1 #LyX 1.2 created this file. For more info see http://www.lyx.org/
15 \use_numerical_citations 0
16 \paperorientation portrait
19 \paragraph_separation indent
21 \quotes_language swedish
29 Please note: double dashed longoptions (e.g.
30 --version) need three dashes in this document to be visable in html and
34 SDCC Compiler User Guide
38 \begin_inset LatexCommand \tableofcontents{}
55 is a Freeware, retargettable, optimizing ANSI-C compiler by
59 designed for 8 bit Microprocessors.
60 The current version targets Intel MCS51 based Microprocessors(8051,8052,
61 etc), Zilog Z80 based MCUs, and the Dallas DS80C390 variant.
62 It can be retargetted for other microprocessors, support for PIC, AVR and
63 186 is under development.
64 The entire source code for the compiler is distributed under GPL.
65 SDCC uses ASXXXX & ASLINK, a Freeware, retargettable assembler & linker.
66 SDCC has extensive language extensions suitable for utilizing various microcont
67 rollers and underlying hardware effectively.
72 In addition to the MCU specific optimizations SDCC also does a host of standard
76 global sub expression elimination,
79 loop optimizations (loop invariant, strength reduction of induction variables
83 constant folding & propagation,
99 For the back-end SDCC uses a global register allocation scheme which should
100 be well suited for other 8 bit MCUs.
105 The peep hole optimizer uses a rule based substitution mechanism which is
111 Supported data-types are:
114 char (8 bits, 1 byte),
117 short and int (16 bits, 2 bytes),
120 long (32 bit, 4 bytes)
127 The compiler also allows
129 inline assembler code
131 to be embedded anywhere in a function.
132 In addition, routines developed in assembly can also be called.
136 SDCC also provides an option (--cyclomatic) to report the relative complexity
138 These functions can then be further optimized, or hand coded in assembly
144 SDCC also comes with a companion source level debugger SDCDB, the debugger
145 currently uses ucSim a freeware simulator for 8051 and other micro-controllers.
150 The latest version can be downloaded from
151 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/}
163 All packages used in this compiler system are
171 ; source code for all the sub-packages (pre-processor, assemblers, linkers
172 etc) is distributed with the package.
173 This documentation is maintained using a freeware word processor (LyX).
175 This program is free software; you can redistribute it and/or modify it
176 under the terms of the GNU General Public License as published by the Free
177 Software Foundation; either version 2, or (at your option) any later version.
178 This program is distributed in the hope that it will be useful, but WITHOUT
179 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
180 FOR A PARTICULAR PURPOSE.
181 See the GNU General Public License for more details.
182 You should have received a copy of the GNU General Public License along
183 with this program; if not, write to the Free Software Foundation, 59 Temple
184 Place - Suite 330, Boston, MA 02111-1307, USA.
185 In other words, you are welcome to use, share and improve this program.
186 You are forbidden to forbid anyone else to use, share and improve what
188 Help stamp out software-hoarding!
191 Typographic conventions
194 Throughout this manual, we will use the following convention.
195 Commands you have to type in are printed in
203 Code samples are printed in
208 Interesting items and new terms are printed in
213 Compatibility with previous versions
216 This version has numerous bug fixes compared with the previous version.
217 But we also introduced some incompatibilities with older versions.
218 Not just for the fun of it, but to make the compiler more stable, efficient
225 short is now equivalent to int (16 bits), it used to be equivalent to char
226 (8 bits) which is not ANSI compliant
229 the default directory for gcc-builds where include, library and documention
230 files are stored is now in /usr/local/share
233 char type parameters to vararg functions are casted to int unless explicitly
250 will push a as an int and as a char resp.
253 option ---regextend has been removed
256 option ---noregparms has been removed
259 option ---stack-after-data has been removed
264 <pending: more incompatibilities?>
270 What do you need before you start installation of SDCC? A computer, and
272 The preferred method of installation is to compile SDCC from source using
274 For Windows some pre-compiled binary distributions are available for your
276 You should have some experience with command line tools and compiler use.
282 The SDCC home page at
283 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net/}
287 is a great place to find distribution sets.
288 You can also find links to the user mailing lists that offer help or discuss
289 SDCC with other SDCC users.
290 Web links to other SDCC related sites can also be found here.
291 This document can be found in the DOC directory of the source package as
293 Some of the other tools (simulator and assembler) included with SDCC contain
294 their own documentation and can be found in the source distribution.
295 If you want the latest unreleased software, the complete source package
296 is available directly by anonymous CVS on cvs.sdcc.sourceforge.net.
299 Wishes for the future
302 There are (and always will be) some things that could be done.
303 Here are some I can think of:
310 char KernelFunction3(char p) at 0x340;
316 If you can think of some more, please send them to the list.
322 <pending: And then of course a proper index-table
323 \begin_inset LatexCommand \index{index}
336 The install paths, search paths and other options are defined when running
338 The defaults can be overriden by:
340 \labelwidthstring 00.00.0000
342 ---prefix see tabel below
344 \labelwidthstring 00.00.0000
346 ---exec_prefix see tabel below
348 \labelwidthstring 00.00.0000
350 ---bindir see tabel below
352 \labelwidthstring 00.00.0000
354 ---datadir see tabel below
356 \labelwidthstring 00.00.0000
358 docdir environment variable, see tabel below
360 \labelwidthstring 00.00.0000
362 include_dir_suffix environment variable, see tabel below
364 \labelwidthstring 00.00.0000
366 lib_dir_suffix environment variable, see tabel below
368 \labelwidthstring 00.00.0000
370 sdccconf_h_dir_separator environment variable, either / or
375 This character will only be used in sdccconf.h; don't forget it's a C-header,
376 therefore a double-backslash is needed there.
378 \labelwidthstring 00.00.0000
380 ---disable-mcs51-port Excludes the Intel mcs51 port
382 \labelwidthstring 00.00.0000
384 ---disable-gbz80-port Excludes the Gameboy gbz80 port
386 \labelwidthstring 00.00.0000
388 ---disable-z80-port Excludes the z80 port
390 \labelwidthstring 00.00.0000
392 ---disable-avr-port Excludes the AVR port
394 \labelwidthstring 00.00.0000
396 ---disable-ds390-port Excludes the DS390 port
398 \labelwidthstring 00.00.0000
400 ---disable-pic-port Excludes the PIC port
402 \labelwidthstring 00.00.0000
404 ---disable-xa51-port Excludes the XA51 port
406 \labelwidthstring 00.00.0000
408 ---disable-ucsim Disables configuring and building of ucsim
410 \labelwidthstring 00.00.0000
412 ---disable-device-lib-build Disables automatically building device libraries
414 \labelwidthstring 00.00.0000
416 ---disable-packihx Disables building packihx
418 \labelwidthstring 00.00.0000
420 ---enable-libgc Use the Bohem memory allocator.
421 Lower runtime footprint.
424 Furthermore the environment variables CC, CFLAGS, ...
425 the tools and their arguments can be influenced.
426 Please see `configure ---help` and the man/info pages of `configure` for
431 The names of the standard libraries STD_LIB, STD_INT_LIB, STD_LONG_LIB,
432 STD_FP_LIB, STD_DS390_LIB, STD_XA51_LIB and the environment variables SDCC_DIR_
433 NAME, SDCC_INCLUDE_NAME, SDCC_LIB_NAME are defined by `configure` too.
434 At the moment it's not possible to change the default settings (it was
435 simply never required.
439 These configure options are compiled into the binaries, and can only be
440 changed by rerunning 'configure' and recompiling SDCC.
441 The configure options are written in
445 to distinguish them from run time environment variables (see section search
451 \begin_inset Quotes sld
455 \begin_inset Quotes srd
458 are used by the SDCC team to build the official Win32 binaries.
459 The SDCC team uses Mingw32 to build the official Windows binaries, because
466 a gcc compiler and last but not least
469 the binaries can be built by cross compiling on Sourceforge's compile farm.
472 See the examples, how to pass the Win32 settings to 'configure'.
473 The other Win32 builds using Borland, VC or whatever don't use 'configure',
474 but a header file sdcc_vc_in.h is the same as sdccconf.h built by 'configure'
485 <lyxtabular version="3" rows="8" columns="3">
487 <column alignment="left" valignment="top" leftline="true" width="0in">
488 <column alignment="left" valignment="top" leftline="true" width="0in">
489 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0in">
490 <row topline="true" bottomline="true">
491 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
499 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
507 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
517 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
527 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
535 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
547 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
557 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
567 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
579 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
589 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
601 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
617 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
627 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
639 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
651 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
661 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
673 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
689 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
699 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
707 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
716 <row topline="true" bottomline="true">
717 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
727 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
735 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
754 'configure' also computes relative paths.
755 This is needed for full relocatability of a binary package and to complete
756 search paths (see section search paths below):
762 <lyxtabular version="3" rows="4" columns="3">
764 <column alignment="left" valignment="top" leftline="true" width="0in">
765 <column alignment="left" valignment="top" leftline="true" width="0in">
766 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0in">
767 <row topline="true" bottomline="true">
768 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
776 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
784 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
793 <row topline="true" bottomline="true">
794 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
804 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
812 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
823 <row bottomline="true">
824 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
834 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
842 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
851 <row bottomline="true">
852 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
862 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
870 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
892 ./configure ---prefix=
893 \begin_inset Quotes srd
897 \begin_inset Quotes srd
901 \begin_inset Quotes srd
905 \begin_inset Quotes srd
911 ./configure ---disable-avr-port ---disable-xa51-port
914 To crosscompile on linux for Mingw32 (see also 'sdcc/support/scripts/sdcc_mingw3
924 \begin_inset Quotes srd
928 \begin_inset Quotes srd
932 \begin_inset Quotes srd
936 \begin_inset Quotes srd
945 \begin_inset Quotes srd
948 i586-mingw32msvc-ranlib
949 \begin_inset Quotes srd
958 \begin_inset Quotes srd
961 i586-mingw32msvc-strip
962 \begin_inset Quotes srd
971 \begin_inset Quotes srd
975 \begin_inset Quotes srd
984 \begin_inset Quotes srd
988 \begin_inset Quotes srd
997 \begin_inset Quotes srd
1001 \begin_inset Quotes srd
1010 \begin_inset Quotes srd
1014 \begin_inset Quotes srd
1023 \begin_inset Quotes srd
1027 \begin_inset Quotes srd
1035 sdccconf_h_dir_separator=
1036 \begin_inset Quotes srd
1048 \begin_inset Quotes srd
1056 ---disable-device-lib-build
1067 \begin_inset Quotes srd
1071 \begin_inset Quotes srd
1079 ---host=i586-mingw32msvc ---build=unknown-unknown-linux-gnu
1083 \begin_inset Quotes sld
1087 \begin_inset Quotes srd
1090 compile on Cygwin for Mingw32(see also sdcc/support/scripts/sdcc_cygwin_mingw32)
1100 \begin_inset Quotes srd
1104 \begin_inset Quotes srd
1113 \begin_inset Quotes srd
1117 \begin_inset Quotes srd
1126 \begin_inset Quotes srd
1130 \begin_inset Quotes srd
1139 \begin_inset Quotes srd
1143 \begin_inset Quotes srd
1152 \begin_inset Quotes srd
1156 \begin_inset Quotes srd
1165 \begin_inset Quotes srd
1169 \begin_inset Quotes srd
1178 \begin_inset Quotes srd
1182 \begin_inset Quotes srd
1190 sdccconf_h_dir_separator=
1191 \begin_inset Quotes srd
1203 \begin_inset Quotes srd
1214 'configure' is quite slow on Cygwin (at least on windows before Win2000/XP).
