1 #LyX 1.3 created this file. For more info see http://www.lyx.org/
7 pdftitle={SDCC Compiler User Guide},
8 pdfauthor={SDCC development team},
9 pdfsubject={installation, user manual},
10 pdfkeywords={8032, 8051, ansi, c, compiler, CPU, DS390,
11 embedded, GPL, HC08, manual, mcs51, PIC, Z80},
13 linkcolor=blue] {hyperref}
17 \emergencystretch=30pt
22 \inputencoding default
25 \paperfontsize default
27 \papersize letterpaper
32 \use_numerical_citations 0
33 \paperorientation portrait
40 \paragraph_separation indent
42 \quotes_language swedish
50 Please note: double dashed longoptions (e.g.
51 --version) are written this way: -
65 three consecutive dashes simply result in a long resp.
69 Architecture specific stuff (like memory models, code examples) should maybe
73 into seperate sections/chapters/appendices (it is hard to document PIC or
77 a 8051 centered document) - for now simply add.
80 SDCC Compiler User Guide
94 The above strings enclosed in $ are automatically updated by cvs
98 \begin_inset LatexCommand \tableofcontents{}
131 ompiler) is an open source, retargettable, optimizing ANSI-C compiler by
136 designed for 8 bit Microprocessors.
137 The current version targets Intel MCS51 based Microprocessors (8031, 8032,
139 \begin_inset LatexCommand \index{8031, 8032, 8051, 8052, mcs51 CPU}
143 , etc.), Dallas DS80C390 variants, Motorola HC08 and Zilog Z80 based MCUs.
144 It can be retargetted for other microprocessors, support for Microchip
145 PIC, Atmel AVR is under development.
146 The entire source code for the compiler is distributed under GPL.
148 \begin_inset LatexCommand \index{asXXXX (as-gbz80, as-hc08, asx8051, as-z80)}
153 \begin_inset LatexCommand \index{aslink}
157 , an open source retargettable assembler & linker.
158 SDCC has extensive language extensions suitable for utilizing various microcont
159 rollers and underlying hardware effectively.
164 In addition to the MCU specific optimizations SDCC also does a host of standard
168 global sub expression elimination,
171 loop optimizations (loop invariant, strength reduction of induction variables
175 constant folding & propagation,
181 dead code elimination
191 For the back-end SDCC uses a global register allocation scheme which should
192 be well suited for other 8 bit MCUs.
197 The peep hole optimizer uses a rule based substitution mechanism which is
203 Supported data-types are:
206 char (8 bits, 1 byte),
209 short and int (16 bits, 2 bytes),
212 long (32 bit, 4 bytes)
219 The compiler also allows
221 inline assembler code
223 to be embedded anywhere in a function.
224 In addition, routines developed in assembly can also be called.
228 SDCC also provides an option (-
238 -cyclomatic) to report the relative complexity of a function.
239 These functions can then be further optimized, or hand coded in assembly
245 SDCC also comes with a companion source level debugger SDCDB, the debugger
246 currently uses ucSim a freeware simulator for 8051 and other micro-controllers.
251 The latest version can be downloaded from
252 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/snap.php}
262 Please note: the compiler will probably always be some steps ahead of this
267 \begin_inset LatexCommand \index{Status of documentation}
277 Obviously this has pros and cons
286 All packages used in this compiler system are
294 ; source code for all the sub-packages (pre-processor, assemblers, linkers
295 etc) is distributed with the package.
296 This documentation is maintained using a freeware word processor (LyX).
298 This program is free software; you can redistribute it and/or modify it
299 under the terms of the GNU General Public License
300 \begin_inset LatexCommand \index{GNU General Public License, GPL}
304 as published by the Free Software Foundation; either version 2, or (at
305 your option) any later version.
306 This program is distributed in the hope that it will be useful, but WITHOUT
307 ANY WARRANTY; without even the implied warranty
308 \begin_inset LatexCommand \index{warranty}
312 of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
313 See the GNU General Public License for more details.
314 You should have received a copy of the GNU General Public License along
315 with this program; if not, write to the Free Software Foundation, 59 Temple
316 Place - Suite 330, Boston, MA 02111-1307, USA.
317 In other words, you are welcome to use, share and improve this program.
318 You are forbidden to forbid anyone else to use, share and improve what
320 Help stamp out software-hoarding!
323 Typographic conventions
324 \begin_inset LatexCommand \index{Typographic conventions}
331 Throughout this manual, we will use the following convention.
332 Commands you have to type in are printed in
340 Code samples are printed in
345 Interesting items and new terms are printed in
350 Compatibility with previous versions
353 This version has numerous bug fixes compared with the previous version.
354 But we also introduced some incompatibilities with older versions.
355 Not just for the fun of it, but to make the compiler more stable, efficient
357 \begin_inset LatexCommand \index{ANSI-compliance}
362 \begin_inset LatexCommand \ref{sub:ANSI-Compliance}
366 for ANSI-Compliance).
372 short is now equivalent to int (16 bits), it used to be equivalent to char
373 (8 bits) which is not ANSI compliant
376 the default directory for gcc-builds where include, library and documentation
377 files are stored is now in /usr/local/share
380 char type parameters to vararg functions are casted to int unless explicitly
397 will push a as an int and as a char resp.
410 -regextend has been removed
423 -noregparms has been removed
436 -stack-after-data has been removed
441 <pending: more incompatibilities?>
447 What do you need before you start installation of SDCC? A computer, and
449 The preferred method of installation is to compile SDCC from source using
451 For Windows some pre-compiled binary distributions are available for your
453 You should have some experience with command line tools and compiler use.
459 The SDCC home page at
460 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/}
464 is a great place to find distribution sets.
465 You can also find links to the user mailing lists that offer help or discuss
466 SDCC with other SDCC users.
467 Web links to other SDCC related sites can also be found here.
468 This document can be found in the DOC directory of the source package as
470 A pdf version of this document is available at
471 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/doc/sdccman.pdf}
476 Some of the other tools (simulator and assembler) included with SDCC contain
477 their own documentation and can be found in the source distribution.
478 If you want the latest unreleased software, the complete source package
479 is available directly by anonymous CVS on cvs.sdcc.sourceforge.net.
482 Wishes for the future
485 There are (and always will be) some things that could be done.
486 Here are some I can think of:
493 char KernelFunction3(char p) at 0x340;
501 \begin_inset LatexCommand \index{code banking (not supported)}
511 If you can think of some more, please see the section
512 \begin_inset LatexCommand \ref{sub:Requesting-Features}
516 about filing feature requests
517 \begin_inset LatexCommand \index{Requesting features}
522 \begin_inset LatexCommand \index{Feature request}
532 \begin_inset LatexCommand \index{Installation}
539 For most users it is sufficient to skip to either section
540 \begin_inset LatexCommand \ref{sub:Building-SDCC-on-Linux}
545 \begin_inset LatexCommand \ref{sub:Windows-Install}
550 More detailled instructions follow below.
554 \begin_inset LatexCommand \index{Options SDCC configuration}
561 The install paths, search paths and other options are defined when running
563 The defaults can be overridden by:
565 \labelwidthstring 00.00.0000
577 -prefix see table below
579 \labelwidthstring 00.00.0000
591 -exec_prefix see table below
593 \labelwidthstring 00.00.0000
605 -bindir see table below
607 \labelwidthstring 00.00.0000
619 -datadir see table below
621 \labelwidthstring 00.00.0000
623 docdir environment variable, see table below
625 \labelwidthstring 00.00.0000
627 include_dir_suffix environment variable, see table below
629 \labelwidthstring 00.00.0000
631 lib_dir_suffix environment variable, see table below
633 \labelwidthstring 00.00.0000
635 sdccconf_h_dir_separator environment variable, either / or
640 This character will only be used in sdccconf.h; don't forget it's a C-header,
641 therefore a double-backslash is needed there.
643 \labelwidthstring 00.00.0000
655 -disable-mcs51-port Excludes the Intel mcs51 port
657 \labelwidthstring 00.00.0000
669 -disable-gbz80-port Excludes the Gameboy gbz80 port
671 \labelwidthstring 00.00.0000
683 -disable-z80-port Excludes the z80 port
685 \labelwidthstring 00.00.0000
697 -disable-avr-port Excludes the AVR port
699 \labelwidthstring 00.00.0000
711 -disable-ds390-port Excludes the DS390 port
713 \labelwidthstring 00.00.0000
725 -disable-hc08-port Excludes the HC08 port
727 \labelwidthstring 00.00.0000
739 -disable-pic-port Excludes the PIC port
741 \labelwidthstring 00.00.0000
753 -disable-xa51-port Excludes the XA51 port
755 \labelwidthstring 00.00.0000
767 -disable-ucsim Disables configuring and building of ucsim
769 \labelwidthstring 00.00.0000
781 -disable-device-lib-build Disables automatically building device libraries
783 \labelwidthstring 00.00.0000
795 -disable-packihx Disables building packihx
797 \labelwidthstring 00.00.0000
809 -enable-libgc Use the Bohem memory allocator.
810 Lower runtime footprint.
813 Furthermore the environment variables CC, CFLAGS, ...
814 the tools and their arguments can be influenced.
815 Please see `configure -
825 -help` and the man/info pages of `configure` for details.
829 The names of the standard libraries STD_LIB, STD_INT_LIB, STD_LONG_LIB,
830 STD_FP_LIB, STD_DS390_LIB, STD_XA51_LIB and the environment variables SDCC_DIR_
831 NAME, SDCC_INCLUDE_NAME, SDCC_LIB_NAME are defined by `configure` too.
832 At the moment it's not possible to change the default settings (it was
833 simply never required).
837 These configure options are compiled into the binaries, and can only be
838 changed by rerunning 'configure' and recompiling SDCC.
839 The configure options are written in
843 to distinguish them from run time environment variables (see section search
849 \begin_inset Quotes sld
853 \begin_inset Quotes srd
856 are used by the SDCC team to build the official Win32 binaries.
857 The SDCC team uses Mingw32 to build the official Windows binaries, because
864 a gcc compiler and last but not least
867 the binaries can be built by cross compiling on Sourceforge's compile farm.
870 See the examples, how to pass the Win32 settings to 'configure'.
871 The other Win32 builds using Borland, VC or whatever don't use 'configure',
872 but a header file sdcc_vc_in.h is the same as sdccconf.h built by 'configure'
884 <lyxtabular version="3" rows="8" columns="3">
886 <column alignment="block" valignment="top" leftline="true" width="0in">
887 <column alignment="block" valignment="top" leftline="true" width="0in">
888 <column alignment="block" valignment="top" leftline="true" rightline="true" width="0in">
889 <row topline="true" bottomline="true">
890 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
898 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
906 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
916 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
926 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
934 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
946 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
956 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
966 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
978 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
988 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1000 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1015 <row topline="true">
1016 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1026 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1038 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1049 <row topline="true">
1050 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1060 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1072 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1087 <row topline="true">
1088 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1098 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1106 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1115 <row topline="true" bottomline="true">
1116 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1126 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1134 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1152 'configure' also computes relative paths.
1153 This is needed for full relocatability of a binary package and to complete
1154 search paths (see section search paths below):
1160 \begin_inset Tabular
1161 <lyxtabular version="3" rows="4" columns="3">
1163 <column alignment="block" valignment="top" leftline="true" width="0in">
1164 <column alignment="block" valignment="top" leftline="true" width="0in">
1165 <column alignment="block" valignment="top" leftline="true" rightline="true" width="0in">
1166 <row topline="true" bottomline="true">
1167 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1175 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1183 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1192 <row topline="true" bottomline="true">
1193 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1203 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1211 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1220 <row bottomline="true">
1221 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1231 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1239 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1248 <row bottomline="true">
1249 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1259 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1267 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1300 \begin_inset Quotes srd
1304 \begin_inset Quotes srd
1318 \begin_inset Quotes srd
1322 \begin_inset Quotes srd
1350 To cross compile on linux for Mingw32 (see also 'sdcc/support/scripts/sdcc_mingw
1359 \begin_inset Quotes srd
1362 i586-mingw32msvc-gcc
1363 \begin_inset Quotes srd
1367 \begin_inset Quotes srd
1370 i586-mingw32msvc-g++
1371 \begin_inset Quotes srd
1379 \begin_inset Quotes srd
1382 i586-mingw32msvc-ranlib
1383 \begin_inset Quotes srd
1391 \begin_inset Quotes srd
1394 i586-mingw32msvc-strip
1395 \begin_inset Quotes srd
1413 \begin_inset Quotes srd
1417 \begin_inset Quotes srd
1435 \begin_inset Quotes srd
1439 \begin_inset Quotes srd
1447 \begin_inset Quotes srd
1451 \begin_inset Quotes srd
1459 \begin_inset Quotes srd
1463 \begin_inset Quotes srd
1471 \begin_inset Quotes srd
1475 \begin_inset Quotes srd
1482 sdccconf_h_dir_separator=
1483 \begin_inset Quotes srd
1495 \begin_inset Quotes srd
1512 -disable-device-lib-build
1540 -host=i586-mingw32msvc -
1550 -build=unknown-unknown-linux-gnu
1554 \begin_inset Quotes sld
1558 \begin_inset Quotes srd
1561 compile on Cygwin for Mingw32 (see also sdcc/support/scripts/sdcc_cygwin_mingw32
1570 \begin_inset Quotes srd
1574 \begin_inset Quotes srd
1582 \begin_inset Quotes srd
1586 \begin_inset Quotes srd
1604 \begin_inset Quotes srd
1608 \begin_inset Quotes srd
1626 \begin_inset Quotes srd
1630 \begin_inset Quotes srd
1638 \begin_inset Quotes srd
1642 \begin_inset Quotes srd
1650 \begin_inset Quotes srd
1654 \begin_inset Quotes srd
1662 \begin_inset Quotes srd
1666 \begin_inset Quotes srd
1673 sdccconf_h_dir_separator=
1674 \begin_inset Quotes srd
1686 \begin_inset Quotes srd
1706 'configure' is quite slow on Cygwin (at least on windows before Win2000/XP).
1717 -C' turns on caching, which gives a little bit extra speed.
1718 However if options are changed, it can be necessary to delete the config.cache
1723 \begin_inset LatexCommand \label{sub:Install-paths}
1728 \begin_inset LatexCommand \index{Install paths}
1734 \added_space_top medskip \align center
1736 \begin_inset Tabular
1737 <lyxtabular version="3" rows="5" columns="4">
1739 <column alignment="center" valignment="top" leftline="true" width="0">
1740 <column alignment="center" valignment="top" leftline="true" width="0">
1741 <column alignment="center" valignment="top" leftline="true" width="0">
1742 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
1743 <row topline="true" bottomline="true">
1744 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1754 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1764 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1774 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1785 <row topline="true">
1786 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1794 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1804 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1812 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1825 <row topline="true">
1826 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1834 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1841 $DATADIR/ $INCLUDE_DIR_SUFFIX
1844 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1849 /usr/local/share/sdcc/include
1852 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1865 <row topline="true">
1866 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1874 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1881 $DATADIR/$LIB_DIR_SUFFIX
1884 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1889 /usr/local/share/sdcc/lib
1892 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1905 <row topline="true" bottomline="true">
1906 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1914 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1924 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1929 /usr/local/share/sdcc/doc
1932 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1954 *compiler, preprocessor, assembler, and linker
1960 is auto-appended by the compiler, e.g.
1961 small, large, z80, ds390 etc
1964 The install paths can still be changed during `make install` with e.g.:
1967 make install prefix=$(HOME)/local/sdcc
1970 Of course this doesn't change the search paths compiled into the binaries.
1974 Moreover the install path can be changed by defining DESTDIR
1975 \begin_inset LatexCommand \index{DESTDIR}
1982 make install DESTDIR=$(HOME)/sdcc.rpm/
1985 Please note that DESTDIR must have a trailing slash!
1989 \begin_inset LatexCommand \label{sub:Search-Paths}
1994 \begin_inset LatexCommand \index{Search path}
2001 Some search paths or parts of them are determined by configure variables
2006 , see section above).
2007 Further search paths are determined by environment variables during runtime.
2010 The paths searched when running the compiler are as follows (the first catch
2016 Binary files (preprocessor, assembler and linker)
2022 \begin_inset Tabular
2023 <lyxtabular version="3" rows="4" columns="3">
2025 <column alignment="block" valignment="top" leftline="true" width="0in">
2026 <column alignment="block" valignment="top" leftline="true" width="0in">
2027 <column alignment="block" valignment="top" leftline="true" rightline="true" width="0in">
2028 <row topline="true" bottomline="true">
2029 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2037 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2045 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2054 <row topline="true">
2055 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2065 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2073 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2084 <row topline="true">
2085 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2090 Path of argv[0] (if available)
2093 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2101 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2110 <row topline="true" bottomline="true">
2111 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2119 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2127 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2152 \begin_inset Tabular
2153 <lyxtabular version="3" rows="6" columns="3">
2155 <column alignment="block" valignment="top" leftline="true" width="1.5in">
2156 <column alignment="block" valignment="top" leftline="true" width="1.5in">
2157 <column alignment="block" valignment="top" leftline="true" rightline="true" width="0in">
2158 <row topline="true" bottomline="true">
2159 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2167 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2175 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2184 <row topline="true">
2185 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2203 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2221 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2240 <row topline="true">
2241 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2249 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2257 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2266 <row topline="true">
2267 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2281 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2293 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2304 <row topline="true">
2305 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2323 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2373 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2386 <row topline="true" bottomline="true">
2387 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2403 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2408 /usr/local/share/sdcc/
2413 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2441 -nostdinc disables the last two search paths.
2451 With the exception of
2452 \begin_inset Quotes sld
2466 \begin_inset Quotes srd
2473 is auto-appended by the compiler (e.g.
2474 small, large, z80, ds390 etc.).
2481 \begin_inset Tabular
2482 <lyxtabular version="3" rows="6" columns="3">
2484 <column alignment="block" valignment="top" leftline="true" width="1.7in">
2485 <column alignment="block" valignment="top" leftline="true" width="1.2in">
2486 <column alignment="block" valignment="top" leftline="true" rightline="true" width="1.2in">
2487 <row topline="true" bottomline="true">
2488 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2496 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2504 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2513 <row topline="true">
2514 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2532 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2550 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2569 <row topline="true">
2570 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2582 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2594 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2609 <row topline="true">
2610 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2621 $LIB_DIR_SUFFIX/<model>
2624 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2638 <cell alignment="left" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2655 <row topline="true">
2656 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2671 $LIB_DIR_SUFFIX/<model>
2674 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2727 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2783 <row topline="true" bottomline="true">
2784 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2793 $LIB_DIR_SUFFIX/<model>
2796 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2801 /usr/local/share/sdcc/
2808 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2826 Don't delete any of the stray spaces in the table above without checking
2827 the HTML output (last line)!
2843 -nostdlib disables the last two search paths.
2847 \begin_inset LatexCommand \index{Building SDCC}
2854 Building SDCC on Linux
2855 \begin_inset LatexCommand \label{sub:Building-SDCC-on-Linux}
2864 Download the source package
2866 either from the SDCC CVS repository or from the nightly snapshots
2868 , it will be named something like sdcc
2879 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/snap.php}
2888 Bring up a command line terminal, such as xterm.
2893 Unpack the file using a command like:
2896 "tar -xvzf sdcc.src.tar.gz
2901 , this will create a sub-directory called sdcc with all of the sources.
2904 Change directory into the main SDCC directory, for example type:
2921 This configures the package for compilation on your system.
2937 All of the source packages will compile, this can take a while.
2953 This copies the binary executables, the include files, the libraries and
2954 the documentation to the install directories.
2955 Proceed with section
2956 \begin_inset LatexCommand \ref{sec:Testing-the-SDCC}
2963 Building SDCC on OSX 2.x
2966 Follow the instruction for Linux.
2970 On OSX 2.x it was reported, that the default gcc (version 3.1 20020420 (prerelease
2971 )) fails to compile SDCC.
2972 Fortunately there's also gcc 2.9.x installed, which works fine.
2973 This compiler can be selected by running 'configure' with:
2976 ./configure CC=gcc2 CXX=g++2
2979 Cross compiling SDCC on Linux for Windows
2982 With the Mingw32 gcc cross compiler it's easy to compile SDCC for Win32.
2983 See section 'Configure Options'.
2986 Building SDCC on Windows
2989 With the exception of Cygwin the SDCC binaries uCsim and sdcdb can't be
2991 They use Unix-sockets, which are not available on Win32.
2994 Building SDCC using Cygwin and Mingw32
2997 For building and installing a Cygwin executable follow the instructions
3003 \begin_inset Quotes sld
3007 \begin_inset Quotes srd
3010 Win32-binary can be built, which will not need the Cygwin-DLL.
3011 For the necessary 'configure' options see section 'configure options' or
3012 the script 'sdcc/support/scripts/sdcc_cygwin_mingw32'.
3016 In order to install Cygwin on Windows download setup.exe from
3017 \begin_inset LatexCommand \url[www.cygwin.com]{http://www.cygwin.com/}
3023 \begin_inset Quotes sld
3026 default text file type
3027 \begin_inset Quotes srd
3031 \begin_inset Quotes sld
3035 \begin_inset Quotes srd
3038 and download/install at least the following packages.
3039 Some packages are selected by default, others will be automatically selected
3040 because of dependencies with the manually selected packages.
3041 Never deselect these packages!
3050 gcc ; version 3.x is fine, no need to use the old 2.9x
3053 binutils ; selected with gcc
3059 rxvt ; a nice console, which makes life much easier under windoze (see below)
3062 man ; not really needed for building SDCC, but you'll miss it sooner or
3066 less ; not really needed for building SDCC, but you'll miss it sooner or
3070 cvs ; only if you use CVS access
3073 If you want to develop something you'll need:
3076 python ; for the regression tests
3079 gdb ; the gnu debugger, together with the nice GUI
3080 \begin_inset Quotes sld
3084 \begin_inset Quotes srd
3090 openssh ; to access the CF or commit changes
3093 autoconf and autoconf-devel ; if you want to fight with 'configure', don't
3094 use autoconf-stable!
3097 rxvt is a nice console with history.
3098 Replace in your cygwin.bat the line
3117 rxvt -sl 1000 -fn "Lucida Console-12" -sr -cr red
3120 -bg black -fg white -geometry 100x65 -e bash -
3133 Text selected with the mouse is automatically copied to the clipboard, pasting
3134 works with shift-insert.
3138 The other good tip is to make sure you have no //c/-style paths anywhere,
3139 use /cygdrive/c/ instead.
3140 Using // invokes a network lookup which is very slow.
3142 \begin_inset Quotes sld
3146 \begin_inset Quotes srd
3149 is too long, you can change it with e.g.
3155 SDCC sources use the unix line ending LF.
3156 Life is much easier, if you store the source tree on a drive which is mounted
3158 And use an editor which can handle LF-only line endings.
3159 Make sure not to commit files with windows line endings.
3160 The tabulator spacing
3161 \begin_inset LatexCommand \index{tabulator spacing (8 columns)}
3165 used in the project is 8.
3166 Although a tabulator spacing of 8 is a sensible choice for programmers
3167 (it's a power of 2 and allows to display 8/16 bit signed variables without
3168 loosing columns) the plan is to move towards using only spaces in the source.
3171 Building SDCC Using Microsoft Visual C++ 6.0/NET (MSVC)
3176 Download the source package
3178 either from the SDCC CVS repository or from the
3179 \begin_inset LatexCommand \url[nightly snapshots]{http://sdcc.sourceforge.net/snap.php}
3185 , it will be named something like sdcc
3192 SDCC is distributed with all the projects, workspaces, and files you need
3193 to build it using Visual C++ 6.0/NET (except for sdcdb.exe which currently
3194 doesn't build under MSVC).
3195 The workspace name is 'sdcc.dsw'.
3196 Please note that as it is now, all the executables are created in a folder
3200 Once built you need to copy the executables from sdcc
3204 bin before running SDCC.
3209 WARNING: Visual studio is very picky with line terminations; it expects
3210 the 0x0d, 0x0a DOS style line endings, not the 0x0a Unix style line endings.
3211 If you are getting a message such as "This makefile was not generated by
3212 Developer Studio etc.
3214 \begin_inset Quotes srd
3217 when opening the sdcc.dsw workspace or any of the *.dsp projects, then you
3218 need to convert the Unix style line endings to DOS style line endings.
3219 To do so you can use the
3220 \begin_inset Quotes sld
3224 \begin_inset Quotes srd
3227 utility freely available on the internet.
3228 Doug Hawkins reported in the sdcc-user list that this works:
3236 SDCC> unix2dos sdcc.dsw
3242 SDCC> for /R %I in (*.dsp) do @unix2dos "%I"
3246 In order to build SDCC with MSVC you need win32 executables of bison.exe,
3247 flex.exe, and gawk.exe.
3248 One good place to get them is
3249 \begin_inset LatexCommand \url[here]{http://unxutils.sourceforge.net}
3257 Download the file UnxUtils
3258 \begin_inset LatexCommand \index{UnxUtils}
3263 Now you have to install the utilities and setup MSVC so it can locate the
3265 Here there are two alternatives (choose one!):
3272 a) Extract UnxUtils.zip to your C:
3274 hard disk PRESERVING the original paths, otherwise bison won't work.
3275 (If you are using WinZip make certain that 'Use folder names' is selected)
3279 b) In the Visual C++ IDE click Tools, Options, select the Directory tab,
3280 in 'Show directories for:' select 'Executable files', and in the directories
3281 window add a new path: 'C:
3291 (As a side effect, you get a bunch of Unix utilities that could be useful,
3292 such as diff and patch.)
3299 This one avoids extracting a bunch of files you may not use, but requires
3304 a) Create a directory were to put the tools needed, or use a directory already
3312 b) Extract 'bison.exe', 'bison.hairy', 'bison.simple', 'flex.exe', and gawk.exe
3313 to such directory WITHOUT preserving the original paths.
3314 (If you are using WinZip make certain that 'Use folder names' is not selected)
3318 c) Rename bison.exe to '_bison.exe'.
3322 d) Create a batch file 'bison.bat' in 'C:
3326 ' and add these lines:
3346 _bison %1 %2 %3 %4 %5 %6 %7 %8 %9
3350 Steps 'c' and 'd' are needed because bison requires by default that the
3351 files 'bison.simple' and 'bison.hairy' reside in some weird Unix directory,
3352 '/usr/local/share/' I think.
3353 So it is necessary to tell bison where those files are located if they
3354 are not in such directory.
3355 That is the function of the environment variables BISON_SIMPLE and BISON_HAIRY.
3359 e) In the Visual C++ IDE click Tools, Options, select the Directory tab,
3360 in 'Show directories for:' select 'Executable files', and in the directories
3361 window add a new path: 'c:
3364 Note that you can use any other path instead of 'c:
3366 util', even the path where the Visual C++ tools are, probably: 'C:
3370 Microsoft Visual Studio
3375 So you don't have to execute step 'e' :)
3379 Open 'sdcc.dsw' in Visual Studio, click 'build all', when it finishes copy
3380 the executables from sdcc
3384 bin, and you can compile using SDCC.
3387 Building SDCC Using Borland
3390 From the sdcc directory, run the command "make -f Makefile.bcc".
3391 This should regenerate all the .exe files in the bin directory except for
3392 sdcdb.exe (which currently doesn't build under Borland C++).
3395 If you modify any source files and need to rebuild, be aware that the dependenci
3396 es may not be correctly calculated.
