1 7 Declaration Of Mirct ON 530s seeds a rE E E NEN 104 B 1 8 Function for Specifying near and far in Multiple Declarations cccesssesscessesesseessesceeesesssseneseseeseees 105 B 1 9 Notes on near and far Attributes ccsscsssssssssssssssssssesssssssscssssssesssssscscsssssssssssescacssesesessseeseaesseaeasesessaeseesenes B2 asm Functioneren ENE aS a a AN tascesstalaceasstasoasesassdsdecessea Wessesaasactheasecepuisiseestateicees B 2 1 Overview Of asm PUNIO ergiini an Ean EE cua ENR OEN EEN EE A EE R B 2 2 Specifying FB Offset Value of auto Variable cccccscssssesssesssescssscsescsssssescssscsescssssesescsesssescsescseacsesceeseesess B 2 3 Specifying Register Name of register Variable B 2 4 Specifying Symbol Name of extern and static Variable ccccscsscesssessssssssscsssssscessesse
318. ssseaesaeaeeeseees 130 B 7 2 Using Memory Mapping Extended FUunctions cccccsssssssssscsesssssssscsessessscsssesesscsesesesssesseecseseeeeseecsesesesens 134 B 7 3 Using Extended Functions for Target Devices ccccscsssssssssssssssscsssescsesesescsescsescssscsescseessescenssesesesesesceeaees 144 B 7 4 Use of the other extension function ceccssssssssessssssssssssssssessesssesessssesssussssesessssescsecssseseesssuscsusessnsesseaeeeseres 152 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS R32C 100 Series C Compiler Package V 1 02 Contents B8 assembler Macro Function iiiiacuscauiiadiiiaaiiaduaiiaauiadiiia ausisiad daa diana suiiiacdaauiaasuss 156 B 8 1 Outline of Assembler Macro Function ccssssssssssssssssssssessessssssssescscscssssesessesescsseseeeseseseeseasaseseaeeneasseesenes 156 B82 Description Example of Assembler Macro FUunction ccscscssssssssssesssescsescsesesescsssescscssssssseeseeseeseseacsesees 156 B83 Commands that Can be Written by Assembler Macro Function ccccsssssssssesssessscscsessecscssseeceeseesees 157 Appendix C Overview of C Language Specifications c ccccccssssssssssscsssssescscsssssessssssessescsescsescsescssscsescseesssesesescsescseseassesees 163 GA Performance SpeciiCa tions ee ereinen ir a iE TAA A dis sevas lg se E sd savacabucdelarehsdzasadaisdatesciad A RENEO 163 C 1 1 Overview of Standard Specifications cccccccsscssssesssescsesesescssscsescsssssescssscssseeseseseseses
319. ssseseseseseassnseensceseees 61 BA ORC E Soha sen co dh savas cde shot San obec ste abba sess eda aoksnelavcststiel Meesheasde eect tetas MN E A AN 62 3 4 1 Precautions on Transporting between NC Series Compilers c cccsccsssssessssscscssssssssesssescsessscscsesescseecseeees 62 Appendix A Command Option Reference c cccssssssssssssssssssescssssesesessssesesssussescssesesesessssesessssesesesscsesessesesuesesesuseeseseeeeseseseeseseaes 63 ALL NCLOO Command Forniiat cccecccccicseeiecesecassesssecetsessoacacsstacacsscseactesssseactenssacactetacacacsetacacsssetaciereracactecacaetseasactetiedetesceanae 63 A 2 nc1l00 Command Line Options c cescsssssssssssssssssssssssscsessssssesssusscsessesssesususscsesessssesscusscseasessseauceseeseeeeseassusseeeaeeeses 64 A2 1 Options for Controlling Compile Driver ccccsscssssssssesssescescsescsssesescsescsescssscsnscsssssscsssessscsesessscessescseeeeneees 64 A 2 2 Options Specifying Output Files wo cccccscsssssssssscsssesssescsescsescsescssscsescseesseesseecsecesesssesesesssessseecseessesceneseseees 67 A 2 3 Version Information Display Option ccccsssssssesssscsssessscscscscsesescsescscssssssscsssesssesesescsesesescssscseesssescseseseaees 68 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS R32C 100 Series C Compiler Package V 1 02 Contents A 2 4 Options for Debug etnies ccicsaicceiscisntadisia dui anieia N NAA ciuisisd diana dieciaaunas 69 A 2 5 Optimization OPTIOWS 5 0 5 dsc cased e
320. st char _far extern int fprintf FILE _ far const char _far extern int sprintf char _far const char _far omitted extern int init_dev FILE _ far int extern int speediint int int int extern int init_prn void extern int _sget void extern int _ sput int extern int _pput int extern const char _far _print int const char _ far int _far _far int _far void mainvoid S if int flag l flag 0 3 printf flag d n flag 4 l l a Figure 1 9 Example Preprocessed C Source File 2 Let s look at the contents of the preprocessed C source file Items 1 to 4 correspond to 1 to 4 in Figure 1 8 and Figure 1 9 1 Shows the expansion of header file stdio h specified in include 2 Shows the C source program resulting from expanding the macro 3 Shows that CLR specified in define is expanded as 0 4 Shows that because PRN specified in define is 1 the compile condition is satisfied and the printf function is output REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 9 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 1 5 3 Assembly Language Source Files The assembly language source file is a file that can be processed by AS100 as a result of the compiler ccom100 converting the preprocess result C source file The output files are assembly language source files with the extension a30 F
321. stored in each argument Returns EOF if null character 0 is input as data Description Converts the characters input as specified in format and stores them in the variables shown in the arguments Argument must be a far pointer to the respective variable Format is specified in the same way as in scanf REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 272 R32C 100 Series C Compiler Package V 1 02 E Standard Library strcat String Handling Functions Function Concatenates character strings Format include lt string h gt char _far strcat s1 s2 Method function Argument char _far s1 Pointer to the character string to be concatenated to const char _far s2 Pointer to the character string to be concatenated from ReturnValue Returns a pointer to the concatenated character string area s1 Description Concatenates character strings s1 and s2 in the sequence s1 s21 The concatenated string ends with NULL When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization strchr String Handling Functions Function Searches the specified character beginning with the top of the character string Format include lt string h gt char far strchr s Cc Method function Argument const char _far s Pointer to the character string to be searched in int c Character to be searched for ReturnValue Returns the position of character
322. strings n characters O strnicmp Compares character strings n characters All alphabets O are handled as upper case letters Search strchr Searches the specified character beginning with the top of O the character string strespn Calculates the length number of unspecified characters O that are not found in the other character string strpbrk Searches the specified character in a character string from O the other character string strrchr Searches the specified character from the end of a character O string strspn Calculates the length number of specified characters that O are found in the other character string strstr Searches the specified character from a character string O strtok Divides some character string from a character string into X tokens Length strlen Calculates the number of characters in a character string O Convert strerror Converts an error number into a character string X strxfrm Converts a character string using locale information O REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 217 R32C 100 Series C Compiler Package V 1 02 E Standard Library 2 Character Handling Functions The following lists character handling functions Table E 3 Character Handling Functions Function Contents Reentrant isalnum Checks whether the character is an alphabet or numeral O isalpha Checks whether the character is an alphabet O iscntrl Checks whether the character
323. strings are equal ReturnValue gt 0 The first character string s1 is greater than the other ReturnValue lt 0 The second character string s2 is greater than the other Description Compares each byte of two character strings ending with NULL However all letters are treated as uppercase letters strerror Function Format Method Argument ReturnValue Description REJ10J2009 0200 Rev 2 00 2010 04 01 276 String Handling Functions Converts an error number into a character string include lt string h gt char _far strerror errcode function int errcode error code Returns a pointer to a message character string for the error code stderr returns the pointer for a static array 7tENESAS R32C 100 Series C Compiler Package V 1 02 E Standard Library strlen Function Calculates the number of characters in a character string Format include lt string h gt size_t strlen s Method function Argument const char _far s Pointer to the character string to be operated on to calculate length ReturnValue Returns the length of the character string Description Determines the length of character string s to NULL strncat String Handling Functions Function Concatenates character strings n characters Format include lt string h gt char _far strncat sl s2 n Method function Argument char _far s1 Pointer to the character string to be concatenated to const char _far