1215 The option '--C' turns on caching, which gives a little bit extra speed.
1216 However if options are changed, it can be necessary to delete the config.cache
1224 Binary files (preprocessor, assembler and linker)
1228 \begin_inset Tabular
1229 <lyxtabular version="3" rows="2" columns="3">
1231 <column alignment="left" valignment="top" leftline="true" width="0in">
1232 <column alignment="left" valignment="top" leftline="true" width="0in">
1233 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0in">
1234 <row topline="true" bottomline="true">
1235 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1243 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1251 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1260 <row topline="true" bottomline="true">
1261 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1271 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1279 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1305 \begin_inset Tabular
1306 <lyxtabular version="3" rows="2" columns="3">
1308 <column alignment="block" valignment="top" leftline="true" width="1.6in">
1309 <column alignment="left" valignment="top" leftline="true" width="0in">
1310 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0in">
1311 <row topline="true" bottomline="true">
1312 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1320 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1328 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1337 <row topline="true" bottomline="true">
1338 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1350 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1355 /usr/local/share/sdcc/include
1358 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1384 is auto-appended by the compiler, e.g.
1385 small, large, z80, ds390 etc.)
1389 \begin_inset Tabular
1390 <lyxtabular version="3" rows="2" columns="3">
1392 <column alignment="left" valignment="top" leftline="true" width="0in">
1393 <column alignment="left" valignment="top" leftline="true" width="0in">
1394 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0in">
1395 <row topline="true" bottomline="true">
1396 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1404 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1412 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1421 <row topline="true" bottomline="true">
1422 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1429 $DATADIR/$LIB_DIR_SUFFIX
1432 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1437 /usr/local/share/sdcc/lib
1440 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1466 \begin_inset Tabular
1467 <lyxtabular version="3" rows="2" columns="3">
1469 <column alignment="left" valignment="top" leftline="true" width="0in">
1470 <column alignment="left" valignment="top" leftline="true" width="0in">
1471 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0in">
1472 <row topline="true" bottomline="true">
1473 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1481 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1489 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1498 <row topline="true" bottomline="true">
1499 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1509 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1514 /usr/local/share/sdcc/doc
1517 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1538 The install paths can still be changed during `make install` with e.g.:
1541 make install prefix=$(HOME)/local/sdcc
1544 Of course this doesn't change the search paths compiled into the binaries.
1550 Some search paths or parts of them are determined by configure variables
1555 , see section above).
1556 Further search paths are determined by environment variables during runtime.
1559 The paths searched when running the compiler are as follows (the first catch
1565 Binary files (preprocessor, assembler and linker)
1568 \begin_inset Tabular
1569 <lyxtabular version="3" rows="4" columns="3">
1571 <column alignment="left" valignment="top" leftline="true" width="0in">
1572 <column alignment="left" valignment="top" leftline="true" width="0in">
1573 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0in">
1574 <row topline="true" bottomline="true">
1575 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1583 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1591 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1600 <row topline="true">
1601 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1611 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1619 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1630 <row topline="true">
1631 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1636 Path of argv[0] (if available)
1639 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1647 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1656 <row topline="true" bottomline="true">
1657 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1665 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1673 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1694 \begin_inset Tabular
1695 <lyxtabular version="3" rows="6" columns="3">
1697 <column alignment="block" valignment="top" leftline="true" width="1.5in">
1698 <column alignment="block" valignment="top" leftline="true" width="1.5in">
1699 <column alignment="left" valignment="top" leftline="true" rightline="true" width="0in">
1700 <row topline="true" bottomline="true">
1701 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1709 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1717 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1726 <row topline="true">
1727 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1735 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1743 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1752 <row topline="true">
1753 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1761 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1769 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1778 <row topline="true">
1779 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1793 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1805 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1816 <row topline="true">
1817 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1835 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
1885 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1898 <row topline="true" bottomline="true">
1899 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1915 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1920 /usr/local/share/sdcc/
1925 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1942 The option ---nostdinc disables the last two search paths.
1949 With the exception of
1950 \begin_inset Quotes sld
1954 \begin_inset Quotes srd
1961 is auto-appended by the compiler (e.g.
1962 small, large, z80, ds390 etc.).
1966 \begin_inset Tabular
1967 <lyxtabular version="3" rows="6" columns="3">
1969 <column alignment="block" valignment="top" leftline="true" width="1.7in">
1970 <column alignment="left" valignment="top" leftline="true" width="1.2in">
1971 <column alignment="block" valignment="top" leftline="true" rightline="true" width="1.2in">
1972 <row topline="true" bottomline="true">
1973 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1981 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
1989 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
1998 <row topline="true">
1999 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2007 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2015 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2024 <row topline="true">
2025 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2037 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2049 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2064 <row topline="true">
2065 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2076 $LIB_DIR_SUFFIX/<model>
2079 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2093 <cell alignment="left" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2110 <row topline="true">
2111 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2126 $LIB_DIR_SUFFIX/<model>
2129 <cell alignment="left" valignment="top" topline="true" leftline="true" usebox="none">
2182 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2238 <row topline="true" bottomline="true">
2239 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2248 $LIB_DIR_SUFFIX/<model>
2251 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
2256 /usr/local/share/sdcc/
2263 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
2279 Don't delete any of the stray spaces in the table above without checking
2280 the HTML output (last line)!
2286 The option ---nostdlib disables the last two search paths.
2290 \layout Subsubsection
2292 Building SDCC on Linux
2297 Download the source package
2299 either from the SDCC CVS repository or from the
2300 \begin_inset LatexCommand \url[nightly snapshots]{http://sdcc.sourceforge.net/snap.php}
2306 , it will be named something like sdcc
2319 Bring up a command line terminal, such as xterm.
2324 Unpack the file using a command like:
2327 "tar -xzf sdcc.src.tar.gz
2332 , this will create a sub-directory called sdcc with all of the sources.
2335 Change directory into the main SDCC directory, for example type:
2352 This configures the package for compilation on your system.
2368 All of the source packages will compile, this can take a while.
2384 This copies the binary executables, the include files, the libraries and
2385 the documentation to the install directories.
2386 \layout Subsubsection
2388 Building SDCC on OSX 2.x
2391 Follow the instruction for Linux.
2395 On OSX 2.x it was reported, that the default gcc (version 3.1 20020420 (prerelease
2396 )) fails to compile SDCC.
2397 Fortunately there's also gcc 2.9.x installed, which works fine.
2398 This compiler can be selected by running 'configure' with:
2401 ./configure CC=gcc2 CXX=g++2
2402 \layout Subsubsection
2404 Crosscompiling SDCC on Linux for Windows
2407 With the Mingw32 gcc crosscompiler it's easy to compile SDCC for Win32.
2408 See section 'Configure Options'.
2409 \layout Subsubsection
2411 Building SDCC on Windows
2414 With the exception of Cygwin the SDCC binaries uCsim and sdcdb can't be
2416 They use Unix-sockets, which are not available on Win32.
2417 \layout Subsubsection
2419 Windows Install Using a Binary Package
2422 Download the binary package and unpack it using your favorite unpacking
2423 tool (gunzip, WinZip, etc).
2424 This should unpack to a group of sub-directories.
2425 An example directory structure after unpacking the mingw32 package is:
2430 bin for the executables, c:
2438 lib for the include and libraries.
2441 Adjust your environment variable PATH to include the location of the bin
2442 directory or start sdcc using the full path.
2443 \layout Subsubsection
2445 Building SDCC using Cygwin and Mingw32
2448 For building and installing a Cygwin executable follow the instructions
2454 \begin_inset Quotes sld
2458 \begin_inset Quotes srd
2461 Win32-binary can be built, which will not need the Cygwin-DLL.
2462 For the necessary 'configure' options see section 'configure options' or
2463 the script 'sdcc/support/scripts/sdcc_cygwinmingw32'.
2467 In order to install Cygwin on Windows download setup.exe from
2468 \begin_inset LatexCommand \url[www.cygwin.com]{http://www.cygwin.com/}
2474 \begin_inset Quotes sld
2477 default text file type
2478 \begin_inset Quotes srd
2482 \begin_inset Quotes sld
2486 \begin_inset Quotes srd
2489 and download/install at least the following packages.
2490 Some packages are selected by default, others will be automatically selected
2491 because of dependencies with the manually selected packages.
2492 Never deselect these packages!
2501 gcc ; version 3.x is fine, no need to use the old 2.9x
2504 binutils ; selected with gcc
2510 rxvt ; a nice console, which makes life much easier under windoze (see below)
2513 man ; not really needed for building SDCC, but you'll miss it sooner or
2517 less ; not really needed for building SDCC, but you'll miss it sooner or
2521 cvs ; only if you use CVS access
2524 If you want to develop something you'll need:
2527 python ; for the regression tests
2530 gdb ; the gnu debugger, together with the nice GUI
2531 \begin_inset Quotes sld
2535 \begin_inset Quotes srd
2541 openssh ; to access the CF or commit changes
2544 autoconf and autoconf-devel ; if you want to fight with 'configure', don't
2545 use autoconf-stable!
2548 rxvt is a nice console with history.
2549 Replace in your cygwin.bat the line
2558 rxvt -sl 1000 -fn "Lucida Console-12" -sr -cr red
2561 -bg black -fg white -geometry 100x65 -e bash --login
2564 Text selected with the mouse is automatically copied to the clipboard, pasting
2565 works with shift-insert.
2569 The other good tip is to make sure you have no //c/-style paths anywhere,
2570 use /cygdrive/c/ instead.
2571 Using // invokes a network lookup which is very slow.
2573 \begin_inset Quotes sld
2577 \begin_inset Quotes srd
2580 is too long, you can change it with e.g.
2586 SDCC sources use the unix line ending LF.
2587 Life is much easier, if you store the source tree on a drive, which is
2588 mount in binary mode.
2589 And use an editor which can handle LF-only line endings.
2590 Make sure not to commit files with windows line endings.
2591 \layout Subsubsection
2593 Windows Install Using Microsoft Visual C++ 6.0/NET
2598 Download the source package
2600 either from the SDCC CVS repository or from the
2601 \begin_inset LatexCommand \url[nightly snapshots]{http://sdcc.sourceforge.net/snap.php}
2607 , it will be named something like sdcc
2614 SDCC is distributed with all the projects, workspaces, and files you need
2615 to build it using Visual C++ 6.0/NET.
2616 The workspace name is 'sdcc.dsw'.
2617 Please note that as it is now, all the executables are created in a folder
2621 Once built you need to copy the executables from sdcc
2625 bin before runnng SDCC.