3397 The safest option is to delete all .obj files and run the build again.
3398 From a Cygwin BASH prompt, this can easily be done with the command (be
3399 sure you are in the sdcc directory):
3409 ( -name '*.obj' -o -name '*.lib' -o -name '*.rul'
3411 ) -print -exec rm {}
3420 or on Windows NT/2000/XP from the command prompt with the command:
3427 del /s *.obj *.lib *.rul
3430 from the sdcc directory.
3433 Windows Install Using a Binary Package
3434 \begin_inset LatexCommand \label{sub:Windows-Install}
3441 Download the binary package from
3442 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/snap.php}
3446 and unpack it using your favorite unpacking tool (gunzip, WinZip, etc).
3447 This should unpack to a group of sub-directories.
3448 An example directory structure after unpacking the mingw32 package is:
3453 bin for the executables, c:
3461 lib for the include and libraries.
3464 Adjust your environment variable PATH to include the location of the bin
3465 directory or start sdcc using the full path.
3468 Building the Documentation
3471 If the necessary tools (LyX, LaTeX, LaTeX2HTML) are installed it is as easy
3472 as changing into the doc directory and typing
3476 \begin_inset Quotes srd
3480 \begin_inset Quotes srd
3487 You're invited to make changes and additions to this manual (sdcc/doc/sdccman.ly
3490 \begin_inset LatexCommand \url{www.lyx.org}
3494 as editor this is straightforward.
3495 Prebuilt documentation in html and pdf format is available from
3496 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/snap.php}
3503 Reading the Documentation
3506 Currently reading the document in pdf format is recommended, as for unknown
3507 reason the hyperlinks are working there whereas in the html version they
3514 If you should know why please drop us a note
3518 You'll find the pdf version at
3519 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/doc/sdccman.pdf}
3525 This documentation is in some aspects different from a commercial documentation:
3529 It tries to document SDCC for several processor architectures in one document
3530 (commercially these probably would be separate documents/products).
3532 \begin_inset LatexCommand \index{Status of documentation}
3536 currently matches SDCC for mcs51 and DS390 best and does give too few informati
3538 Z80, PIC14, PIC16 and HC08.
3541 There are many references pointing away from this documentation.
3542 Don't let this distract you.
3544 was a reference like
3545 \begin_inset LatexCommand \url{www.opencores.org}
3549 together with a statement
3550 \begin_inset Quotes sld
3553 some processors which are targetted by SDCC can be implemented in a
3570 \begin_inset LatexCommand \index{fpga (field programmable gate array)}
3575 \begin_inset Quotes srd
3578 we expect you to have a quick look there and come back.
3579 If you read this you are on the right track.
3582 Some sections attribute more space to problems, restrictions and warnings
3583 than to the solution.
3586 The installation section and the section about the debugger is intimidating.
3589 There are still lots of typos and there are more different writing styles
3593 Testing the SDCC Compiler
3594 \begin_inset LatexCommand \label{sec:Testing-the-SDCC}
3601 The first thing you should do after installing your SDCC compiler is to
3617 \begin_inset LatexCommand \index{version}
3624 at the prompt, and the program should run and tell you the version.
3625 If it doesn't run, or gives a message about not finding sdcc program, then
3626 you need to check over your installation.
3627 Make sure that the sdcc bin directory is in your executable search path
3628 defined by the PATH environment setting (
3633 \begin_inset LatexCommand \ref{sub:Install-Trouble-shooting}
3640 Install trouble-shooting for suggestions
3643 Make sure that the sdcc program is in the bin folder, if not perhaps something
3644 did not install correctly.
3652 is commonly installed as described in section
3653 \begin_inset Quotes sld
3656 Install and search paths
3657 \begin_inset Quotes srd
3666 Make sure the compiler works on a very simple example.
3667 Type in the following test.c program using your favorite
3693 Compile this using the following command:
3702 If all goes well, the compiler will generate a test.asm and test.rel file.
3703 Congratulations, you've just compiled your first program with SDCC.
3704 We used the -c option to tell SDCC not to link the generated code, just
3705 to keep things simple for this step.
3713 The next step is to try it with the linker.
3723 If all goes well the compiler will link with the libraries and produce
3724 a test.ihx output file.
3729 (no test.ihx, and the linker generates warnings), then the problem is most
3738 usr/local/share/sdcc/lib directory
3745 \begin_inset LatexCommand \ref{sub:Install-Trouble-shooting}
3752 Install trouble-shooting for suggestions).
3760 The final test is to ensure
3768 header files and libraries.
3769 Edit test.c and change it to the following:
3786 strcpy(str1, "testing");
3793 Compile this by typing
3800 This should generate a test.ihx output file, and it should give no warnings
3801 such as not finding the string.h file.
3802 If it cannot find the string.h file, then the problem is that
3806 cannot find the /usr/local/share/sdcc/include directory
3813 \begin_inset LatexCommand \ref{sub:Install-Trouble-shooting}
3820 Install trouble-shooting section for suggestions).
3838 \begin_inset LatexCommand \index{-\/-print-search-dirs}
3842 to find exactly where SDCC is looking for the include and lib files.
3845 Install Trouble-shooting
3846 \begin_inset LatexCommand \label{sub:Install-Trouble-shooting}
3851 \begin_inset LatexCommand \index{Install trouble-shooting}
3858 If SDCC does not build correctly
3861 A thing to try is starting from scratch by unpacking the .tgz source package
3862 again in an empty directory.
3870 ./configure 2>&1 | tee configure.log
3884 make 2>&1 | tee make.log
3891 If anything goes wrong, you can review the log files to locate the problem.
3892 Or a relevant part of this can be attached to an email that could be helpful
3893 when requesting help from the mailing list.
3897 \begin_inset Quotes sld
3901 \begin_inset Quotes srd
3908 \begin_inset Quotes sld
3912 \begin_inset Quotes srd
3915 command is a script that analyzes your system and performs some configuration
3916 to ensure the source package compiles on your system.
3917 It will take a few minutes to run, and will compile a few tests to determine
3918 what compiler features are installed.
3922 \begin_inset Quotes sld
3926 \begin_inset Quotes srd
3932 This runs the GNU make tool, which automatically compiles all the source
3933 packages into the final installed binary executables.
3937 \begin_inset Quotes sld
3941 \begin_inset Quotes erd
3947 This will install the compiler, other executables libraries and include
3948 files into the appropriate directories.
3950 \begin_inset LatexCommand \ref{sub:Install-paths}
3956 \begin_inset LatexCommand \ref{sub:Search-Paths}
3961 about install and search paths.
3963 On most systems you will need super-user privileges to do this.
3969 SDCC is not just a compiler, but a collection of tools by various developers.
3970 These include linkers, assemblers, simulators and other components.
3971 Here is a summary of some of the components.
3972 Note that the included simulator and assembler have separate documentation
3973 which you can find in the source package in their respective directories.
3974 As SDCC grows to include support for other processors, other packages from
3975 various developers are included and may have their own sets of documentation.
3979 You might want to look at the files which are installed in <installdir>.
3980 At the time of this writing, we find the following programs for gcc-builds:
3984 In <installdir>/bin:
3987 sdcc - The compiler.
3990 sdcpp - The C preprocessor.
3993 asx8051 - The assembler for 8051 type processors.
4000 as-gbz80 - The Z80 and GameBoy Z80 assemblers.
4003 aslink -The linker for 8051 type processors.
4010 link-gbz80 - The Z80 and GameBoy Z80 linkers.
4013 s51 - The ucSim 8051 simulator.
4016 sdcdb - The source debugger.
4019 packihx - A tool to pack (compress) Intel hex files.
4022 In <installdir>/share/sdcc/include
4028 In <installdir>/share/sdcc/lib
4031 the subdirs src and small, large, z80, gbz80 and ds390 with the precompiled
4035 In <installdir>/share/sdcc/doc
4041 As development for other processors proceeds, this list will expand to include
4042 executables to support processors like AVR, PIC, etc.
4048 This is the actual compiler, it in turn uses the c-preprocessor and invokes
4049 the assembler and linkage editor.
4052 sdcpp - The C-Preprocessor
4056 \begin_inset LatexCommand \index{sdcpp (preprocessor)}
4060 is a modified version of the GNU preprocessor.
4061 The C preprocessor is used to pull in #include sources, process #ifdef
4062 statements, #defines and so on.
4073 - The Assemblers and Linkage Editors
4076 This is retargettable assembler & linkage editor, it was developed by Alan
4078 John Hartman created the version for 8051, and I (Sandeep) have made some
4079 enhancements and bug fixes for it to work properly with SDCC.
4086 \begin_inset LatexCommand \index{s51}
4090 is a freeware, opensource simulator developed by Daniel Drotos (
4091 \begin_inset LatexCommand \url{mailto:drdani@mazsola.iit.uni-miskolc.hu}
4096 The simulator is built as part of the build process.
4097 For more information visit Daniel's web site at:
4098 \begin_inset LatexCommand \url{http://mazsola.iit.uni-miskolc.hu/~drdani/embedded/s51}
4103 It currently supports the core mcs51, the Dallas DS80C390 and the Phillips
4107 sdcdb - Source Level Debugger
4111 \begin_inset LatexCommand \index{sdcdb (debugger)}
4115 is the companion source level debugger.
4116 More about sdcdb in section
4117 \begin_inset LatexCommand \ref{cha:Debugging-with-SDCDB}
4122 The current version of the debugger uses Daniel's Simulator S51
4123 \begin_inset LatexCommand \index{s51}
4127 , but can be easily changed to use other simulators.
4137 Single Source File Projects
4140 For single source file 8051 projects the process is very simple.
4141 Compile your programs with the following command
4144 "sdcc sourcefile.c".
4148 This will compile, assemble and link your source file.
4149 Output files are as follows:
4153 \begin_inset LatexCommand \index{<file>.asm}
4158 \begin_inset LatexCommand \index{Assembler source}
4162 file created by the compiler
4166 \begin_inset LatexCommand \index{<file>.lst}
4171 \begin_inset LatexCommand \index{Assembler listing}
4175 file created by the Assembler
4179 \begin_inset LatexCommand \index{<file>.rst}
4184 \begin_inset LatexCommand \index{Assembler listing}
4188 file updated with linkedit information, created by linkage editor
4192 \begin_inset LatexCommand \index{<file>.sym}
4197 \begin_inset LatexCommand \index{Symbol listing}
4201 for the sourcefile, created by the assembler
4205 \begin_inset LatexCommand \index{<file>.rel}
4210 \begin_inset LatexCommand \index{<file>.o}
4215 \begin_inset LatexCommand \index{Object file}
4219 created by the assembler, input to Linkage editor
4223 \begin_inset LatexCommand \index{<file>.map}
4228 \begin_inset LatexCommand \index{Memory map}
4232 for the load module, created by the Linker
4236 \begin_inset LatexCommand \index{<file>.mem}
4240 - A file with a summary of the memory usage
4244 \begin_inset LatexCommand \index{<file>.ihx}
4248 - The load module in Intel hex format
4249 \begin_inset LatexCommand \index{Intel hex format}
4253 (you can select the Motorola S19 format
4254 \begin_inset LatexCommand \index{Motorola S19 format}
4269 \begin_inset LatexCommand \index{-\/-out-fmt-s19}
4274 If you need another format you might want to use
4281 \begin_inset LatexCommand \index{objdump (tool)}
4292 \begin_inset LatexCommand \index{srecord (tool)}
4297 Both formats are documented in the documentation of srecord
4298 \begin_inset LatexCommand \index{srecord (tool)}
4306 \begin_inset LatexCommand \index{<file>.adb}
4310 - An intermediate file containing debug information needed to create the
4322 \begin_inset LatexCommand \index{-\/-debug}
4330 \begin_inset LatexCommand \index{<file>.cdb}
4334 - An optional file (with -
4344 -debug) containing debug information.
4345 The format is documented in cdbfileformat.pdf.
4350 \begin_inset LatexCommand \index{<file> (no extension)}
4354 An optional AOMF or AOMF51
4355 \begin_inset LatexCommand \index{AOMF, AOMF51}
4359 file containing debug information (generated with option -
4386 ormat is commonly used by third party tools (debuggers
4387 \begin_inset LatexCommand \index{Debugger}
4391 , simulators, emulators)
4395 \begin_inset LatexCommand \index{<file>.dump*}
4399 - Dump file to debug the compiler it self (generated with option -
4409 -dumpall) (see section
4410 \begin_inset LatexCommand \ref{sub:Intermediate-Dump-Options}
4416 \begin_inset LatexCommand \ref{sub:The-anatomy-of}
4422 \begin_inset Quotes sld
4425 Anatomy of the compiler
4426 \begin_inset Quotes srd
4432 Projects with Multiple Source Files
4435 SDCC can compile only ONE file at a time.
4436 Let us for example assume that you have a project containing the following
4441 foo1.c (contains some functions)
4443 foo2.c (contains some more functions)
4445 foomain.c (contains more functions and the function main)
4453 The first two files will need to be compiled separately with the commands:
4485 Then compile the source file containing the
4490 \begin_inset LatexCommand \index{Linker}
4494 the files together with the following command:
4502 foomain.c\SpecialChar ~
4503 foo1.rel\SpecialChar ~
4508 \begin_inset LatexCommand \index{<file>.rel}
4520 can be separately compiled as well:
4531 sdcc foomain.rel foo1.rel foo2.rel
4538 The file containing the
4553 file specified in the command line, since the linkage editor processes
4554 file in the order they are presented to it.
4555 The linker is invoked from SDCC using a script file with extension .lnk
4556 \begin_inset LatexCommand \index{<file>.lnk}
4561 You can view this file to troubleshoot linking problems such as those arising
4562 from missing libraries.
4565 Projects with Additional Libraries
4566 \begin_inset LatexCommand \index{Libraries}
4573 Some reusable routines may be compiled into a library, see the documentation
4574 for the assembler and linkage editor (which are in <installdir>/share/sdcc/doc)
4578 \begin_inset LatexCommand \index{<file>.lib}
4585 Libraries created in this manner can be included in the command line.
4586 Make sure you include the -L <library-path> option to tell the linker where
4587 to look for these files if they are not in the current directory.
4588 Here is an example, assuming you have the source file
4600 (if that is not the same as your current project):
4607 sdcc foomain.c foolib.lib -L mylib
4618 must be an absolute path name.
4622 The most efficient way to use libraries is to keep separate modules in separate
4624 The lib file now should name all the modules.rel
4625 \begin_inset LatexCommand \index{<file>.rel}
4630 For an example see the standard library file
4634 in the directory <installdir>/share/lib/small.
4637 Using sdcclib to Create and Manage Libraries
4638 \begin_inset LatexCommand \index{sdcclib}
4645 Alternatively, instead of having a .rel file for each entry on the library
4646 file as described in the preceding section, sdcclib can be used to embed
4647 all the modules belonging to such library in the library file itself.
4648 This results in a larger library file, but it greatly reduces the number
4649 of disk files accessed by the linker.
4650 Additionally, the packed library file contains an index of all include
4651 modules and symbols that significantly speeds up the linking process.
4652 To display a list of options supported by sdcclib type:
4661 \begin_inset LatexCommand \index{sdcclib}
4672 To create a new library file, start by compiling all the required modules.
4710 This will create files _divsint.rel, _divuint.rel, _modsint.rel, _moduint.rel,
4712 The next step is to add the .rel files to the library file:
4720 sdcclib libint.lib _divsint.rel
4723 \begin_inset LatexCommand \index{sdcclib}
4733 sdcclib libint.lib _divuint.rel
4739 sdcclib libint.lib _modsint.rel
4745 sdcclib libint.lib _moduint.rel
4751 sdcclib libint.lib _mulint.rel
4758 If the file already exists in the library, it will be replaced.
4759 To see what modules and symbols are included in the library, options -s
4760 and -m are available.
4768 sdcclib -s libint.lib
4771 \begin_inset LatexCommand \index{sdcclib}
4881 If the source files are compiled using -
4892 \begin_inset LatexCommand \index{-\/-debug}
4896 , the corresponding debug information file .adb will be include in the library
4898 The library files created with sdcclib are plain text files, so they can
4899 be viewed with a text editor.
4900 It is not recomended to modify a library file created with sdcclib using
4901 a text editor, as there are file indexes numbers located accross the file
4902 used by the linker to quickly locate the required module to link.
4903 Once a .rel file (as well as a .adb file) is added to a library using sdcclib,
4904 it can be safely deleted, since all the information required for linking
4905 is embedded in the library file itself.
4906 Library files created using sdcclib are used as described in the preceding
4910 Command Line Options
4911 \begin_inset LatexCommand \index{Command Line Options}
4918 Processor Selection Options
4919 \begin_inset LatexCommand \index{Options processor selection}
4924 \begin_inset LatexCommand \index{Processor selection options}
4930 \labelwidthstring 00.00.0000
4935 \begin_inset LatexCommand \index{-mmcs51}
4941 Generate code for the Intel MCS51
4942 \begin_inset LatexCommand \index{MCS51}
4946 family of processors.
4947 This is the default processor target.
4949 \labelwidthstring 00.00.0000
4954 \begin_inset LatexCommand \index{-mds390}
4960 Generate code for the Dallas DS80C390
4961 \begin_inset LatexCommand \index{DS80C390}
4967 \labelwidthstring 00.00.0000
4972 \begin_inset LatexCommand \index{-mds400}
4978 Generate code for the Dallas DS80C400
4979 \begin_inset LatexCommand \index{DS80C400}
4985 \labelwidthstring 00.00.0000
4990 \begin_inset LatexCommand \index{-mhc08}
4996 Generate code for the Motorola HC08
4997 \begin_inset LatexCommand \index{HC08}
5001 family of processors (added Oct 2003).
5003 \labelwidthstring 00.00.0000
5008 \begin_inset LatexCommand \index{-mz80}
5014 Generate code for the Zilog Z80
5015 \begin_inset LatexCommand \index{Z80}
5019 family of processors.
5021 \labelwidthstring 00.00.0000
5026 \begin_inset LatexCommand \index{-mgbz80}
5032 Generate code for the GameBoy Z80
5033 \begin_inset LatexCommand \index{gbz80 (GameBoy Z80)}
5037 processor (Not actively maintained).
5039 \labelwidthstring 00.00.0000
5044 \begin_inset LatexCommand \index{-mavr}
5050 Generate code for the Atmel AVR
5051 \begin_inset LatexCommand \index{AVR}
5055 processor (In development, not complete).
5056 AVR users should probably have a look at avr-gcc
5057 \begin_inset LatexCommand \url{ http://savannah.nongnu.org/download/avr-libc/snapshots/}
5062 \begin_inset LatexCommand \url{http://winavr.sourceforge.net}
5069 I think it is fair to direct users there for now.
5070 Open source is also about avoiding unnecessary work .
5071 But I didn't find the 'official' link.
5073 \labelwidthstring 00.00.0000
5078 \begin_inset LatexCommand \index{-mpic14}
5084 Generate code for the Microchip PIC 14
5085 \begin_inset LatexCommand \index{PIC14}
5089 -bit processors (p16f84 and variants.
5090 In development, not complete).
5093 p16f627 p16f628 p16f84 p16f873 p16f877?
5095 \labelwidthstring 00.00.0000
5100 \begin_inset LatexCommand \index{-mpic16}
5106 Generate code for the Microchip PIC 16
5107 \begin_inset LatexCommand \index{PIC16}
5111 -bit processors (p18f452 and variants.
5112 In development, not complete).
5114 \labelwidthstring 00.00.0000
5120 Generate code for the Toshiba TLCS-900H
5121 \begin_inset LatexCommand \index{TLCS-900H}
5125 processor (Not maintained, not complete).
5127 \labelwidthstring 00.00.0000
5132 \begin_inset LatexCommand \index{-mxa51}
5138 Generate code for the Phillips XA51
5139 \begin_inset LatexCommand \index{XA51}
5143 processor (Not maintained, not complete).
5146 Preprocessor Options
5147 \begin_inset LatexCommand \index{Options preprocessor}
5152 \begin_inset LatexCommand \index{Preprocessor options}
5157 \begin_inset LatexCommand \index{sdcpp (preprocessor)}
5163 \labelwidthstring 00.00.0000
5168 \begin_inset LatexCommand \index{-I<path>}
5174 The additional location where the pre processor will look for <..h> or
5175 \begin_inset Quotes eld
5179 \begin_inset Quotes erd
5184 \labelwidthstring 00.00.0000
5189 \begin_inset LatexCommand \index{-D<macro[=value]>}
5195 Command line definition of macros.
5196 Passed to the preprocessor.
5198 \labelwidthstring 00.00.0000
5203 \begin_inset LatexCommand \index{-M}
5209 Tell the preprocessor to output a rule suitable for make describing the
5210 dependencies of each object file.
5211 For each source file, the preprocessor outputs one make-rule whose target
5212 is the object file name for that source file and whose dependencies are
5213 all the files `#include'd in it.
5214 This rule may be a single line or may be continued with `
5216 '-newline if it is long.
5217 The list of rules is printed on standard output instead of the preprocessed
5220 \begin_inset LatexCommand \index{-E}
5226 \labelwidthstring 00.00.0000
5231 \begin_inset LatexCommand \index{-C}
5237 Tell the preprocessor not to discard comments.
5238 Used with the `-E' option.
5240 \labelwidthstring 00.00.0000
5245 \begin_inset LatexCommand \index{-MM}
5256 Like `-M' but the output mentions only the user header files included with
5258 \begin_inset Quotes eld
5262 System header files included with `#include <file>' are omitted.
5264 \labelwidthstring 00.00.0000
5269 \begin_inset LatexCommand \index{-Aquestion(answer)}
5275 Assert the answer answer for question, in case it is tested with a preprocessor
5276 conditional such as `#if #question(answer)'.
5277 `-A-' disables the standard assertions that normally describe the target
5280 \labelwidthstring 00.00.0000
5285 \begin_inset LatexCommand \index{-Umacro}
5291 Undefine macro macro.
5292 `-U' options are evaluated after all `-D' options, but before any `-include'
5293 and `-imacros' options.
5295 \labelwidthstring 00.00.0000
5300 \begin_inset LatexCommand \index{-dM}
5306 Tell the preprocessor to output only a list of the macro definitions that
5307 are in effect at the end of preprocessing.
5308 Used with the `-E' option.
5310 \labelwidthstring 00.00.0000
5315 \begin_inset LatexCommand \index{-dD}
5321 Tell the preprocessor to pass all macro definitions into the output, in
5322 their proper sequence in the rest of the output.
5324 \labelwidthstring 00.00.0000
5329 \begin_inset LatexCommand \index{-dN}
5340 Like `-dD' except that the macro arguments and contents are omitted.
5341 Only `#define name' is included in the output.
5343 \labelwidthstring 00.00.0000
5348 preprocessorOption[,preprocessorOption]
5351 \begin_inset LatexCommand \index{-Wp preprocessorOption[,preprocessorOption]}
5356 Pass the preprocessorOption to the preprocessor
5361 \begin_inset LatexCommand \index{sdcpp (preprocessor)}
5366 SDCC uses an adapted version of the preprocessor cpp of the GNU Compiler
5367 Collection (gcc), if you need more dedicated options please refer to the
5369 \begin_inset LatexCommand \htmlurl{http://www.gnu.org/software/gcc/onlinedocs/}
5377 \begin_inset LatexCommand \index{Options linker}
5382 \begin_inset LatexCommand \index{Linker options}
5388 \labelwidthstring 00.00.0000
5408 \begin_inset LatexCommand \index{-\/-lib-path <path>}
5413 \begin_inset LatexCommand \index{-L -\/-lib-path}
5420 <absolute path to additional libraries> This option is passed to the linkage
5421 editor's additional libraries
5422 \begin_inset LatexCommand \index{Libraries}
5427 The path name must be absolute.
5428 Additional library files may be specified in the command line.
5429 See section Compiling programs for more details.
5431 \labelwidthstring 00.00.0000
5448 \begin_inset LatexCommand \index{-\/-xram-loc <Value>}
5453 <Value> The start location of the external ram
5454 \begin_inset LatexCommand \index{xdata (mcs51, ds390 storage class)}
5458 , default value is 0.
5459 The value entered can be in Hexadecimal or Decimal format, e.g.: -
5469 -xram-loc 0x8000 or -
5481 \labelwidthstring 00.00.0000
5498 \begin_inset LatexCommand \index{-\/-code-loc <Value>}
5503 <Value> The start location of the code
5504 \begin_inset LatexCommand \index{code}
5508 segment, default value 0.
5509 Note when this option is used the interrupt vector table is also relocated
5510 to the given address.
5511 The value entered can be in Hexadecimal or Decimal format, e.g.: -
5521 -code-loc 0x8000 or -
5533 \labelwidthstring 00.00.0000
5550 \begin_inset LatexCommand \index{-\/-stack-loc <Value>}
5555 <Value> By default the stack
5556 \begin_inset LatexCommand \index{stack}
5560 is placed after the data segment.
5561 Using this option the stack can be placed anywhere in the internal memory
5563 The value entered can be in Hexadecimal or Decimal format, e.g.
5574 -stack-loc 0x20 or -
5585 Since the sp register is incremented before a push or call, the initial
5586 sp will be set to one byte prior the provided value.
5587 The provided value should not overlap any other memory areas such as used
5588 register banks or the data segment and with enough space for the current
5591 \labelwidthstring 00.00.0000
5608 \begin_inset LatexCommand \index{-\/-data-loc <Value>}
5613 <Value> The start location of the internal ram data
5614 \begin_inset LatexCommand \index{data (mcs51, ds390 storage class)}
5619 The value entered can be in Hexadecimal or Decimal format, eg.
5641 (By default, the start location of the internal ram data segment is set
5642 as low as possible in memory, taking into account the used register banks
5643 and the bit segment at address 0x20.
5644 For example if register banks 0 and 1 are used without bit variables, the
5645 data segment will be set, if -
5655 -data-loc is not used, to location 0x10.)
5657 \labelwidthstring 00.00.0000
5674 \begin_inset LatexCommand \index{-\/-idata-loc <Value>}
5679 <Value> The start location of the indirectly addressable internal ram
5680 \begin_inset LatexCommand \index{idata (mcs51, ds390 storage class)}
5684 of the 8051, default value is 0x80.
5685 The value entered can be in Hexadecimal or Decimal format, eg.
5696 -idata-loc 0x88 or -
5708 \labelwidthstring 00.00.0000
5725 <Value> The start location of the bit
5726 \begin_inset LatexCommand \index{bit}
5730 addressable internal ram of the 8051.