324. suitable to perform this operation when there is an order to the consecutive bit operations as in I O bit fields use this option to suppress optimization Notes This option is only valid if you specify option O 8 to 5 OR or OS Ono_break_source_debug ONBSD Suppression of optimization Function Suppresses optimization that affects source line data Supplement Specifying the OR or O option performs the following optimization which may affect source line data This option ONBSD is used to suppress such optimization Notes This option is valid only when the OR or O option is specified REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 78 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Ono_float_const_fold ONFCF Function Suppresses the constant folding processing of floating point numbers Supplement By default NC100 folds constants Following is an example before optimization val 1000e250 50 0 after optimization val 20e250 In this case if the application uses the full dynamic range of floating points the results of calculation differ as the order of calculation is changed This option suppresses the constant folding in floating point numbers so that the calculation sequence in the C source file is preserved Ono_logical_or_combine ONLOC Suppression of optimization Function Suppresses the optimization that puts logical OR together Sup
325. t include lt stdlib h gt void _far bsearch key base nelem size cmp Method function Variable const void _far key Search key const void _far base Start address of array size_t nelem Element number size_t size Element size int cmp0 Compare function ReturnValue Returns a pointer to an array element that equals the search key Returns a NULL pointer if no elements matched Note The specified item is searched from the array after it has been sorted in ascending order REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 227 R32C 100 Series C Compiler Package V 1 02 E Standard Library bzero Function Initializes a memory area by clearing it to zero Format include lt string h gt void bzero top size Method function Variable char _far top Start address of the memory area to be cleared to zero unsigned long size Number of bytes to be cleared to zero ReturnValue No value is returned Description Initializes to 0 the number of bytes specified in size from the starting address of the area specified in top REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 228 R32C 100 Series C Compiler Package V 1 02 E Standard Library calloc Memory Management Functions Function Allocates a memory area and initializes it to zero 0 Format include lt stdlib h gt void _far calloc n size Method function Variable size_t n3 Number of elements size_t size Value indicating the elemen
326. t 01H _ port Optimization is suppressed Figure B 33 Example of Suppressing Optimization by Dummy asm B 2 7 Notes on the asm Function a Extended Features Concerning asm functions When using the asm function for the following processing be sure to use the format shown in the coding examples 1 For variables with storage class auto arguments and 1 bit bit fields Do not specify auto variables or parameters or 1 bit bit fields using the offset from the frame base register FB Use the format shown in Figure B 34 to specify auto variables and parameters asm MOV W 01H FB i lt Format for referencing auto variables asm BSET FB s bit0 lt Format for checking auto bit fields Figure B 34 Example Coding of asm Function 1 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 116 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference 2 Specification of the register storage class You can specify the register storage class in NC100 When register class variables are compiled with option fenable_register fER use the format shown in Figure B 35 for register variables in asm functions asm MOV W 0 i lt Format for checking register variables Figure B 35 Example Coding of asm Function 2 Note that when you specify option O 1 5 OR OS OR_MAX or OS_MAX parameters passed via the registers may to improve code efficiency be processed
327. t 123 software int 124 software int 125 software int 126 software int 127 Sora Figure 2 16 REJ10J2009 0200 Rev 2 00 2010 04 01 31 Startup Program List 11 Sect100 inc 2tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler word dummy_int software int 133 word dummy_int software int 134 word dummy_int software int 135 omitted word dummy_int software int 253 word dummy_int software int 254 word dummy_int software int 255 endif FIXED VECTOR SECTION section __ fvector romdata org OFFFFFFDCH UDI word dummy_int OVER_FLOW word dummy_int BRKI word dummy_int word OFFFFFFFFH word OFFFFFFFFH WDT word dummy_int word dummy_int NMI word dummy_int RESET word start ID code DEFINITION ID code check function id CodeChk J ake eee keke dee ke kee eke eede IK eede IK eeke dede keke ke ekeke eke deke kek eke deede keke eeke kek ekeke kekke kk kkk kk kkk kkk a End of R32C 100 start up 3 IRI TIKI KIKI RIK IK II IK IIIA III IAI IAAI eke IK III IIIA IIIA IK IAI IKI a 3 Figure 2 17 REJ10J2009 0200 Rev 2 00 2010 04 01 32 Startup Program List 12 Sect100 inc 2tENESAS R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler 2 2 2 Customizing the Startup Program a Overview of Startup Program Processing 1 About ncrt0 a30 This program is run at the start of the program or imme
328. t size in bytes ReturnValue Returns NULL if a memory area of the specified size could not be allocated Description After allocating the specified memory it is cleared to zero The size of the memory area is the product of the two parameters Rule The rules for securing memory are the same as for malloc ceil Function Format Method Argument ReturnValue REJ10J2009 0200 Rev 2 00 2010 04 01 229 Mathematical Functions Calculates an integer carry value include lt math h gt double ceil x function double x arbitrary real number Returns the minimum integer value from among integers larger than given real number x 7tENESAS R32C 100 Series C Compiler Package V 1 02 E Standard Library clearerr Function Initializes clears error status specifiers Format include lt stdio h gt void clearerr stream Method function Argument FILE _far stream Pointer of stream ReturnValue No value is returned Description Resets the error designator and end of file designator to their normal values Cos Mathematical Functions Function Calculates cosine Format include lt math h gt double cos x Method function Argument double x arbitrary real number ReturnValue Returns the cosine of given real number x handled in units of radian cosh Function Calculates hyperbolic cosine Format include lt math h gt double cosh x Method function Argument double x arbitrar
329. te module file x30 will be generated REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 3 R32C 100 Series C Compiler Package V 1 02 1 Introduction to NC100 1 3 Notes To use the technical contents shown in product data diagrams or tables or the programs or algorithms presented herein for your system please carefully evaluate their suitability as part of the entire system not singly as technical content program or algorithm alone to determine in advance whether they are actually suitable for your system Renesas Electronics Corporation and Renesas Resolutions Corporation will not assume responsibility for the suitability of said items in user systems 1 3 1 Notes about Version up of compiler The machine language instructions assembly language generated by NC100 vary in contents depending on the startup options specified when compiling contents of version up etc Therefore when you have changed the startup options or upgraded the compiler version be sure to reevaluate the operation of your application program Furthermore when the same RAM data is referenced and its contents changed between interrupt handling and non interrupt handling routines or between tasks under realtime OS always be sure to use exclusive control such as volatile specification Also use exclusive control for bit field structures which have different member names but are mapped into the same RAM 1 3 2 Notes about the R32C s Type Dependent Part When writing t