2630 In order to build SDCC with Visual C++ 6.0/NET you need win32 executables
2631 of bison.exe, flex.exe, and gawk.exe.
2632 One good place to get them is
2633 \begin_inset LatexCommand \url[here]{http://unxutils.sourceforge.net}
2641 Download the file UnxUtils.zip.
2642 Now you have to install the utilities and setup Visual C++ so it can locate
2643 the required programs.
2644 Here there are two alternatives (choose one!):
2651 a) Extract UnxUtils.zip to your C:
2653 hard disk PRESERVING the original paths, otherwise bison won't work.
2654 (If you are using WinZip make certain that 'Use folder names' is selected)
2658 b) In the Visual C++ IDE click Tools, Options, select the Directory tab,
2659 in 'Show directories for:' select 'Executable files', and in the directories
2660 window add a new path: 'C:
2670 (As a side effect, you get a bunch of Unix utilities that could be useful,
2671 such as diff and patch.)
2678 This one avoids extracting a bunch of files you may not use, but requires
2683 a) Create a directory were to put the tools needed, or use a directory already
2691 b) Extract 'bison.exe', 'bison.hairy', 'bison.simple', 'flex.exe', and gawk.exe
2692 to such directory WITHOUT preserving the original paths.
2693 (If you are using WinZip make certain that 'Use folder names' is not selected)
2697 c) Rename bison.exe to '_bison.exe'.
2701 d) Create a batch file 'bison.bat' in 'C:
2705 ' and add these lines:
2725 _bison %1 %2 %3 %4 %5 %6 %7 %8 %9
2729 Steps 'c' and 'd' are needed because bison requires by default that the
2730 files 'bison.simple' and 'bison.hairy' reside in some weird Unix directory,
2731 '/usr/local/share/' I think.
2732 So it is necessary to tell bison where those files are located if they
2733 are not in such directory.
2734 That is the function of the environment variables BISON_SIMPLE and BISON_HAIRY.
2738 e) In the Visual C++ IDE click Tools, Options, select the Directory tab,
2739 in 'Show directories for:' select 'Executable files', and in the directories
2740 window add a new path: 'c:
2743 Note that you can use any other path instead of 'c:
2745 util', even the path where the Visual C++ tools are, probably: 'C:
2749 Microsoft Visual Studio
2754 So you don't have to execute step 'e' :)
2758 Open 'sdcc.dsw' in Visual Studio, click 'build all', when it finishes copy
2759 the executables from sdcc
2763 bin, and you can compile using sdcc.
2764 \layout Subsubsection
2766 Windows Install Using Borland
2769 From the sdcc directory, run the command "make -f Makefile.bcc".
2770 This should regenerate all the .exe files in the bin directory except for
2771 sdcdb.exe (which currently doesn't build under Borland C++).
2774 If you modify any source files and need to rebuild, be aware that the dependanci
2775 es may not be correctly calculated.
2776 The safest option is to delete all .obj files and run the build again.
2777 From a Cygwin BASH prompt, this can easily be done with the commmand:
2787 ( -name '*.obj' -o -name '*.lib' -o -name '*.rul'
2789 ) -print -exec rm {}
2798 or on Windows NT/2000/XP from the command prompt with the commmand:
2805 del /s *.obj *.lib *.rul
2808 from the sdcc directory.
2811 Building the Documentation
2818 Testing out the SDCC Compiler
2821 The first thing you should do after installing your SDCC compiler is to
2829 at the prompt, and the program should run and tell you the version.
2830 If it doesn't run, or gives a message about not finding sdcc program, then
2831 you need to check over your installation.
2832 Make sure that the sdcc bin directory is in your executable search path
2833 defined by the PATH environment setting (see the Trouble-shooting section
2835 Make sure that the sdcc program is in the bin folder, if not perhaps something
2836 did not install correctly.
2844 is commonly installed as described in section
2845 \begin_inset Quotes sld
2848 Install and search paths
2849 \begin_inset Quotes srd
2858 Make sure the compiler works on a very simple example.
2859 Type in the following test.c program using your favorite
2894 Compile this using the following command:
2903 If all goes well, the compiler will generate a test.asm and test.rel file.
2904 Congratulations, you've just compiled your first program with SDCC.
2905 We used the -c option to tell SDCC not to link the generated code, just
2906 to keep things simple for this step.
2914 The next step is to try it with the linker.
2924 If all goes well the compiler will link with the libraries and produce
2925 a test.ihx output file.
2930 (no test.ihx, and the linker generates warnings), then the problem is most
2931 likely that sdcc cannot find the
2935 usr/local/share/sdcc/lib directory
2939 (see the Install trouble-shooting section for suggestions).
2947 The final test is to ensure sdcc can use the
2951 header files and libraries.
2952 Edit test.c and change it to the following:
2972 strcpy(str1, "testing");
2981 Compile this by typing
2988 This should generate a test.ihx output file, and it should give no warnings
2989 such as not finding the string.h file.
2990 If it cannot find the string.h file, then the problem is that sdcc cannot
2991 find the /usr/local/share/sdcc/include directory
2995 (see the Install trouble-shooting section for suggestions).
2998 Install Trouble-shooting
2999 \layout Subsubsection
3001 SDCC does not build correctly.
3004 A thing to try is starting from scratch by unpacking the .tgz source package
3005 again in an empty directory.
3013 ./configure 2>&1 | tee configure.log
3027 make 2>&1 | tee make.log
3034 If anything goes wrong, you can review the log files to locate the problem.
3035 Or a relevant part of this can be attached to an email that could be helpful
3036 when requesting help from the mailing list.
3037 \layout Subsubsection
3040 \begin_inset Quotes sld
3044 \begin_inset Quotes srd
3051 \begin_inset Quotes sld
3055 \begin_inset Quotes srd
3058 command is a script that analyzes your system and performs some configuration
3059 to ensure the source package compiles on your system.
3060 It will take a few minutes to run, and will compile a few tests to determine
3061 what compiler features are installed.
3062 \layout Subsubsection
3065 \begin_inset Quotes sld
3069 \begin_inset Quotes srd
3075 This runs the GNU make tool, which automatically compiles all the source
3076 packages into the final installed binary executables.
3077 \layout Subsubsection
3080 \begin_inset Quotes sld
3084 \begin_inset Quotes erd
3090 This will install the compiler, other executables libraries and include
3091 files in to the appropriate directories.
3093 \begin_inset Quotes sld
3096 Install and Search PATHS
3097 \begin_inset Quotes srd
3102 On most systems you will need super-user privilages to do this.
3108 SDCC is not just a compiler, but a collection of tools by various developers.
3109 These include linkers, assemblers, simulators and other components.
3110 Here is a summary of some of the components.
3111 Note that the included simulator and assembler have separate documentation
3112 which you can find in the source package in their respective directories.
3113 As SDCC grows to include support for other processors, other packages from
3114 various developers are included and may have their own sets of documentation.
3118 You might want to look at the files which are installed in <installdir>.
3119 At the time of this writing, we find the following programs for gcc-builds:
3123 In <installdir>/bin:
3126 sdcc - The compiler.
3129 sdcpp - The C preprocessor.
3132 asx8051 - The assembler for 8051 type processors.
3139 as-gbz80 - The Z80 and GameBoy Z80 assemblers.
3142 aslink -The linker for 8051 type processors.
3149 link-gbz80 - The Z80 and GameBoy Z80 linkers.
3152 s51 - The ucSim 8051 simulator.
3155 sdcdb - The source debugger.
3158 packihx - A tool to pack (compress) Intel hex files.
3161 In <installdir>/share/sdcc/include
3167 In <installdir>/share/sdcc/lib
3170 the subdirs src and small, large, z80, gbz80 and ds390 with the precompiled
3174 In <installdir>/share/sdcc/doc
3180 As development for other processors proceeds, this list will expand to include
3181 executables to support processors like AVR, PIC, etc.
3182 \layout Subsubsection
3187 This is the actual compiler, it in turn uses the c-preprocessor and invokes
3188 the assembler and linkage editor.
3189 \layout Subsubsection
3191 sdcpp - The C-Preprocessor
3194 The preprocessor is a modified version of the GNU preprocessor.
3195 The C preprocessor is used to pull in #include sources, process #ifdef
3196 statements, #defines and so on.
3197 \layout Subsubsection
3199 asx8051, as-z80, as-gbz80, aslink, link-z80, link-gbz80 - The Assemblers
3203 This is retargettable assembler & linkage editor, it was developed by Alan
3205 John Hartman created the version for 8051, and I (Sandeep) have made some
3206 enhancements and bug fixes for it to work properly with the SDCC.
3207 \layout Subsubsection
3212 S51 is a freeware, opensource simulator developed by Daniel Drotos (
3213 \begin_inset LatexCommand \url{mailto:drdani@mazsola.iit.uni-miskolc.hu}
3218 The simulator is built as part of the build process.
3219 For more information visit Daniel's website at:
3220 \begin_inset LatexCommand \url{http://mazsola.iit.uni-miskolc.hu/~drdani/embedded/s51}
3225 It currently support the core mcs51, the Dallas DS80C390 and the Philips
3227 \layout Subsubsection
3229 sdcdb - Source Level Debugger
3235 <todo: is this thing still alive?>
3242 Sdcdb is the companion source level debugger.
3243 The current version of the debugger uses Daniel's Simulator S51, but can
3244 be easily changed to use other simulators.
3251 \layout Subsubsection
3253 Single Source File Projects
3256 For single source file 8051 projects the process is very simple.
3257 Compile your programs with the following command
3260 "sdcc sourcefile.c".
3264 This will compile, assemble and link your source file.
3265 Output files are as follows
3269 sourcefile.asm - Assembler source file created by the compiler
3271 sourcefile.lst - Assembler listing file created by the Assembler
3273 sourcefile.rst - Assembler listing file updated with linkedit information,
3274 created by linkage editor
3276 sourcefile.sym - symbol listing for the sourcefile, created by the assembler
3278 sourcefile.rel - Object file created by the assembler, input to Linkage editor
3280 sourcefile.map - The memory map for the load module, created by the Linker
3282 sourcefile.ihx - The load module in Intel hex format (you can select the
3283 Motorola S19 format with ---out-fmt-s19)
3285 sourcefile.cdb - An optional file (with ---debug) containing debug information
3287 sourcefile.dump* - Dump file to debug the compiler it self (with ---dumpall)
3289 \begin_inset Quotes sld
3292 Anatomy of the compiler
3293 \begin_inset Quotes srd
3297 \layout Subsubsection
3299 Projects with Multiple Source Files
3302 SDCC can compile only ONE file at a time.
3303 Let us for example assume that you have a project containing the following
3308 foo1.c (contains some functions)
3310 foo2.c (contains some more functions)
3312 foomain.c (contains more functions and the function main)
3320 The first two files will need to be compiled separately with the commands:
3352 Then compile the source file containing the
3356 function and link the files together with the following command:
3364 foomain.c\SpecialChar ~
3365 foo1.rel\SpecialChar ~
3377 can be separately compiled as well:
3388 sdcc foomain.rel foo1.rel foo2.rel
3395 The file containing the
3410 file specified in the command line, since the linkage editor processes
3411 file in the order they are presented to it.