5736 Instead an option can be passed directly to the linker: -Wl\SpecialChar ~
5739 \labelwidthstring 00.00.0000
5754 \begin_inset LatexCommand \index{-\/-out-fmt-ihx}
5763 The linker output (final object code) is in Intel Hex format.
5764 \begin_inset LatexCommand \index{Intel hex format}
5768 This is the default option.
5769 The format itself is documented in the documentation of srecord
5770 \begin_inset LatexCommand \index{srecord (tool)}
5776 \labelwidthstring 00.00.0000
5791 \begin_inset LatexCommand \index{-\/-out-fmt-s19}
5800 The linker output (final object code) is in Motorola S19 format
5801 \begin_inset LatexCommand \index{Motorola S19 format}
5806 The format itself is documented in the documentation of srecord.
5808 \labelwidthstring 00.00.0000
5813 linkOption[,linkOption]
5816 \begin_inset LatexCommand \index{-Wl linkOption[,linkOption]}
5821 Pass the linkOption to the linker.
5822 See file sdcc/as/doc/asxhtm.html for more on linker options.
5826 \begin_inset LatexCommand \index{Options MCS51}
5831 \begin_inset LatexCommand \index{MCS51 options}
5837 \labelwidthstring 00.00.0000
5852 \begin_inset LatexCommand \index{-\/-model-small}
5863 Generate code for Small Model programs, see section Memory Models for more
5865 This is the default model.
5867 \labelwidthstring 00.00.0000
5882 \begin_inset LatexCommand \index{-\/-model-large}
5888 Generate code for Large model programs, see section Memory Models for more
5890 If this option is used all source files in the project have to be compiled
5893 \labelwidthstring 00.00.0000
5908 \begin_inset LatexCommand \index{-\/-xstack}
5914 Uses a pseudo stack in the first 256 bytes in the external ram for allocating
5915 variables and passing parameters.
5917 \begin_inset LatexCommand \ref{sub:External-Stack}
5922 External Stack for more details.
5924 \labelwidthstring 00.00.0000
5942 \begin_inset LatexCommand \index{-\/-iram-size <Value>}
5946 Causes the linker to check if the internal ram usage is within limits of
5949 \labelwidthstring 00.00.0000
5967 \begin_inset LatexCommand \index{-\/-xram-size <Value>}
5971 Causes the linker to check if the external ram usage is within limits of
5974 \labelwidthstring 00.00.0000
5992 \begin_inset LatexCommand \index{-\/-code-size <Value>}
5996 Causes the linker to check if the code memory usage is within limits of
5999 \labelwidthstring 00.00.0000
6017 \begin_inset LatexCommand \index{-\/-stack-size <Value>}
6021 Causes the linker to check if there is at minimum <Value> bytes for stack.
6023 \labelwidthstring 00.00.0000
6041 \begin_inset LatexCommand \index{-\/-pack-iram}
6045 Causes the linker use unused register banks for data variables or stack.
6048 DS390 / DS400 Options
6049 \begin_inset LatexCommand \index{Options DS390}
6054 \begin_inset LatexCommand \index{DS390 options}
6060 \labelwidthstring 00.00.0000
6077 \begin_inset LatexCommand \index{-\/-model-flat24}
6087 Generate 24-bit flat mode code.
6088 This is the one and only that the ds390 code generator supports right now
6089 and is default when using
6094 See section Memory Models for more details.
6096 \labelwidthstring 00.00.0000
6111 \begin_inset LatexCommand \index{-\/-protect-sp-update}
6117 disable interrupts during ESP:SP updates.
6119 \labelwidthstring 00.00.0000
6136 \begin_inset LatexCommand \index{-\/-stack-10bit}
6140 Generate code for the 10 bit stack mode of the Dallas DS80C390 part.
6141 This is the one and only that the ds390 code generator supports right now
6142 and is default when using
6147 In this mode, the stack is located in the lower 1K of the internal RAM,
6148 which is mapped to 0x400000.
6149 Note that the support is incomplete, since it still uses a single byte
6150 as the stack pointer.
6151 This means that only the lower 256 bytes of the potential 1K stack space
6152 will actually be used.
6153 However, this does allow you to reclaim the precious 256 bytes of low RAM
6154 for use for the DATA and IDATA segments.
6155 The compiler will not generate any code to put the processor into 10 bit
6157 It is important to ensure that the processor is in this mode before calling
6158 any re-entrant functions compiled with this option.
6159 In principle, this should work with the
6172 \begin_inset LatexCommand \index{-\/-stack-auto}
6178 option, but that has not been tested.
6179 It is incompatible with the
6192 \begin_inset LatexCommand \index{-\/-xstack}
6199 It also only makes sense if the processor is in 24 bit contiguous addressing
6212 -model-flat24 option
6216 \labelwidthstring 00.00.0000
6231 \begin_inset LatexCommand \index{-\/-stack-probe}
6237 insert call to function __stack_probe at each function prologue.
6239 \labelwidthstring 00.00.0000
6254 \begin_inset LatexCommand \index{-\/-tini-libid}
6260 <nnnn> LibraryID used in -mTININative.
6263 \labelwidthstring 00.00.0000
6278 \begin_inset LatexCommand \index{-\/-use-accelerator}
6284 generate code for DS390 Arithmetic Accelerator.
6289 \begin_inset LatexCommand \index{Options Z80}
6294 \begin_inset LatexCommand \index{Z80 options}
6300 \labelwidthstring 00.00.0000
6317 \begin_inset LatexCommand \index{-\/-callee-saves-bc}
6327 Force a called function to always save BC.
6329 \labelwidthstring 00.00.0000
6346 \begin_inset LatexCommand \index{-\/-no-std-crt0}
6350 When linking, skip the standard crt0.o object file.
6351 You must provide your own crt0.o for your system when linking.
6355 Optimization Options
6356 \begin_inset LatexCommand \index{Options optimization}
6361 \begin_inset LatexCommand \index{Optimization options}
6367 \labelwidthstring 00.00.0000
6382 \begin_inset LatexCommand \index{-\/-nogcse}
6388 Will not do global subexpression elimination, this option may be used when
6389 the compiler creates undesirably large stack/data spaces to store compiler
6391 A warning message will be generated when this happens and the compiler
6392 will indicate the number of extra bytes it allocated.
6393 It is recommended that this option NOT be used, #pragma\SpecialChar ~
6395 \begin_inset LatexCommand \index{\#pragma nogcse}
6399 can be used to turn off global subexpression elimination
6400 \begin_inset LatexCommand \index{Subexpression elimination}
6404 for a given function only.
6406 \labelwidthstring 00.00.0000
6421 \begin_inset LatexCommand \index{-\/-noinvariant}
6427 Will not do loop invariant optimizations, this may be turned off for reasons
6428 explained for the previous option.
6429 For more details of loop optimizations performed see Loop Invariants in
6431 \begin_inset LatexCommand \ref{sub:Loop-Optimizations}
6436 It is recommended that this option NOT be used, #pragma\SpecialChar ~
6438 \begin_inset LatexCommand \index{\#pragma noinvariant}
6442 can be used to turn off invariant optimizations for a given function only.
6444 \labelwidthstring 00.00.0000
6459 \begin_inset LatexCommand \index{-\/-noinduction}
6465 Will not do loop induction optimizations, see section strength reduction
6467 It is recommended that this option is NOT used, #pragma\SpecialChar ~
6469 \begin_inset LatexCommand \index{\#pragma noinduction}
6473 can be used to turn off induction optimizations for a given function only.
6475 \labelwidthstring 00.00.0000
6490 \begin_inset LatexCommand \index{-\/-nojtbound}
6501 Will not generate boundary condition check when switch statements
6502 \begin_inset LatexCommand \index{switch statement}
6506 are implemented using jump-tables.
6508 \begin_inset LatexCommand \ref{sub:'switch'-Statements}
6513 Switch Statements for more details.
6514 It is recommended that this option is NOT used, #pragma\SpecialChar ~
6516 \begin_inset LatexCommand \index{\#pragma nojtbound}
6520 can be used to turn off boundary checking for jump tables for a given function
6523 \labelwidthstring 00.00.0000
6538 \begin_inset LatexCommand \index{-\/-noloopreverse}
6547 Will not do loop reversal
6548 \begin_inset LatexCommand \index{Loop reversing}
6554 \labelwidthstring 00.00.0000
6571 \begin_inset LatexCommand \index{-\/-nolabelopt }
6575 Will not optimize labels (makes the dumpfiles more readable).
6577 \labelwidthstring 00.00.0000
6592 \begin_inset LatexCommand \index{-\/-no-xinit-opt}
6598 Will not memcpy initialized data from code space into xdata space.
6599 This saves a few bytes in code space if you don't have initialized data.
6601 \labelwidthstring 00.00.0000
6616 \begin_inset LatexCommand \index{-\/-nooverlay}
6622 The compiler will not overlay parameters and local variables of any function,
6623 see section Parameters and local variables for more details.
6625 \labelwidthstring 00.00.0000
6640 \begin_inset LatexCommand \index{-\/-no-peep}
6646 Disable peep-hole optimization.
6648 \labelwidthstring 00.00.0000
6665 \begin_inset LatexCommand \index{-\/-peep-file}
6670 <filename> This option can be used to use additional rules to be used by
6671 the peep hole optimizer.
6673 \begin_inset LatexCommand \ref{sub:Peephole-Optimizer}
6678 Peep Hole optimizations for details on how to write these rules.
6680 \labelwidthstring 00.00.0000
6695 \begin_inset LatexCommand \index{-\/-peep-asm}
6701 Pass the inline assembler code through the peep hole optimizer.
6702 This can cause unexpected changes to inline assembler code, please go through
6703 the peephole optimizer
6704 \begin_inset LatexCommand \index{Peephole optimizer}
6708 rules defined in the source file tree '<target>/peeph.def' before using
6713 \begin_inset LatexCommand \index{Options other}
6719 \labelwidthstring 00.00.0000
6735 \begin_inset LatexCommand \index{-\/-compile-only}
6740 \begin_inset LatexCommand \index{-c -\/-compile-only}
6746 will compile and assemble the source, but will not call the linkage editor.
6748 \labelwidthstring 00.00.0000
6767 \begin_inset LatexCommand \index{-\/-c1mode}
6773 reads the preprocessed source from standard input and compiles it.
6774 The file name for the assembler output must be specified using the -o option.
6776 \labelwidthstring 00.00.0000
6781 \begin_inset LatexCommand \index{-E}
6787 Run only the C preprocessor.
6788 Preprocess all the C source files specified and output the results to standard
6791 \labelwidthstring 00.00.0000
6797 \begin_inset LatexCommand \index{-o <path/file>}
6803 The output path resp.
6804 file where everything will be placed.
6805 If the parameter is a path, it must have a trailing slash (or backslash
6806 for the Windows binaries) to be recognized as a path.
6809 \labelwidthstring 00.00.0000
6824 \begin_inset LatexCommand \index{-\/-stack-auto}
6835 All functions in the source file will be compiled as
6840 \begin_inset LatexCommand \index{reentrant}
6845 the parameters and local variables will be allocated on the stack
6846 \begin_inset LatexCommand \index{stack}
6852 \begin_inset LatexCommand \ref{sec:Parameters-and-Local-Variables}
6856 Parameters and Local Variables for more details.
6857 If this option is used all source files in the project should be compiled
6861 \labelwidthstring 00.00.0000
6876 \begin_inset LatexCommand \index{-\/-callee-saves}
6880 function1[,function2][,function3]....
6883 The compiler by default uses a caller saves convention for register saving
6884 across function calls, however this can cause unnecessary register pushing
6885 & popping when calling small functions from larger functions.
6886 This option can be used to switch the register saving convention for the
6887 function names specified.
6888 The compiler will not save registers when calling these functions, no extra
6889 code will be generated at the entry & exit (function prologue
6892 \begin_inset LatexCommand \index{function prologue}
6901 \begin_inset LatexCommand \index{function epilogue}
6907 ) for these functions to save & restore the registers used by these functions,
6908 this can SUBSTANTIALLY reduce code & improve run time performance of the
6910 In the future the compiler (with inter procedural analysis) will be able
6911 to determine the appropriate scheme to use for each function call.
6912 DO NOT use this option for built-in functions such as _mulint..., if this
6913 option is used for a library function the appropriate library function
6914 needs to be recompiled with the same option.
6915 If the project consists of multiple source files then all the source file
6916 should be compiled with the same -
6926 -callee-saves option string.
6927 Also see #pragma\SpecialChar ~
6929 \begin_inset LatexCommand \index{\#pragma callee\_saves}
6935 \labelwidthstring 00.00.0000
6950 \begin_inset LatexCommand \index{-\/-debug}
6959 When this option is used the compiler will generate debug information.
6960 The debug information collected in a file with .cdb extension can be used
6962 For more information see documentation for SDCDB.
6963 Another file with no extension contains debug information in AOMF or AOMF51
6964 \begin_inset LatexCommand \index{AOMF, AOMF51}
6968 format which is commonly used by third party tools.
6970 \labelwidthstring 00.00.0000
6975 \begin_inset LatexCommand \index{-S}
6986 Stop after the stage of compilation proper; do not assemble.
6987 The output is an assembler code file for the input file specified.
6989 \labelwidthstring 00.00.0000
7004 \begin_inset LatexCommand \index{-\/-int-long-reent}
7010 Integer (16 bit) and long (32 bit) libraries have been compiled as reentrant.
7011 Note by default these libraries are compiled as non-reentrant.
7012 See section Installation for more details.
7014 \labelwidthstring 00.00.0000
7029 \begin_inset LatexCommand \index{-\/-cyclomatic}
7038 This option will cause the compiler to generate an information message for
7039 each function in the source file.
7040 The message contains some
7044 information about the function.
7045 The number of edges and nodes the compiler detected in the control flow
7046 graph of the function, and most importantly the
7048 cyclomatic complexity
7049 \begin_inset LatexCommand \index{Cyclomatic complexity}
7055 see section on Cyclomatic Complexity for more details.
7057 \labelwidthstring 00.00.0000
7072 \begin_inset LatexCommand \index{-\/-float-reent}
7078 Floating point library is compiled as reentrant
7079 \begin_inset LatexCommand \index{reentrant}
7084 See section Installation for more details.
7086 \labelwidthstring 00.00.0000
7101 \begin_inset LatexCommand \index{-\/-main-return}
7107 This option can be used if the code generated is called by a monitor program
7108 or if the main routine includes an endless loop.
7109 This option might result in slightly smaller code and save two bytes of
7111 The return from the 'main'
7112 \begin_inset LatexCommand \index{main return}
7116 function will return to the function calling main.
7117 The default setting is to lock up i.e.
7124 \labelwidthstring 00.00.0000
7139 \begin_inset LatexCommand \index{-\/-nostdincl}
7145 This will prevent the compiler from passing on the default include path
7146 to the preprocessor.
7148 \labelwidthstring 00.00.0000
7163 \begin_inset LatexCommand \index{-\/-nostdlib}
7169 This will prevent the compiler from passing on the default library
7170 \begin_inset LatexCommand \index{Libraries}
7176 \labelwidthstring 00.00.0000
7191 \begin_inset LatexCommand \index{-\/-verbose}
7197 Shows the various actions the compiler is performing.
7199 \labelwidthstring 00.00.0000
7204 \begin_inset LatexCommand \index{-V}
7210 Shows the actual commands the compiler is executing.
7212 \labelwidthstring 00.00.0000
7227 \begin_inset LatexCommand \index{-\/-no-c-code-in-asm}
7233 Hides your ugly and inefficient c-code from the asm file, so you can always
7234 blame the compiler :)
7236 \labelwidthstring 00.00.0000
7251 \begin_inset LatexCommand \index{-\/-no-peep-comments}
7257 Will not include peep-hole comments in the generated files.
7259 \labelwidthstring 00.00.0000
7274 \begin_inset LatexCommand \index{-\/-i-code-in-asm}
7280 Include i-codes in the asm file.
7281 Sounds like noise but is most helpful for debugging the compiler itself.
7283 \labelwidthstring 00.00.0000
7298 \begin_inset LatexCommand \index{-\/-less-pedantic}
7304 Disable some of the more pedantic warnings
7305 \begin_inset LatexCommand \index{Warnings}
7309 (jwk burps: please be more specific here, please!).
7310 If you want rather more than less warnings you should consider using a
7311 separate tool dedicated to syntax checking like splint
7312 \begin_inset LatexCommand \url{www.splint.org}
7318 \labelwidthstring 00.00.0000
7333 \begin_inset LatexCommand \index{-\/-print-search-dirs}
7339 Display the directories in the compiler's search path
7341 \labelwidthstring 00.00.0000
7356 \begin_inset LatexCommand \index{-\/-vc}
7362 Display errors and warnings using MSVC style, so you can use SDCC with
7365 \labelwidthstring 00.00.0000
7380 \begin_inset LatexCommand \index{-\/-use-stdout}
7386 Send errors and warnings to stdout instead of stderr.
7388 \labelwidthstring 00.00.0000
7393 asmOption[,asmOption]
7396 \begin_inset LatexCommand \index{-Wa asmOption[,asmOption]}
7401 Pass the asmOption to the assembler
7402 \begin_inset LatexCommand \index{Options assembler}
7407 \begin_inset LatexCommand \index{Assembler options}
7412 See file sdcc/as/doc/asxhtm.html for assembler options.
7415 Intermediate Dump Options
7416 \begin_inset LatexCommand \label{sub:Intermediate-Dump-Options}
7421 \begin_inset LatexCommand \index{Options intermediate dump}
7426 \begin_inset LatexCommand \index{Intermediate dump options}
7433 The following options are provided for the purpose of retargetting and debugging
7435 These provided a means to dump the intermediate code (iCode
7436 \begin_inset LatexCommand \index{iCode}
7440 ) generated by the compiler in human readable form at various stages of
7441 the compilation process.
7442 More on iCodes see chapter
7443 \begin_inset LatexCommand \ref{sub:The-anatomy-of}
7448 \begin_inset Quotes srd
7451 The anatomy of the compiler
7452 \begin_inset Quotes srd
7457 \labelwidthstring 00.00.0000
7472 \begin_inset LatexCommand \index{-\/-dumpraw}
7478 This option will cause the compiler to dump the intermediate code into
7481 <source filename>.dumpraw
7483 just after the intermediate code has been generated for a function, i.e.
7484 before any optimizations are done.
7486 \begin_inset LatexCommand \index{Basic blocks}
7490 at this stage ordered in the depth first number, so they may not be in
7491 sequence of execution.
7493 \labelwidthstring 00.00.0000
7508 \begin_inset LatexCommand \index{-\/-dumpgcse}
7514 Will create a dump of iCode's, after global subexpression elimination
7515 \begin_inset LatexCommand \index{Global subexpression elimination}
7521 <source filename>.dumpgcse.
7523 \labelwidthstring 00.00.0000
7538 \begin_inset LatexCommand \index{-\/-dumpdeadcode}
7544 Will create a dump of iCode's, after deadcode elimination
7545 \begin_inset LatexCommand \index{Dead-code elimination}
7551 <source filename>.dumpdeadcode.
7553 \labelwidthstring 00.00.0000
7568 \begin_inset LatexCommand \index{-\/-dumploop}
7577 Will create a dump of iCode's, after loop optimizations
7578 \begin_inset LatexCommand \index{Loop optimization}
7584 <source filename>.dumploop.
7586 \labelwidthstring 00.00.0000
7601 \begin_inset LatexCommand \index{-\/-dumprange}
7610 Will create a dump of iCode's, after live range analysis
7611 \begin_inset LatexCommand \index{Live range analysis}
7617 <source filename>.dumprange.
7619 \labelwidthstring 00.00.0000
7634 \begin_inset LatexCommand \index{-\/-dumlrange}
7640 Will dump the life ranges
7641 \begin_inset LatexCommand \index{Live range analysis}
7647 \labelwidthstring 00.00.0000
7662 \begin_inset LatexCommand \index{-\/-dumpregassign}
7671 Will create a dump of iCode's, after register assignment
7672 \begin_inset LatexCommand \index{Register assignment}
7678 <source filename>.dumprassgn.
7680 \labelwidthstring 00.00.0000
7695 \begin_inset LatexCommand \index{-\/-dumplrange}
7701 Will create a dump of the live ranges of iTemp's
7703 \labelwidthstring 00.00.0000
7718 \begin_inset LatexCommand \index{-\/-dumpall}
7729 Will cause all the above mentioned dumps to be created.
7732 Redirecting output on Windows Shells
7735 By default SDCC writes it's error messages to
7736 \begin_inset Quotes sld
7740 \begin_inset Quotes srd
7744 To force all messages to
7745 \begin_inset Quotes sld
7749 \begin_inset Quotes srd
7773 \begin_inset LatexCommand \index{-\/-use-stdout}
7778 Additionally, if you happen to have visual studio installed in your windows
7779 machine, you can use it to compile your sources using a custom build and
7795 \begin_inset LatexCommand \index{-\/-vc}
7800 Something like this should work:
7844 -model-large -c $(InputPath)
7847 Environment variables
7848 \begin_inset LatexCommand \index{Environment variables}
7855 SDCC recognizes the following environment variables:
7857 \labelwidthstring 00.00.0000
7862 \begin_inset LatexCommand \index{SDCC\_LEAVE\_SIGNALS}
7868 SDCC installs a signal handler
7869 \begin_inset LatexCommand \index{signal handler}
7873 to be able to delete temporary files after an user break (^C) or an exception.
7874 If this environment variable is set, SDCC won't install the signal handler
7875 in order to be able to debug SDCC.
7877 \labelwidthstring 00.00.0000
7884 \begin_inset LatexCommand \index{TMP, TEMP, TMPDIR}
7890 Path, where temporary files will be created.
7891 The order of the variables is the search order.
7892 In a standard *nix environment these variables are not set, and there's
7893 no need to set them.
7894 On Windows it's recommended to set one of them.
7896 \labelwidthstring 00.00.0000
7901 \begin_inset LatexCommand \index{SDCC\_HOME}
7908 \begin_inset LatexCommand \ref{sub:Install-paths}
7914 \begin_inset Quotes sld
7918 \begin_inset Quotes srd
7923 \labelwidthstring 00.00.0000
7928 \begin_inset LatexCommand \index{SDCC\_INCLUDE}
7935 \begin_inset LatexCommand \ref{sub:Search-Paths}
7941 \begin_inset Quotes sld
7945 \begin_inset Quotes srd
7950 \labelwidthstring 00.00.0000
7955 \begin_inset LatexCommand \index{SDCC\_LIB}
7962 \begin_inset LatexCommand \ref{sub:Search-Paths}
7968 \begin_inset Quotes sld
7972 \begin_inset Quotes srd
7978 There are some more environment variables recognized by SDCC, but these
7979 are solely used for debugging purposes.
7980 They can change or disappear very quickly, and will never be documented.
7983 Storage Class Language Extensions
7986 MCS51/DS390 Storage Class
7987 \begin_inset LatexCommand \index{Storage class}
7994 In addition to the ANSI storage classes SDCC allows the following MCS51
7995 specific storage classes:
7996 \layout Subsubsection
7999 \begin_inset LatexCommand \index{data (mcs51, ds390 storage class)}
8004 \begin_inset LatexCommand \index{near (storage class)}
8015 storage class for the Small Memory model (
8023 can be used synonymously).
8024 Variables declared with this storage class will be allocated in the directly
8025 addressable portion of the internal RAM of a 8051, e.g.:
8030 data unsigned char test_data;
8033 Writing 0x01 to this variable generates the assembly code:
8038 75*00 01\SpecialChar ~
8044 \layout Subsubsection
8047 \begin_inset LatexCommand \index{xdata (mcs51, ds390 storage class)}
8052 \begin_inset LatexCommand \index{far (storage class)}
8059 Variables declared with this storage class will be placed in the external
8065 storage class for the Large Memory model, e.g.:
8070 xdata unsigned char test_xdata;
8073 Writing 0x01 to this variable generates the assembly code:
8078 90s00r00\SpecialChar ~
8107 \layout Subsubsection
8110 \begin_inset LatexCommand \index{idata (mcs51, ds390 storage class)}
8117 Variables declared with this storage class will be allocated into the indirectly
8118 addressable portion of the internal ram of a 8051, e.g.:
8123 idata unsigned char test_idata;
8126 Writing 0x01 to this variable generates the assembly code:
8155 Please note, the first 128 byte of idata physically access the same RAM
8157 The original 8051 had 128 byte idata memory, nowadays most devices have
8158 256 byte idata memory.
8160 \begin_inset LatexCommand \index{stack}
8164 is located in idata memory.
8165 \layout Subsubsection
8168 \begin_inset LatexCommand \index{pdata (mcs51, ds390 storage class)}
8175 Paged xdata access is currently not as straightforward as using the other
8176 addressing modes of a 8051.
8177 The following example writes 0x01 to the address pointed to.
8178 Please note, pdata access physically accesses xdata memory.
8179 The high byte of the address is determined by port P2
8180 \begin_inset LatexCommand \index{P2 (mcs51 sfr)}
8184 (or in case of some 8051 variants by a separate Special Function Register,
8186 \begin_inset LatexCommand \ref{sub:MCS51-variants}
8195 pdata unsigned char *test_pdata_ptr;
8207 test_pdata_ptr = (pdata *)0xfe;
8213 *test_pdata_ptr = 1;
8218 Generates the assembly code:
8223 75*01 FE\SpecialChar ~
8227 _test_pdata_ptr,#0xFE
8259 Be extremely carefull if you use pdata together with the -
8270 \begin_inset LatexCommand \index{-\/-xstack}
8275 \layout Subsubsection
8278 \begin_inset LatexCommand \index{code}
8285 'Variables' declared with this storage class will be placed in the code
8291 code unsigned char test_code;
8294 Read access to this variable generates the assembly code:
8299 90s00r6F\SpecialChar ~
8302 mov dptr,#_test_code
8331 indexed arrays of characters in code memory can be accessed efficiently:
8336 code char test_array[] = {'c','h','e','a','p'};
8339 Read access to this array using an 8-bit unsigned index generates the assembly
8356 90s00r41\SpecialChar ~
8359 mov dptr,#_test_array
8374 \layout Subsubsection
8377 \begin_inset LatexCommand \index{bit}
8384 This is a data-type and a storage class specifier.
8385 When a variable is declared as a bit, it is allocated into the bit addressable
8386 memory of 8051, e.g.:
8394 Writing 1 to this variable generates the assembly code:
8410 The bit addressable memory consists of 128 bits which are located from 0x20
8411 to 0x2f in data memory.
8414 Apart from this 8051 specific storage class most architectures support ANSI-C
8416 \begin_inset LatexCommand \index{bitfields}
8426 Not really meant as examples, but nevertheless showing what bitfields are
8427 about: device/include/mc68hc908qy.h and support/regression/tests/bitfields.c
8431 In accordance with ISO/IEC 9899 bits and bitfields without an explicit
8432 signed modifier are implemented as unsigned.
8433 \layout Subsubsection
8436 \begin_inset LatexCommand \index{sfr}
8441 \begin_inset LatexCommand \index{sbit}
8448 Like the bit keyword,
8452 signifies both a data-type and storage class, they are used to describe
8473 variables of a 8051, eg:
8479 \begin_inset LatexCommand \index{at}
8483 0x80 P0;\SpecialChar ~
8484 /* special function register P0 at location 0x80 */
8486 sbit at 0xd7 CY; /* CY (Carry Flag
8487 \begin_inset LatexCommand \index{Flags}
8492 \begin_inset LatexCommand \index{Carry flag}
8499 Special function registers which are located on an address dividable by
8500 8 are bit-addressable, an
8504 addresses a specific bit within these sfr.
8505 \layout Subsubsection
8508 \begin_inset LatexCommand \index{Pointer}
8512 to MCS51/DS390 specific memory spaces
8515 SDCC allows (via language extensions) pointers to explicitly point to any
8516 of the memory spaces
8517 \begin_inset LatexCommand \index{Memory model}
8522 In addition to the explicit pointers, the compiler uses (by default) generic
8523 pointers which can be used to point to any of the memory spaces.