330. ternal variable initialized change to e An initialization expression is specified for an public y extern declared variable extern will be ignored Delete extern REJ10J2009 0200 Rev 2 00 2010 04 01 323 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 23 ccom100 Warning Messages 4 9 Warning message Description and countermeasure far pointer implicitly casted by near The far pointer was converted into the near pointer pointer Check the data types near or far function must be far The function is declared with the near type Write it correctly function function yname has _ no used argument variable name ey The variable declared in the argument to the function is not used Check the variables used handler function called The function specified by pragma HANDLER is called Be careful not to call a handler handler function can t return value eu ey y The function specified by pragma HANDLER is using a returned value The function specified by pragma HANDLER cannot use a returned value Delete the return value handler function has argument The function specified by pragma HANDLER is using an argument The function specified by pragma HANDLER cannot use an argument Delete the argument hex character is out of range The hex character in a character constant is excessively long Also some character that is not a
331. the expressions on the left and right sides of the operator invalid gt gt operands The gt gt operation contains an error Check the expressions on the left and right sides of the operator invalid gt gt operands The gt gt operation contains an error Check the expressions on the left and right sides of the operator invalid operands The left side expression of is not array type or pointer type Use an array or pointer type to write the left side expression of invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid operands The operation contains an error Check the expressions on the left and right sides of the operator REJ10J2009 0200 Rev 2 00 2010 04 01 313 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 13 ccom100Error message 7 13 Error message Description and countermeasure invalid operands The operation contains an error Check the expressions on the left and right sides of the operator invalid operands The unary operator or postfix operator contains an error For the unary operator check the right side expression For the postfix operator check the lefts
332. then inline e The inline declared function name is called recursively is ignored The inline declaration will be ignored Correct the statement not to call such a function name recursively REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 320 R32C 100 Series C Compiler Package V 1 02 F Error Messages F 6 cccom100 Warning Messages Table F 20 to Table F28 list the ccom100 compiler warning messages and their countermeasures Table F 20 ccom100 Warning Messages 1 9 Warning message Description and countermeasure pragma pragma name amp HANDLER both specified Both pragma pragma name and pragma HANDLER are specified in one function Specify pragma pragma name and pragma HANDLER exclusive to each other pragma pragma name amp INTERRUPT both specified Both pragma pragma name and pragma INTERRUPT are specified in one function gt Specify pragma pragma name and pragma INTERRUPT exclusive to each other pragma pragma name amp TASK both e Both pragma pragma name and pragma TASK are specified specified in one function gt Specify pragma pragma name and pragma TASK exclusive to each other pragma pragma name format error e The pragma pragma name is erroneously written Processing will be continued gt Write it correctly pragma pragma name format error e The pragma pragma name is erroneously written ignored This line will be ignored gt
333. thod function Argument char _far pointer Pointer of the location to be stored const char _far format Pointer of the format specifying character string ReturnValue Returns the number of characters output Description Converts argument to a character string as specified in format and stores them from the pointer Format is specified in the same way as in printf sqrt Mathematical Functions Function Calculates the square root of a numeric value Format include lt math h gt double sqrt x Method function Argument double x arbitrary real number ReturnValue Returns the square root of given real number x REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 271 R32C 100 Series C Compiler Package V 1 02 E Standard Library srand Function Imparts seed to a pseudo random number generating routine Format include lt stdlib h gt void srand seed Method function Argument unsigned int seed Series value of random number ReturnValue No value is returned Description Initializes seeds the pseudo random number series produced by rand using seed sscant Function Reads data with format from a character string Format include lt stdio h gt int sscanf string format argument Method function Argument const char _far string Pointer of the input character string const char _far format Pointer of the format specifying character string ReturnValue Returns the number of data entries
334. tion Options Option Short form Function O 1 5 None Optimization of speed and ROM size O50A None Inhibits code generation based on bit manipulating instructions when the optimization option 05 is selected OR None Optimization of ROM size followed by speed OS None Optimization of speed followed by ROM size OR_MAX ORM Maximum optimization of ROM size followed by speed OS_MAX OSM Maximum optimization of speed followed by ROM size Ocompare_byte_to_word OCBTW_ Compares consecutive bytes of data at contiguous addresses in words Oconst OC Performs optimization by replacing references to the const qualified external variables with constants Ofile_inline OFI All inline functions are expanded inline Oinline_line OIL This option changes the size number of lines of the function to be inline expanded Oglb_jmp OGJ Global jump is optimized Oglobal_to_inline OGTI Handles global functions as inline declared Oloop_unroll oop count OLU Unrolls code as many times as the loop count without revolving the loop statement The loop count can be omitted When omitted this option is applied to a loop count of up to 5 Ono_bit ONB Suppresses optimization based on grouping of bit manipulations Ono_break_source_debug ONBSD Suppresses optimization based on grouping of bit manipulations Ono_float_const_fold ONFCF Suppresses the constant folding processing of
335. tion section This saves the ROM size by stopping unwanted library items from being linked 3 Notes on Writing Initialization Programs Note the following when writing your own initialization programs to be added to the startup program 1 If your initialization program changes the U or B flags return these flags to the original state where you exit the initialization program Do not change the contents of the SB register 2 If your initialization program calls a subroutine written in C note the following two points Call the C subroutine only after clearing them B and D flags e Call the C subroutine only after setting the U flag d Setting the Stack Section Size A stack section has the domain used for user stacks and the domain used for interruption stacks Since stack is surely used please surely secure a domain stack size should set up the greatest size to be used Stack size is calculated to use the stack size calculation utility Call Walker 1 The stack is used within the startup program as well Although the initial values are reloaded before calling the main function consideration is required if the stack size used by the main function etc is insufficient REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 35 R32C 100 Series C Compiler Package V 1 02 2 Basic Method for Using the Compiler e Heap Section Size Set the heap to the maximum amount of memory allocated using the memory management functions calloc and