3412 \layout Subsubsection
3414 Projects with Additional Libraries
3417 Some reusable routines may be compiled into a library, see the documentation
3418 for the assembler and linkage editor (which are in <installdir>/share/sdcc/doc)
3424 Libraries created in this manner can be included in the command line.
3425 Make sure you include the -L <library-path> option to tell the linker where
3426 to look for these files if they are not in the current directory.
3427 Here is an example, assuming you have the source file
3439 (if that is not the same as your current project):
3446 sdcc foomain.c foolib.lib -L mylib
3457 must be an absolute path name.
3461 The most efficient way to use libraries is to keep seperate modules in seperate
3463 The lib file now should name all the modules.rel files.
3464 For an example see the standard library file
3468 in the directory <installdir>/share/lib/small.
3471 Command Line Options
3472 \layout Subsubsection
3474 Processor Selection Options
3476 \labelwidthstring 00.00.0000
3482 Generate code for the MCS51 (8051) family of processors.
3483 This is the default processor target.
3485 \labelwidthstring 00.00.0000
3491 Generate code for the DS80C390 processor.
3493 \labelwidthstring 00.00.0000
3499 Generate code for the Z80 family of processors.
3501 \labelwidthstring 00.00.0000
3507 Generate code for the GameBoy Z80 processor.
3509 \labelwidthstring 00.00.0000
3515 Generate code for the Atmel AVR processor (In development, not complete).
3517 \labelwidthstring 00.00.0000
3523 Generate code for the PIC 14-bit processors (In development, not complete).
3525 \labelwidthstring 00.00.0000
3531 Generate code for the Toshiba TLCS-900H processor (In development, not
3534 \labelwidthstring 00.00.0000
3540 Generate code for the Philips XA51 processor (In development, not complete).
3541 \layout Subsubsection
3543 Preprocessor Options
3545 \labelwidthstring 00.00.0000
3551 The additional location where the pre processor will look for <..h> or
3552 \begin_inset Quotes eld
3556 \begin_inset Quotes erd
3561 \labelwidthstring 00.00.0000
3567 Command line definition of macros.
3568 Passed to the pre processor.
3570 \labelwidthstring 00.00.0000
3576 Tell the preprocessor to output a rule suitable for make describing the
3577 dependencies of each object file.
3578 For each source file, the preprocessor outputs one make-rule whose target
3579 is the object file name for that source file and whose dependencies are
3580 all the files `#include'd in it.
3581 This rule may be a single line or may be continued with `
3583 '-newline if it is long.
3584 The list of rules is printed on standard output instead of the preprocessed
3588 \labelwidthstring 00.00.0000
3594 Tell the preprocessor not to discard comments.
3595 Used with the `-E' option.
3597 \labelwidthstring 00.00.0000
3608 Like `-M' but the output mentions only the user header files included with
3610 \begin_inset Quotes eld
3614 System header files included with `#include <file>' are omitted.
3616 \labelwidthstring 00.00.0000
3622 Assert the answer answer for question, in case it is tested with a preprocessor
3623 conditional such as `#if #question(answer)'.
3624 `-A-' disables the standard assertions that normally describe the target
3627 \labelwidthstring 00.00.0000
3633 (answer) Assert the answer answer for question, in case it is tested with
3634 a preprocessor conditional such as `#if #question(answer)'.
3635 `-A-' disables the standard assertions that normally describe the target
3638 \labelwidthstring 00.00.0000
3644 Undefine macro macro.
3645 `-U' options are evaluated after all `-D' options, but before any `-include'
3646 and `-imacros' options.
3648 \labelwidthstring 00.00.0000
3654 Tell the preprocessor to output only a list of the macro definitions that
3655 are in effect at the end of preprocessing.
3656 Used with the `-E' option.
3658 \labelwidthstring 00.00.0000
3664 Tell the preprocessor to pass all macro definitions into the output, in
3665 their proper sequence in the rest of the output.
3667 \labelwidthstring 00.00.0000
3678 Like `-dD' except that the macro arguments and contents are omitted.
3679 Only `#define name' is included in the output.
3680 \layout Subsubsection
3684 \labelwidthstring 00.00.0000
3694 <absolute path to additional libraries> This option is passed to the linkage
3695 editor's additional libraries search path.
3696 The path name must be absolute.
3697 Additional library files may be specified in the command line.
3698 See section Compiling programs for more details.
3700 \labelwidthstring 00.00.0000
3706 <Value> The start location of the external ram, default value is 0.
3707 The value entered can be in Hexadecimal or Decimal format, e.g.: ---xram-loc
3708 0x8000 or ---xram-loc 32768.
3710 \labelwidthstring 00.00.0000
3716 <Value> The start location of the code segment, default value 0.
3717 Note when this option is used the interrupt vector table is also relocated
3718 to the given address.
3719 The value entered can be in Hexadecimal or Decimal format, e.g.: ---code-loc
3720 0x8000 or ---code-loc 32768.
3722 \labelwidthstring 00.00.0000
3728 <Value> By default the stack is placed after the data segment.
3729 Using this option the stack can be placed anywhere in the internal memory
3731 The value entered can be in Hexadecimal or Decimal format, e.g.
3732 ---stack-loc 0x20 or ---stack-loc 32.
3733 Since the sp register is incremented before a push or call, the initial
3734 sp will be set to one byte prior the provided value.
3735 The provided value should not overlap any other memory areas such as used
3736 register banks or the data segment and with enough space for the current
3739 \labelwidthstring 00.00.0000
3745 <Value> The start location of the internal ram data segment.
3746 The value entered can be in Hexadecimal or Decimal format, eg.
3747 ---data-loc 0x20 or ---data-loc 32.
3748 (By default, the start location of the internal ram data segment is set
3749 as low as possible in memory, taking into account the used register banks
3750 and the bit segment at address 0x20.
3751 For example if register banks 0 and 1 are used without bit variables, the
3752 data segment will be set, if ---data-loc is not used, to location 0x10.)
3754 \labelwidthstring 00.00.0000
3760 <Value> The start location of the indirectly addressable internal ram, default
3762 The value entered can be in Hexadecimal or Decimal format, eg.
3763 ---idata-loc 0x88 or ---idata-loc 136.
3765 \labelwidthstring 00.00.0000
3774 The linker output (final object code) is in Intel Hex format.
3775 (This is the default option).
3777 \labelwidthstring 00.00.0000
3786 The linker output (final object code) is in Motorola S19 format.
3787 \layout Subsubsection
3791 \labelwidthstring 00.00.0000
3797 Generate code for Large model programs see section Memory Models for more
3799 If this option is used all source files in the project should be compiled
3801 In addition the standard library routines are compiled with small model,
3802 they will need to be recompiled.
3804 \labelwidthstring 00.00.0000
3815 Generate code for Small Model programs see section Memory Models for more
3817 This is the default model.
3818 \layout Subsubsection
3822 \labelwidthstring 00.00.0000
3833 Generate 24-bit flat mode code.
3834 This is the one and only that the ds390 code generator supports right now
3835 and is default when using
3840 See section Memory Models for more details.
3842 \labelwidthstring 00.00.0000
3848 Generate code for the 10 bit stack mode of the Dallas DS80C390 part.
3849 This is the one and only that the ds390 code generator supports right now
3850 and is default when using
3855 In this mode, the stack is located in the lower 1K of the internal RAM,
3856 which is mapped to 0x400000.
3857 Note that the support is incomplete, since it still uses a single byte
3858 as the stack pointer.
3859 This means that only the lower 256 bytes of the potential 1K stack space
3860 will actually be used.
3861 However, this does allow you to reclaim the precious 256 bytes of low RAM
3862 for use for the DATA and IDATA segments.
3863 The compiler will not generate any code to put the processor into 10 bit
3865 It is important to ensure that the processor is in this mode before calling
3866 any re-entrant functions compiled with this option.
3867 In principle, this should work with the
3871 option, but that has not been tested.
3872 It is incompatible with the
3877 It also only makes sense if the processor is in 24 bit contiguous addressing
3880 ---model-flat24 option
3883 \layout Subsubsection
3885 Optimization Options
3887 \labelwidthstring 00.00.0000
3893 Will not do global subexpression elimination, this option may be used when
3894 the compiler creates undesirably large stack/data spaces to store compiler
3896 A warning message will be generated when this happens and the compiler
3897 will indicate the number of extra bytes it allocated.
3898 It recommended that this option NOT be used, #pragma\SpecialChar ~
3900 to turn off global subexpression elimination for a given function only.
3902 \labelwidthstring 00.00.0000
3908 Will not do loop invariant optimizations, this may be turned off for reasons
3909 explained for the previous option.
3910 For more details of loop optimizations performed see section Loop Invariants.It
3911 recommended that this option NOT be used, #pragma\SpecialChar ~
3912 NOINVARIANT can be used
3913 to turn off invariant optimizations for a given function only.
3915 \labelwidthstring 00.00.0000
3921 Will not do loop induction optimizations, see section strength reduction
3922 for more details.It is recommended that this option is NOT used, #pragma\SpecialChar ~
3924 ION can be used to turn off induction optimizations for a given function
3927 \labelwidthstring 00.00.0000
3938 Will not generate boundary condition check when switch statements are implement
3939 ed using jump-tables.
3940 See section Switch Statements for more details.
3941 It is recommended that this option is NOT used, #pragma\SpecialChar ~
3943 used to turn off boundary checking for jump tables for a given function
3946 \labelwidthstring 00.00.0000
3955 Will not do loop reversal optimization.
3957 \labelwidthstring 00.00.0000
3963 Will not optimize labels (makes the dumpfiles more readable).
3965 \labelwidthstring 00.00.0000
3971 Will not memcpy initialized data in far space from code space.
3972 This saves a few bytes in code space if you don't have initialized data.
3973 \layout Subsubsection
3977 \labelwidthstring 00.00.0000
3984 will compile and assemble the source, but will not call the linkage editor.
3986 \labelwidthstring 00.00.0000
3992 reads the preprocessed source from standard input and compiles it.
3993 The file name for the assembler output must be specified using the -o option.
3995 \labelwidthstring 00.00.0000
4001 Run only the C preprocessor.
4002 Preprocess all the C source files specified and output the results to standard
4005 \labelwidthstring 00.00.0000
4012 The output path resp.
4013 file where everything will be placed.
4014 If the parameter is a path, it must have a trailing slash (or backslash
4015 for the Windows binaries) to be recognized as a path.
4018 \labelwidthstring 00.00.0000
4029 All functions in the source file will be compiled as
4034 the parameters and local variables will be allocated on the stack.
4035 see section Parameters and Local Variables for more details.
4036 If this option is used all source files in the project should be compiled
4040 \labelwidthstring 00.00.0000
4046 Uses a pseudo stack in the first 256 bytes in the external ram for allocating
4047 variables and passing parameters.
4048 See section on external stack for more details.
4050 \labelwidthstring 00.00.0000
4054 ---callee-saves function1[,function2][,function3]....