8527 Pointer declaration examples:
8532 /* pointer physically in internal ram pointing to object in external ram
8535 xdata unsigned char * data p;
8539 /* pointer physically in external ram pointing to object in internal ram
8542 data unsigned char * xdata p;
8546 /* pointer physically in code rom pointing to data in xdata space */
8548 xdata unsigned char * code p;
8552 /* pointer physically in code space pointing to data in code space */
8554 code unsigned char * code p;
8558 /* the following is a generic pointer physically located in xdata space
8564 Well you get the idea.
8569 All unqualified pointers are treated as 3-byte (4-byte for the ds390)
8582 The highest order byte of the
8586 pointers contains the data space information.
8587 Assembler support routines are called whenever data is stored or retrieved
8593 These are useful for developing reusable library
8594 \begin_inset LatexCommand \index{Libraries}
8599 Explicitly specifying the pointer type will generate the most efficient
8601 \layout Subsubsection
8603 Notes on MCS51 memory
8604 \begin_inset LatexCommand \index{MCS51 memory}
8611 The 8051 family of microcontrollers have a minimum of 128 bytes of internal
8612 RAM memory which is structured as follows:
8616 - Bytes 00-1F - 32 bytes to hold up to 4 banks of the registers R0 to R7,
8619 - Bytes 20-2F - 16 bytes to hold 128 bit
8620 \begin_inset LatexCommand \index{bit}
8626 - Bytes 30-7F - 80 bytes for general purpose use.
8631 Additionally some members of the MCS51 family may have up to 128 bytes of
8632 additional, indirectly addressable, internal RAM memory (
8637 \begin_inset LatexCommand \index{idata (mcs51, ds390 storage class)}
8642 Furthermore, some chips may have some built in external memory (
8647 \begin_inset LatexCommand \index{xdata (mcs51, ds390 storage class)}
8651 ) which should not be confused with the internal, directly addressable RAM
8657 \begin_inset LatexCommand \index{data (mcs51, ds390 storage class)}
8662 Sometimes this built in
8666 memory has to be activated before using it (you can probably find this
8667 information on the datasheet of the microcontroller your are using, see
8669 \begin_inset LatexCommand \ref{sub:Startup-Code}
8677 Normally SDCC will only use the first bank
8678 \begin_inset LatexCommand \index{register bank (mcs51, ds390)}
8682 of registers (register bank 0), but it is possible to specify that other
8683 banks of registers should be used in interrupt
8684 \begin_inset LatexCommand \index{interrupt}
8689 By default, the compiler will place the stack after the last byte of allocated
8690 memory for variables.
8691 For example, if the first 2 banks of registers are used, and only four
8696 variables, it will position the base of the internal stack at address 20
8698 This implies that as the stack
8699 \begin_inset LatexCommand \index{stack}
8703 grows, it will use up the remaining register banks, and the 16 bytes used
8704 by the 128 bit variables, and 80 bytes for general purpose use.
8705 If any bit variables are used, the data variables will be placed after
8706 the byte holding the last bit variable.
8707 For example, if register banks 0 and 1 are used, and there are 9 bit variables
8712 variables will be placed starting at address 0x22.
8724 \begin_inset LatexCommand \index{-\/-data-loc<Value>}
8728 to specify the start address of the
8742 -iram-size to specify the size of the total internal RAM (
8754 By default the 8051 linker will place the stack after the last byte of data
8767 \begin_inset LatexCommand \index{-\/-stack-loc<Value>}
8771 allows you to specify the start of the stack, i.e.
8772 you could start it after any data in the general purpose area.
8773 If your microcontroller has additional indirectly addressable internal
8778 ) you can place the stack on it.
8779 You may also need to use -
8790 \begin_inset LatexCommand \index{-\/-xdata-loc<Value>}
8794 to set the start address of the external RAM (
8809 \begin_inset LatexCommand \index{-\/-data-loc}
8813 to specify its size.
8814 Same goes for the code memory, using -
8825 \begin_inset LatexCommand \index{-\/-data-loc}
8840 \begin_inset LatexCommand \index{-\/-data-loc}
8845 If in doubt, don't specify any options and see if the resulting memory
8846 layout is appropriate, then you can adjust it.
8849 The linker generates two files with memory allocation information.
8850 The first, with extension .map
8851 \begin_inset LatexCommand \index{<file>.map}
8855 shows all the variables and segments.
8856 The second with extension .mem
8857 \begin_inset LatexCommand \index{<file>.mem}
8861 shows the final memory layout.
8862 The linker will complain either if memory segments overlap, there is not
8863 enough memory, or there is not enough space for stack.
8864 If you get any linking warnings and/or errors related to stack or segments
8865 allocation, take a look at either the .map or .mem files to find out what
8867 The .mem file may even suggest a solution to the problem.
8870 Z80/Z180 Storage Class
8871 \begin_inset LatexCommand \index{Storage class}
8876 \layout Subsubsection
8879 \begin_inset LatexCommand \index{sfr}
8883 (in/out to 8-bit addresses)
8887 \begin_inset LatexCommand \index{Z80}
8891 family has separate address spaces for memory and
8901 \begin_inset LatexCommand \index{I/O memory (Z80, Z180)}
8905 is accessed with special instructions, e.g.:
8910 sfr at 0x78 IoPort;\SpecialChar ~
8912 /* define a var in I/O space at 78h called IoPort */
8916 Writing 0x01 to this variable generates the assembly code:
8936 \layout Subsubsection
8939 \begin_inset LatexCommand \index{sfr}
8943 (in/out to 16-bit addresses)
8950 is used to support 16 bit addresses in I/O memory e.g.:
8956 \begin_inset LatexCommand \index{at}
8963 Writing 0x01 to this variable generates the assembly code:
8968 01 23 01\SpecialChar ~
8988 \layout Subsubsection
8991 \begin_inset LatexCommand \index{sfr}
8995 (in0/out0 to 8 bit addresses on Z180
8996 \begin_inset LatexCommand \index{Z180}
9001 \begin_inset LatexCommand \index{HD64180}
9008 The compiler option -
9018 -portmode=180 (80) and a compiler #pragma\SpecialChar ~
9020 \begin_inset LatexCommand \index{\#pragma portmode}
9024 =z180 (z80) is used to turn on (off) the Z180/HD64180 port addressing instructio
9034 If you include the file z180.h this will be set automatically.
9038 \begin_inset LatexCommand \index{Storage class}
9043 \layout Subsubsection
9046 \begin_inset LatexCommand \index{data (mcs51, ds390 storage class)}
9053 The data storage class declares a variable that resides in the first 256
9054 bytes of memory (the direct page).
9055 The HC08 is most efficient at accessing variables (especially pointers)
9057 \layout Subsubsection
9060 \begin_inset LatexCommand \index{xdata (mcs51, ds390 storage class)}
9067 The xdata storage class declares a variable that can reside anywhere in
9069 This is the default if no storage class is specified.
9074 \begin_inset LatexCommand \index{Absolute addressing}
9081 Data items can be assigned an absolute address with the
9084 \begin_inset LatexCommand \index{at}
9090 keyword, in addition to a storage class, e.g.:
9096 \begin_inset LatexCommand \index{xdata (mcs51, ds390 storage class)}
9101 \begin_inset LatexCommand \index{at}
9105 0x7ffe unsigned int chksum;
9108 In the above example the variable chksum will be located at 0x7ffe and 0x7fff
9109 of the external ram.
9114 reserve any space for variables declared in this way (they are implemented
9115 with an equate in the assembler).
9116 Thus it is left to the programmer to make sure there are no overlaps with
9117 other variables that are declared without the absolute address.
9118 The assembler listing file (.lst
9119 \begin_inset LatexCommand \index{<file>.lst}
9123 ) and the linker output files (.rst
9124 \begin_inset LatexCommand \index{<file>.rst}
9129 \begin_inset LatexCommand \index{<file>.map}
9133 ) are good places to look for such overlaps.
9134 Variables with an absolute address are
9139 \begin_inset LatexCommand \index{Variable initialization}
9146 In case of memory mapped I/O devices the keyword
9150 has to be used to tell the compiler that accesses might not be removed:
9156 \begin_inset LatexCommand \index{volatile}
9161 \begin_inset LatexCommand \index{xdata (mcs51, ds390 storage class)}
9166 \begin_inset LatexCommand \index{at}
9170 0x8000 unsigned char PORTA_8255;
9173 For some architectures (mcs51) array accesses are more efficient if an (xdata/fa
9178 \begin_inset LatexCommand \index{Aligned array}
9185 starts at a block (256 byte) boundary
9186 \begin_inset LatexCommand \index{block boundary}
9191 \begin_inset LatexCommand \ref{sub:A-Step-by Assembler Introduction}
9197 Absolute addresses can be specified for variables in all storage classes,
9204 \begin_inset LatexCommand \index{bit}
9209 \begin_inset LatexCommand \index{at}
9216 The above example will allocate the variable at offset 0x02 in the bit-addressab
9218 There is no real advantage to assigning absolute addresses to variables
9219 in this manner, unless you want strict control over all the variables allocated.
9220 One possible use would be to write hardware portable code.
9221 For example, if you have a routine that uses one or more of the microcontroller
9222 I/O pins, and such pins are different for two different hardwares, you
9223 can declare the I/O pins in your routine using:
9228 extern volatile bit MOSI;\SpecialChar ~
9232 /* master out, slave in */
9234 extern volatile bit MISO;\SpecialChar ~
9238 /* master in, slave out */
9240 extern volatile bit MCLK;\SpecialChar ~
9248 /* Input and Output of a byte on a 3-wire serial bus.
9253 If needed adapt polarity of clock, polarity of data and bit order
9258 unsigned char spi_io(unsigned char out_byte)
9282 MOSI = out_byte & 0x80;
9312 /* _asm nop _endasm; */\SpecialChar ~
9320 /* for slow peripherals */
9371 Then, someplace in the code for the first hardware you would use
9376 bit at 0x80 MOSI;\SpecialChar ~
9380 /* I/O port 0, bit 0 */
9382 bit at 0x81 MISO;\SpecialChar ~
9386 /* I/O port 0, bit 1 */
9388 bit at 0x82 MCLK;\SpecialChar ~
9392 /* I/O port 0, bit 2 */
9395 Similarly, for the second hardware you would use
9400 bit at 0x83 MOSI;\SpecialChar ~
9404 /* I/O port 0, bit 3 */
9406 bit at 0x91 MISO;\SpecialChar ~
9410 /* I/O port 1, bit 1 */
9413 \begin_inset LatexCommand \index{bit}
9417 at 0x92 MCLK;\SpecialChar ~
9421 /* I/O port 1, bit 2 */
9424 and you can use the same hardware dependent routine without changes, as
9425 for example in a library.
9426 This is somehow similar to sbit, but only one absolute address has to be
9427 specified in the whole project.
9431 \begin_inset LatexCommand \index{Parameters}
9436 \begin_inset LatexCommand \index{function parameter}
9441 \begin_inset LatexCommand \index{local variables}
9446 \begin_inset LatexCommand \label{sec:Parameters-and-Local-Variables}
9453 Automatic (local) variables and parameters to functions can either be placed
9454 on the stack or in data-space.
9455 The default action of the compiler is to place these variables in the internal
9456 RAM (for small model) or external RAM (for large model).
9457 This in fact makes them similar to
9460 \begin_inset LatexCommand \index{static}
9466 so by default functions are non-reentrant
9467 \begin_inset LatexCommand \index{reentrant}
9476 They can be placed on the stack
9477 \begin_inset LatexCommand \index{stack}
9494 \begin_inset LatexCommand \index{-\/-stack-auto}
9502 #pragma\SpecialChar ~
9506 \begin_inset LatexCommand \index{\#pragma stackauto}
9513 \begin_inset LatexCommand \index{reentrant}
9519 keyword in the function declaration, e.g.:
9524 unsigned char foo(char i) reentrant
9538 Since stack space on 8051 is limited, the
9556 option should be used sparingly.
9557 Note that the reentrant keyword just means that the parameters & local
9558 variables will be allocated to the stack, it
9562 mean that the function is register bank
9563 \begin_inset LatexCommand \index{register bank (mcs51, ds390)}
9572 \begin_inset LatexCommand \index{local variables}
9576 can be assigned storage classes and absolute
9577 \begin_inset LatexCommand \index{Absolute addressing}
9594 xdata unsigned char i;
9607 \begin_inset LatexCommand \index{at}
9611 0x31 unsigned char j;
9623 In the above example the variable
9627 will be allocated in the external ram,
9631 in bit addressable space and
9650 or when a function is declared as
9654 this should only be done for static variables.
9658 \begin_inset LatexCommand \index{function parameter}
9662 however are not allowed any storage class
9663 \begin_inset LatexCommand \index{Storage class}
9667 , (storage classes for parameters will be ignored), their allocation is
9668 governed by the memory model in use, and the reentrancy options.
9672 \begin_inset LatexCommand \label{sub:Overlaying}
9677 \begin_inset LatexCommand \index{Overlaying}
9685 \begin_inset LatexCommand \index{reentrant}
9689 functions SDCC will try to reduce internal ram space usage by overlaying
9690 parameters and local variables of a function (if possible).
9691 Parameters and local variables
9692 \begin_inset LatexCommand \index{local variables}
9696 of a function will be allocated to an overlayable segment if the function
9699 no other function calls and the function is non-reentrant and the memory
9701 \begin_inset LatexCommand \index{Memory model}
9708 If an explicit storage class
9709 \begin_inset LatexCommand \index{Storage class}
9713 is specified for a local variable, it will NOT be overlayed.
9716 Note that the compiler (not the linkage editor) makes the decision for overlayin
9718 Functions that are called from an interrupt service routine should be preceded
9719 by a #pragma\SpecialChar ~
9721 \begin_inset LatexCommand \index{\#pragma nooverlay}
9725 if they are not reentrant.
9728 Also note that the compiler does not do any processing of inline assembler
9729 code, so the compiler might incorrectly assign local variables and parameters
9730 of a function into the overlay segment if the inline assembler code calls
9731 other c-functions that might use the overlay.
9732 In that case the #pragma\SpecialChar ~
9733 nooverlay should be used.
9736 Parameters and local variables of functions that contain 16 or 32 bit multiplica
9738 \begin_inset LatexCommand \index{Multiplication}
9743 \begin_inset LatexCommand \index{Division}
9747 will NOT be overlayed since these are implemented using external functions,
9756 \begin_inset LatexCommand \index{\#pragma nooverlay}
9762 void set_error(unsigned char errcd)
9778 void some_isr () interrupt
9779 \begin_inset LatexCommand \index{interrupt}
9809 In the above example the parameter
9817 would be assigned to the overlayable segment if the #pragma\SpecialChar ~
9819 not present, this could cause unpredictable runtime behavior when called
9820 from an interrupt service routine.
9821 The #pragma\SpecialChar ~
9822 nooverlay ensures that the parameters and local variables for
9823 the function are NOT overlayed.
9826 Interrupt Service Routines
9827 \begin_inset LatexCommand \label{sub:Interrupt-Service-Routines}
9849 outines to be coded in C, with some extended keywords.
9854 void timer_isr (void) interrupt 1 using 1
9868 The optional number following the
9871 \begin_inset LatexCommand \index{interrupt}
9877 keyword is the interrupt number this routine will service.
9878 When present, the compiler will insert a call to this routine in the interrupt
9879 vector table for the interrupt number specified.
9880 If you have multiple source files in your project, interrupt service routines
9881 can be present in any of them, but a prototype of the isr MUST be present
9882 or included in the file that contains the function
9891 keyword can be used to tell the compiler to use the specified register
9892 bank (8051 specific) when generating code for this function.
9898 Interrupt service routines open the door for some very interesting bugs:
9900 If an interrupt service routine changes variables which are accessed by
9901 other functions these variables have to be declared
9906 \begin_inset LatexCommand \index{volatile}
9914 If the access to these variables is not
9917 \begin_inset LatexCommand \index{atomic}
9924 the processor needs more than one instruction for the access and could
9925 be interrupted while accessing the variable) the interrupt must be disabled
9926 during the access to avoid inconsistent data.
9927 Access to 16 or 32 bit variables is obviously not atomic on 8 bit CPUs
9928 and should be protected by disabling interrupts.
9929 You're not automatically on the safe side if you use 8 bit variables though.
9930 We need an example here: f.e.
9931 on the 8051 the harmless looking
9932 \begin_inset Quotes srd
9942 \begin_inset Quotes sld
9951 \begin_inset Quotes srd
9961 \begin_inset Quotes sld
9964 from within an interrupt routine might get lost if the interrupt occurs
9967 \begin_inset Quotes sld
9972 counter\SpecialChar ~
9977 \begin_inset Quotes srd
9980 is not atomic on the 8051 even if
9984 is located in data memory.
9985 Bugs like these are hard to reproduce and can cause a lot of trouble.
9989 The return address and the registers used in the interrupt service routine
9990 are saved on the stack
9991 \begin_inset LatexCommand \index{stack}
9995 so there must be sufficient stack space.
9996 If there isn't variables or registers (or even the return address itself)
10003 \begin_inset LatexCommand \index{stack overflow}
10007 is most likely to happen if the interrupt occurs during the
10008 \begin_inset Quotes sld
10012 \begin_inset Quotes srd
10015 subroutine when the stack is already in use for f.e.
10016 many return addresses.
10019 A special note here, int (16 bit) and long (32 bit) integer division
10020 \begin_inset LatexCommand \index{Division}
10025 \begin_inset LatexCommand \index{Multiplication}
10030 \begin_inset LatexCommand \index{Modulus}
10035 \begin_inset LatexCommand \index{Floating point support}
10039 operations are implemented using external support routines developed in
10041 If an interrupt service routine needs to do any of these operations then
10042 the support routines (as mentioned in a following section) will have to
10043 be recompiled using the
10056 \begin_inset LatexCommand \index{-\/-stack-auto}
10062 option and the source file will need to be compiled using the
10077 \begin_inset LatexCommand \index{-\/-int-long-reent}
10084 Calling other functions from an interrupt service routine is not recommended,
10085 avoid it if possible.
10086 Note that when some function is called from an interrupt service routine
10087 it should be preceded by a #pragma\SpecialChar ~
10089 \begin_inset LatexCommand \index{\#pragma nooverlay}
10093 if it is not reentrant.
10094 Furthermore nonreentrant functions should not be called from the main program
10095 while the interrupt service routine might be active.
10101 \begin_inset LatexCommand \ref{sub:Overlaying}
10106 about Overlaying and section
10107 \begin_inset LatexCommand \ref{sub:Functions-using-private-banks}
10112 about Functions using private register banks.
10115 MCS51/DS390 Interrupt Service Routines
10118 Interrupt numbers and the corresponding address & descriptions for the Standard
10119 8051/8052 are listed below.
10120 SDCC will automatically adjust the interrupt vector table to the maximum
10121 interrupt number specified.
10127 \begin_inset Tabular
10128 <lyxtabular version="3" rows="7" columns="3">
10130 <column alignment="center" valignment="top" leftline="true" width="0in">
10131 <column alignment="center" valignment="top" leftline="true" width="0in">
10132 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0in">
10133 <row topline="true" bottomline="true">
10134 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10142 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10150 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
10159 <row topline="true">
10160 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10168 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10176 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
10185 <row topline="true">
10186 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10194 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10202 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
10211 <row topline="true">
10212 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10220 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10228 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
10237 <row topline="true">
10238 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10246 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10254 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
10263 <row topline="true">
10264 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10272 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10280 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
10289 <row topline="true" bottomline="true">
10290 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10298 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
10306 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
10324 If the interrupt service routine is defined without
10327 \begin_inset LatexCommand \index{using (mcs51, ds390 register bank)}
10333 a register bank or with register bank 0 (
10337 0), the compiler will save the registers used by itself on the stack upon
10338 entry and restore them at exit, however if such an interrupt service routine
10339 calls another function then the entire register bank will be saved on the
10341 This scheme may be advantageous for small interrupt service routines which
10342 have low register usage.
10345 If the interrupt service routine is defined to be using a specific register
10350 & psw are saved and restored, if such an interrupt service routine calls
10351 another function (using another register bank) then the entire register
10352 bank of the called function will be saved on the stack.
10353 This scheme is recommended for larger interrupt service routines.
10356 HC08 Interrupt Service Routines
10359 Since the number of interrupts available is chip specific and the interrupt
10360 vector table always ends at the last byte of memory, the interrupt numbers
10361 corresponds to the interrupt vectors in reverse order of address.
10362 For example, interrupt 1 will use the interrupt vector at 0xfffc, interrupt
10363 2 will use the interrupt vector at 0xfffa, and so on.
10364 However, interrupt 0 (the reset vector at 0xfffe) is not redefinable in
10365 this way; instead see section
10366 \begin_inset LatexCommand \ref{sub:Startup-Code}
10370 for details on customizing startup.
10373 Z80 Interrupt Service Routines
10376 The Z80 uses several different methods for determining the correct interrupt
10377 vector depending on the hardware implementation.
10378 Therefore, SDCC ignores the optional interrupt number and does not attempt
10379 to generate an interrupt vector table.
10382 By default, SDCC generates code for a maskable interrupt, which uses an
10383 RETI instruction to return from the interrupt.
10384 To write an interrupt handler for the non-maskable interrupt, which needs
10385 an RETN instruction instead, add the
10394 void nmi_isr (void) critical interrupt
10408 Enabling and Disabling Interrupts
10411 Critical Functions and Critical Statements
10414 A special keyword may be associated with a block or a function declaring
10420 SDCC will generate code to disable all interrupts
10421 \begin_inset LatexCommand \index{interrupt}
10425 upon entry to a critical function and restore the interrupt enable to the
10426 previous state before returning.
10427 Nesting critical functions will need one additional byte on the stack
10428 \begin_inset LatexCommand \index{stack}
10437 int foo () critical
10438 \begin_inset LatexCommand \index{critical}
10463 The critical attribute maybe used with other attributes like
10473 may also be used to disable interrupts more locally:
10481 More than one statement could have been included in the block.
10484 Enabling and Disabling Interrupts directly
10488 \begin_inset LatexCommand \index{interrupt}
10492 can also be disabled and enabled directly (8051):
10497 EA = 0;\SpecialChar ~
10560 EA = 1;\SpecialChar ~
10627 On other architectures which have seperate opcodes for enabling and disabling
10628 interrupts you might want to make use of defines with inline assembly
10629 \begin_inset LatexCommand \index{Assembler routines}
10639 \begin_inset LatexCommand \index{\_asm}
10648 \begin_inset LatexCommand \index{\_endasm}
10657 #define SEI _asm\SpecialChar ~
10669 Note: it is sometimes sufficient to disable only a specific interrupt source
10671 a timer or serial interrupt by manipulating an
10674 \begin_inset LatexCommand \index{interrupt mask}
10684 Usually the time during which interrupts are disabled should be kept as
10686 This minimizes both
10691 \begin_inset LatexCommand \index{interrupt latency}
10695 (the time between the occurrence of the interrupt and the execution of
10696 the first code in the interrupt routine) and
10701 \begin_inset LatexCommand \index{interrupt jitter}
10705 (the difference between the shortest and the longest interrupt latency).
10706 These really are something different, f.e.
10707 a serial interrupt has to be served before its buffer overruns so it cares
10708 for the maximum interrupt latency, whereas it does not care about jitter.
10709 On a loudspeaker driven via a digital to analog converter which is fed
10710 by an interrupt a latency of a few milliseconds might be tolerable, whereas
10711 a much smaller jitter will be very audible.
10714 You can reenable interrupts within an interrupt routine and on some architecture
10715 s you can make use of two (or more) levels of
10717 interrupt priorities
10720 \begin_inset LatexCommand \index{interrupt priority}
10725 On some architectures which don't support interrupt priorities these can
10726 be implemented by manipulating the interrupt mask and reenabling interrupts
10727 within the interrupt routine.
10728 Check there is sufficient space on the stack
10729 \begin_inset LatexCommand \index{stack}
10733 and don't add complexity unless you have to.
10738 \begin_inset LatexCommand \index{semaphore}
10742 locking (mcs51/ds390)
10745 Some architectures (mcs51/ds390) have an atomic
10746 \begin_inset LatexCommand \index{atomic}
10759 These type of instructions are typically used in preemptive multitasking
10760 systems, where a routine f.e.
10761 claims the use of a data structure ('acquires a lock
10762 \begin_inset LatexCommand \index{lock}
10766 on it'), makes some modifications and then releases the lock when the data
10767 structure is consistent again.
10768 The instruction may also be used if interrupt and non-interrupt code have
10769 to compete for a resource.
10770 With the atomic bit test and clear instruction interrupts
10771 \begin_inset LatexCommand \index{interrupt}
10775 don't have to be disabled for the locking operation.
10779 SDCC generates this instruction if the source follows this pattern:
10784 volatile bit resource_is_free;
10788 if (resource_is_free)
10798 resource_is_free=0;
10811 resource_is_free=1;
10818 Note, mcs51 and ds390 support only an atomic
10819 \begin_inset LatexCommand \index{atomic}
10827 instruction (as opposed to atomic bit test and
10832 Functions using private register banks
10833 \begin_inset LatexCommand \label{sub:Functions-using-private-banks}
10840 Some architectures have support for quickly changing register sets.
10841 SDCC supports this feature with the
10844 \begin_inset LatexCommand \index{using (mcs51, ds390 register bank)}
10850 attribute (which tells the compiler to use a register bank
10851 \begin_inset LatexCommand \index{register bank (mcs51, ds390)}
10855 other than the default bank zero).
10856 It should only be applied to
10859 \begin_inset LatexCommand \index{interrupt}
10865 functions (see footnote below).
10866 This will in most circumstances make the generated ISR code more efficient
10867 since it will not have to save registers on the stack.
10874 attribute will have no effect on the generated code for a
10878 function (but may occasionally be useful anyway
10884 possible exception: if a function is called ONLY from 'interrupt' functions
10885 using a particular bank, it can be declared with the same 'using' attribute
10886 as the calling 'interrupt' functions.
10887 For instance, if you have several ISRs using bank one, and all of them
10888 call memcpy(), it might make sense to create a specialized version of memcpy()
10889 'using 1', since this would prevent the ISR from having to save bank zero
10890 to the stack on entry and switch to bank zero before calling the function
10897 (pending: I don't think this has been done yet)
10904 function using a non-zero bank will assume that it can trash that register
10905 bank, and will not save it.
10906 Since high-priority interrupts
10907 \begin_inset LatexCommand \index{interrupt priority}
10911 can interrupt low-priority ones on the 8051 and friends, this means that
10912 if a high-priority ISR
10916 a particular bank occurs while processing a low-priority ISR
10920 the same bank, terrible and bad things can happen.
10921 To prevent this, no single register bank should be
10925 by both a high priority and a low priority ISR.
10926 This is probably most easily done by having all high priority ISRs use
10927 one bank and all low priority ISRs use another.
10928 If you have an ISR which can change priority at runtime, you're on your
10929 own: I suggest using the default bank zero and taking the small performance
10933 It is most efficient if your ISR calls no other functions.
10934 If your ISR must call other functions, it is most efficient if those functions
10935 use the same bank as the ISR (see note 1 below); the next best is if the
10936 called functions use bank zero.
10937 It is very inefficient to call a function using a different, non-zero bank
10943 \begin_inset LatexCommand \label{sub:Startup-Code}
10948 \begin_inset LatexCommand \index{Startup code}
10955 MCS51/DS390 Startup Code
10958 The compiler inserts a call to the C routine
10960 _sdcc_external_startup()
10961 \begin_inset LatexCommand \index{\_sdcc\_external\_startup()}
10970 at the start of the CODE area.