336. tions Function Checks whether the character is printable including a blank 0x20 Ox7e Format include lt ctype h gt int isprint c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if printable Returns 0 if not printable Description Determines the type of character in the parameter REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 245 R32C 100 Series C Compiler Package V 1 02 E Standard Library isounct Character Handling Functions Function Checks whether the character is a punctuation character Format include lt ctype h gt int ispunct c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if a punctuation character Returns 0 if not a punctuation character Description Determines the type of character in the parameter isspace Character Handling Functions Function Checks whether the character is a blank tab or new line Format include lt ctype h gt int isspace c Method macro Argument int c Character to be checked ReturnValue Returns any value other than 0 if a blank tab or new line Returns 0 if not a blank tab or new line Description Determines the type of character in the parameter REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 246 R32C 100 Series C Compiler Package V 1 02 E Standard Library isupper Character Handling Functions Function Che
337. tions with good code efficiency by optimization strcoll String Handling Functions Function Compares character strings using locale information Format include lt string h gt int strcoll sl s2 Method function Argument const char _far s1 Pointer to the first character string to be compared const char _far s2 Pointer to the second character string to be compared ReturnValue ReturnValue 0 The two character strings are equal ReturnValue gt 0 The first character string s1 is greater than the other ReturnValue lt 0 The second character string s2 is greater than the other Description When you specify options O 3 5 or OS the system may selects another functions with good code efficiency by optimization REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 274 R32C 100 Series C Compiler Package V 1 02 E Standard Library strcp String Handling Functions Function Copies a character string Format include lt string h gt char _far strcpy s1 s2 Method macro or function Argument char _far s1 Pointer to the character string to be copied to const char _far s2 Pointer to the character string to be copied from ReturnValue Returns a pointer to the character string at the destination of copy Description Usually the program code described by macro is used for this function In using the function in a library please describe it as undef strcpy after description of include lt s
338. tored using val as the data to store p as the address to from val address which to transfer and count as the number of times to transfer data There is no return value Syntax include lt asmmacro h gt When calculated in 8 bits static void sstr_b usigned char val void _far p unsigned long int count When calculated in 16 bits static void sstr_w usigned short int val void_far p unsigned long int count When calculated in 32 bits static void sstr_1 usigned long int val void _far p unsigned long int count REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 161 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference SUNTIL Function Searches the file in the address increment direction from the comparison address indicated by from as many times as specified by count until the data that matches val is encountered Syntax include lt asmmacro h gt When calculated in 8 bits static void _far suntil_b unsigned char val const void _far from unsigned long int count When calculated in 16 bits static void _far suntil_w unsigned short int val const void _far from unsigned long int count When calculated in 32 bits static void _far suntil_l unsigned long iont val const void _far from unsigned long int count SWHILE Function Searches continually in the address incrementing direction from the comparison address indicated by from as many time
339. tring h gt Copies character string s2 ending with NULL to area s1 After copying the character string ends with NULL When you specify options O 3 5 OR or OS the system may selects functions with good code efficiency by optimization strespn String Handling Functions Function Calculates the length number of unspecified characters that are not found in the other character string Format include lt string h gt size_t strcspn sl s2 Method function Argument const char _far s1 Pointer to the character string to be searched in const char _far s2 Pointer to the character string to be searched for ReturnValue Returns the length number of unspecified characters Description Calculates the size of the first character string consisting of characters other than those in s2 from area s1 and searches the characters from the beginning of oy You cannot search for 0 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 275 R32C 100 Series C Compiler Package V 1 02 E Standard Library stricmp String Handling Functions Function Compares character strings All alphabets are handled as upper case letters Format include lt string h gt int stricmp s1 s2 Method function Argument char _far s1 Pointer to the first character string to be compared char _far s2 Pointer to the second character string to be compared ReturnValue ReturnValue 0 The two character
340. tted ncl00A l amp lenameA lt C source file name gt 1 In file specification the extension can be omitted If the extension of a file is omitted it is processed assuming an extension lib 2 Ifyou specify a file extension be sure to specify lib 8 NC100 links by default the library nc100lib lib that is present in the directory specified by the environment variable LIB100 The table below lists the library files to be linked for each compile option specified compile option compile option reference library fint_16 fuse_FPU None None nc100lib lib Specify None nc100i16 lib None Specify nc100fpu lib Specify Specify nc100i116fpu lib 29 4 If multiple libraries are specified references to nclOOlib lib are assigned the lowest priority REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 88 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 8 Warning Options Table A 9 shows the command line options for outputting warning messages for contraventions of nc100 language specifications Table A 9 Warning Options Option Short form Function Wall None Displays message for all detectable warnings however not including alarms output by Wlarge_to_small and Wno_used_argument Wccom_max_warnings WCMW This option allows you to specify an upper limit for the Warning Count number
341. turns EOF if an error or the end of the stream is encountered Description Reads one character from the stream Interprets code 0x1A as the end code and ignores any subsequent data fgets Input Output Functions Function Reads one line from the stream Format include lt stdio h gt char _far fgets buffer n stream Method function Argument char _far buffer Pointer of the location to be stored in int n Maximum number of characters FILE _far stream Pointer of stream ReturnValue Returns the pointer of the location to be stored the same pointer as given by the argument if normally input Returns the NULL pointer if an error or the end of the stream is encountered Description Reads character string from the specified stream and stores it in the buffer Input ends at the input of any of the following 1 new line character n 2 n 1 characters 3 endof stream Anull character 0 is appended to the end of the input character string The new line character n is stored as is Interprets code 0x1A as the end code and ignores any subsequent data REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 234 R32C 100 Series C Compiler Package V 1 02 E Standard Library floor Mathematical Functions Function Calculates an integer borrow value Format include lt math h gt double floor x Method function Argument double x arbitrary real number ReturnValue The real value