4057 The compiler by default uses a caller saves convention for register saving
4058 across function calls, however this can cause unneccessary register pushing
4059 & popping when calling small functions from larger functions.
4060 This option can be used to switch the register saving convention for the
4061 function names specified.
4062 The compiler will not save registers when calling these functions, no extra
4063 code will be generated at the entry & exit for these functions to save
4064 & restore the registers used by these functions, this can SUBSTANTIALLY
4065 reduce code & improve run time performance of the generated code.
4066 In the future the compiler (with interprocedural analysis) will be able
4067 to determine the appropriate scheme to use for each function call.
4068 DO NOT use this option for built-in functions such as _muluint..., if this
4069 option is used for a library function the appropriate library function
4070 needs to be recompiled with the same option.
4071 If the project consists of multiple source files then all the source file
4072 should be compiled with the same ---callee-saves option string.
4073 Also see #pragma\SpecialChar ~
4076 \labelwidthstring 00.00.0000
4085 When this option is used the compiler will generate debug information, that
4086 can be used with the SDCDB.
4087 The debug information is collected in a file with .cdb extension.
4088 For more information see documentation for SDCDB.
4090 \labelwidthstring 00.00.0000
4096 <filename> This option can be used to use additional rules to be used by
4097 the peep hole optimizer.
4098 See section Peep Hole optimizations for details on how to write these rules.
4100 \labelwidthstring 00.00.0000
4111 Stop after the stage of compilation proper; do not assemble.
4112 The output is an assembler code file for the input file specified.
4114 \labelwidthstring 00.00.0000
4118 -Wa_asmOption[,asmOption]
4121 Pass the asmOption to the assembler.
4123 \labelwidthstring 00.00.0000
4127 -Wl_linkOption[,linkOption]
4130 Pass the linkOption to the linker.
4132 \labelwidthstring 00.00.0000
4141 Integer (16 bit) and long (32 bit) libraries have been compiled as reentrant.
4142 Note by default these libraries are compiled as non-reentrant.
4143 See section Installation for more details.
4145 \labelwidthstring 00.00.0000
4154 This option will cause the compiler to generate an information message for
4155 each function in the source file.
4156 The message contains some
4160 information about the function.
4161 The number of edges and nodes the compiler detected in the control flow
4162 graph of the function, and most importantly the
4164 cyclomatic complexity
4166 see section on Cyclomatic Complexity for more details.
4168 \labelwidthstring 00.00.0000
4177 Floating point library is compiled as reentrant.See section Installation
4180 \labelwidthstring 00.00.0000
4186 The compiler will not overlay parameters and local variables of any function,
4187 see section Parameters and local variables for more details.
4189 \labelwidthstring 00.00.0000
4195 This option can be used when the code generated is called by a monitor
4197 The compiler will generate a 'ret' upon return from the 'main' function.
4198 The default option is to lock up i.e.
4201 \labelwidthstring 00.00.0000
4207 Disable peep-hole optimization.
4209 \labelwidthstring 00.00.0000
4215 Pass the inline assembler code through the peep hole optimizer.
4216 This can cause unexpected changes to inline assembler code, please go through
4217 the peephole optimizer rules defined in the source file tree '<target>/peeph.def
4218 ' before using this option.
4220 \labelwidthstring 00.00.0000
4226 <Value> Causes the linker to check if the internal ram usage is within limits
4229 \labelwidthstring 00.00.0000
4235 <Value> Causes the linker to check if the external ram usage is within limits
4238 \labelwidthstring 00.00.0000
4244 <Value> Causes the linker to check if the code usage is within limits of
4247 \labelwidthstring 00.00.0000
4253 This will prevent the compiler from passing on the default include path
4254 to the preprocessor.
4256 \labelwidthstring 00.00.0000
4262 This will prevent the compiler from passing on the default library path
4265 \labelwidthstring 00.00.0000
4271 Shows the various actions the compiler is performing.
4273 \labelwidthstring 00.00.0000
4279 Shows the actual commands the compiler is executing.
4281 \labelwidthstring 00.00.0000
4287 Hides your ugly and inefficient c-code from the asm file, so you can always
4288 blame the compiler :).
4290 \labelwidthstring 00.00.0000
4296 Include i-codes in the asm file.
4297 Sounds like noise but is most helpfull for debugging the compiler itself.
4299 \labelwidthstring 00.00.0000
4305 Disable some of the more pedantic warnings (jwk burps: please be more specific
4307 \layout Subsubsection
4309 Intermediate Dump Options
4312 The following options are provided for the purpose of retargetting and debugging
4314 These provided a means to dump the intermediate code (iCode) generated
4315 by the compiler in human readable form at various stages of the compilation
4319 \labelwidthstring 00.00.0000
4325 This option will cause the compiler to dump the intermediate code into
4328 <source filename>.dumpraw
4330 just after the intermediate code has been generated for a function, i.e.
4331 before any optimizations are done.
4332 The basic blocks at this stage ordered in the depth first number, so they
4333 may not be in sequence of execution.
4335 \labelwidthstring 00.00.0000
4341 Will create a dump of iCode's, after global subexpression elimination,
4344 <source filename>.dumpgcse.
4346 \labelwidthstring 00.00.0000
4352 Will create a dump of iCode's, after deadcode elimination, into a file
4355 <source filename>.dumpdeadcode.
4357 \labelwidthstring 00.00.0000
4366 Will create a dump of iCode's, after loop optimizations, into a file named
4369 <source filename>.dumploop.
4371 \labelwidthstring 00.00.0000
4380 Will create a dump of iCode's, after live range analysis, into a file named
4383 <source filename>.dumprange.
4385 \labelwidthstring 00.00.0000
4391 Will dump the life ranges for all symbols.
4393 \labelwidthstring 00.00.0000
4402 Will create a dump of iCode's, after register assignment, into a file named
4405 <source filename>.dumprassgn.
4407 \labelwidthstring 00.00.0000
4413 Will create a dump of the live ranges of iTemp's
4415 \labelwidthstring 00.00.0000
4426 Will cause all the above mentioned dumps to be created.
4429 Environment variables
4432 SDCC recognizes the following environment variables:
4434 \labelwidthstring 00.00.0000
4440 SDCC installs a signal handler to be able to delete temporary files after
4441 an user break (^C) or an exception.
4442 If this environment variable is set, SDCC won't install the signal handler
4443 in order to be able to debug SDCC.
4445 \labelwidthstring 00.00.0000
4453 Path, where temporary files will be created.
4454 The order of the variables is the search order.
4455 In a standard *nix environment these variables are not set, and there's
4456 no need to set them.
4457 On Windows it's recommended to set one of them.
4459 \labelwidthstring 00.00.0000
4466 \begin_inset Quotes sld
4469 2.3 Install and search paths
4470 \begin_inset Quotes srd
4475 \labelwidthstring 00.00.0000
4482 \begin_inset Quotes sld
4485 2.3 Install and search paths
4486 \begin_inset Quotes srd
4491 \labelwidthstring 00.00.0000
4498 \begin_inset Quotes sld
4501 2.3 Install and search paths
4502 \begin_inset Quotes srd
4508 There are some more environment variables recognized by SDCC, but these
4509 are solely used for debugging purposes.
4510 They can change or disappear very quickly, and will never be documentated.
4513 MCS51/DS390 Storage Class Language Extensions
4516 In addition to the ANSI storage classes SDCC allows the following MCS51
4517 specific storage classes.
4518 \layout Subsubsection
4523 Variables declared with this storage class will be placed in the extern
4529 storage class for Large Memory model, e.g.:
4535 xdata unsigned char xduc;
4536 \layout Subsubsection
4545 storage class for Small Memory model.
4546 Variables declared with this storage class will be allocated in the internal
4554 \layout Subsubsection
4559 Variables declared with this storage class will be allocated into the indirectly
4560 addressable portion of the internal ram of a 8051, e.g.:
4567 \layout Subsubsection
4572 This is a data-type and a storage class specifier.
4573 When a variable is declared as a bit, it is allocated into the bit addressable
4574 memory of 8051, e.g.:
4581 \layout Subsubsection
4586 Like the bit keyword,
4590 signifies both a data-type and storage class, they are used to describe
4591 the special function registers and special bit variables of a 8051, eg:
4597 sfr at 0x80 P0; /* special function register P0 at location 0x80 */
4599 sbit at 0xd7 CY; /* CY (Carry Flag) */
4605 SDCC allows (via language extensions) pointers to explicitly point to any
4606 of the memory spaces of the 8051.
4607 In addition to the explicit pointers, the compiler uses (by default) generic
4608 pointers which can be used to point to any of the memory spaces.
4612 Pointer declaration examples:
4621 /* pointer physically in xternal ram pointing to object in internal ram
4624 data unsigned char * xdata p;
4628 /* pointer physically in code rom pointing to data in xdata space */
4630 xdata unsigned char * code p;
4634 /* pointer physically in code space pointing to data in code space */
4636 code unsigned char * code p;
4640 /* the folowing is a generic pointer physically located in xdata space */
4651 Well you get the idea.
4656 All unqualified pointers are treated as 3-byte (4-byte for the ds390)
4669 The highest order byte of the
4673 pointers contains the data space information.
4674 Assembler support routines are called whenever data is stored or retrieved
4680 These are useful for developing reusable library routines.
4681 Explicitly specifying the pointer type will generate the most efficient
4685 Parameters & Local Variables
4688 Automatic (local) variables and parameters to functions can either be placed
4689 on the stack or in data-space.
4690 The default action of the compiler is to place these variables in the internal
4691 RAM (for small model) or external RAM (for large model).
4692 This in fact makes them
4696 so by default functions are non-reentrant.
4700 They can be placed on the stack either by using the
4704 option or by using the
4708 keyword in the function declaration, e.g.:
4717 unsigned char foo(char i) reentrant
4730 Since stack space on 8051 is limited, the
4738 option should be used sparingly.
4739 Note that the reentrant keyword just means that the parameters & local
4740 variables will be allocated to the stack, it
4744 mean that the function is register bank independent.
4748 Local variables can be assigned storage classes and absolute addresses,
4755 unsigned char foo() {
4761 xdata unsigned char i;
4773 data at 0x31 unsiged char j;
4788 In the above example the variable
4792 will be allocated in the external ram,
4796 in bit addressable space and
4805 or when a function is declared as
4809 this should only be done for static variables.
4812 Parameters however are not allowed any storage class, (storage classes for
4813 parameters will be ignored), their allocation is governed by the memory
4814 model in use, and the reentrancy options.
4820 For non-reentrant functions SDCC will try to reduce internal ram space usage
4821 by overlaying parameters and local variables of a function (if possible).
4822 Parameters and local variables of a function will be allocated to an overlayabl
4823 e segment if the function has
4825 no other function calls and the function is non-reentrant and the memory
4829 If an explicit storage class is specified for a local variable, it will
4833 Note that the compiler (not the linkage editor) makes the decision for overlayin
4835 Functions that are called from an interrupt service routine should be preceded
4836 by a #pragma\SpecialChar ~
4837 NOOVERLAY if they are not reentrant.
4840 Also note that the compiler does not do any processing of inline assembler
4841 code, so the compiler might incorrectly assign local variables and parameters
4842 of a function into the overlay segment if the inline assembler code calls
4843 other c-functions that might use the overlay.