10971 This routine is in the runtime library
10972 \begin_inset LatexCommand \index{Runtime library}
10977 By default this routine returns 0, if this routine returns a non-zero value,
10978 the static & global variable initialization will be skipped and the function
10979 main will be invoked.
10980 Otherwise static & global variables will be initialized before the function
10984 _sdcc_external_startup()
10986 routine to your program to override the default if you need to setup hardware
10987 or perform some other critical operation prior to static & global variable
10989 On some mcs51 variants xdata has to be explicitly enabled before it can
10990 be accessed, this is the place to do it.
10991 The startup code clears the complete 256 byte of idata memory, this might
10992 cause problems for 128 byte devices (endless loop reported for Chipcon
10996 See also the compiler option
11015 \begin_inset LatexCommand \index{-\/-no-xinit-opt}
11020 \begin_inset LatexCommand \ref{sub:MCS51-variants}
11025 about MCS51-variants.
11031 The HC08 startup code follows the same scheme as the MCS51 startup code.
11037 On the Z80 the startup code is inserted by linking with crt0.o which is generated
11038 from sdcc/device/lib/z80/crt0.s.
11039 If you need a different startup code you can use the compiler option
11060 \begin_inset LatexCommand \index{-\/-no-std-crt0}
11064 and provide your own crt0.o.
11068 Inline Assembler Code
11069 \begin_inset LatexCommand \index{Assembler routines}
11076 A Step by Step Introduction
11077 \begin_inset LatexCommand \label{sub:A-Step-by Assembler Introduction}
11084 Starting from a small snippet of c-code this example shows for the MCS51
11085 how to use inline assembly, access variables, a function parameter and
11086 an array in xdata memory.
11087 The example uses an MCS51 here but is easily adapted for other architectures.
11088 This is a buffer routine which should be optimized:
11095 \begin_inset LatexCommand \index{far (storage class)}
11100 \begin_inset LatexCommand \index{at}
11105 \begin_inset LatexCommand \index{Aligned array}
11111 unsigned char head,tail;
11115 void to_buffer( unsigned char c )
11123 if( head != tail-1 )
11133 buf[ head++ ] = c;\SpecialChar ~
11137 /* access to a 256 byte aligned array */
11142 If the code snippet (assume it is saved in buffer.c) is compiled with SDCC
11143 then a corresponding buffer.asm file is generated.
11144 We define a new function
11148 in file buffer.c in which we cut and paste the generated code, removing
11149 unwanted comments and some ':'.
11151 \begin_inset Quotes sld
11155 \begin_inset Quotes srd
11159 \begin_inset Quotes sld
11163 \begin_inset Quotes srd
11166 to the beginning and the end of the function body:
11172 /* With a cut and paste from the .asm file, we have something to start with.
11177 The function is not yet OK! (registers aren't saved) */
11179 void to_buffer_asm( unsigned char c )
11188 \begin_inset LatexCommand \index{\_asm}
11202 ;buffer.c if( head != tail-1 )
11250 ;buffer.c buf[ head++ ] = c; /* access to a 256 byte aligned array */
11251 \begin_inset LatexCommand \index{Aligned array}
11320 The new file buffer.c should compile with only one warning about the unreferenced
11321 function argument 'c'.
11322 Now we hand-optimize the assembly code and insert an #define USE_ASSEMBLY
11323 (1) and finally have:
11329 unsigned char far at 0x7f00 buf[0x100];
11331 unsigned char head,tail;
11333 #define USE_ASSEMBLY (1)
11341 void to_buffer( unsigned char c )
11349 if( head != tail-1 )
11369 void to_buffer( unsigned char c )
11377 c; // to avoid warning: unreferenced function argument
11384 \begin_inset LatexCommand \index{\_asm}
11398 ; save used registers here.
11409 ; If we were still using r2,r3 we would have to push them here.
11412 ; if( head != tail-1 )
11455 ; we could do an ANL a,#0x0f here to use a smaller buffer (see below)
11479 ; buf[ head++ ] = c;
11490 a,dpl \SpecialChar ~
11497 ; dpl holds lower byte of function argument
11508 dpl,_head \SpecialChar ~
11511 ; buf is 0x100 byte aligned so head can be used directly
11553 ; we could do an ANL _head,#0x0f here to use a smaller buffer (see above)
11565 ; restore used registers here
11578 The inline assembler code can contain any valid code understood by the assembler
11579 , this includes any assembler directives and comment lines
11585 The assembler does not like some characters like ':' or ''' in comments.
11586 You'll find an 100+ pages assembler manual in sdcc/as/doc/asxhtm.html
11590 The compiler does not do any validation of the code within the
11593 \begin_inset LatexCommand \index{\_asm}
11601 Specifically it will not know which registers are used and thus register
11603 \begin_inset LatexCommand \index{push/pop}
11607 has to be done manually.
11611 It is recommended that each assembly instruction (including labels) be placed
11612 in a separate line (as the example shows).
11626 \begin_inset LatexCommand \index{-\/-peep-asm}
11632 command line option is used, the inline assembler code will be passed through
11633 the peephole optimizer
11634 \begin_inset LatexCommand \index{Peephole optimizer}
11639 There are only a few (if any) cases where this option makes sense, it might
11640 cause some unexpected changes in the inline assembler code.
11641 Please go through the peephole optimizer rules defined in file
11645 before using this option.
11649 \begin_inset LatexCommand \label{sub:Naked-Functions}
11654 \begin_inset LatexCommand \index{Naked functions}
11661 A special keyword may be associated with a function declaring it as
11664 \begin_inset LatexCommand \index{\_naked}
11675 function modifier attribute prevents the compiler from generating prologue
11676 \begin_inset LatexCommand \index{function prologue}
11681 \begin_inset LatexCommand \index{function epilogue}
11685 code for that function.
11686 This means that the user is entirely responsible for such things as saving
11687 any registers that may need to be preserved, selecting the proper register
11688 bank, generating the
11692 instruction at the end, etc.
11693 Practically, this means that the contents of the function must be written
11694 in inline assembler.
11695 This is particularly useful for interrupt functions, which can have a large
11696 (and often unnecessary) prologue/epilogue.
11697 For example, compare the code generated by these two functions:
11703 \begin_inset LatexCommand \index{volatile}
11707 data unsigned char counter;
11711 void simpleInterrupt(void) interrupt
11712 \begin_inset LatexCommand \index{interrupt}
11730 void nakedInterrupt(void) interrupt 2 _naked
11739 \begin_inset LatexCommand \index{\_asm}
11756 _counter ; does not change flags, no need to save psw
11768 ; MUST explicitly include ret or reti in _naked function.
11775 \begin_inset LatexCommand \index{\_endasm}
11784 For an 8051 target, the generated simpleInterrupt looks like:
11925 whereas nakedInterrupt looks like:
11940 _counter ; does not change flags, no need to save psw
11958 ; MUST explicitly include ret or reti in _naked function
11961 The related directive #pragma exclude
11962 \begin_inset LatexCommand \index{\#pragma exclude}
11966 allows a more fine grained control over pushing & popping
11967 \begin_inset LatexCommand \index{push/pop}
11974 While there is nothing preventing you from writing C code inside a
11978 function, there are many ways to shoot yourself in the foot doing this,
11979 and it is recommended that you stick to inline assembler.
11982 Use of Labels within Inline Assembler
11985 SDCC allows the use of in-line assembler with a few restrictions regarding
11987 In older versions of the compiler all labels defined within inline assembler
11996 where nnnn is a number less than 100 (which implies a limit of utmost 100
11997 inline assembler labels
12011 \begin_inset LatexCommand \index{\_asm}
12041 \begin_inset LatexCommand \index{\_endasm}
12048 Inline assembler code cannot reference any C-Labels, however it can reference
12050 \begin_inset LatexCommand \index{Labels}
12054 defined by the inline assembler, e.g.:
12079 ; some assembler code
12099 /* some more c code */
12101 clabel:\SpecialChar ~
12103 /* inline assembler cannot reference this label */
12115 $0003: ;label (can be referenced by inline assembler only)
12127 /* some more c code */
12132 In other words inline assembly code can access labels defined in inline
12133 assembly within the scope of the function.
12134 The same goes the other way, i.e.
12135 labels defines in inline assembly can not be accessed by C statements.
12138 Interfacing with Assembler Code
12139 \begin_inset LatexCommand \index{Assembler routines}
12146 Global Registers used for Parameter Passing
12147 \begin_inset LatexCommand \index{Parameter passing}
12154 The compiler always uses the global registers
12157 \begin_inset LatexCommand \index{DPTR, DPH, DPL}
12162 \begin_inset LatexCommand \index{DPTR}
12167 \begin_inset LatexCommand \index{B (mcs51, ds390 register)}
12176 \begin_inset LatexCommand \index{ACC (mcs51, ds390 register)}
12182 to pass the first parameter to a routine.
12183 The second parameter onwards is either allocated on the stack (for reentrant
12194 -stack-auto is used) or in data / xdata memory (depending on the memory
12199 Assembler Routine (non-reentrant)
12202 In the following example
12203 \begin_inset LatexCommand \index{reentrant}
12208 \begin_inset LatexCommand \index{Assembler routines (non-reentrant)}
12212 the function c_func calls an assembler routine asm_func, which takes two
12214 \begin_inset LatexCommand \index{function parameter}
12223 extern int asm_func(unsigned char, unsigned char);
12227 int c_func (unsigned char i, unsigned char j)
12235 return asm_func(i,j);
12249 return c_func(10,9);
12254 The corresponding assembler function is:
12259 .globl _asm_func_PARM_2
12360 \begin_inset LatexCommand \index{DPTR, DPH, DPL}
12377 Note here that the return values
12378 \begin_inset LatexCommand \index{return value}
12382 are placed in 'dpl' - One byte return value, 'dpl' LSB & 'dph' MSB for
12384 'dpl', 'dph' and 'b' for three byte values (generic pointers) and 'dpl','dph','
12385 b' & 'acc' for four byte values.
12388 The parameter naming convention is _<function_name>_PARM_<n>, where n is
12389 the parameter number starting from 1, and counting from the left.
12390 The first parameter is passed in
12391 \begin_inset Quotes eld
12395 \begin_inset Quotes erd
12398 for a one byte parameter,
12399 \begin_inset Quotes eld
12403 \begin_inset Quotes erd
12407 \begin_inset Quotes eld
12411 \begin_inset Quotes erd
12414 for three bytes and
12415 \begin_inset Quotes eld
12419 \begin_inset Quotes erd
12422 for a four bytes parameter.
12423 The variable name for the second parameter will be _<function_name>_PARM_2.
12427 Assemble the assembler routine with the following command:
12434 asx8051 -losg asmfunc.asm
12441 Then compile and link the assembler routine to the C source file with the
12449 sdcc cfunc.c asmfunc.rel
12452 Assembler Routine (reentrant)
12456 \begin_inset LatexCommand \index{reentrant}
12461 \begin_inset LatexCommand \index{Assembler routines (reentrant)}
12465 the second parameter
12466 \begin_inset LatexCommand \index{function parameter}
12470 onwards will be passed on the stack, the parameters are pushed from right
12472 after the call the leftmost parameter will be on the top of the stack.
12473 Here is an example:
12478 extern int asm_func(unsigned char, unsigned char);
12482 int c_func (unsigned char i, unsigned char j) reentrant
12490 return asm_func(i,j);
12504 return c_func(10,9);
12509 The corresponding assembler routine is:
12609 The compiling and linking procedure remains the same, however note the extra
12610 entry & exit linkage required for the assembler code, _bp is the stack
12611 frame pointer and is used to compute the offset into the stack for parameters
12612 and local variables.
12616 \begin_inset LatexCommand \index{int (16 bit)}
12621 \begin_inset LatexCommand \index{long (32 bit)}
12628 For signed & unsigned int (16 bit) and long (32 bit) variables, division,
12629 multiplication and modulus operations are implemented by support routines.
12630 These support routines are all developed in ANSI-C to facilitate porting
12631 to other MCUs, although some model specific assembler optimizations are
12633 The following files contain the described routines, all of them can be
12634 found in <installdir>/share/sdcc/lib.
12640 \begin_inset Tabular
12641 <lyxtabular version="3" rows="11" columns="2">
12643 <column alignment="center" valignment="top" leftline="true" width="0">
12644 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
12645 <row topline="true" bottomline="true">
12646 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12656 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12667 <row topline="true">
12668 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12676 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12681 16 bit multiplication
12685 <row topline="true">
12686 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12694 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12699 signed 16 bit division (calls _divuint)
12703 <row topline="true">
12704 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12712 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12717 unsigned 16 bit division
12721 <row topline="true">
12722 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12730 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12735 signed 16 bit modulus (calls _moduint)
12739 <row topline="true">
12740 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12748 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12753 unsigned 16 bit modulus
12757 <row topline="true">
12758 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12766 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12771 32 bit multiplication
12775 <row topline="true">
12776 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12784 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12789 signed 32 division (calls _divulong)
12793 <row topline="true">
12794 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12802 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12807 unsigned 32 division
12811 <row topline="true">
12812 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12820 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12825 signed 32 bit modulus (calls _modulong)
12829 <row topline="true" bottomline="true">
12830 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12838 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12843 unsigned 32 bit modulus
12856 Since they are compiled as
12861 \begin_inset LatexCommand \index{reentrant}
12866 \begin_inset LatexCommand \index{interrupt}
12870 service routines should not do any of the above operations.
12871 If this is unavoidable then the above routines will need to be compiled
12885 \begin_inset LatexCommand \index{-\/-stack-auto}
12891 option, after which the source program will have to be compiled with
12904 \begin_inset LatexCommand \index{-\/-int-long-reent}
12911 Notice that you don't have to call these routines directly.
12912 The compiler will use them automatically every time an integer operation
12916 Floating Point Support
12917 \begin_inset LatexCommand \index{Floating point support}
12924 SDCC supports IEEE (single precision 4 bytes) floating point numbers.The
12925 floating point support routines are derived from gcc's floatlib.c and consist
12926 of the following routines:
12934 \begin_inset Tabular
12935 <lyxtabular version="3" rows="17" columns="2">
12937 <column alignment="center" valignment="top" leftline="true" width="0">
12938 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
12939 <row topline="true" bottomline="true">
12940 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12957 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12966 <row topline="true">
12967 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
12984 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
12998 add floating point numbers
13002 <row topline="true">
13003 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13020 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13034 subtract floating point numbers
13038 <row topline="true">
13039 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13056 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13070 divide floating point numbers
13074 <row topline="true">
13075 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13092 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13106 multiply floating point numbers
13110 <row topline="true">
13111 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13128 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13142 convert floating point to unsigned char
13146 <row topline="true">
13147 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13164 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13178 convert floating point to signed char
13182 <row topline="true">
13183 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13200 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13214 convert floating point to unsigned int
13218 <row topline="true">
13219 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13236 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13250 convert floating point to signed int
13254 <row topline="true">
13255 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13281 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13295 convert floating point to unsigned long
13299 <row topline="true">
13300 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13317 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13331 convert floating point to signed long
13335 <row topline="true">
13336 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13353 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13367 convert unsigned char to floating point
13371 <row topline="true">
13372 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13389 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13403 convert char to floating point number
13407 <row topline="true">
13408 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13425 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13439 convert unsigned int to floating point
13443 <row topline="true">
13444 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13461 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13475 convert int to floating point numbers
13479 <row topline="true">
13480 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13497 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13511 convert unsigned long to floating point number
13515 <row topline="true" bottomline="true">
13516 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
13533 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
13547 convert long to floating point number
13560 These support routines are developed in ANSI-C so there is room for space
13561 and speed improvement.
13562 Note if all these routines are used simultaneously the data space might
13564 For serious floating point usage it is recommended that the large model
13566 Also notice that you don't have to call this routines directly.
13567 The compiler will use them automatically every time a floating point operation
13572 \begin_inset LatexCommand \index{Libraries}
13581 <pending: this is messy and incomplete - a little more information is in
13582 sdcc/doc/libdoc.txt
13587 Compiler support routines (_gptrget, _mulint etc.)
13590 Stdclib functions (puts, printf, strcat etc.)
13591 \layout Subsubsection
13597 \begin_inset LatexCommand \index{<stdio.h>}
13601 As usual on embedded systems you have to provide your own
13604 \begin_inset LatexCommand \index{getchar()}
13613 \begin_inset LatexCommand \index{putchar()}
13620 SDCC does not know whether the system connects to a serial line with or
13621 without handshake, LCD, keyboard or other device.
13622 You'll find examples for serial routines f.e.
13623 in sdcc/device/lib.
13626 If you're short on memory you might want to use
13637 \begin_inset LatexCommand \index{printf()}
13644 For the mcs51 there is an assembly version
13648 which should fit the requirements of many embedded systems (by unsetting
13649 #defines it can be customized to
13653 support long variables and field widths).
13656 Math functions (sin, pow, sqrt etc.)
13663 \begin_inset LatexCommand \index{Libraries}
13667 included in SDCC should have a license at least as liberal as the GNU Lesser
13668 General Public License
13669 \begin_inset LatexCommand \index{GNU Lesser General Public License, LGPL}
13680 license statements for the libraries are missing.
13681 sdcc/device/lib/ser_ir.c
13685 come with a GPL (as opposed to LGPL) License - this will not be liberal
13686 enough for many embedded programmers.
13689 If you have ported some library or want to share experience about some code
13691 falls into any of these categories Busses (I
13692 \begin_inset Formula $^{\textrm{2}}$
13695 C, CAN, Ethernet, Profibus, Modbus, USB, SPI, JTAG ...), Media (IDE, Memory
13696 cards, eeprom, flash...), En-/Decryption, Remote debugging, Realtime kernel,
13697 Keyboard, LCD, RTC, FPGA, PID then the sdcc-user mailing list
13698 \begin_inset LatexCommand \url{http://sourceforge.net/mail/?group_id=599}
13703 would certainly like to hear about it.
13704 Programmers coding for embedded systems are not especially famous for being
13705 enthusiastic, so don't expect a big hurray but as the mailing list is searchabl
13706 e these references are very valuable.
13707 Let's help to create a climate where information is shared.
13713 MCS51 Memory Models
13714 \begin_inset LatexCommand \index{Memory model}
13719 \begin_inset LatexCommand \index{MCS51 memory model}
13724 \layout Subsubsection
13729 SDCC allows two memory models for MCS51 code,
13738 Modules compiled with different memory models should
13742 be combined together or the results would be unpredictable.
13743 The library routines supplied with the compiler are compiled as both small
13745 The compiled library modules are contained in separate directories as small
13746 and large so that you can link to either set.
13750 When the large model is used all variables declared without a storage class
13751 will be allocated into the external ram, this includes all parameters and
13752 local variables (for non-reentrant
13753 \begin_inset LatexCommand \index{reentrant}
13758 When the small model is used variables without storage class are allocated
13759 in the internal ram.
13762 Judicious usage of the processor specific storage classes
13763 \begin_inset LatexCommand \index{Storage class}
13767 and the 'reentrant' function type will yield much more efficient code,
13768 than using the large model.
13769 Several optimizations are disabled when the program is compiled using the
13770 large model, it is therefore recommended that the small model be used unless
13771 absolutely required.
13772 \layout Subsubsection
13775 \begin_inset LatexCommand \label{sub:External-Stack}
13780 \begin_inset LatexCommand \index{stack}
13785 \begin_inset LatexCommand \index{External stack (mcs51)}
13796 : this option wasn't maintained for a long time and is quite buggy.
13797 Small programs might work.
13798 You've been warned!
13801 The external stack (-
13812 \begin_inset LatexCommand \index{-\/-xstack}
13816 ) is located in pdata
13817 \begin_inset LatexCommand \index{pdata (mcs51, ds390 storage class)}
13821 memory (usually at the start of the external ram segment) and is 256 bytes
13833 -xstack option is used to compile the program, the parameters and local
13835 \begin_inset LatexCommand \index{local variables}
13839 of all reentrant functions are allocated in this area.
13840 This option is provided for programs with large stack space requirements.
13841 When used with the -
13852 \begin_inset LatexCommand \index{-\/-stack-auto}
13856 option, all parameters and local variables are allocated on the external
13857 stack (note: support libraries will need to be recompiled with the same
13861 The compiler outputs the higher order address byte of the external ram segment
13863 \begin_inset LatexCommand \index{P2 (mcs51 sfr)}
13868 \begin_inset LatexCommand \ref{sub:MCS51-variants}
13872 ), therefore when using the External Stack option, this port
13876 be used by the application program.
13880 \begin_inset LatexCommand \index{Memory model}
13885 \begin_inset LatexCommand \index{DS390 memory model}
13892 The only model supported is Flat 24
13893 \begin_inset LatexCommand \index{Flat 24 (DS390 memory model)}
13898 This generates code for the 24 bit contiguous addressing mode of the Dallas
13900 In this mode, up to four meg of external RAM or code space can be directly
13902 See the data sheets at www.dalsemi.com for further information on this part.
13906 Note that the compiler does not generate any code to place the processor
13907 into 24 bitmode (although
13911 in the ds390 libraries will do that for you).
13917 \begin_inset LatexCommand \index{Tinibios (DS390)}
13921 , the boot loader or similar code must ensure that the processor is in 24
13922 bit contiguous addressing mode before calling the SDCC startup code.
13940 option, variables will by default be placed into the XDATA segment.
13945 Segments may be placed anywhere in the 4 meg address space using the usual
13957 Note that if any segments are located above 64K, the -r flag must be passed
13958 to the linker to generate the proper segment relocations, and the Intel
13959 HEX output format must be used.
13960 The -r flag can be passed to the linker by using the option
13964 on the SDCC command line.
13965 However, currently the linker can not handle code segments > 64k.
13969 \begin_inset LatexCommand \index{Pragmas}
13976 SDCC supports the following #pragma directives:
13980 \begin_inset LatexCommand \index{\#pragma save}
13984 - this will save all current options to the save/restore stack.
13985 See #pragma\SpecialChar ~
13990 \begin_inset LatexCommand \index{\#pragma restore}
13994 - will restore saved options from the last save.
13995 saves & restores can be nested.
13996 SDCC uses a save/restore stack: save pushes current options to the stack,
13997 restore pulls current options from the stack.
13998 See #pragma\SpecialChar ~
14005 \begin_inset LatexCommand \index{\#pragma callee\_saves}
14010 \begin_inset LatexCommand \index{function prologue}
14014 function1[,function2[,function3...]] - The compiler by default uses a caller
14015 saves convention for register saving across function calls, however this
14016 can cause unnecessary register pushing & popping
14017 \begin_inset LatexCommand \index{push/pop}
14021 when calling small functions from larger functions.
14022 This option can be used to switch off the register saving convention for
14023 the function names specified.
14024 The compiler will not save registers when calling these functions, extra
14025 code need to be manually inserted at the entry & exit for these functions
14026 to save & restore the registers used by these functions, this can SUBSTANTIALLY
14027 reduce code & improve run time performance of the generated code.
14028 In the future the compiler (with inter procedural analysis) may be able
14029 to determine the appropriate scheme to use for each function call.
14040 -callee-saves command line option is used, the function names specified
14041 in #pragma\SpecialChar ~
14043 \begin_inset LatexCommand \index{\#pragma callee\_saves}
14047 is appended to the list of functions specified in the command line.
14051 \begin_inset LatexCommand \index{\#pragma exclude}
14055 none | {acc[,b[,dpl[,dph]]] - The exclude pragma disables the generation
14056 of pairs of push/pop
14057 \begin_inset LatexCommand \index{push/pop}
14066 \begin_inset LatexCommand \index{interrupt}
14079 The directive should be placed immediately before the ISR function definition
14080 and it affects ALL ISR functions following it.
14081 To enable the normal register saving for ISR functions use #pragma\SpecialChar ~
14082 exclude\SpecialChar ~
14084 \begin_inset LatexCommand \index{\#pragma exclude}
14089 See also the related keyword _naked
14090 \begin_inset LatexCommand \index{\_naked}
14098 \begin_inset LatexCommand \index{\#pragma less\_pedantic}
14102 - the compiler will not warn you anymore for obvious mistakes, you'r on
14107 \begin_inset LatexCommand \index{\#pragma nogcse}
14111 - will stop global common subexpression elimination.
14115 \begin_inset LatexCommand \index{\#pragma noinduction}
14119 - will stop loop induction optimizations.
14123 \begin_inset LatexCommand \index{\#pragma noinvariant}
14127 - will not do loop invariant optimizations.
14128 For more details see Loop Invariants in section
14129 \begin_inset LatexCommand \ref{sub:Loop-Optimizations}
14137 \begin_inset LatexCommand \index{\#pragma noiv}
14141 - Do not generate interrupt
14142 \begin_inset LatexCommand \index{interrupt}
14146 vector table entries for all ISR functions defined after the pragma.
14147 This is useful in cases where the interrupt vector table must be defined
14148 manually, or when there is a secondary, manually defined interrupt vector
14150 for the autovector feature of the Cypress EZ-USB FX2).
14151 More elegantly this can be achieved by obmitting the optional interrupt
14152 number after the interrupt keyword, see section
14153 \begin_inset LatexCommand \ref{sub:Interrupt-Service-Routines}
14162 \begin_inset LatexCommand \index{\#pragma nojtbound}
14166 - will not generate code for boundary value checking, when switch statements
14167 are turned into jump-tables (dangerous).
14168 For more details see section
14169 \begin_inset LatexCommand \ref{sub:'switch'-Statements}
14177 \begin_inset LatexCommand \index{\#pragma noloopreverse}
14181 - Will not do loop reversal optimization
14185 \begin_inset LatexCommand \index{\#pragma nooverlay}
14189 - the compiler will not overlay the parameters and local variables of a
14194 \begin_inset LatexCommand \index{\#pragma stackauto}
14209 \begin_inset LatexCommand \index{-\/-stack-auto}
14214 \begin_inset LatexCommand \ref{sec:Parameters-and-Local-Variables}
14218 Parameters and Local Variables.
14221 SDCPP supports the following #pragma directives:
14225 \begin_inset LatexCommand \index{\#pragma preproc\_asm}
14229 (+ | -) - switch _asm _endasm block preprocessing on / off.
14233 The pragma's are intended to be used to turn-on or off certain optimizations
14234 which might cause the compiler to generate extra stack / data space to
14235 store compiler generated temporary variables.
14236 This usually happens in large functions.
14237 Pragma directives should be used as shown in the following example, they
14238 are used to control options & optimizations for a given function; pragmas
14239 should be placed before and/or after a function, placing pragma's inside
14240 a function body could have unpredictable results.
14246 \begin_inset LatexCommand \index{\#pragma save}
14257 /* save the current settings */
14260 \begin_inset LatexCommand \index{\#pragma nogcse}
14269 /* turnoff global subexpression elimination */
14271 #pragma noinduction
14272 \begin_inset LatexCommand \index{\#pragma noinduction}
14276 /* turn off induction optimizations */
14299 \begin_inset LatexCommand \index{\#pragma restore}
14303 /* turn the optimizations back on */
14306 The compiler will generate a warning message when extra space is allocated.
14307 It is strongly recommended that the save and restore pragma's be used when
14308 changing options for a function.