342. uage programs to be burned into ROM NC100 comes with a sample startup program written in the assembly language to initial set the hardware R32C 100 locate sections and set up interrupt vector address tables etc This startup program needs to be modified to suit the system in which it will be installed The following explains about the startup program and describes how to customize it 2 2 1 Sample of Startup Program The NC100 startup program consists of the following two files nert0 a30 Write a program which is executed immediately after reset e sect100 inc Included from ncrt0 a30 this file defines section locations memory mapping Figure 2 6 to Figure 2 11 show the ncrt0 a30 source program list Figure 2 12 to Figure 2 17 show the sect100 inc source program list AIK IKK KK III KIER dee IK III de e kekeke eke e ee III IK IKI III IER IKI III IK kkk kkk kkk kk 4 C COMPILER for R32C 100 Copyright C XXXX Renesas Electronics Corp and Renesas Solutions Corp All rights reserved This program is applicable when using the basic I O library ncrt0 a30 startup program ld ncrt0 a30 v X XX XXXX XX XX XX XX XX XXXXX Exp RIK eeke ee kehee dee RIK IK eke ke ede kekeke eke e ee kee e keke dede keke ee keke k eke keee kek kekeke kekeke k kekk kkk kkk kkk u J HEAPSIZE equ Oh else if HEAPSIZE HEAPSIZE equ 300h else for HEW HEAPSIZE equ __HEAPSIZE _ endif endif 1 defines the heap size
343. ude lt stdlib h gt int rand void Method function Argument No argument used ReturnValue Returns the seed random number series specified in srand The generated random number is a value between 0 and RAND_MAX realloc Memory Management Functions Function Changes the size of an allocated memory area Format include lt stdlib h gt void _far realloc cp nbytes Method function Argument void _far cp Pointer to the memory area before change size_t nbytes Size of memory area in bytes to be changed ReturnValue Returns the pointer of the memory area which has had its size changed Returns NULL if a memory area of the specified size could not be secured Description Changes the size of an area already secured using malloc or calloc Specify a previously secured pointer in parameter cp and specify the number of bytes to change in nbytes REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 266 R32C 100 Series C Compiler Package V 1 02 E Standard Library scanf Input Output Functions Function Reads characters with format from stdin Format include lt stdio h gt include lt ctype h gt int scanf format argument Method function Argument const char _far format Pointer of format specifying character string The part after the percent sign in the character string given in format has the following meaning The part between and is optional Details of the format are sh
344. ue No value is returned pow Function Calculates the power of a number Format include lt math h gt double pow x y Method function Argument double x multiplicand double y power of a numbe Oe ReturnValue Returns the multiplicand x raised to the power of y REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 259 R32C 100 Series C Compiler Package V 1 02 E Standard Library printf Input Output Functions Function Outputs characters with format to stdout Format include lt stdio h gt int printf format argument Method function Argument const char _far format Pointer of the format specifying character string The part after the percent sign in the character string given in format has the following meaning The part between and is optional Details of the format are shown below Format fflag minimum field width precision modifier I L or h conversion specification character Example format 05 8ld ReturnValue Returns the number of characters output Returns EOF if a hardware error occurs Description Converts argument to a character string as specified in format and outputs the character string to stdout When giving a pointer to argument it is necessary to be a far type pointer 1 Conversion specification symbol e d i Converts the integer in the parameter to a signed decimal e u Converts the integer in the parameter to an unsigned de
345. uments to Function NC30 Type of argument First argument Second argument Third and following arguments _Bool RiL register R2 register Stack char int R1 register Stack Stack near pointer D 3 3 Rules for Converting Functions into Assembly Language Symbols The function names in which functions are defined in a C language source file are used as the start labels of functions in an assembler source file The beginning label of a function in an assembler source file consists of the function name in the C language source file that is prefixed by an underbar _ or dollar mark or the function name itself The appended strings and the conditions under which strings are appended are shown in Table D 11 Table D 11 Conditions Under Which Character Strings Are Added to Function Added character string Condition dollar Functions where any one of arguments is passed via register _ underbar Functions that do not belong to the above Shown in Figure D 10 is a sample program where a function has register arguments and where a function has its arguments passed via only a stack 3 However function names are not output for the functions that are specified by pragma INTCALL REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 194 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules int func_proto int int int lt 1 lint func_proto intiintjintk SS 12 l return k j i i l Sa D S a A a a a
346. ummy_int TIMER B2 software int 23 word dummy_int TIMER B3 software int 24 word dummy_int TIMER B4 software int 25 word dummy_int INT5 software int 26 word dummy_int INT4 software int 27 word dummy_int INT3 software int 28 word dummy_int INT2 software int 29 word dummy_int INT1 software int 30 word dummy_int INTO software int 31 word dummy_int TIMER B5 software int 32 word dummy_int uart2 trance NACK software int 33 word dummy_int uart2 receive ACK software int 34 word dummy_int uart3 trance NACK software int 35 lword dummy_int uart3 receive ACK software int 36 word dummy_int uart4 trance NACK software int 37 word dummy_int uart4 receive ACK software int 38 word dummy Lint uart2 bus collision software int 39 word dummy_int uart3 uartO bus collision software int 40 word dummy_int uart4 uart1 bus collision software int 41 word dummy_int A D Convert software int 42 word dummy_int input key software int 43 word dummy_int intelligent O O software int 44 word dummy_int intelligent O 1 software int 45 word dummy_int intelligent O 2 software int 46 word dummy_int inteligent YO 3 software int 47 word dummy_int intelligent O 4 software int 48 word dummy_int intelligent I O 5 software int 49 word dummy_int intelligent O 6 software int 50 word dummy_int intelligent YO 7 software int 51 word dummy_int intell
347. up file included with the product alter the content of the vector section before use For details on how to alter it refer to Preparing the Startup Program 2 When using a register on the back side be careful that the back register is not corrupted by a nesting of interrupts REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 150 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference oragma PARAMETER Declare assembler function that passed arguments via register Function Declares an assembler function that passes parameters via registers Syntax pragma PARAMETERAassembler funct ion name register name register name Description This extended function declares that when calling an assembler function its parameters are passed via registers double types long long types 64 bit registers R8R1R2RO R7RS5R7R4 A1A0 A8A2 float types long types int types far far pointer 32 bit registers R2RO0 R3R1 R6R4 R7R5 AO Al A2 A3 short types int types fint_16 option use 16 bit registers RO R1 R2 R3 R4 R5 R6 R7 char types _Bool types 8 bit registers ROL ROH R1L R1H R2L R2H R3L R3H There is no differentiation between uppercase and lowercase letters in register names Structure and union types cannot be declared Rules 1 Always put the prototype declaration for the assembler function before the pragma PARAMETER declaration If you fail
348. urns the common logarithm of given real number x REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 250 R32C 100 Series C Compiler Package V 1 02 E Standard Library longjmp Execution Control Functions Function Restores the environment when making a function call Format include lt setjmp h gt void longjmp env val Method function Argument jmp_buf env Pointer to the area where environment is restored int val Value returned as a result of setjmp ReturnValue No value is returned Description Restores the environment from the area indicated in env Program control is passed to the statement following that from which setjmp was called The value specified in val is returned as the result of setjmp However if val is 0 it is converted to 1 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 251 R32C 100 Series C Compiler Package V 1 02 E Standard Library malloc Memory Management Functions Function Allocates a memory area Format include lt stdlib h gt void _far malloc nbytes Method function Argument size_t nbytes Size of memory area in bytes to be allocated ReturnValue Returns NULLif a memory area of the specified size could not be allocated Description Dynamically allocates memory areas Rule malloc performs the following two checks to secure memory in the appropriate location 1 Ifmemory areas have been freed with free If the amount of memory to be secured is smaller