4844 In that case the #pragma\SpecialChar ~
4845 NOOVERLAY should be used.
4848 Parameters and Local variables of functions that contain 16 or 32 bit multiplica
4849 tion or division will NOT be overlayed since these are implemented using
4850 external functions, e.g.:
4860 void set_error(unsigned char errcd)
4876 void some_isr () interrupt 2 using 1
4905 In the above example the parameter
4913 would be assigned to the overlayable segment if the #pragma\SpecialChar ~
4915 not present, this could cause unpredictable runtime behavior when called
4917 The #pragma\SpecialChar ~
4918 NOOVERLAY ensures that the parameters and local variables for
4919 the function are NOT overlayed.
4922 Interrupt Service Routines
4925 SDCC allows interrupt service routines to be coded in C, with some extended
4932 void timer_isr (void) interrupt 2 using 1
4945 The number following the
4949 keyword is the interrupt number this routine will service.
4950 The compiler will insert a call to this routine in the interrupt vector
4951 table for the interrupt number specified.
4956 keyword is used to tell the compiler to use the specified register bank
4957 (8051 specific) when generating code for this function.
4958 Note that when some function is called from an interrupt service routine
4959 it should be preceded by a #pragma\SpecialChar ~
4960 NOOVERLAY if it is not reentrant.
4961 A special note here, int (16 bit) and long (32 bit) integer division, multiplic
4962 ation & modulus operations are implemented using external support routines
4963 developed in ANSI-C, if an interrupt service routine needs to do any of
4964 these operations then the support routines (as mentioned in a following
4965 section) will have to be recompiled using the
4969 option and the source file will need to be compiled using the
4976 If you have multiple source files in your project, interrupt service routines
4977 can be present in any of them, but a prototype of the isr MUST be present
4978 or included in the file that contains the function
4985 Interrupt Numbers and the corresponding address & descriptions for the Standard
4986 8051 are listed below.
4987 SDCC will automatically adjust the interrupt vector table to the maximum
4988 interrupt number specified.
4994 \begin_inset Tabular
4995 <lyxtabular version="3" rows="6" columns="3">
4997 <column alignment="center" valignment="top" leftline="true" width="0in">
4998 <column alignment="center" valignment="top" leftline="true" width="0in">
4999 <column alignment="center" valignment="top" leftline="true" rightline="true" width="0in">
5000 <row topline="true" bottomline="true">
5001 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5009 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5017 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
5026 <row topline="true">
5027 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5035 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5043 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
5052 <row topline="true">
5053 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5061 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5069 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
5078 <row topline="true">
5079 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5087 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5095 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
5104 <row topline="true">
5105 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5113 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5121 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
5130 <row topline="true" bottomline="true">
5131 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5139 <cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
5147 <cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
5164 If the interrupt service routine is defined without
5168 a register bank or with register bank 0 (using 0), the compiler will save
5169 the registers used by itself on the stack upon entry and restore them at
5170 exit, however if such an interrupt service routine calls another function
5171 then the entire register bank will be saved on the stack.
5172 This scheme may be advantageous for small interrupt service routines which
5173 have low register usage.
5176 If the interrupt service routine is defined to be using a specific register
5181 are save and restored, if such an interrupt service routine calls another
5182 function (using another register bank) then the entire register bank of
5183 the called function will be saved on the stack.
5184 This scheme is recommended for larger interrupt service routines.
5187 Calling other functions from an interrupt service routine is not recommended,
5188 avoid it if possible.
5192 Also see the _naked modifier.
5200 <TODO: this isn't implemented at all!>
5206 A special keyword may be associated with a function declaring it as
5211 SDCC will generate code to disable all interrupts upon entry to a critical
5212 function and enable them back before returning.
5213 Note that nesting critical functions may cause unpredictable results.
5238 The critical attribute maybe used with other attributes like
5246 A special keyword may be associated with a function declaring it as
5255 function modifier attribute prevents the compiler from generating prologue
5256 and epilogue code for that function.
5257 This means that the user is entirely responsible for such things as saving
5258 any registers that may need to be preserved, selecting the proper register
5259 bank, generating the
5263 instruction at the end, etc.
5264 Practically, this means that the contents of the function must be written
5265 in inline assembler.
5266 This is particularly useful for interrupt functions, which can have a large
5267 (and often unnecessary) prologue/epilogue.
5268 For example, compare the code generated by these two functions:
5274 data unsigned char counter;
5276 void simpleInterrupt(void) interrupt 1
5290 void nakedInterrupt(void) interrupt 2 _naked
5323 ; MUST explicitly include ret in _naked function.
5337 For an 8051 target, the generated simpleInterrupt looks like:
5482 whereas nakedInterrupt looks like:
5507 ; MUST explicitly include ret(i) in _naked function.
5513 While there is nothing preventing you from writing C code inside a _naked
5514 function, there are many ways to shoot yourself in the foot doing this,
5515 and it is recommended that you stick to inline assembler.
5518 Functions using private banks
5525 attribute (which tells the compiler to use a register bank other than the
5526 default bank zero) should only be applied to
5530 functions (see note 1 below).
5531 This will in most circumstances make the generated ISR code more efficient
5532 since it will not have to save registers on the stack.
5539 attribute will have no effect on the generated code for a
5543 function (but may occasionally be useful anyway
5549 possible exception: if a function is called ONLY from 'interrupt' functions
5550 using a particular bank, it can be declared with the same 'using' attribute
5551 as the calling 'interrupt' functions.
5552 For instance, if you have several ISRs using bank one, and all of them
5553 call memcpy(), it might make sense to create a specialized version of memcpy()
5554 'using 1', since this would prevent the ISR from having to save bank zero
5555 to the stack on entry and switch to bank zero before calling the function
5562 (pending: I don't think this has been done yet)
5569 function using a non-zero bank will assume that it can trash that register
5570 bank, and will not save it.
5571 Since high-priority interrupts can interrupt low-priority ones on the 8051
5572 and friends, this means that if a high-priority ISR
5576 a particular bank occurs while processing a low-priority ISR
5580 the same bank, terrible and bad things can happen.
5581 To prevent this, no single register bank should be
5585 by both a high priority and a low priority ISR.
5586 This is probably most easily done by having all high priority ISRs use
5587 one bank and all low priority ISRs use another.
5588 If you have an ISR which can change priority at runtime, you're on your
5589 own: I suggest using the default bank zero and taking the small performance
5593 It is most efficient if your ISR calls no other functions.
5594 If your ISR must call other functions, it is most efficient if those functions
5595 use the same bank as the ISR (see note 1 below); the next best is if the
5596 called functions use bank zero.
5597 It is very inefficient to call a function using a different, non-zero bank
5605 Data items can be assigned an absolute address with the
5609 keyword, in addition to a storage class, e.g.:
5615 xdata at 0x8000 unsigned char PORTA_8255 ;
5621 In the above example the PORTA_8255 will be allocated to the location 0x8000
5622 of the external ram.
5623 Note that this feature is provided to give the programmer access to
5627 devices attached to the controller.
5628 The compiler does not actually reserve any space for variables declared
5629 in this way (they are implemented with an equate in the assembler).
5630 Thus it is left to the programmer to make sure there are no overlaps with
5631 other variables that are declared without the absolute address.
5632 The assembler listing file (.lst) and the linker output files (.rst) and
5633 (.map) are a good places to look for such overlaps.
5637 Absolute address can be specified for variables in all storage classes,
5650 The above example will allocate the variable at offset 0x02 in the bit-addressab
5652 There is no real advantage to assigning absolute addresses to variables
5653 in this manner, unless you want strict control over all the variables allocated.
5659 The compiler inserts a call to the C routine
5661 _sdcc__external__startup()
5666 at the start of the CODE area.
5667 This routine is in the runtime library.
5668 By default this routine returns 0, if this routine returns a non-zero value,
5669 the static & global variable initialization will be skipped and the function
5670 main will be invoked Other wise static & global variables will be initialized
5671 before the function main is invoked.
5674 _sdcc__external__startup()
5676 routine to your program to override the default if you need to setup hardware
5677 or perform some other critical operation prior to static & global variable
5681 Inline Assembler Code
5684 SDCC allows the use of in-line assembler with a few restriction as regards
5686 All labels defined within inline assembler code
5694 where nnnn is a number less than 100 (which implies a limit of utmost 100
5695 inline assembler labels
5703 It is strongly recommended that each assembly instruction (including labels)
5704 be placed in a separate line (as the example shows).
5709 command line option is used, the inline assembler code will be passed through
5710 the peephole optimizer.
5711 This might cause some unexpected changes in the inline assembler code.
5712 Please go throught the peephole optimizer rules defined in file
5716 carefully before using this option.
5756 The inline assembler code can contain any valid code understood by the assembler
5757 , this includes any assembler directives and comment lines.
5758 The compiler does not do any validation of the code within the
5768 Inline assembler code cannot reference any C-Labels, however it can reference
5769 labels defined by the inline assembler, e.g.:
5795 ; some assembler code
5815 /* some more c code */
5817 clabel:\SpecialChar ~
5819 /* inline assembler cannot reference this label */
5831 $0003: ;label (can be reference by inline assembler only)
5843 /* some more c code */
5851 In other words inline assembly code can access labels defined in inline
5852 assembly within the scope of the funtion.
5856 The same goes the other way, ie.
5857 labels defines in inline assembly CANNOT be accessed by C statements.
5860 int (16 bit) and long (32 bit) Support
5863 For signed & unsigned int (16 bit) and long (32 bit) variables, division,
5864 multiplication and modulus operations are implemented by support routines.
5865 These support routines are all developed in ANSI-C to facilitate porting
5866 to other MCUs, although some model specific assembler optimations are used.
5867 The following files contain the described routine, all of them can be found
5868 in <installdir>/share/sdcc/lib.
5874 <pending: tabularise this>
5880 _mulsint.c - signed 16 bit multiplication (calls _muluint)
5882 _muluint.c - unsigned 16 bit multiplication
5884 _divsint.c - signed 16 bit division (calls _divuint)
5886 _divuint.c - unsigned 16 bit division
5888 _modsint.c - signed 16 bit modulus (call _moduint)
5890 _moduint.c - unsigned 16 bit modulus
5892 _mulslong.c - signed 32 bit multiplication (calls _mululong)
5894 _mululong.c - unsigned32 bit multiplication
5896 _divslong.c - signed 32 division (calls _divulong)
5898 _divulong.c - unsigned 32 division
5900 _modslong.c - signed 32 bit modulus (calls _modulong)
5902 _modulong.c - unsigned 32 bit modulus
5910 Since they are compiled as
5914 , interrupt service routines should not do any of the above operations.