14311 Defines Created by the Compiler
14314 The compiler creates the following #defines
14315 \begin_inset LatexCommand \index{\#defines}
14320 \begin_inset LatexCommand \index{Defines created by the compiler}
14330 \begin_inset Tabular
14331 <lyxtabular version="3" rows="10" columns="2">
14333 <column alignment="center" valignment="top" leftline="true" width="0">
14334 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
14335 <row topline="true" bottomline="true">
14336 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14346 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14357 <row topline="true">
14358 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14364 \begin_inset LatexCommand \index{SDCC}
14371 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14376 this Symbol is always defined
14380 <row topline="true">
14381 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14387 \begin_inset LatexCommand \index{SDCC\_mcs51}
14392 \begin_inset LatexCommand \index{SDCC\_ds390}
14397 \begin_inset LatexCommand \index{SDCC\_z80}
14404 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14409 depending on the model used (e.g.: -mds390
14413 <row topline="true">
14414 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14420 \begin_inset LatexCommand \index{\_\_mcs51}
14425 \begin_inset LatexCommand \index{\_\_ds390}
14430 \begin_inset LatexCommand \index{\_\_hc08}
14435 \begin_inset LatexCommand \index{\_\_z80}
14442 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14447 depending on the model used (e.g.
14452 <row topline="true">
14453 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14459 \begin_inset LatexCommand \index{SDCC\_STACK\_AUTO}
14466 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14489 <row topline="true">
14490 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14496 \begin_inset LatexCommand \index{SDCC\_MODEL\_SMALL}
14503 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14526 <row topline="true">
14527 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14533 \begin_inset LatexCommand \index{SDCC\_MODEL\_LARGE}
14540 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14563 <row topline="true">
14564 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14570 \begin_inset LatexCommand \index{SDCC\_USE\_XSTACK}
14577 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14600 <row topline="true">
14601 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14607 \begin_inset LatexCommand \index{SDCC\_STACK\_TENBIT}
14614 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14627 <row topline="true" bottomline="true">
14628 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
14634 \begin_inset LatexCommand \index{SDCC\_MODEL\_FLAT24}
14641 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
14661 Notes on supported Processors
14665 \begin_inset LatexCommand \label{sub:MCS51-variants}
14670 \begin_inset LatexCommand \index{MCS51 variants}
14677 MCS51 processors are available from many vendors and come in many different
14679 While they might differ considerably in respect to Special Function Registers
14680 the core MCS51 is usually not modified or is kept compatible.
14684 pdata access by SFR
14687 With the upcome of devices with internal xdata and flash memory devices
14689 \begin_inset LatexCommand \index{P2 (mcs51 sfr)}
14693 as dedicated I/O port is becoming more popular.
14694 Switching the high byte for pdata
14695 \begin_inset LatexCommand \index{pdata (mcs51, ds390 storage class)}
14699 access which was formerly done by port P2 is then achieved by a Special
14701 \begin_inset LatexCommand \index{sfr}
14706 In well-established MCS51 tradition the address of this
14710 is where the chip designers decided to put it.
14711 Needless to say that they didn't agree on a common name either.
14712 So that the startup code can correctly initialize xdata variables, you
14713 should define an sfr with the name _XPAGE
14716 \begin_inset LatexCommand \index{\_XPAGE (mcs51)}
14722 at the appropriate location if the default, port P2, is not used for this.
14728 sfr at 0x92 _XPAGE; /* Cypress EZ-USB family */
14733 sfr at 0xaf _XPAGE; /* some Silicon Labs (Cygnal) chips */
14738 sfr at 0xaa _XPAGE; /* some Silicon Labs (Cygnal) chips */
14741 For more exotic implementations further customizations may be needed.
14743 \begin_inset LatexCommand \ref{sub:Startup-Code}
14747 for other possibilities.
14750 Other Features available by SFR
14753 Some MCS51 variants offer features like Double DPTR
14754 \begin_inset LatexCommand \index{DPTR}
14758 , multiple DPTR, decrementing DPTR, 16x16 Multiply.
14759 These are currently not used for the MCS51 port.
14760 If you absolutely need them you can fall back to inline assembly or submit
14767 The DS80C400 microcontroller has a rich set of peripherals.
14768 In its built-in ROM library it includes functions to access some of the
14769 features, among them is a TCP stack with IP4 and IP6 support.
14770 Library headers (currently in beta status) and other files are provided
14774 \begin_inset LatexCommand \url{ftp://ftp.dalsemi.com/pub/tini/ds80c400/c_libraries/sdcc/index.html}
14782 The Z80 and gbz80 port
14785 SDCC can target both the Zilog
14786 \begin_inset LatexCommand \index{Z80}
14790 and the Nintendo Gameboy's Z80-like gbz80
14791 \begin_inset LatexCommand \index{gbz80 (GameBoy Z80)}
14796 The Z80 port is passed through the same
14799 \begin_inset LatexCommand \index{Regression test}
14805 as the MCS51 and DS390 ports, so floating point support, support for long
14806 variables and bitfield support is fine.
14807 See mailing lists and forums about interrupt routines.
14810 As always, the code is the authoritative reference - see z80/ralloc.c and
14813 \begin_inset LatexCommand \index{stack}
14817 frame is similar to that generated by the IAR Z80 compiler.
14818 IX is used as the base pointer, HL and IY are used as a temporary registers,
14819 and BC and DE are available for holding variables.
14821 \begin_inset LatexCommand \index{return value}
14825 for the Z80 port are stored in L (one byte), HL (two bytes), or DEHL (four
14827 The gbz80 port use the same set of registers for the return values, but
14828 in a different order of significance: E (one byte), DE (two bytes), or
14835 The port to the Motorola HC08
14836 \begin_inset LatexCommand \index{HC08}
14840 family has been added in October 2003, and is still undergoing some basic
14842 The code generator is complete, but the register allocation is still quite
14844 Some of the SDCC's standard C library functions have embedded non-HC08
14845 inline assembly and so are not yet usable.
14856 \begin_inset LatexCommand \index{PIC14}
14860 port still requires a major effort from the development community.
14861 However it can work for very simple code.
14864 C code and 14bit PIC code page
14865 \begin_inset LatexCommand \index{code page (pic14)}
14870 \begin_inset LatexCommand \index{RAM bank (pic14)}
14877 The linker organizes allocation for the code page and RAM banks.
14878 It does not have intimate knowledge of the code flow.
14879 It will put all the code section of a single asm file into a single code
14881 In order to make use of multiple code pages, separate asm files must be
14883 The compiler treats all functions of a single C file as being in the same
14884 code page unless it is non static.
14885 The compiler treats all local variables of a single C file as being in
14886 the same RAM bank unless it is an extern.
14890 To get the best follow these guide lines:
14893 make local functions static, as non static functions require code page selection
14897 Make local variables static as extern variables require RAM bank selection
14901 For devices that have multiple code pages it is more efficient to use the
14902 same number of files as pages, i.e.
14903 for the 16F877 use 4 separate files and i.e.
14904 for the 16F874 use 2 separate files.
14905 This way the linker can put the code for each file into different code
14906 pages and the compiler can allocate reusable variables more efficiently
14907 and there's less page selection overhead.
14908 And as for any 8 bit micro (especially for PIC 14 as they have a very simple
14909 instruction set) use 'unsigned char' whereever possible instead of 'int'.
14912 Creating a device include file
14915 For generating a device include file use the support perl script inc2h.pl
14916 kept in directory support/script.
14922 For the interrupt function, use the keyword 'interrupt'
14923 \begin_inset LatexCommand \index{interrupt}
14927 with level number of 0 (PIC14 only has 1 interrupt so this number is only
14928 there to avoid a syntax error - it ought to be fixed).
14934 void Intr(void) interrupt 0
14940 T0IF = 0; /* Clear timer interrupt */
14945 Linking and assembling
14948 For assembling you can use either GPUTILS'
14949 \begin_inset LatexCommand \index{gputils (pic tools)}
14953 gpasm.exe or MPLAB's mpasmwin.exe.
14954 GPUTILS is available from
14955 \begin_inset LatexCommand \url{http://gputils.sourceforge.net/}
14960 For linking you can use either GPUTIL's gplink or MPLAB's mplink.exe.
14961 If you use MPLAB and an interrupt function then the linker script file
14962 vectors section will need to be enlarged to link with mplink.
14985 sdcc -S -V -mpic14 -p16F877 $<
14999 $(PRJ).hex: $(OBJS)
15009 gplink -m -s $(PRJ).lkr -o $(PRJ).hex $(OBJS)
15031 sdcc -S -V -mpic14 -p16F877 $<
15041 mpasmwin /q /o $*.asm
15045 $(PRJ).hex: $(OBJS)
15055 mplink /v $(PRJ).lkr /m $(PRJ).map /o $(PRJ).hex $(OBJS)
15058 Please note that indentations within a
15062 have to be done with a tabulator character.
15066 \begin_inset LatexCommand \index{PIC16}
15074 \begin_inset LatexCommand \index{PIC16}
15078 port is the portion of SDCC that is responsible to produce code for the
15080 \begin_inset LatexCommand \index{Microchip}
15084 (TM) microcontrollers with 16 bit core.
15085 Currently this family of microcontrollers contains the PIC18Fxxx and PIC18Fxxxx.
15091 PIC16 port supports the standard command line arguments as supposed, with
15092 the exception of certain cases that will be mentioned in the following
15095 \labelwidthstring 00.00.0000
15107 -stack-auto Auto variables that are function parameters, will be saved on
15111 There is no need to specify this in the command line.
15113 \labelwidthstring 00.00.0000
15125 -float-reent All floating point functions are reentrant by default.
15128 There is no need to specifiy this in the command line.
15130 \labelwidthstring 00.00.0000
15142 -callee-saves See -
15154 \labelwidthstring 00.00.0000
15166 -all-callee-saves All function arguments are passed on stack by default.
15169 There is no need to specify this in the command line.
15171 \labelwidthstring 00.00.0000
15183 -fommit-frame-pointer Frame pointer will be omitted when the function uses
15184 no local variables.
15187 Port Specific Options
15188 \begin_inset LatexCommand \index{Options PIC16}
15195 The port specific options appear after the global options in the sdcc --help
15197 \layout Subsubsection
15202 General options enable certain port features and optimizations.
15204 \labelwidthstring 00.00.0000
15216 -pgen-bank Instructs the port to insert BANKSEL directives before instructions
15217 that use the Bank Select Register (BSR).
15219 \labelwidthstring 00.00.0000
15231 -pomit-config-words Instructs the port to omit the generation of the configurati
15234 \labelwidthstring 00.00.0000
15246 -pomit-ivt Instructs the port to omit the generation of the interrupt vectors
15248 \labelwidthstring 00.00.0000
15260 -pleave-reset-vector Used in conjuction with the previous command, instructs
15261 the port NOT to omit the reset vector.
15263 \labelwidthstring 00.00.0000
15275 -stack-model=[model] Used in conjuction with the command above.
15276 Defines the stack model to be used, valid stack models are :
15279 \labelwidthstring 00.00.0000
15285 Selects small stack model.
15286 8 bit stack and frame pointers.
15287 Supports 256 bytes stack size.
15289 \labelwidthstring 00.00.0000
15295 Selects large stack model.
15296 16 bit stack and frame pointers.
15297 Supports 65536 bytes stack size.
15300 \labelwidthstring 00.00.0000
15312 -preplace-udata-with=[kword] Replaces the default udata keyword for allocating
15313 unitialized data variables with [kword].
15314 Valid keywords are: "udata_acs", "udata_shr", "udata_ovr".
15316 \labelwidthstring 00.00.0000
15328 -ivt-loc <nnnn> positions the Interrupt Vector Table at location <nnnn>.
15329 Useful for bootloaders.
15331 \labelwidthstring 00.00.0000
15343 -asm= sets the full path and name of an external assembler to call.
15345 \labelwidthstring 00.00.0000
15357 -link= sets the full path and name of an external linker to call.
15358 \layout Subsubsection
15363 Debugging options enable extra debugging information in the output files.
15365 \labelwidthstring 00.00.0000
15377 -debug-xtra Similar to -
15388 \begin_inset LatexCommand \index{-\/-debug}
15392 , but dumps more information.
15394 \labelwidthstring 00.00.0000
15406 -debug-ralloc Force register allocator to dump <source>.d file with debugging
15408 <source> is the name of the file compiled.
15410 \labelwidthstring 00.00.0000
15422 -pcode-verbose Enable pcode debugging information in translation.
15425 Preprocessor Macros
15428 PIC16 port defines the following preprocessor macros while translating a
15433 \begin_inset Tabular
15434 <lyxtabular version="3" rows="2" columns="2">
15436 <column alignment="center" valignment="top" leftline="true" width="0">
15437 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
15438 <row topline="true" bottomline="true">
15439 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15447 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15456 <row topline="true" bottomline="true">
15457 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15465 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15470 MCU Identification.
15475 is the microcontrol identification number, i.e.
15491 \begin_inset LatexCommand \index{PIC16}
15495 port uses the following directories for searching header files and libraries.
15499 \begin_inset Tabular
15500 <lyxtabular version="3" rows="3" columns="4">
15502 <column alignment="center" valignment="top" leftline="true" width="0">
15503 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
15504 <column alignment="center" valignment="top" width="0">
15505 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
15506 <row topline="true" bottomline="true">
15507 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15515 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15523 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15531 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15540 <row topline="true">
15541 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15546 PREFIX/sdcc/include/pic16
15549 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15554 PIC16 specific headers
15557 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15565 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15574 <row topline="true" bottomline="true">
15575 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15580 PREFIX/sdcc/lib/pic16
15583 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15588 PIC16 specific libraries
15591 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15599 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15616 \begin_inset LatexCommand \label{sub:PIC16_Pragmas}
15623 PIC16 port currently supports the following pragmas:
15625 \labelwidthstring 00.00.0000
15627 stack pragma stack forces the code generator to initialize the stack & frame
15628 pointers at a specific address.
15629 This is an adhoc solution since gplink does not support yet stack.
15630 When the gplink issue is resolved the pragma will be deprecated
15638 It is important to initialize the stack, otherwise strange things can happen.
15639 Stack is not initialized by default because there are some sources that
15641 (like library sources)
15646 The stack pragma should be used only once in a project.
15647 Multiple pragmas may result in indeterminate behaviour of the program.
15655 /* initializes stack at RAM address 0x5ff */
15658 #pragma stack 0x5ff
15660 \labelwidthstring 00.00.0000
15662 udata pragma udata instructs the compiler to emit code so that linker will
15663 place a variable at a specific memory bank
15671 /* places variable foo at bank2 */
15674 #pragma udata bank2 foo
15680 In order for this pragma to work there are some changes that must be made
15681 in the .lkr script used in link stage.
15682 In the following example a sample .lkr file is shown:
15687 // Sample linker script for the PIC18F452 processor
15693 CODEPAGE NAME=vectors START=0x0 END=0x29 PROTECTED
15696 CODEPAGE NAME=page START=0x2A END=0x7FFF
15699 CODEPAGE NAME=idlocs START=0x200000 END=0x200007 PROTECTED
15702 CODEPAGE NAME=config START=0x300000 END=0x30000D PROTECTED
15705 CODEPAGE NAME=devid START=0x3FFFFE END=0x3FFFFF PROTECTED
15708 CODEPAGE NAME=eedata START=0xF00000 END=0xF000FF PROTECTED
15711 ACCESSBANK NAME=accessram START=0x0 END=0x7F
15716 DATABANK NAME=gpr0 START=0x80 END=0xFF
15719 DATABANK NAME=gpr1 START=0x100 END=0x1FF
15722 DATABANK NAME=gpr2 START=0x200 END=0x2FF
15725 DATABANK NAME=gpr3 START=0x300 END=0x3FF
15728 DATABANK NAME=gpr4 START=0x400 END=0x4FF
15731 DATABANK NAME=gpr5 START=0x500 END=0x5FF
15734 ACCESSBANK NAME=accesssfr START=0xF80 END=0xFFF PROTECTED
15739 SECTION NAME=CONFIG ROM=config
15744 SECTION NAME=bank0 RAM=gpr0
15747 SECTION NAME=bank1 RAM=gpr1
15750 SECTION NAME=bank2 RAM=gpr2
15753 SECTION NAME=bank3 RAM=gpr3
15756 SECTION NAME=bank4 RAM=gpr4
15759 SECTION NAME=bank5 RAM=gpr5
15762 The linker will recognise the section name set in the pragma statement and
15763 will position the variable at the memory bank set with the RAM field at
15764 the SECTION line in the linker script file.
15768 \begin_inset LatexCommand \label{sub:PIC16_Header-Files}
15775 There is one main header file that can be included to the source files using
15782 This header file contains the definitions for the processor special registers,
15783 so it is necessary if the source accesses them.
15784 It can be included by adding the following line in the beginning of the
15788 #include <pic18fregs.h>
15791 The specific microcontroller is selected within the pic18fregs.h automatically,
15792 so the same source can be used with a variety of devices.
15798 The libraries that PIC16
15799 \begin_inset LatexCommand \index{PIC16}
15803 port depends on are the microcontroller device libraries which contain
15804 the symbol definitions for the microcontroller special function registers.
15805 These libraries have the format pic18fxxxx.lib, where
15809 is the microcontroller identification number.
15810 The specific library is selected automatically by the compiler at link
15811 stage according to the selected device.
15814 Libraries are created with gplib which is part of the gputils package
15815 \begin_inset LatexCommand \url{http://gputils.sourceforge.net}
15825 The following memory models are supported by the PIC16 port:
15834 Memory model affects the default size of pointers within the source.
15835 The sizes are shown in the next table:
15839 \begin_inset Tabular
15840 <lyxtabular version="3" rows="3" columns="3">
15842 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
15843 <column alignment="center" valignment="top" leftline="true" width="0">
15844 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
15845 <row topline="true" bottomline="true">
15846 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15851 Pointer sizes according to memory model
15854 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15862 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15871 <row topline="true" bottomline="true">
15872 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15880 <cell multicolumn="1" alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15888 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15897 <row topline="true" bottomline="true">
15898 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
15906 <cell multicolumn="1" alignment="center" valignment="top" topline="true" bottomline="true" leftline="true" usebox="none">
15914 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
15930 It is advisable that all sources within a project are compiled with the
15932 If one wants to override the default memory model, this can be done by
15933 declaring a pointer as
15942 Far selects large memory model's pointers, while near selects small memory
15946 The standard device libraries (see
15947 \begin_inset LatexCommand \ref{sub:PIC16_Header-Files}
15951 ) contain no reference to pointers, so they can be used with both memory
15958 The stack implementation for the PIC16 port uses two indirect registers,
15961 \labelwidthstring 00.00.0000
15963 FSR1 is assigned as stack pointer
15965 \labelwidthstring 00.00.0000
15967 FSR2 is assigned as frame pointer
15970 The following stack models are supported by the PIC16 port
15983 model means that only the FSRxL byte is used to access stack and frame,
15990 uses both FSRxL and FSRxH registers.
15991 The following table shows the stack/frame pointers sizes according to stack
15992 model and the maximum space they can address:
15996 \begin_inset Tabular
15997 <lyxtabular version="3" rows="3" columns="3">
15999 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
16000 <column alignment="center" valignment="top" leftline="true" width="0">
16001 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
16002 <row topline="true" bottomline="true">
16003 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16008 Stack & Frame pointer sizes according to stack model
16011 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16019 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16028 <row topline="true">
16029 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16037 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16045 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16054 <row topline="true" bottomline="true">
16055 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16063 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16071 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16089 Currently stack and frame pointers should be initialized explicit by the
16090 user at the desired Data RAM position (see
16091 \begin_inset LatexCommand \ref{sub:PIC16_Pragmas}
16096 Uninitialized stack and frame pointers can result in unexpected behavior
16097 of the resulting binary.
16100 Function return values
16103 Return values from functions are placed to the appropriate registers following
16104 a modified Microchip policy optimized for SDCC.
16105 The following table shows these registers:
16109 \begin_inset Tabular
16110 <lyxtabular version="3" rows="6" columns="2">
16112 <column alignment="center" valignment="top" leftline="true" width="0">
16113 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0">
16114 <row topline="true" bottomline="true">
16115 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16123 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
16128 destination register
16132 <row topline="true">
16133 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16141 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
16150 <row topline="true">
16151 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16159 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
16168 <row topline="true">
16169 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16177 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
16186 <row topline="true">
16187 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16195 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
16200 FSR0L:PRODH:PRODL:WREG
16204 <row topline="true" bottomline="true">
16205 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
16213 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
16218 on stack, FSR0 points to the beginning
16232 When entering an interrupt, currently the PIC16
16233 \begin_inset LatexCommand \index{PIC16}
16237 port automatically saves the following registers:
16249 PROD (PRODL and PRODH)
16252 FSR0 (FSR0L and FSR0H)
16255 These registers are restored upon return from the interrupt routine.
16259 When entering a high priority interrupt WREG, STATUS and BSR are not explicit
16261 The hardware shadow registers for WREG, STATUS and BSR are used in these
16271 NOTE that when the _naked attribute is specified for an interrupt routine,
16272 then NO registers are stored or restored.
16278 Debugging with SDCDB
16279 \begin_inset LatexCommand \label{cha:Debugging-with-SDCDB}
16284 \begin_inset LatexCommand \index{sdcdb (debugger)}
16291 SDCC is distributed with a source level debugger
16292 \begin_inset LatexCommand \index{Debugger}
16297 The debugger uses a command line interface, the command repertoire of the
16298 debugger has been kept as close to gdb
16299 \begin_inset LatexCommand \index{gdb}
16303 (the GNU debugger) as possible.
16304 The configuration and build process is part of the standard compiler installati
16305 on, which also builds and installs the debugger in the target directory
16306 specified during configuration.
16307 The debugger allows you debug BOTH at the C source and at the ASM source
16309 Sdcdb is available on Unix platforms only.
16312 Compiling for Debugging
16326 \begin_inset LatexCommand \index{-\/-debug}
16330 option must be specified for all files for which debug information is to
16332 The complier generates a .adb file for each of these files.
16333 The linker creates the .cdb
16334 \begin_inset LatexCommand \index{<file>.cdb}
16339 \begin_inset LatexCommand \index{<file>.adb}
16343 files and the address information.
16344 This .cdb is used by the debugger.
16347 How the Debugger Works
16360 -debug option is specified the compiler generates extra symbol information
16361 some of which are put into the assembler source and some are put into the
16363 Then the linker creates the .cdb file from the individual .adb files with
16364 the address information for the symbols.
16365 The debugger reads the symbolic information generated by the compiler &
16366 the address information generated by the linker.
16367 It uses the SIMULATOR (Daniel's S51) to execute the program, the program
16368 execution is controlled by the debugger.
16369 When a command is issued for the debugger, it translates it into appropriate
16370 commands for the simulator.
16373 Starting the Debugger
16376 The debugger can be started using the following command line.
16377 (Assume the file you are debugging has the file name foo).
16391 The debugger will look for the following files.
16394 foo.c - the source file.
16397 foo.cdb - the debugger symbol information file.
16400 foo.ihx - the Intel hex format
16401 \begin_inset LatexCommand \index{Intel hex format}
16408 Command Line Options.
16421 -directory=<source file directory> this option can used to specify the directory
16423 The debugger will look into the directory list specified for source, cdb
16425 The items in the directory list must be separated by ':', e.g.
16426 if the source files can be in the directories /home/src1 and /home/src2,
16437 -directory option should be -
16447 -directory=/home/src1:/home/src2.
16448 Note there can be no spaces in the option.
16452 -cd <directory> - change to the <directory>.
16455 -fullname - used by GUI front ends.
16458 -cpu <cpu-type> - this argument is passed to the simulator please see the
16459 simulator docs for details.
16462 -X <Clock frequency > this options is passed to the simulator please see
16463 the simulator docs for details.
16466 -s <serial port file> passed to simulator see the simulator docs for details.
16469 -S <serial in,out> passed to simulator see the simulator docs for details.
16472 -k <port number> passed to simulator see the simulator docs for details.
16478 As mentioned earlier the command interface for the debugger has been deliberatel
16479 y kept as close the GNU debugger gdb, as possible.
16480 This will help the integration with existing graphical user interfaces
16481 (like ddd, xxgdb or xemacs) existing for the GNU debugger.
16482 If you use a graphical user interface for the debugger you can skip the
16484 \layout Subsubsection*
16486 break [line | file:line | function | file:function]
16489 Set breakpoint at specified line or function:
16498 sdcdb>break foo.c:100
16500 sdcdb>break funcfoo
16502 sdcdb>break foo.c:funcfoo
16503 \layout Subsubsection*
16505 clear [line | file:line | function | file:function ]
16508 Clear breakpoint at specified line or function:
16517 sdcdb>clear foo.c:100
16519 sdcdb>clear funcfoo
16521 sdcdb>clear foo.c:funcfoo
16522 \layout Subsubsection*
16527 Continue program being debugged, after breakpoint.
16528 \layout Subsubsection*
16533 Execute till the end of the current function.
16534 \layout Subsubsection*
16539 Delete breakpoint number 'n'.
16540 If used without any option clear ALL user defined break points.
16541 \layout Subsubsection*
16543 info [break | stack | frame | registers ]
16546 info break - list all breakpoints
16549 info stack - show the function call stack.
16552 info frame - show information about the current execution frame.
16555 info registers - show content of all registers.
16556 \layout Subsubsection*
16561 Step program until it reaches a different source line.
16562 Note: pressing <return> repeats the last command.
16563 \layout Subsubsection*
16568 Step program, proceeding through subroutine calls.
16569 \layout Subsubsection*
16574 Start debugged program.
16575 \layout Subsubsection*
16580 Print type information of the variable.
16581 \layout Subsubsection*
16586 print value of variable.
16587 \layout Subsubsection*
16592 load the given file name.
16593 Note this is an alternate method of loading file for debugging.
16594 \layout Subsubsection*
16599 print information about current frame.
16600 \layout Subsubsection*
16605 Toggle between C source & assembly source.
16606 \layout Subsubsection*
16608 ! simulator command
16611 Send the string following '!' to the simulator, the simulator response is
16613 Note the debugger does not interpret the command being sent to the simulator,
16614 so if a command like 'go' is sent the debugger can loose its execution
16615 context and may display incorrect values.
16616 \layout Subsubsection*
16623 My name is Bobby Brown"
16626 Interfacing with XEmacs
16627 \begin_inset LatexCommand \index{XEmacs}
16632 \begin_inset LatexCommand \index{Emacs}
16639 Two files (in emacs lisp) are provided for the interfacing with XEmacs,
16640 sdcdb.el and sdcdbsrc.el.
16641 These two files can be found in the $(prefix)/bin directory after the installat
16643 These files need to be loaded into XEmacs for the interface to work.
16644 This can be done at XEmacs startup time by inserting the following into
16645 your '.xemacs' file (which can be found in your HOME directory):
16651 (load-file sdcdbsrc.el)
16657 .xemacs is a lisp file so the () around the command is REQUIRED.
16658 The files can also be loaded dynamically while XEmacs is running, set the
16659 environment variable 'EMACSLOADPATH' to the installation bin directory
16660 (<installdir>/bin), then enter the following command ESC-x load-file sdcdbsrc.
16661 To start the interface enter the following command:
16675 You will prompted to enter the file name to be debugged.
16680 The command line options that are passed to the simulator directly are bound
16681 to default values in the file sdcdbsrc.el.