349. ut Files Option Function dirdirectory name Specifies the destination directory of the file s absolute module file map file etc generated by In100 ofile name Specifies the name s of the file s absolute module file map file etc generated by In100 This option can also be used to specify the destination directory This option can also be used to specify the file name includes the path Do not specify the filename extension dirdirectory name Output file specification Function This option allows you to specify an output destination directory for the output file Syntax ncl00A dirdirectory name Notes The source file information used for debugging is generated starting from the directory from which the compiler was invoked the current directory Therefore if output files were generated in different directories the debugger etc must be notified of the directory from which the compiler was invoked ofile name Output file specification Function Specifies the name s of the file s absolute module file map file etc generated by In100 This option can also be used to specify the file name includes the path You must not specify the filename extension Syntax ncl00A ofi e nameA lt C source file gt REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 67 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference A 2 3 Version Information Display Option Table A 3 shows the comman
350. v l R2R0 4 FB ans lt 7 ine 10 H HC_SRC exitd E1 Figure D 15 Assembly language sample program 1 REJ10J2009 0200 Rev 2 00 2010 04 01 200 2tENESAS R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules HHH FUNCTION func HHH FRAME AUTO u size 4 offset 4 H REGISTERARG x size 4 REGISTER R2RO H REGISTERARG y size 4 REGISTER R3R1 REGISTERARG z size 4 REGISTER R6R4 HH ARG Size 0 Auto Size 4 Context Size 8 align line 13 H HC_SRC glib func func enter 04H lt 4 ine 15 H C_ SRC int s 0x4444 mov l 00004444H R7R5 s line 16 HHHC SRC int t 0x5555 mov 00005555H A0 t ine 17 H C_ SRC int u 0x6666 mov l 00006666H 4 FB u line 18 H C_SRC suM S t U X Y Z add AO R7R5 t add l 4 FB R7R5 u add l R7R5 R2R0 add l R2R0 R3R1 mov l R6R4 R2RO ZZ add l R3R1 R2R0 sum 55 line 19 HHHC SRC return sum exitd lt 6 E2 Figure D 16 Assembly language sample program 2 Figure D 17 and Figure D 18 show the stack and register behaviors during the processes 1 2 and 3 in Figure D 15 Ge process at entry to the function main and process to call the function func and during the processes 4 5 6 and 7 Ge process to build the stack frame used in the function func and process to return from the function func to the function main respectively REJ10J2009 0200 Rev 2 00 2010 04 01 RENES
351. valid types other than the integer and enumerated types are used for the expression of a switch statement Use the integer type or enumerated type syntax error This is a syntax error Write the description correctly System Error eiu ey It does not normally occur This is an internal error This error may occur pursuant to one of errors that occurred before it If this error occurs even after eliminating all errors that occurred before it please send the content of the error message to Renesas Solutions Corp as you contact too big data length e An attempt is made to get an address exceeding the 32 bit range Make sure the set values are within the address range of the microcomputer used too big address e Anattempt is made to set an address exceeding the 32 bit range Make sure the set values are within the address range of the microcomputer used too many storage class of typedef Storage class specifiers such as extern typedef static auto register are described more than twice in declaration Do not describe a storage class specifier more than twice REJ10J2009 0200 Rev 2 00 2010 04 01 318 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 18 ccom100Error message 12 13 Error message Description and countermeasure type redeclaration of variable name The variable is defined with different types each time Always use the same type when dec
352. variable string terminator not added Since the variable to be initialized and the size of the initialize expression are equal O cannot be affixed to the character string Increase a element number of array struct or union member s address can t has no near far information near or far is used as arrangement position information of members variables of a struct or union Do not specify near and far for members REJ10J2009 0200 Rev 2 00 2010 04 01 327 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 27 ccom100 Warning Messages 8 9 Warning message Description and countermeasure task function called The function specified by pragma TASK is called Be careful not to call a task function task function can t return value The function specified by pragma TASK is using a return value The function specified by pragma TASK cannot use return values Delete the return value task function has invalid argument y The function specified with pragma TASK uses arguments Any function specified with pragma TASK cannot use arguments Delete the arguments this comparison is always false Comparison is made that always results in false Check the conditional expression this comparison is always true Comparison is made that always results in true Check the conditional expression this feature not supported now ignored bo
353. ve defined members Define complete structs or unions first incomplete struct member An attempt is made to reference an struct member of incomplete Define complete structs or unions first incomplete struct initialized An attempt is made to initialize an array of incomplete structs or unions that do not have defined members Define complete structs or unions first incomplete struct return function call An attempt is made to call a function that has as a return value the of incomplete struct or union that does not have defined members Define a complete struct or union first incomplete struct union s member access An attempt is made to reference members of an incomplete struct or union that do not have defined members Define a complete struct or union first incomplete struct union tagname s member access An attempt is made to reference members of an incomplete struct or union that do not have defined members Define a complete struct or union first inline function have invalid argument or return code inline function has an invalid argument or an invalid return value Write the argument or an invalid return value correctly inline function is called as normal function before The function declared in storage class inline is called as an ordinary function Always be sure to define an inline function before using it REJ10J2009 0200 Rev 2 00 2010 04 01 311 7tENESAS
354. vel 3 I O function s Set the stream C Compile the modified source program s Figure E 2 Example Sequence of Modifying I O Functions a Modifying Level 3 I O Function The level 3 I O functions perform 1 byte I O via the R32C 100 series I O ports The level 3 I O functions include _sget and _sput which perform I O via the serial communications circuits UART and _pput which performs I O via the Centronics communications circuit 1 Circuit settings o Clock frequency 20MHz 2 Initial serial communications settings e Use UART1 Baud rate 9600bps Data size 8 bits Parity None Stop bits 2 bits The initial serial communications settings are made in the init function init c REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 292 R32C 100 Series C Compiler Package V 1 02 E Standard Library The level 3 I O functions are written in the C library source file device c Table E 14 lists the specifications of these functions Table E 14 Specifications of Level 3 Functions Input functions Parameters Return value int type _sget None If no error occurs returns the input character Returns EOF if an _sput error occurs _pput Output unctions Parameters int type Return value int type _sput Character to Tf no error occurs returns 1 _pput output Returns EOF if an error occurs Serial communication is set to UART1 in the R32C 100 series s two UARTs device c is written so that the