5915 If this is unavoidable then the above routines will need to be compiled
5920 option, after which the source program will have to be compiled with
5927 Floating Point Support
5930 SDCC supports IEEE (single precision 4bytes) floating point numbers.The floating
5931 point support routines are derived from gcc's floatlib.c and consists of
5932 the following routines:
5938 <pending: tabularise this>
5944 _fsadd.c - add floating point numbers
5946 _fssub.c - subtract floating point numbers
5948 _fsdiv.c - divide floating point numbers
5950 _fsmul.c - multiply floating point numbers
5952 _fs2uchar.c - convert floating point to unsigned char
5954 _fs2char.c - convert floating point to signed char
5956 _fs2uint.c - convert floating point to unsigned int
5958 _fs2int.c - convert floating point to signed int
5960 _fs2ulong.c - convert floating point to unsigned long
5962 _fs2long.c - convert floating point to signed long
5964 _uchar2fs.c - convert unsigned char to floating point
5966 _char2fs.c - convert char to floating point number
5968 _uint2fs.c - convert unsigned int to floating point
5970 _int2fs.c - convert int to floating point numbers
5972 _ulong2fs.c - convert unsigned long to floating point number
5974 _long2fs.c - convert long to floating point number
5982 Note if all these routines are used simultaneously the data space might
5984 For serious floating point usage it is strongly recommended that the large
5991 SDCC allows two memory models for MCS51 code, small and large.
5992 Modules compiled with different memory models should
5996 be combined together or the results would be unpredictable.
5997 The library routines supplied with the compiler are compiled as both small
5999 The compiled library modules are contained in seperate directories as small
6000 and large so that you can link to either set.
6004 When the large model is used all variables declared without a storage class
6005 will be allocated into the external ram, this includes all parameters and
6006 local variables (for non-reentrant functions).
6007 When the small model is used variables without storage class are allocated
6008 in the internal ram.
6011 Judicious usage of the processor specific storage classes and the 'reentrant'
6012 function type will yield much more efficient code, than using the large
6014 Several optimizations are disabled when the program is compiled using the
6015 large model, it is therefore strongly recommdended that the small model
6016 be used unless absolutely required.
6022 The only model supported is Flat 24.
6023 This generates code for the 24 bit contiguous addressing mode of the Dallas
6025 In this mode, up to four meg of external RAM or code space can be directly
6027 See the data sheets at www.dalsemi.com for further information on this part.
6031 In older versions of the compiler, this option was used with the MCS51 code
6037 Now, however, the '390 has it's own code generator, selected by the
6046 Note that the compiler does not generate any code to place the processor
6047 into 24 bitmode (although
6051 in the ds390 libraries will do that for you).
6056 , the boot loader or similar code must ensure that the processor is in 24
6057 bit contiguous addressing mode before calling the SDCC startup code.
6065 option, variables will by default be placed into the XDATA segment.
6070 Segments may be placed anywhere in the 4 meg address space using the usual
6072 Note that if any segments are located above 64K, the -r flag must be passed
6073 to the linker to generate the proper segment relocations, and the Intel
6074 HEX output format must be used.
6075 The -r flag can be passed to the linker by using the option
6079 on the sdcc command line.
6080 However, currently the linker can not handle code segments > 64k.
6083 Defines Created by the Compiler
6086 The compiler creates the following #defines.
6089 SDCC - this Symbol is always defined.
6092 SDCC_mcs51 or SDCC_ds390 or SDCC_z80, etc - depending on the model used
6096 __mcs51 or __ds390 or __z80, etc - depending on the model used (e.g.
6100 SDCC_STACK_AUTO - this symbol is defined when
6107 SDCC_MODEL_SMALL - when
6114 SDCC_MODEL_LARGE - when
6121 SDCC_USE_XSTACK - when
6128 SDCC_STACK_TENBIT - when
6135 SDCC_MODEL_FLAT24 - when
6148 SDCC performs a host of standard optimizations in addition to some MCU specific
6151 \layout Subsubsection
6153 Sub-expression Elimination
6156 The compiler does local and global common subexpression elimination, e.g.:
6171 will be translated to
6187 Some subexpressions are not as obvious as the above example, e.g.:
6201 In this case the address arithmetic a->b[i] will be computed only once;
6202 the equivalent code in C would be.
6218 The compiler will try to keep these temporary variables in registers.
6219 \layout Subsubsection
6221 Dead-Code Elimination
6236 i = 1; \SpecialChar ~
6241 global = 1;\SpecialChar ~
6254 global = 3;\SpecialChar ~
6269 int global; void f ()
6282 \layout Subsubsection
6343 Note: the dead stores created by this copy propagation will be eliminated
6344 by dead-code elimination.
6345 \layout Subsubsection
6350 Two types of loop optimizations are done by SDCC loop invariant lifting
6351 and strength reduction of loop induction variables.
6352 In addition to the strength reduction the optimizer marks the induction
6353 variables and the register allocator tries to keep the induction variables
6354 in registers for the duration of the loop.
6355 Because of this preference of the register allocator, loop induction optimizati
6356 on causes an increase in register pressure, which may cause unwanted spilling
6357 of other temporary variables into the stack / data space.
6358 The compiler will generate a warning message when it is forced to allocate
6359 extra space either on the stack or data space.
6360 If this extra space allocation is undesirable then induction optimization
6361 can be eliminated either for the entire source file (with ---noinduction
6362 option) or for a given function only using #pragma\SpecialChar ~
6373 for (i = 0 ; i < 100 ; i ++)
6391 for (i = 0; i < 100; i++)
6401 As mentioned previously some loop invariants are not as apparent, all static
6402 address computations are also moved out of the loop.
6406 Strength Reduction, this optimization substitutes an expression by a cheaper
6413 for (i=0;i < 100; i++)
6433 for (i=0;i< 100;i++) {
6437 ar[itemp1] = itemp2;
6453 The more expensive multiplication is changed to a less expensive addition.
6454 \layout Subsubsection
6459 This optimization is done to reduce the overhead of checking loop boundaries
6460 for every iteration.
6461 Some simple loops can be reversed and implemented using a
6462 \begin_inset Quotes eld
6465 decrement and jump if not zero
6466 \begin_inset Quotes erd
6470 SDCC checks for the following criterion to determine if a loop is reversible
6471 (note: more sophisticated compilers use data-dependency analysis to make
6472 this determination, SDCC uses a more simple minded analysis).
6475 The 'for' loop is of the form
6481 for (<symbol> = <expression> ; <sym> [< | <=] <expression> ; [<sym>++ |
6491 The <for body> does not contain
6492 \begin_inset Quotes eld
6496 \begin_inset Quotes erd
6500 \begin_inset Quotes erd
6506 All goto's are contained within the loop.
6509 No function calls within the loop.
6512 The loop control variable <sym> is not assigned any value within the loop
6515 The loop control variable does NOT participate in any arithmetic operation
6519 There are NO switch statements in the loop.
6520 \layout Subsubsection
6522 Algebraic Simplifications
6525 SDCC does numerous algebraic simplifications, the following is a small sub-set
6526 of these optimizations.
6532 i = j + 0 ; /* changed to */ i = j;
6534 i /= 2; /* changed to */ i >>= 1;
6536 i = j - j ; /* changed to */ i = 0;
6538 i = j / 1 ; /* changed to */ i = j;
6544 Note the subexpressions given above are generally introduced by macro expansions
6545 or as a result of copy/constant propagation.
6546 \layout Subsubsection
6551 SDCC changes switch statements to jump tables when the following conditions
6556 The case labels are in numerical sequence, the labels need not be in order,
6557 and the starting number need not be one or zero.
6563 switch(i) {\SpecialChar ~
6670 Both the above switch statements will be implemented using a jump-table.
6673 The number of case labels is at least three, since it takes two conditional
6674 statements to handle the boundary conditions.
6677 The number of case labels is less than 84, since each label takes 3 bytes
6678 and a jump-table can be utmost 256 bytes long.
6682 Switch statements which have gaps in the numeric sequence or those that
6683 have more that 84 case labels can be split into more than one switch statement
6684 for efficient code generation, e.g.:
6722 If the above switch statement is broken down into two switch statements
6756 case 9: \SpecialChar ~
6766 case 12:\SpecialChar ~
6776 then both the switch statements will be implemented using jump-tables whereas
6777 the unmodified switch statement will not be.
6778 \layout Subsubsection
6780 Bit-shifting Operations.
6783 Bit shifting is one of the most frequently used operation in embedded programmin
6785 SDCC tries to implement bit-shift operations in the most efficient way
6805 generates the following code:
6823 In general SDCC will never setup a loop if the shift count is known.
6863 Note that SDCC stores numbers in little-endian format (i.e.
6864 lowest order first).
6865 \layout Subsubsection
6870 A special case of the bit-shift operation is bit rotation, SDCC recognizes
6871 the following expression to be a left bit-rotation:
6882 i = ((i << 1) | (i >> 7));
6890 will generate the following code:
6906 SDCC uses pattern matching on the parse tree to determine this operation.Variatio
6907 ns of this case will also be recognized as bit-rotation, i.e.:
6913 i = ((i >> 7) | (i << 1)); /* left-bit rotation */
6914 \layout Subsubsection
6919 It is frequently required to obtain the highest order bit of an integral
6920 type (long, int, short or char types).
6921 SDCC recognizes the following expression to yield the highest order bit
6922 and generates optimized code for it, e.g.:
6943 hob = (gint >> 15) & 1;
6956 will generate the following code:
6995 000A E5*01\SpecialChar ~
7023 000C 33\SpecialChar ~
7054 000D E4\SpecialChar ~
7085 000E 13\SpecialChar ~
7116 000F F5*02\SpecialChar ~
7146 Variations of this case however will
7151 It is a standard C expression, so I heartily recommend this be the only
7152 way to get the highest order bit, (it is portable).
7153 Of course it will be recognized even if it is embedded in other expressions,
7160 xyz = gint + ((gint >> 15) & 1);
7166 will still be recognized.
7167 \layout Subsubsection
7172 The compiler uses a rule based, pattern matching and re-writing mechanism
7173 for peep-hole optimization.
7178 a peep-hole optimizer by Christopher W.
7179 Fraser (cwfraser@microsoft.com).
7180 A default set of rules are compiled into the compiler, additional rules
7181 may be added with the
7183 ---peep-file <filename>
7186 The rule language is best illustrated with examples.
7214 The above rule will change the following assembly sequence:
7244 Note: All occurrences of a
7248 (pattern variable) must denote the same string.
7249 With the above rule, the assembly sequence:
7267 will remain unmodified.
7271 Other special case optimizations may be added by the user (via
7277 some variants of the 8051 MCU allow only
7286 The following two rules will change all
7308 replace { lcall %1 } by { acall %1 }
7310 replace { ljmp %1 } by { ajmp %1 }
7318 inline-assembler code
7320 is also passed through the peep hole optimizer, thus the peephole optimizer
7321 can also be used as an assembly level macro expander.
7322 The rules themselves are MCU dependent whereas the rule language infra-structur
7323 e is MCU independent.
7324 Peephole optimization rules for other MCU can be easily programmed using
7329 The syntax for a rule is as follows:
7335 rule := replace [ restart ] '{' <assembly sequence> '
7373 <assembly sequence> '
7391 '}' [if <functionName> ] '
7399 <assembly sequence> := assembly instruction (each instruction including
7400 labels must be on a separate line).