16682 The variables are listed below, these values maybe changed as required.
16685 sdcdbsrc-cpu-type '51
16688 sdcdbsrc-frequency '11059200
16691 sdcdbsrc-serial nil
16694 The following is a list of key mapping for the debugger interface.
16705 ;;key\SpecialChar ~
16719 binding\SpecialChar ~
16743 ;;---\SpecialChar ~
16757 -------\SpecialChar ~
16799 sdcdb-next-from-src\SpecialChar ~
16827 sdcdb-back-from-src\SpecialChar ~
16855 sdcdb-cont-from-src\SpecialChar ~
16865 SDCDB continue command
16883 sdcdb-step-from-src\SpecialChar ~
16911 sdcdb-whatis-c-sexp\SpecialChar ~
16921 SDCDB ptypecommand for data at
16988 sdcdbsrc-delete\SpecialChar ~
17002 SDCDB Delete all breakpoints if no arg
17051 given or delete arg (C-u arg x)
17069 sdcdbsrc-frame\SpecialChar ~
17084 SDCDB Display current frame if no arg,
17133 given or display frame arg
17200 sdcdbsrc-goto-sdcdb\SpecialChar ~
17210 Goto the SDCDB output buffer
17228 sdcdb-print-c-sexp\SpecialChar ~
17239 SDCDB print command for data at
17306 sdcdbsrc-goto-sdcdb\SpecialChar ~
17316 Goto the SDCDB output buffer
17334 sdcdbsrc-mode\SpecialChar ~
17350 Toggles Sdcdbsrc mode (turns it off)
17365 sdcdb-finish-from-src\SpecialChar ~
17373 SDCDB finish command
17388 sdcdb-break\SpecialChar ~
17406 Set break for line with point
17421 sdcdbsrc-mode\SpecialChar ~
17437 Toggle Sdcdbsrc mode
17452 sdcdbsrc-srcmode\SpecialChar ~
17475 Here are a few guidelines that will help the compiler generate more efficient
17476 code, some of the tips are specific to this compiler others are generally
17477 good programming practice.
17480 Use the smallest data type to represent your data-value.
17481 If it is known in advance that the value is going to be less than 256 then
17482 use an 'unsigned char' instead of a 'short' or 'int'.
17483 Please note, that ANSI C requires both signed and unsigned chars to be
17484 promoted to 'signed int' before doing any operation.
17485 This promotion can be omitted, if the result is the same.
17486 The effect of the promotion rules together with the sign-extension is often
17493 unsigned char uc = 0xfe;
17495 if (uc * uc < 0) /* this is true! */
17514 (int) uc * (int) uc = (int) 0xfe * (int) 0xfe = (int) 0xfc04 = -1024
17524 (unsigned char) -12 / (signed char) -3 = ...
17527 No, the result is not 4:
17532 (int) (unsigned char) -12 / (int) (signed char) -3 =
17534 (int) (unsigned char) 0xf4 / (int) (signed char) 0xfd =
17536 (int) 0x00f4 / (int) 0xfffd =
17538 (int) 0x00f4 / (int) 0xfffd =
17540 (int) 244 / (int) -3 =
17542 (int) -81 = (int) 0xffaf;
17545 Don't complain, that gcc gives you a different result.
17546 gcc uses 32 bit ints, while SDCC uses 16 bit ints.
17547 Therefore the results are different.
17550 \begin_inset Quotes sld
17554 \begin_inset Quotes srd
17560 If well-defined overflow characteristics are important and negative values
17561 are not, or if you want to steer clear of sign-extension problems when
17562 manipulating bits or bytes, use one of the corresponding unsigned types.
17563 (Beware when mixing signed and unsigned values in expressions, though.)
17565 Although character types (especially unsigned char) can be used as "tiny"
17566 integers, doing so is sometimes more trouble than it's worth, due to unpredicta
17567 ble sign extension and increased code size.
17571 Use unsigned when it is known in advance that the value is not going to
17573 This helps especially if you are doing division or multiplication, bit-shifting
17574 or are using an array index.
17577 NEVER jump into a LOOP.
17580 Declare the variables to be local
17581 \begin_inset LatexCommand \index{local variables}
17585 whenever possible, especially loop control variables (induction).
17588 Since the compiler does not always do implicit integral promotion, the programme
17589 r should do an explicit cast when integral promotion is required.
17592 Reducing the size of division, multiplication & modulus operations can reduce
17593 code size substantially.
17594 Take the following code for example.
17600 foobar(unsigned int p1, unsigned char ch)
17608 unsigned char ch1 = p1 % ch ;
17619 For the modulus operation the variable ch will be promoted to unsigned int
17620 first then the modulus operation will be performed (this will lead to a
17621 call to support routine _moduint()), and the result will be casted to a
17623 If the code is changed to
17628 foobar(unsigned int p1, unsigned char ch)
17636 unsigned char ch1 = (unsigned char)p1 % ch ;
17647 It would substantially reduce the code generated (future versions of the
17648 compiler will be smart enough to detect such optimization opportunities).
17652 Have a look at the assembly listing to get a
17653 \begin_inset Quotes sld
17657 \begin_inset Quotes srd
17660 for the code generation.
17664 \begin_inset LatexCommand \index{Tools}
17668 included in the distribution
17672 \begin_inset Tabular
17673 <lyxtabular version="3" rows="12" columns="3">
17675 <column alignment="center" valignment="top" leftline="true" width="0pt">
17676 <column alignment="center" valignment="top" leftline="true" width="0pt">
17677 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0pt">
17678 <row topline="true" bottomline="true">
17679 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17687 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17695 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17704 <row topline="true">
17705 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17713 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17718 Simulator for various architectures
17721 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17730 <row topline="true">
17731 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17739 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17744 header file conversion
17747 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17752 sdcc/support/scripts
17756 <row topline="true">
17757 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17765 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17770 header file conversion
17773 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17778 sdcc/support/scripts
17782 <row topline="true">
17783 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17791 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17799 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17817 <row topline="true">
17818 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17826 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17834 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17852 <row topline="true">
17853 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17861 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17869 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17887 <row topline="true">
17888 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17896 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17904 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17922 <row topline="true">
17923 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17931 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17939 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17957 <row topline="true">
17958 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17966 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
17974 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
17992 <row topline="true">
17993 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18001 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18009 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18027 <row topline="true" bottomline="true">
18028 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18036 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18044 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18072 \begin_inset LatexCommand \index{Documentation}
18076 included in the distribution
18080 \begin_inset Tabular
18081 <lyxtabular version="3" rows="10" columns="2">
18083 <column alignment="left" valignment="top" leftline="true" width="0">
18084 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0pt">
18085 <row topline="true" bottomline="true">
18086 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18094 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18099 Where to get / filename
18103 <row topline="true">
18104 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18109 SDCC Compiler User Guide
18112 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18117 You're reading it right now
18121 <row topline="true">
18122 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18130 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18139 <row topline="true">
18140 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18146 \begin_inset LatexCommand \index{asXXXX (as-gbz80, as-hc08, asx8051, as-z80)}
18151 \begin_inset LatexCommand \index{Assembler documentation}
18155 Assemblers and ASLINK
18156 \begin_inset LatexCommand \index{aslink}
18161 \begin_inset LatexCommand \index{Linker documentation}
18168 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18173 sdcc/as/doc/asxhtm.html
18177 <row topline="true">
18178 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18183 SDCC regression test
18184 \begin_inset LatexCommand \index{Regression test}
18191 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18196 sdcc/doc/test_suite_spec.pdf
18200 <row topline="true">
18201 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18209 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18218 <row topline="true">
18219 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18224 Notes on debugging with sdcdb
18225 \begin_inset LatexCommand \index{sdcdb (debugger)}
18232 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18237 sdcc/debugger/README
18241 <row topline="true">
18242 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18247 Software simulator for microcontrollers
18250 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18277 <row topline="true">
18278 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18283 Temporary notes on the pic16
18284 \begin_inset LatexCommand \index{PIC16}
18291 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18296 sdcc/src/pic16/NOTES
18300 <row topline="true" bottomline="true">
18301 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18306 SDCC internal documentation (debugging file format)
18309 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18345 Related open source tools
18346 \begin_inset LatexCommand \index{Related tools}
18354 \begin_inset Tabular
18355 <lyxtabular version="3" rows="11" columns="3">
18357 <column alignment="center" valignment="top" leftline="true" width="0pt">
18358 <column alignment="block" valignment="top" leftline="true" width="30line%">
18359 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0pt">
18360 <row topline="true" bottomline="true">
18361 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18369 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18377 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18386 <row topline="true">
18387 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18393 \begin_inset LatexCommand \index{gpsim (pic simulator)}
18400 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18408 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18414 \begin_inset LatexCommand \url{http://www.dattalo.com/gnupic/gpsim.html}
18422 <row topline="true">
18423 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18429 \begin_inset LatexCommand \index{gputils (pic tools)}
18436 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18444 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18450 \begin_inset LatexCommand \url{http://gputils.sourceforge.net/}
18458 <row topline="true">
18459 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18467 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18475 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18481 \begin_inset LatexCommand \url{http://digilander.libero.it/fbradasc/FLP5.html}
18489 <row topline="true">
18490 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18496 \begin_inset LatexCommand \index{indent (source formatting tool)}
18503 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18508 Formats C source - Master of the white spaces
18511 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18517 \begin_inset LatexCommand \url{http://home.hccnet.nl/d.ingamells/beautify.html}
18525 <row topline="true">
18526 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18532 \begin_inset LatexCommand \index{srecord (tool)}
18539 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18544 Object file conversion, checksumming, ...
18547 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18553 \begin_inset LatexCommand \url{http://srecord.sourceforge.net/}
18561 <row topline="true">
18562 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18568 \begin_inset LatexCommand \index{objdump (tool)}
18575 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18580 Object file conversion, ...
18583 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18588 Part of binutils (should be there anyway)
18592 <row topline="true">
18593 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18599 \begin_inset LatexCommand \index{doxygen (source documentation tool)}
18606 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18611 Source code documentation system
18614 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18620 \begin_inset LatexCommand \url{http://www.doxygen.org}
18628 <row topline="true">
18629 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18637 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18642 IDE (has anyone tried integrating SDCC & sdcdb? Unix only)
18645 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18651 \begin_inset LatexCommand \url{http://www.kdevelop.org}
18659 <row topline="true">
18660 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18666 \begin_inset LatexCommand \index{splint (syntax checking tool)}
18673 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18678 Statically checks c sources (has anyone adapted splint for SDCC?)
18681 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18687 \begin_inset LatexCommand \url{http://www.splint.org}
18695 <row topline="true" bottomline="true">
18696 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18702 \begin_inset LatexCommand \index{ddd (debugger)}
18709 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18714 Debugger, serves nicely as GUI to sdcdb
18715 \begin_inset LatexCommand \index{sdcdb (debugger)}
18722 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18728 \begin_inset LatexCommand \url{http://www.gnu.org/software/ddd/}
18745 Related documentation / recommended reading
18749 \begin_inset Tabular
18750 <lyxtabular version="3" rows="6" columns="3">
18752 <column alignment="center" valignment="top" leftline="true" width="0pt">
18753 <column alignment="block" valignment="top" leftline="true" width="30line%">
18754 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0pt">
18755 <row topline="true" bottomline="true">
18756 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18764 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18772 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18781 <row topline="true">
18782 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18799 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18805 \begin_inset LatexCommand \index{C Reference card}
18812 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18818 \begin_inset LatexCommand \url{http://www.refcards.com/about/c.html}
18826 <row topline="true">
18827 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18835 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18843 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18849 \begin_inset LatexCommand \url{http://www.eskimo.com/~scs/C-faq/top.html}
18857 <row topline="true">
18858 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18865 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18870 Latest datasheet of the target CPU
18873 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18882 <row topline="true">
18883 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18890 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18895 Revision history of datasheet
18898 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18907 <row topline="true" bottomline="true">
18908 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18918 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
18923 Advanced Compiler Design and Implementation
18926 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
18931 bookstore (very dedicated, probably read other books first)
18947 Some questions answered, some pointers given - it might be time to in turn
18955 can you solve your project with the selected microcontroller? Would you
18956 find out early or rather late that your target is too small/slow/whatever?
18957 Can you switch to a slightly better device if it doesn't fit?
18960 should you solve the problem with an 8 bit CPU? Or would a 16/32 bit CPU
18961 and/or another programming language be more adequate? Would an operating
18962 system on the target device help?
18965 if you solved the problem, will the marketing department be happy?
18968 if the marketing department is happy, will customers be happy?
18971 if you're the project manager, marketing department and maybe even the customer
18972 in one person, have you tried to see the project from the outside?
18975 is the project done if you think it is done? Or is just that other interface/pro
18976 tocol/feature/configuration/option missing? How about website, manual(s),
18977 internationali(z|s)ation, packaging, labels, 2nd source for components,
18978 electromagnetic compatability/interference, documentation for production,
18979 production test software, update mechanism, patent issues?
18982 is your project adequately positioned in that magic triangle: fame, fortune,
18986 Maybe not all answers to these questions are known and some answers may
18991 , nevertheless knowing these questions may help you to avoid burnout
18997 burnout is bad for electronic devices, programmers and motorcycle tyres
19001 Chances are you didn't want to hear some of them...
19005 \begin_inset LatexCommand \index{Support}
19012 SDCC has grown to be a large project.
19013 The compiler alone (without the preprocessor, assembler and linker) is
19014 well over 100,000 lines of code (blank stripped).
19015 The open source nature of this project is a key to its continued growth
19017 You gain the benefit and support of many active software developers and
19019 Is SDCC perfect? No, that's why we need your help.
19020 The developers take pride in fixing reported bugs.
19021 You can help by reporting the bugs and helping other SDCC users.
19022 There are lots of ways to contribute, and we encourage you to take part
19023 in making SDCC a great software package.
19027 The SDCC project is hosted on the SDCC sourceforge site at
19028 \begin_inset LatexCommand \htmlurl{http://sourceforge.net/projects/sdcc}
19033 You'll find the complete set of mailing lists
19034 \begin_inset LatexCommand \index{Mailing list(s)}
19038 , forums, bug reporting system, patch submission
19039 \begin_inset LatexCommand \index{Patch submission}
19044 \begin_inset LatexCommand \index{download}
19048 area and cvs code repository
19049 \begin_inset LatexCommand \index{cvs code repository}
19057 \begin_inset LatexCommand \index{Bug reporting}
19062 \begin_inset LatexCommand \index{Reporting bugs}
19069 The recommended way of reporting bugs is using the infrastructure of the
19071 You can follow the status of bug reports there and have an overview about
19075 Bug reports are automatically forwarded to the developer mailing list and
19076 will be fixed ASAP.
19077 When reporting a bug, it is very useful to include a small test program
19078 (the smaller the better) which reproduces the problem.
19079 If you can isolate the problem by looking at the generated assembly code,
19080 this can be very helpful.
19081 Compiling your program with the -
19092 \begin_inset LatexCommand \index{-\/-dumpall}
19096 option can sometimes be useful in locating optimization problems.
19097 When reporting a bug please maker sure you:
19100 Attach the code you are compiling with SDCC.
19104 Specify the exact command you use to run SDCC, or attach your Makefile.
19108 Specify the SDCC version (type "
19114 "), your platform, and operating system.
19118 Provide an exact copy of any error message or incorrect output.
19122 Put something meaningful in the subject of your message.
19125 Please attempt to include these 5 important parts, as applicable, in all
19126 requests for support or when reporting any problems or bugs with SDCC.
19127 Though this will make your message lengthy, it will greatly improve your
19128 chance that SDCC users and developers will be able to help you.
19129 Some SDCC developers are frustrated by bug reports without code provided
19130 that they can use to reproduce and ultimately fix the problem, so please
19131 be sure to provide sample code if you are reporting a bug!
19134 Please have a short check that you are using a recent version of SDCC and
19135 the bug is not yet known.
19136 This is the link for reporting bugs:
19137 \begin_inset LatexCommand \htmlurl{http://sourceforge.net/tracker/?group_id=599&atid=100599}
19144 Requesting Features
19145 \begin_inset LatexCommand \label{sub:Requesting-Features}
19150 \begin_inset LatexCommand \index{Feature request}
19155 \begin_inset LatexCommand \index{Requesting features}
19162 Like bug reports feature requests are forwarded to the developer mailing
19164 This is the link for requesting features:
19165 \begin_inset LatexCommand \htmlurl{http://sourceforge.net/tracker/?group_id=599&atid=350599}
19175 Like bug reports contributed patches are forwarded to the developer mailing
19177 This is the link for submitting patches
19178 \begin_inset LatexCommand \index{Patch submission}
19183 \begin_inset LatexCommand \url{http://sourceforge.net/tracker/?group_id=599&atid=300599}
19190 You need to specify some parameters to the
19194 command for the patches to be useful.
19195 If you modified more than one file a patch created f.e.
19200 \begin_inset Quotes sld
19203 diff -Naur unmodified_directory modified_directory >my_changes.patch
19204 \begin_inset Quotes srd
19210 will be fine, otherwise
19214 \begin_inset Quotes sld
19217 diff -u sourcefile.c.orig sourcefile.c >my_changes.patch
19218 \begin_inset Quotes srd
19231 These links should take you directly to the
19232 \begin_inset LatexCommand \url[Mailing lists]{http://sourceforge.net/mail/?group_id=599}
19242 Traffic on sdcc-devel and sdcc-user is about 100 mails/month each not counting
19243 automated messages (mid 2003)
19247 \begin_inset LatexCommand \url[Forums]{http://sourceforge.net/forum/?group_id=599}
19252 \begin_inset LatexCommand \index{Mailing list(s)}
19256 and forums are archived and searchable so if you are lucky someone already
19257 had a similar problem.
19258 While mails to the lists themselves are delivered promptly their web front
19259 end on sourceforge sometimes shows a severe time lag (up to several weeks),
19260 if you're seriously using SDCC please consider subscribing to the lists.
19266 You can follow the status of the cvs version
19267 \begin_inset LatexCommand \index{version}
19271 of SDCC by watching the Changelog
19272 \begin_inset LatexCommand \index{Changelog}
19276 in the cvs-repository
19281 \begin_inset LatexCommand \htmlurl{http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/sdcc/sdcc/ChangeLog?rev=HEAD&content-type=text/plain}
19289 \begin_inset LatexCommand \index{Release policy}
19296 Historically there often were long delays between official releases and
19297 the sourceforge download area tends to get not updated at all.
19298 Excuses in the past might have referred to problems with live range analysis,
19299 but as this was fixed a while ago, the current problem is that another
19300 excuse has to be found.
19301 Kidding aside, we have to get better there! On the other hand there are
19302 daily snapshots available at
19303 \begin_inset LatexCommand \htmlurl[snap]{http://sdcc.sourceforge.net/snap.php}
19307 , and you can always build the very last version (hopefully with many bugs
19308 fixed, and features added) from the source code available at
19309 \begin_inset LatexCommand \htmlurl[Source]{http://sdcc.sourceforge.net/snap.php#Source}
19317 \begin_inset LatexCommand \index{Examples}
19324 You'll find some small examples in the directory
19326 sdcc/device/examples/.
19329 More examples and libraries are available at
19331 The SDCC Open Knowledge Resource
19332 \begin_inset LatexCommand \url{http://www.qsl.net/dl9sec/SDCC_OKR.html}
19339 \begin_inset LatexCommand \url{http://www.pjrc.com/tech/8051/}
19346 I did insert a reference to Paul's web site here although it seems rather
19347 dedicated to a specific 8032 board (I think it's okay because it f.e.
19348 shows LCD/Harddisc interface and has a free 8051 monitor.
19349 Independent 8032 board vendors face hard competition of heavily subsidized
19350 development boards anyway).
19353 Maybe we should include some links to real world applications.
19354 Preferably pointer to pointers (one for each architecture) so this stays
19359 \begin_inset LatexCommand \index{Quality control}
19366 The compiler is passed through nightly compile and build checks.
19372 \begin_inset LatexCommand \index{Regression test}
19376 check that SDCC itself compiles flawlessly on several platforms and checks
19377 the quality of the code generated by SDCC by running the code through simulator
19379 There is a separate document
19382 \begin_inset LatexCommand \index{Test suite}
19391 You'll find the test code in the directory
19393 sdcc/support/regression
19396 You can run these tests manually by running
19400 in this directory (or f.e.
19405 \begin_inset Quotes sld
19409 \begin_inset Quotes srd
19415 if you don't want to run the complete tests).
19416 The test code might also be interesting if you want to look for examples
19417 \begin_inset LatexCommand \index{Examples}
19421 checking corner cases of SDCC or if you plan to submit patches
19422 \begin_inset LatexCommand \index{Patch submission}
19429 The pic port uses a different set of regression tests, you'll find them
19432 sdcc/src/regression
19437 SDCC Technical Data
19441 \begin_inset LatexCommand \index{Optimizations}
19448 SDCC performs a host of standard optimizations in addition to some MCU specific
19453 Sub-expression Elimination
19454 \begin_inset LatexCommand \index{Subexpression elimination}
19461 The compiler does local and
19487 will be translated to
19499 Some subexpressions are not as obvious as the above example, e.g.:
19509 In this case the address arithmetic a->b[i] will be computed only once;
19510 the equivalent code in C would be.
19522 The compiler will try to keep these temporary variables in registers.
19525 Dead-Code Elimination
19526 \begin_inset LatexCommand \index{Dead-code elimination}
19547 i = 1; \SpecialChar ~
19556 global = 1;\SpecialChar ~
19569 global = 3;\SpecialChar ~
19594 \begin_inset LatexCommand \index{Copy propagation}
19650 Note: the dead stores created by this copy propagation will be eliminated
19651 by dead-code elimination.
19655 \begin_inset LatexCommand \index{Loop optimization}
19660 \begin_inset LatexCommand \label{sub:Loop-Optimizations}
19667 Two types of loop optimizations are done by SDCC
19675 of loop induction variables.
19676 In addition to the strength reduction the optimizer marks the induction
19677 variables and the register allocator tries to keep the induction variables
19678 in registers for the duration of the loop.
19679 Because of this preference of the register allocator
19680 \begin_inset LatexCommand \index{Register allocation}
19684 , loop induction optimization causes an increase in register pressure, which
19685 may cause unwanted spilling of other temporary variables into the stack
19686 \begin_inset LatexCommand \index{stack}
19691 The compiler will generate a warning message when it is forced to allocate
19692 extra space either on the stack or data space.
19693 If this extra space allocation is undesirable then induction optimization
19694 can be eliminated either for the entire source file (with -
19704 -noinduction option) or for a given function only using #pragma\SpecialChar ~
19706 \begin_inset LatexCommand \index{\#pragma noinduction}
19719 for (i = 0 ; i < 100 ; i ++)
19735 for (i = 0; i < 100; i++)
19744 As mentioned previously some loop invariants are not as apparent, all static
19745 address computations are also moved out of the loop.
19750 \begin_inset LatexCommand \index{Strength reduction}
19754 , this optimization substitutes an expression by a cheaper expression:
19759 for (i=0;i < 100; i++)
19777 for (i=0;i< 100;i++) {
19783 ar[itemp1] = itemp2;
19800 The more expensive multiplication
19801 \begin_inset LatexCommand \index{Multiplication}
19805 is changed to a less expensive addition.
19809 \begin_inset LatexCommand \index{Loop reversing}
19816 This optimization is done to reduce the overhead of checking loop boundaries
19817 for every iteration.
19818 Some simple loops can be reversed and implemented using a
19819 \begin_inset Quotes eld
19822 decrement and jump if not zero
19823 \begin_inset Quotes erd
19827 SDCC checks for the following criterion to determine if a loop is reversible
19828 (note: more sophisticated compilers use data-dependency analysis to make
19829 this determination, SDCC uses a more simple minded analysis).
19832 The 'for' loop is of the form
19838 for(<symbol> = <expression>; <sym> [< | <=] <expression>; [<sym>++ | <sym>
19848 The <for body> does not contain
19849 \begin_inset Quotes eld
19853 \begin_inset Quotes erd
19857 \begin_inset Quotes erd
19863 All goto's are contained within the loop.
19866 No function calls within the loop.
19869 The loop control variable <sym> is not assigned any value within the loop
19872 The loop control variable does NOT participate in any arithmetic operation
19876 There are NO switch statements in the loop.
19879 Algebraic Simplifications
19882 SDCC does numerous algebraic simplifications, the following is a small sub-set
19883 of these optimizations.
19888 i = j + 0;\SpecialChar ~
19892 /* changed to: */\SpecialChar ~
19898 i /= 2;\SpecialChar ~
19905 /* changed to: */\SpecialChar ~
19911 i = j - j;\SpecialChar ~
19915 /* changed to: */\SpecialChar ~
19921 i = j / 1;\SpecialChar ~
19925 /* changed to: */\SpecialChar ~
19932 Note the subexpressions
19933 \begin_inset LatexCommand \index{Subexpression}
19937 given above are generally introduced by macro expansions or as a result
19938 of copy/constant propagation.
19941 'switch' Statements
19942 \begin_inset LatexCommand \label{sub:'switch'-Statements}
19947 \begin_inset LatexCommand \index{switch statement}
19954 SDCC can optimize switch statements to jump tables
19955 \begin_inset LatexCommand \index{jump tables}
19960 It makes the decision based on an estimate of the generated code size.
19961 SDCC is quite liberal in the requirements for jump table generation:
19964 The labels need not be in order, and the starting number need not be one
19965 or zero, the case labels are in numerical sequence or not too many case
19966 labels are missing.
19972 switch(i) {\SpecialChar ~
20003 case 4: ...\SpecialChar ~
20035 case 5: ...\SpecialChar ~
20067 case 3: ...\SpecialChar ~
20098 case 6: ...\SpecialChar ~
20130 case 7: ...\SpecialChar ~
20162 case 8: ...\SpecialChar ~
20194 case 9: ...\SpecialChar ~
20226 case 10: ...\SpecialChar ~
20257 case 11: ...\SpecialChar ~
20324 Both the above switch statements will be implemented using a jump-table.
20325 The example to the right side is slightly more efficient as the check for
20326 the lower boundary of the jump-table is not needed.
20330 The number of case labels is not larger than supported by the target architectur
20334 If the case labels are not in numerical sequence ('gaps' between cases)
20335 SDCC checks whether a jump table with additionally inserted dummy cases
20336 is still attractive.
20340 If the starting number is not zero and a check for the lower boundary of
20341 the jump-table can thus be eliminated SDCC might insert dummy cases 0,
20346 Switch statements which have large gaps in the numeric sequence or those
20347 that have too many case labels can be split into more than one switch statement
20348 for efficient code generation, e.g.:
20428 If the above switch statement is broken down into two switch statements
20518 then both the switch statements will be implemented using jump-tables whereas
20519 the unmodified switch statement will not be.
20522 There might be reasons which SDCC cannot know about to either favour or
20523 not favour jump tables.
20524 If the target system has to be as quick for the last switch case as for
20525 the first (pro jump table), or if the switch argument is known to be zero
20526 in the majority of the cases (contra jump table).
20529 The pragma nojtbound
20530 \begin_inset LatexCommand \index{\#pragma nojtbound}
20534 can be used to turn off checking the
20547 It has no effect if a default label is supplied.
20548 Use of this pragma is dangerous: if the switch argument is not matched
20549 by a case statement the processor will happily jump into Nirvana.