355. volatile b C a b 0 whether a c or a b a b whether a bora b 1 Figure3 7 Example of Ambiguously Interpreted volatile qualifier 3 1 4 Precautions on Using register Variables a register qualification and compile option fenable_register fER If the compile option fenable_register fER is specified the variables that are register qualified so as to satisfy specific conditions can be forcibly assigned to registers This facility is provided for improving generated codes without relying on optimization Because improper use of this facility produces negative effects always be sure to examine generated codes before deciding to use it b About register qualification and optimization options The compiler automatically assigns variables to the registers This assignment facility is unaffected by a register qualification 3 1 5 About Startup Handling Startup may need to be modified depending on the type of microcomputer you are using or depending on your application system For modifications pertinent to the type of microcomputer consult the data book etc for your microcomputer and correct the startup file included with the compiler package before use REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 53 R32C 100 Series C Compiler Package V 1 02 3 Programming Technique 3 2 For Greater Code Efficiency 3 2 1 Programming Techniques for Greater Code Efficiency a Using Prototype declaration Effici
356. with unexpected results when performing the char type calculation shown in Figure D 9 void func void char c1 c2 c3 c1 c2 2 03 Figure D 9 Example of C Program To generate code that maximizes code efficiency and maximizes speed NC100 does not by default extend char types to int types The default can however be overridden using the nc100 compile driver command line option fansi or fextend_to_int fETT to achieve the same sign extension as in standard C If you do not use the fansi or fextend_to_int fETI option and your program assigns the result of a calculation to a char type as in Figure D 9 make sure that the maximum or minimum value for a char type does not result in an overflow in the calculation 1 The ranges of values that can be expressed as char types in NC100 are as follows unsigned char type scccseseceeeeees between 0 and 255 signed Char type eccecsssseseeceeseeseees between 128 and 127 REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 191 R32C 100 Series C Compiler Package V 1 02 D C Language Specification Rules D 3 Function Call Rules D 3 1 Rules of Return Value When returning a return value from a function the system uses a register to return that value for the integer pointer and floating point types Table D 4 shows rules on calls regarding return values Table D 4 Return Value related Calling Rules Type of return value Rules c
357. ws command prompt G 3 1 Input format To start gensni specify an input file name and startup option according to the input format shown below gensniA Command option AAbsolute module file extension x30 Denotes the prompt lt gt Denotes the essential items Denotes the items that need to be written when necessary A Denotes a space When writing multiple startup options separate each with a space Figure G 2 gensni command input format To use gensni specify both of the following in the startup options of this compiler o Inspector information output finfo option e Debug information output to generate absolute module files extension x30 g option An input example is shown below In the input example here the following option is specified in gensni J Information output to a specified file ss o option By default the information is output to a file named after the input file by changing the file extension from x30 to sni Generate an absolute module file nc100 g finfo ncrt0 a30 sample c lt RET gt R32C 100 Series C Compiler V X XX Release XX Copyright C XXXX XXXX XXXX Renesas Electronics Corp and Renesas Solutions Corp All rights reserved nert0 a30 sample c Generate stack information file gensni o sample ncrt0 x30 lt RET gt sample sni is created lt RET gt The input of the return key is shown Figure G 3 gensni com
358. y a Notes on Regarding Standard Header File When using functions in the standard library always be sure to include the specified standard header file If this header file is not included the integrity of arguments and return values will be lost making the program unable to operate normally b Notes on Regarding Optimization of Standard Library If you specify any of optimization options O 3 5 OS or OR the system performs optimization for the standard functions This optimization can be suppressed by specifying Ono_stdlib Such suppression of optimization is necessary when you use a user function that bear the same name as one of the standard library functions 12 Inline padding of functions Regarding functions strepy and memcpy the system performs inline padding of functions if the conditions inTable E 13 are met Table E 13 Optimization Conditions for Standard Library Functions Function Name Optimization Condition Description Example strcpy First argument far pointer strepy str sample Second argument string constant memcpy First argument far pointer memepy str sample 6 Second argument far pointer memcepy lt str fp 6 Third argument constant REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 290 R32C 100 Series C Compiler Package V 1 02 E Standard Library E 3 Modifying Standard Library The NC100 package includes a sophisticated function library which includes functions such as the
359. y a character string or identifier other than the above after pragma it will be ignored By default no warning is output if you specify an unsupported pragma function Warnings are only output if you specify the ncl00 command line option Wunknown_pragma WUP REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 184 R32C 100 Series C Compiler Package V 1 02 C Overview of C Language Specifications undef Function Nullifies an identifier that is defined as macro Format undefAidentifier Description Nullifies an identifier that is defined as macro The following four identifiers are compiler reserved words Because these identifiers must be permanently valid do not undefine them with undef ae tl eee eee eee ere Name of source file AUNE eucsantacuns Current source file line No _ DATE oscscanhaacicaiad Date compiled mm dd yyyy MME _ nssssssssssssssrsrrssenren Time compiled hh mm ss C 3 3 Predefined Macros The following macros are predefined in NC 100 R32C100 e NC100 INT 16_ When compilation option f116 fint_16 is used it is defined _ CHAR SIGNED_ When compilation option fSC fsigned_char is used it is defined C 3 4 Usage of predefined Macros The predefined macros are used to for example use preprocess commands to switch machine dependent code in non NC100 C programs ifdef NC100 pragma ADDRESS port0 2H pragma ADDRESS porti 3H else pragma AD portA 0x5
360. y Copies a character string strncpy Copies a character string n characters Concatenate strcat Concatenates character strings strncat Concatenates character strings n characters Compare strcmp Compares character strings strcoll Compares character strings using locale information stricmp Compares character strings All alphabets are handled as upper case letters strncmp Compares character strings n characters strnicmp Compares character strings n characters All alphabets are handled as upper case letters Search strchr Searches the specified character beginning with the top of the character string strcspn Calculates the length number of unspecified characters that are not found in the other character string strpbrk Searches the specified character in a character string from the other character string strrchr Searches the specified character from the end of a character string strspn Calculates the length number of specified characters that are found in the other character string strstr Searches the specified character from a character string strtok Divides some character string from a character string into tokens Length strlen Calculates the number of characters in a character string Convert strerror Converts an error number into a character string strxfrm Converts a character string using locale information Initialize bzero Initializes a memory area by clearing it to zero Copy bcopy Copies characters f