20552 Bit-shifting Operations
20553 \begin_inset LatexCommand \index{Bit shifting}
20560 Bit shifting is one of the most frequently used operation in embedded programmin
20562 SDCC tries to implement bit-shift operations in the most efficient way
20578 generates the following code:
20595 In general SDCC will never setup a loop if the shift count is known.
20638 \begin_inset LatexCommand \index{Bit rotation}
20645 A special case of the bit-shift operation is bit rotation
20646 \begin_inset LatexCommand \index{rotating bits}
20650 , SDCC recognizes the following expression to be a left bit-rotation:
20660 char i;\SpecialChar ~
20671 /* unsigned is needed for rotation */
20676 i = ((i << 1) | (i >> 7));
20685 will generate the following code:
20704 SDCC uses pattern matching on the parse tree to determine this operation.Variatio
20705 ns of this case will also be recognized as bit-rotation, i.e.:
20710 i = ((i >> 7) | (i << 1)); /* left-bit rotation */
20713 Nibble and Byte Swapping
20716 Other special cases of the bit-shift operations are nibble or byte swapping
20717 \begin_inset LatexCommand \index{swapping nibbles/bytes}
20721 , SDCC recognizes the following expressions:
20744 i = ((i << 4) | (i >> 4));
20750 j = ((j << 8) | (j >> 8));
20753 and generates a swap instruction for the nibble swapping
20754 \begin_inset LatexCommand \index{Nibble swapping}
20758 or move instructions for the byte swapping
20759 \begin_inset LatexCommand \index{Byte swapping}
20765 \begin_inset Quotes sld
20769 \begin_inset Quotes srd
20772 example can be used to convert from little to big-endian or vice versa.
20773 If you want to change the endianness of a
20777 integer you have to cast to
20784 Note that SDCC stores numbers in little-endian
20790 Usually 8-bit processors don't care much about endianness.
20791 This is not the case for the standard 8051 which only has an instruction
20797 \begin_inset LatexCommand \index{DPTR}
20805 so little-endian is the more efficient byte order.
20809 \begin_inset LatexCommand \index{little-endian}
20814 \begin_inset LatexCommand \index{Endianness}
20819 lowest order first).
20823 \begin_inset LatexCommand \index{Highest Order Bit}
20830 It is frequently required to obtain the highest order bit of an integral
20831 type (long, int, short or char types).
20832 SDCC recognizes the following expression to yield the highest order bit
20833 and generates optimized code for it, e.g.:
20855 hob = (gint >> 15) & 1;
20865 will generate the following code:
20898 000A E5*01\SpecialChar ~
20925 000C 23\SpecialChar ~
20956 000D 54 01\SpecialChar ~
20983 000F F5*02\SpecialChar ~
21011 Variations of this case however will
21016 It is a standard C expression, so I heartily recommend this be the only
21017 way to get the highest order bit, (it is portable).
21018 Of course it will be recognized even if it is embedded in other expressions,
21024 xyz = gint + ((gint >> 15) & 1);
21027 will still be recognized.
21031 \begin_inset LatexCommand \label{sub:Peephole-Optimizer}
21036 \begin_inset LatexCommand \index{Peephole optimizer}
21043 The compiler uses a rule based, pattern matching and re-writing mechanism
21044 for peep-hole optimization.
21049 a peep-hole optimizer by Christopher W.
21050 Fraser (cwfraser@microsoft.com).
21051 A default set of rules are compiled into the compiler, additional rules
21052 may be added with the
21065 \begin_inset LatexCommand \index{-\/-peep-file}
21072 The rule language is best illustrated with examples.
21096 The above rule will change the following assembly
21097 \begin_inset LatexCommand \index{Assembler routines}
21119 Note: All occurrences of a
21123 (pattern variable) must denote the same string.
21124 With the above rule, the assembly sequence:
21134 will remain unmodified.
21138 Other special case optimizations may be added by the user (via
21154 some variants of the 8051 MCU
21155 \begin_inset LatexCommand \index{MCS51 variants}
21168 The following two rules will change all
21187 replace { lcall %1 } by { acall %1 }
21189 replace { ljmp %1 } by { ajmp %1 }
21194 inline-assembler code
21196 is also passed through the peep hole optimizer, thus the peephole optimizer
21197 can also be used as an assembly level macro expander.
21198 The rules themselves are MCU dependent whereas the rule language infra-structur
21199 e is MCU independent.
21200 Peephole optimization rules for other MCU can be easily programmed using
21205 The syntax for a rule is as follows:
21210 rule := replace [ restart ] '{' <assembly sequence> '
21248 <assembly sequence> '
21266 '}' [if <functionName> ] '
21271 <assembly sequence> := assembly instruction (each instruction including
21272 labels must be on a separate line).
21276 The optimizer will apply to the rules one by one from the top in the sequence
21277 of their appearance, it will terminate when all rules are exhausted.
21278 If the 'restart' option is specified, then the optimizer will start matching
21279 the rules again from the top, this option for a rule is expensive (performance)
21280 , it is intended to be used in situations where a transformation will trigger
21281 the same rule again.
21282 An example of this (not a good one, it has side effects) is the following
21305 Note that the replace pattern cannot be a blank, but can be a comment line.
21306 Without the 'restart' option only the innermost 'pop' 'push' pair would
21307 be eliminated, i.e.:
21337 the restart option the rule will be applied again to the resulting code
21338 and then all the pop-push pairs will be eliminated to yield:
21348 A conditional function can be attached to a rule.
21349 Attaching rules are somewhat more involved, let me illustrate this with
21376 The optimizer does a look-up of a function name table defined in function
21381 in the source file SDCCpeeph.c, with the name
21386 If it finds a corresponding entry the function is called.
21387 Note there can be no parameters specified for these functions, in this
21392 is crucial, since the function
21396 expects to find the label in that particular variable (the hash table containin
21397 g the variable bindings is passed as a parameter).
21398 If you want to code more such functions, take a close look at the function
21399 labelInRange and the calling mechanism in source file SDCCpeeph.c.
21400 Currently implemented are
21402 labelInRange, labelRefCount, labelIsReturnOnly, operandsNotSame, xramMovcOption,
21403 24bitMode, portIsDS390, 24bitModeAndPortDS390
21412 I know this whole thing is a little kludgey, but maybe some day we will
21413 have some better means.
21414 If you are looking at this file, you will see the default rules that are
21415 compiled into the compiler, you can add your own rules in the default set
21416 there if you get tired of specifying the -
21430 \begin_inset LatexCommand \index{ANSI-compliance}
21435 \begin_inset LatexCommand \label{sub:ANSI-Compliance}
21442 Deviations from the compliance:
21445 functions are not always reentrant
21446 \begin_inset LatexCommand \index{reentrant}
21453 structures cannot be assigned values directly, cannot be passed as function
21454 parameters or assigned to each other and cannot be a return value from
21481 s1 = s2 ; /* is invalid in SDCC although allowed in ANSI */
21492 struct s foo1 (struct s parms) /* invalid in SDCC although allowed in ANSI
21514 return rets;/* is invalid in SDCC although allowed in ANSI */
21521 \begin_inset LatexCommand \index{long long (not supported)}
21526 \begin_inset LatexCommand \index{int (64 bit) (not supported)}
21534 \begin_inset LatexCommand \index{double (not supported)}
21538 ' precision floating point
21539 \begin_inset LatexCommand \index{Floating point support}
21546 No support for setjmp
21547 \begin_inset LatexCommand \index{setjmp (not supported)}
21552 \begin_inset LatexCommand \index{longjmp (not supported)}
21560 \begin_inset LatexCommand \index{K\&R style}
21564 function declarations are NOT allowed.
21570 foo(i,j) /* this old style of function declarations */
21572 int i,j; /* are valid in ANSI but not valid in SDCC */
21587 Cyclomatic Complexity
21588 \begin_inset LatexCommand \index{Cyclomatic complexity}
21595 Cyclomatic complexity of a function is defined as the number of independent
21596 paths the program can take during execution of the function.
21597 This is an important number since it defines the number test cases you
21598 have to generate to validate the function.
21599 The accepted industry standard for complexity number is 10, if the cyclomatic
21600 complexity reported by SDCC exceeds 10 you should think about simplification
21601 of the function logic.
21602 Note that the complexity level is not related to the number of lines of
21603 code in a function.
21604 Large functions can have low complexity, and small functions can have large
21610 SDCC uses the following formula to compute the complexity:
21615 complexity = (number of edges in control flow graph) - (number of nodes
21616 in control flow graph) + 2;
21620 Having said that the industry standard is 10, you should be aware that in
21621 some cases it be may unavoidable to have a complexity level of less than
21623 For example if you have switch statement with more than 10 case labels,
21624 each case label adds one to the complexity level.
21625 The complexity level is by no means an absolute measure of the algorithmic
21626 complexity of the function, it does however provide a good starting point
21627 for which functions you might look at for further optimization.
21630 Retargetting for other Processors
21633 The issues for retargetting the compiler are far too numerous to be covered
21635 What follows is a brief description of each of the seven phases of the
21636 compiler and its MCU dependency.
21639 Parsing the source and building the annotated parse tree.
21640 This phase is largely MCU independent (except for the language extensions).
21641 Syntax & semantic checks are also done in this phase, along with some initial
21642 optimizations like back patching labels and the pattern matching optimizations
21643 like bit-rotation etc.
21646 The second phase involves generating an intermediate code which can be easy
21647 manipulated during the later phases.
21648 This phase is entirely MCU independent.
21649 The intermediate code generation assumes the target machine has unlimited
21650 number of registers, and designates them with the name iTemp.
21651 The compiler can be made to dump a human readable form of the code generated
21665 This phase does the bulk of the standard optimizations and is also MCU independe
21667 This phase can be broken down into several sub-phases:
21671 Break down intermediate code (iCode) into basic blocks.
21673 Do control flow & data flow analysis on the basic blocks.
21675 Do local common subexpression elimination, then global subexpression elimination
21677 Dead code elimination
21681 If loop optimizations caused any changes then do 'global subexpression eliminati
21682 on' and 'dead code elimination' again.
21685 This phase determines the live-ranges; by live range I mean those iTemp
21686 variables defined by the compiler that still survive after all the optimization
21688 Live range analysis
21689 \begin_inset LatexCommand \index{Live range analysis}
21693 is essential for register allocation, since these computation determines
21694 which of these iTemps will be assigned to registers, and for how long.
21697 Phase five is register allocation.
21698 There are two parts to this process.
21702 The first part I call 'register packing' (for lack of a better term).
21703 In this case several MCU specific expression folding is done to reduce
21708 The second part is more MCU independent and deals with allocating registers
21709 to the remaining live ranges.
21710 A lot of MCU specific code does creep into this phase because of the limited
21711 number of index registers available in the 8051.
21714 The Code generation phase is (unhappily), entirely MCU dependent and very
21715 little (if any at all) of this code can be reused for other MCU.
21716 However the scheme for allocating a homogenized assembler operand for each
21717 iCode operand may be reused.
21720 As mentioned in the optimization section the peep-hole optimizer is rule
21721 based system, which can reprogrammed for other MCUs.
21725 \begin_inset LatexCommand \index{Compiler internals}
21732 The anatomy of the compiler
21733 \begin_inset LatexCommand \label{sub:The-anatomy-of}
21742 This is an excerpt from an article published in Circuit Cellar Magazine
21744 It's a little outdated (the compiler is much more efficient now and user/develo
21745 per friendly), but pretty well exposes the guts of it all.
21751 The current version of SDCC can generate code for Intel 8051 and Z80 MCU.
21752 It is fairly easy to retarget for other 8-bit MCU.
21753 Here we take a look at some of the internals of the compiler.
21758 \begin_inset LatexCommand \index{Parsing}
21765 Parsing the input source file and creating an AST (Annotated Syntax Tree
21766 \begin_inset LatexCommand \index{Annotated syntax tree}
21771 This phase also involves propagating types (annotating each node of the
21772 parse tree with type information) and semantic analysis.
21773 There are some MCU specific parsing rules.
21774 For example the storage classes, the extended storage classes are MCU specific
21775 while there may be a xdata storage class for 8051 there is no such storage
21776 class for z80 or Atmel AVR.
21777 SDCC allows MCU specific storage class extensions, i.e.
21778 xdata will be treated as a storage class specifier when parsing 8051 C
21779 code but will be treated as a C identifier when parsing z80 or ATMEL AVR
21784 \begin_inset LatexCommand \index{iCode}
21791 Intermediate code generation.
21792 In this phase the AST is broken down into three-operand form (iCode).
21793 These three operand forms are represented as doubly linked lists.
21794 ICode is the term given to the intermediate form generated by the compiler.
21795 ICode example section shows some examples of iCode generated for some simple
21796 C source functions.
21800 \begin_inset LatexCommand \index{Optimizations}
21807 Bulk of the target independent optimizations is performed in this phase.
21808 The optimizations include constant propagation, common sub-expression eliminati
21809 on, loop invariant code movement, strength reduction of loop induction variables
21810 and dead-code elimination.
21813 Live range analysis
21814 \begin_inset LatexCommand \index{Live range analysis}
21821 During intermediate code generation phase, the compiler assumes the target
21822 machine has infinite number of registers and generates a lot of temporary
21824 The live range computation determines the lifetime of each of these compiler-ge
21825 nerated temporaries.
21826 A picture speaks a thousand words.
21827 ICode example sections show the live range annotations for each of the
21829 It is important to note here, each iCode is assigned a number in the order
21830 of its execution in the function.
21831 The live ranges are computed in terms of these numbers.
21832 The from number is the number of the iCode which first defines the operand
21833 and the to number signifies the iCode which uses this operand last.
21836 Register Allocation
21837 \begin_inset LatexCommand \index{Register allocation}
21844 The register allocation determines the type and number of registers needed
21846 In most MCUs only a few registers can be used for indirect addressing.
21847 In case of 8051 for example the registers R0 & R1 can be used to indirectly
21848 address the internal ram and DPTR to indirectly address the external ram.
21849 The compiler will try to allocate the appropriate register to pointer variables
21851 ICode example section shows the operands annotated with the registers assigned
21853 The compiler will try to keep operands in registers as much as possible;
21854 there are several schemes the compiler uses to do achieve this.
21855 When the compiler runs out of registers the compiler will check to see
21856 if there are any live operands which is not used or defined in the current
21857 basic block being processed, if there are any found then it will push that
21858 operand and use the registers in this block, the operand will then be popped
21859 at the end of the basic block.
21863 There are other MCU specific considerations in this phase.
21864 Some MCUs have an accumulator; very short-lived operands could be assigned
21865 to the accumulator instead of a general-purpose register.
21871 Figure II gives a table of iCode operations supported by the compiler.
21872 The code generation involves translating these operations into corresponding
21873 assembly code for the processor.
21874 This sounds overly simple but that is the essence of code generation.
21875 Some of the iCode operations are generated on a MCU specific manner for
21876 example, the z80 port does not use registers to pass parameters so the
21877 SEND and RECV iCode operations will not be generated, and it also does
21878 not support JUMPTABLES.
21885 <Where is Figure II?>
21888 In the original article Figure II was announced to be downloadable on
21893 Unfortunately it never seemed to have shown up there, so: where is Figure
21898 \begin_inset LatexCommand \index{iCode}
21905 This section shows some details of iCode.
21906 The example C code does not do anything useful; it is used as an example
21907 to illustrate the intermediate code generated by the compiler.
21919 /* This function does nothing useful.
21926 for the purpose of explaining iCode */
21929 short function (data int *x)
21937 short i=10; \SpecialChar ~
21939 /* dead initialization eliminated */
21944 short sum=10; /* dead initialization eliminated */
21957 while (*x) *x++ = *p++;
21971 /* compiler detects i,j to be induction variables */
21975 for (i = 0, j = 10 ; i < 10 ; i++, j
22001 mul += i * 3; \SpecialChar ~
22003 /* this multiplication remains */
22009 gint += j * 3;\SpecialChar ~
22011 /* this multiplication changed to addition */
22025 In addition to the operands each iCode contains information about the filename
22026 and line it corresponds to in the source file.
22027 The first field in the listing should be interpreted as follows:
22032 Filename(linenumber: iCode Execution sequence number : ICode hash table
22033 key : loop depth of the iCode).
22038 Then follows the human readable form of the ICode operation.
22039 Each operand of this triplet form can be of three basic types a) compiler
22040 generated temporary b) user defined variable c) a constant value.
22041 Note that local variables and parameters are replaced by compiler generated
22044 \begin_inset LatexCommand \index{Live range analysis}
22048 are computed only for temporaries (i.e.
22049 live ranges are not computed for global variables).
22051 \begin_inset LatexCommand \index{Register allocation}
22055 are allocated for temporaries only.
22056 Operands are formatted in the following manner:
22061 Operand Name [lr live-from : live-to ] { type information } [ registers
22067 As mentioned earlier the live ranges are computed in terms of the execution
22068 sequence number of the iCodes, for example
22070 the iTemp0 is live from (i.e.
22071 first defined in iCode with execution sequence number 3, and is last used
22072 in the iCode with sequence number 5).
22073 For induction variables such as iTemp21 the live range computation extends
22074 the lifetime from the start to the end of the loop.
22076 The register allocator used the live range information to allocate registers,
22077 the same registers may be used for different temporaries if their live
22078 ranges do not overlap, for example r0 is allocated to both iTemp6 and to
22079 iTemp17 since their live ranges do not overlap.
22080 In addition the allocator also takes into consideration the type and usage
22081 of a temporary, for example itemp6 is a pointer to near space and is used
22082 as to fetch data from (i.e.
22083 used in GET_VALUE_AT_ADDRESS) so it is allocated a pointer register (r0).
22084 Some short lived temporaries are allocated to special registers which have
22085 meaning to the code generator e.g.
22086 iTemp13 is allocated to a pseudo register CC which tells the back end that
22087 the temporary is used only for a conditional jump the code generation makes
22088 use of this information to optimize a compare and jump ICode.
22090 There are several loop optimizations
22091 \begin_inset LatexCommand \index{Loop optimization}
22095 performed by the compiler.
22096 It can detect induction variables iTemp21(i) and iTemp23(j).
22097 Also note the compiler does selective strength reduction
22098 \begin_inset LatexCommand \index{Strength reduction}
22103 the multiplication of an induction variable in line 18 (gint = j * 3) is
22104 changed to addition, a new temporary iTemp17 is allocated and assigned
22105 a initial value, a constant 3 is then added for each iteration of the loop.
22106 The compiler does not change the multiplication
22107 \begin_inset LatexCommand \index{Multiplication}
22111 in line 17 however since the processor does support an 8 * 8 bit multiplication.
22113 Note the dead code elimination
22114 \begin_inset LatexCommand \index{Dead-code elimination}
22118 optimization eliminated the dead assignments in line 7 & 8 to I and sum
22126 Sample.c (5:1:0:0) _entry($9) :
22131 Sample.c(5:2:1:0) proc _function [lr0:0]{function short}
22136 Sample.c(11:3:2:0) iTemp0 [lr3:5]{_near * int}[r2] = recv
22141 Sample.c(11:4:53:0) preHeaderLbl0($11) :
22146 Sample.c(11:5:55:0) iTemp6 [lr5:16]{_near * int}[r0] := iTemp0 [lr3:5]{_near
22152 Sample.c(11:6:5:1) _whilecontinue_0($1) :
22157 Sample.c(11:7:7:1) iTemp4 [lr7:8]{int}[r2 r3] = @[iTemp6 [lr5:16]{_near *
22163 Sample.c(11:8:8:1) if iTemp4 [lr7:8]{int}[r2 r3] == 0 goto _whilebreak_0($3)
22168 Sample.c(11:9:14:1) iTemp7 [lr9:13]{_far * int}[DPTR] := _p [lr0:0]{_far
22174 Sample.c(11:10:15:1) _p [lr0:0]{_far * int} = _p [lr0:0]{_far * int} + 0x2
22180 Sample.c(11:13:18:1) iTemp10 [lr13:14]{int}[r2 r3] = @[iTemp7 [lr9:13]{_far
22186 Sample.c(11:14:19:1) *(iTemp6 [lr5:16]{_near * int}[r0]) := iTemp10 [lr13:14]{int
22192 Sample.c(11:15:12:1) iTemp6 [lr5:16]{_near * int}[r0] = iTemp6 [lr5:16]{_near
22193 * int}[r0] + 0x2 {short}
22198 Sample.c(11:16:20:1) goto _whilecontinue_0($1)
22203 Sample.c(11:17:21:0)_whilebreak_0($3) :
22208 Sample.c(12:18:22:0) iTemp2 [lr18:40]{short}[r2] := 0x0 {short}
22213 Sample.c(13:19:23:0) iTemp11 [lr19:40]{short}[r3] := 0x0 {short}
22218 Sample.c(15:20:54:0)preHeaderLbl1($13) :
22223 Sample.c(15:21:56:0) iTemp21 [lr21:38]{short}[r4] := 0x0 {short}
22228 Sample.c(15:22:57:0) iTemp23 [lr22:38]{int}[r5 r6] := 0xa {int}
22233 Sample.c(15:23:58:0) iTemp17 [lr23:38]{int}[r7 r0] := 0x1e {int}
22238 Sample.c(15:24:26:1)_forcond_0($4) :
22243 Sample.c(15:25:27:1) iTemp13 [lr25:26]{char}[CC] = iTemp21 [lr21:38]{short}[r4]
22249 Sample.c(15:26:28:1) if iTemp13 [lr25:26]{char}[CC] == 0 goto _forbreak_0($7)
22254 Sample.c(16:27:31:1) iTemp2 [lr18:40]{short}[r2] = iTemp2 [lr18:40]{short}[r2]
22255 + ITemp21 [lr21:38]{short}[r4]
22260 Sample.c(17:29:33:1) iTemp15 [lr29:30]{short}[r1] = iTemp21 [lr21:38]{short}[r4]
22266 Sample.c(17:30:34:1) iTemp11 [lr19:40]{short}[r3] = iTemp11 [lr19:40]{short}[r3]
22267 + iTemp15 [lr29:30]{short}[r1]
22272 Sample.c(18:32:36:1:1) iTemp17 [lr23:38]{int}[r7 r0]= iTemp17 [lr23:38]{int}[r7
22278 Sample.c(18:33:37:1) _gint [lr0:0]{int} = _gint [lr0:0]{int} + iTemp17 [lr23:38]{
22284 Sample.c(15:36:42:1) iTemp21 [lr21:38]{short}[r4] = iTemp21 [lr21:38]{short}[r4]
22290 Sample.c(15:37:45:1) iTemp23 [lr22:38]{int}[r5 r6]= iTemp23 [lr22:38]{int}[r5
22296 Sample.c(19:38:47:1) goto _forcond_0($4)
22301 Sample.c(19:39:48:0)_forbreak_0($7) :
22306 Sample.c(20:40:49:0) iTemp24 [lr40:41]{short}[DPTR] = iTemp2 [lr18:40]{short}[r2]
22307 + ITemp11 [lr19:40]{short}[r3]
22312 Sample.c(20:41:50:0) ret iTemp24 [lr40:41]{short}
22317 Sample.c(20:42:51:0)_return($8) :
22322 Sample.c(20:43:52:0) eproc _function [lr0:0]{ ia0 re0 rm0}{function short}
22328 Finally the code generated for this function:
22369 ; ----------------------------------------------
22374 ; function function
22379 ; ----------------------------------------------
22389 ; iTemp0 [lr3:5]{_near * int}[r2] = recv
22401 ; iTemp6 [lr5:16]{_near * int}[r0] := iTemp0 [lr3:5]{_near * int}[r2]
22413 ;_whilecontinue_0($1) :
22423 ; iTemp4 [lr7:8]{int}[r2 r3] = @[iTemp6 [lr5:16]{_near * int}[r0]]
22428 ; if iTemp4 [lr7:8]{int}[r2 r3] == 0 goto _whilebreak_0($3)
22487 ; iTemp7 [lr9:13]{_far * int}[DPTR] := _p [lr0:0]{_far * int}
22506 ; _p [lr0:0]{_far * int} = _p [lr0:0]{_far * int} + 0x2 {short}
22553 ; iTemp10 [lr13:14]{int}[r2 r3] = @[iTemp7 [lr9:13]{_far * int}[DPTR]]
22593 ; *(iTemp6 [lr5:16]{_near * int}[r0]) := iTemp10 [lr13:14]{int}[r2 r3]
22619 ; iTemp6 [lr5:16]{_near * int}[r0] =
22624 ; iTemp6 [lr5:16]{_near * int}[r0] +
22641 ; goto _whilecontinue_0($1)
22653 ; _whilebreak_0($3) :
22663 ; iTemp2 [lr18:40]{short}[r2] := 0x0 {short}
22675 ; iTemp11 [lr19:40]{short}[r3] := 0x0 {short}
22687 ; iTemp21 [lr21:38]{short}[r4] := 0x0 {short}
22699 ; iTemp23 [lr22:38]{int}[r5 r6] := 0xa {int}
22718 ; iTemp17 [lr23:38]{int}[r7 r0] := 0x1e {int}
22747 ; iTemp13 [lr25:26]{char}[CC] = iTemp21 [lr21:38]{short}[r4] < 0xa {short}
22752 ; if iTemp13 [lr25:26]{char}[CC] == 0 goto _forbreak_0($7)
22797 ; iTemp2 [lr18:40]{short}[r2] = iTemp2 [lr18:40]{short}[r2] +
22802 ; iTemp21 [lr21:38]{short}[r4]
22828 ; iTemp15 [lr29:30]{short}[r1] = iTemp21 [lr21:38]{short}[r4] * 0x3 {short}
22861 ; iTemp11 [lr19:40]{short}[r3] = iTemp11 [lr19:40]{short}[r3] +
22866 ; iTemp15 [lr29:30]{short}[r1]
22885 ; iTemp17 [lr23:38]{int}[r7 r0]= iTemp17 [lr23:38]{int}[r7 r0]- 0x3 {short}
22932 ; _gint [lr0:0]{int} = _gint [lr0:0]{int} + iTemp17 [lr23:38]{int}[r7 r0]
22979 ; iTemp21 [lr21:38]{short}[r4] = iTemp21 [lr21:38]{short}[r4] + 0x1 {short}
22991 ; iTemp23 [lr22:38]{int}[r5 r6]= iTemp23 [lr22:38]{int}[r5 r6]- 0x1 {short}
23005 cjne r5,#0xff,00104$
23017 ; goto _forcond_0($4)
23029 ; _forbreak_0($7) :
23039 ; ret iTemp24 [lr40:41]{short}
23082 A few words about basic block successors, predecessors and dominators
23085 Successors are basic blocks
23086 \begin_inset LatexCommand \index{Basic blocks}
23090 that might execute after this basic block.
23092 Predecessors are basic blocks that might execute before reaching this basic
23095 Dominators are basic blocks that WILL execute before reaching this basic
23129 a) succList of [BB2] = [BB4], of [BB3] = [BB4], of [BB1] = [BB2,BB3]
23132 b) predList of [BB2] = [BB1], of [BB3] = [BB1], of [BB4] = [BB2,BB3]
23135 c) domVect of [BB4] = BB1 ...
23136 here we are not sure if BB2 or BB3 was executed but we are SURE that BB1
23144 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net#Who}
23154 Thanks to all the other volunteer developers who have helped with coding,
23155 testing, web-page creation, distribution sets, etc.
23156 You know who you are :-)
23163 This document was initially written by Sandeep Dutta
23166 All product names mentioned herein may be trademarks
23167 \begin_inset LatexCommand \index{Trademarks}
23171 of their respective companies.
23178 To avoid confusion, the installation and building options for SDCC itself
23179 (chapter 2) are not part of the index.
23183 \begin_inset LatexCommand \printindex{}