361. y real number ReturnValue Returns the hyperbolic cosine of given real number x REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 230 R32C 100 Series C Compiler Package V 1 02 E Standard Library div Function Divides an int type integer and calculates the remainder Format include lt stdlib h gt div_t div number denom Method function Argument int number Dividend int denom Divisor ReturnValue Returns the quotient derived by dividing number by denom and the remainder of the division Description Returns the quotient derived by dividing number by denom and the remainder of the division in structure div_t div tis defined in stdlib h This structure consists of members int quot and int rem exp Mathematical Functions Function Calculates exponential function Format include lt math h gt double exp x Method function Argument double x arbitrary real number ReturnValue Returns the calculation result of an exponential function of given real number x REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 231 R32C 100 Series C Compiler Package V 1 02 E Standard Library fabs Function Calculates the absolute value of a double precision floating point number Format include lt math h gt double fabs x Method function Argument double x arbitrary real number ReturnValue Returns the absolute value of a double precision floating point number feof Function Checks EOF End of F
362. y result 3 Inline functions cannot be called indirectly If any indirect call is encountered a compile error may result 4 Inline functions cannot be called recursively If any recursive call is encountered a compile error may result 5 If multiple inline expansion files are specified inline expansion is performed in the order in which the files are specified If the inline expand option is specified as main c Ofile_inline a c b c c c it is processed assuming the file configuration shown below Since processing is performed in the direction of the arrow the forward referenced functions in a c b c and c c each are not inline expanded 6 This option applies to the program section only If section names are changed by pragma SECTION functions are not inline expanded across a file boundary 7 The static functions defined by pragma _ASMMACRO that begin with the underbar _ those defined in asmmacro h and string h are inline expanded REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 75 R32C 100 Series C Compiler Package V 1 02 A Command Option Reference Oinline_line inline expansion line Ol L inline expansion line Inline expansion Function This option changes the size number of lines of the function to be inline expanded for the option Ofile_inline Oglobal_to_inline or Ostatic_to_inline specified in the program When this option is omitted inline expansion is performed on on
363. ype floating point number atoi Converts a character string into an int type integer atol Converts a character string into a long type integer bsearch Performs binary search in an array calloc Allocates a memory area and initializes it to zero 0 div Divides an int type integer and calculates the remainder free Frees the allocated memory area labs Calculates the absolute value of a long type integer Idiv Divides a long type integer and calculates the remainder malloc Allocates a memory area mblen Calculates the length of a multibyte character string mbstowcs Converts a multibyte character string into a wide character string mbtowc Converts a multibyte character into a wide character qsort Sorts elements in an array realloc Changes the size of an allocated memory area strtod Converts a character string into a double type integer strtol Converts a character string into a long type integer strtoul Converts a character string into an unsigned long type integer westombs Converts a wide character string into a multibyte character string wctomb Converts a wide character into a multibyte character REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 214 R32C 100 Series C Compiler Package V 1 02 E Standard Library ps Function Declares the prototypes of string handling functions and memory handling functions Type Type Contents Copy strcp
364. ysis in the ccom100 compiler the EUC codes are then converted for output as specified in the environment variable REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 119 R32C 100 Series C Compiler Package V 1 02 B Extended Functions Reference B 3 3 Japanese Characters in Character Strings Figure B 40 shows the format for including Japanese characters in character strings RFLP Figure B 40 Format of Kanji code Description in Character Strings If you write Japanese using the format L FXI as with normal character strings it is processed as a pointer type to a char type when manipulating the character string You therefore cannot manipulate them as 2 byte characters To process the Japanese as 2 byte characters precede the character string with L and process it as a pointer type to a wchar t type wchar_t types are defined typedef as unsigned short types in the standard header file stdlib h Figure B 41 shows an example of a Japanese character string include lt stdlib h gt void func void wchart JC 4 L XF JI remainder omitted Figure B 41 Example of Japanese Character Strings Description Figure B 42 is a memory map of the character string initialized in 1 in Figure B 41 JC O JCf1 JC 2 Jos address higher Figure B 42 Memory Location of wchar_t Type Character Strings REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 120 R32C 100 Series C Compiler Package V 1 0
365. zation REJ10J2009 0200 Rev 2 00 2010 04 01 RENESAS 280 R32C 100 Series C Compiler Package V 1 02 E Standard Library strstr String Handling Functions Function Searches the specified character from a character string Format include lt string h gt char _far strstr s1 s2 Method function Argument const char _far s1 Pointer to the character string to be searched in const char _far s2 Pointer to the character string of the character to be searched for ReturnValue Returns the position pointer where the specified character is found Returns NULL when the specified character cannot be found Description Returns the location pointer of the first character string s2 from the beginning of area s1 When you specify options O 3 5 OR or OS the system may selects another functions with good code efficiency by optimization strtod Character String Value Convert Functions Function Converts a character string into a double type integer Format include lt string h gt double strtod s endptr Method function Argument const char _far s Pointer to the converted character string char _far _far endptr Pointer to the remaining character strings that have not been converted ReturnValue ReturnValue 0L Does not constitute a number ReturnValue OL Returns the configured number in double type Description When you specify options O 3 5 OR or OS the system may selects another functio
366. zer on char array The initialization expression contains an error Check to see if the number of initialize expressions in the parentheses matches the number of array elements and the number of structure members invalid initializer on scalar The initialization expression contains an error Check to see if the number of initialize expressions in the parentheses matches the number of array elements and the number of structure members invalid initializer on struct The initialization expression contains an error Check to see if the number of initialization expressions in the parentheses matches the number of array elements and the number of structure members invalid initializer too many brace Too many braces are used in a scalar type initialization expression of the auto storage class Reduce the number of braces used invalid Ivalue The left side of the assignment statement is not lvalue Write a substitutable expression on the left side of the statement invalid lvalue at operator The left side of the assignment statement is not lvalue Write a substitutable expression on the left side of the statement REJ10J2009 0200 Rev 2 00 2010 04 01 315 7tENESAS R32C 100 Series C Compiler Package V 1 02 F Error Messages Table F 15 ccom100Error message 9 13 Error message Description and countermeasure invalid member e The member reference contains an error

R32C/100 Series C Compiler Package V.1.02 C Compiler User`s

Contents

Download Pdf Manuals

Related Search

Related Contents