7404 The optimizer will apply to the rules one by one from the top in the sequence
7405 of their appearance, it will terminate when all rules are exhausted.
7406 If the 'restart' option is specified, then the optimizer will start matching
7407 the rules again from the top, this option for a rule is expensive (performance)
7408 , it is intended to be used in situations where a transformation will trigger
7409 the same rule again.
7410 An example of this (not a good one, it has side effects) is the following
7437 Note that the replace pattern cannot be a blank, but can be a comment line.
7438 Without the 'restart' option only the inner most 'pop' 'push' pair would
7439 be eliminated, i.e.:
7491 the restart option the rule will be applied again to the resulting code
7492 and then all the pop-push pairs will be eliminated to yield:
7510 A conditional function can be attached to a rule.
7511 Attaching rules are somewhat more involved, let me illustrate this with
7542 The optimizer does a look-up of a function name table defined in function
7547 in the source file SDCCpeeph.c, with the name
7552 If it finds a corresponding entry the function is called.
7553 Note there can be no parameters specified for these functions, in this
7558 is crucial, since the function
7562 expects to find the label in that particular variable (the hash table containin
7563 g the variable bindings is passed as a parameter).
7564 If you want to code more such functions, take a close look at the function
7565 labelInRange and the calling mechanism in source file SDCCpeeph.c.
7566 I know this whole thing is a little kludgey, but maybe some day we will
7567 have some better means.
7568 If you are looking at this file, you will also see the default rules that
7569 are compiled into the compiler, you can add your own rules in the default
7570 set there if you get tired of specifying the ---peep-file option.
7576 SDCC supports the following #pragma directives.
7579 SAVE - this will save all current options to the SAVE/RESTORE stack.
7583 RESTORE - will restore saved options from the last save.
7584 SAVEs & RESTOREs can be nested.
7585 SDCC uses a SAVE/RESTORE stack: SAVE pushes current options to the stack,
7586 RESTORE pulls current options from the stack.
7590 NOGCSE - will stop global subexpression elimination.
7593 NOINDUCTION - will stop loop induction optimizations.
7596 NOJTBOUND - will not generate code for boundary value checking, when switch
7597 statements are turned into jump-tables.
7600 NOOVERLAY - the compiler will not overlay the parameters and local variables
7604 LESS_PEDANTIC - the compiler will not warn you anymore for obvious mistakes,
7605 you'r on your own now ;-(
7608 NOLOOPREVERSE - Will not do loop reversal optimization
7611 EXCLUDE NONE | {acc[,b[,dpl[,dph]]] - The exclude pragma disables generation
7612 of pair of push/pop instruction in ISR function (using interrupt keyword).
7613 The directive should be placed immediately before the ISR function definition
7614 and it affects ALL ISR functions following it.
7615 To enable the normal register saving for ISR functions use #pragma\SpecialChar ~
7616 EXCLUDE\SpecialChar ~
7620 NOIV - Do not generate interrupt vector table entries for all ISR functions
7621 defined after the pragma.
7622 This is useful in cases where the interrupt vector table must be defined
7623 manually, or when there is a secondary, manually defined interrupt vector
7625 for the autovector feature of the Cypress EZ-USB FX2).
7628 CALLEE-SAVES function1[,function2[,function3...]] - The compiler by default
7629 uses a caller saves convention for register saving across function calls,
7630 however this can cause unneccessary register pushing & popping when calling
7631 small functions from larger functions.
7632 This option can be used to switch off the register saving convention for
7633 the function names specified.
7634 The compiler will not save registers when calling these functions, extra
7635 code need to be manually inserted at the entry & exit for these functions
7636 to save & restore the registers used by these functions, this can SUBSTANTIALLY
7637 reduce code & improve run time performance of the generated code.
7638 In the future the compiler (with interprocedural analysis) may be able
7639 to determine the appropriate scheme to use for each function call.
7640 If ---callee-saves command line option is used, the function names specified
7641 in #pragma\SpecialChar ~
7642 CALLEE-SAVES is appended to the list of functions specified in
7646 The pragma's are intended to be used to turn-off certain optimizations which
7647 might cause the compiler to generate extra stack / data space to store
7648 compiler generated temporary variables.
7649 This usually happens in large functions.
7650 Pragma directives should be used as shown in the following example, they
7651 are used to control options & optimizations for a given function; pragmas
7652 should be placed before and/or after a function, placing pragma's inside
7653 a function body could have unpredictable results.
7659 #pragma SAVE /* save the current settings */
7661 #pragma NOGCSE /* turnoff global subexpression elimination */
7663 #pragma NOINDUCTION /* turn off induction optimizations */
7685 #pragma RESTORE /* turn the optimizations back on */
7691 The compiler will generate a warning message when extra space is allocated.
7692 It is strongly recommended that the SAVE and RESTORE pragma's be used when
7693 changing options for a function.
7698 <pending: this is messy and incomplete>
7703 Compiler support routines (_gptrget, _mulint etc)
7706 Stdclib functions (puts, printf, strcat etc)
7709 Math functions (sin, pow, sqrt etc)
7712 Interfacing with Assembly Routines
7713 \layout Subsubsection
7715 Global Registers used for Parameter Passing
7718 The compiler always uses the global registers
7726 to pass the first parameter to a routine.
7727 The second parameter onwards is either allocated on the stack (for reentrant
7728 routines or if ---stack-auto is used) or in the internal / external ram
7729 (depending on the memory model).
7731 \layout Subsubsection
7733 Assembler Routine(non-reentrant)
7736 In the following example the function cfunc calls an assembler routine asm_func,
7737 which takes two parameters.
7743 extern int asm_func(unsigned char, unsigned char);
7747 int c_func (unsigned char i, unsigned char j)
7755 return asm_func(i,j);
7769 return c_func(10,9);
7777 The corresponding assembler function is:
7783 .globl _asm_func_PARM_2
7847 add a,_asm_func_PARM_2
7883 Note here that the return values are placed in 'dpl' - One byte return value,
7884 'dpl' LSB & 'dph' MSB for two byte values.
7885 'dpl', 'dph' and 'b' for three byte values (generic pointers) and 'dpl','dph','
7886 b' & 'acc' for four byte values.
7889 The parameter naming convention is _<function_name>_PARM_<n>, where n is
7890 the parameter number starting from 1, and counting from the left.
7891 The first parameter is passed in
7892 \begin_inset Quotes eld
7896 \begin_inset Quotes erd
7899 for One bye parameter,
7900 \begin_inset Quotes eld
7904 \begin_inset Quotes erd
7908 \begin_inset Quotes eld
7912 \begin_inset Quotes erd
7916 \begin_inset Quotes eld
7920 \begin_inset Quotes erd
7923 for four bytes, the varible name for the second parameter will be _<function_na
7928 Assemble the assembler routine with the following command:
7935 asx8051 -losg asmfunc.asm
7942 Then compile and link the assembler routine to the C source file with the
7950 sdcc cfunc.c asmfunc.rel
7951 \layout Subsubsection
7953 Assembler Routine(reentrant)
7956 In this case the second parameter onwards will be passed on the stack, the
7957 parameters are pushed from right to left i.e.
7958 after the call the left most parameter will be on the top of the stack.
7965 extern int asm_func(unsigned char, unsigned char);
7969 int c_func (unsigned char i, unsigned char j) reentrant
7977 return asm_func(i,j);
7991 return c_func(10,9);
7999 The corresponding assembler routine is:
8109 The compiling and linking procedure remains the same, however note the extra
8110 entry & exit linkage required for the assembler code, _bp is the stack
8111 frame pointer and is used to compute the offset into the stack for parameters
8112 and local variables.
8118 The external stack is located at the start of the external ram segment,
8119 and is 256 bytes in size.
8120 When ---xstack option is used to compile the program, the parameters and
8121 local variables of all reentrant functions are allocated in this area.
8122 This option is provided for programs with large stack space requirements.
8123 When used with the ---stack-auto option, all parameters and local variables
8124 are allocated on the external stack (note support libraries will need to
8125 be recompiled with the same options).
8128 The compiler outputs the higher order address byte of the external ram segment
8129 into PORT P2, therefore when using the External Stack option, this port
8130 MAY NOT be used by the application program.
8136 Deviations from the compliancy.
8139 functions are not always reentrant.
8142 structures cannot be assigned values directly, cannot be passed as function
8143 parameters or assigned to each other and cannot be a return value from
8170 s1 = s2 ; /* is invalid in SDCC although allowed in ANSI */
8181 struct s foo1 (struct s parms) /* is invalid in SDCC although allowed in
8203 return rets;/* is invalid in SDCC although allowed in ANSI */
8208 'long long' (64 bit integers) not supported.
8211 'double' precision floating point not supported.
8214 No support for setjmp and longjmp (for now).
8217 Old K&R style function declarations are NOT allowed.
8223 foo(i,j) /* this old style of function declarations */
8225 int i,j; /* are valid in ANSI but not valid in SDCC */
8239 functions declared as pointers must be dereferenced during the call.
8250 /* has to be called like this */
8252 (*foo)(); /* ansi standard allows calls to be made like 'foo()' */
8255 Cyclomatic Complexity
8258 Cyclomatic complexity of a function is defined as the number of independent
8259 paths the program can take during execution of the function.
8260 This is an important number since it defines the number test cases you
8261 have to generate to validate the function.
8262 The accepted industry standard for complexity number is 10, if the cyclomatic
8263 complexity reported by SDCC exceeds 10 you should think about simplification
8264 of the function logic.
8265 Note that the complexity level is not related to the number of lines of
8267 Large functions can have low complexity, and small functions can have large
8273 SDCC uses the following formula to compute the complexity:
8278 complexity = (number of edges in control flow graph) - (number of nodes
8279 in control flow graph) + 2;
8283 Having said that the industry standard is 10, you should be aware that in
8284 some cases it be may unavoidable to have a complexity level of less than
8286 For example if you have switch statement with more than 10 case labels,
8287 each case label adds one to the complexity level.
8288 The complexity level is by no means an absolute measure of the algorithmic
8289 complexity of the function, it does however provide a good starting point
8290 for which functions you might look at for further optimization.
8296 Here are a few guidelines that will help the compiler generate more efficient
8297 code, some of the tips are specific to this compiler others are generally
8298 good programming practice.
8301 Use the smallest data type to represent your data-value.
8302 If it is known in advance that the value is going to be less than 256 then
8303 use an 'unsigned char' instead of a 'short' or 'int'.
8306 Use unsigned when it is known in advance that the value is not going to
8308 This helps especially if you are doing division or multiplication.
8311 NEVER jump into a LOOP.
8314 Declare the variables to be local whenever possible, especially loop control
8315 variables (induction).
8318 Since the compiler does not always do implicit integral promotion, the programme
8319 r should do an explicit cast when integral promotion is required.
8322 Reducing the size of division, multiplication & modulus operations can reduce
8323 code size substantially.
8324 Take the following code for example.
8330 foobar(unsigned int p1, unsigned char ch)
8334 unsigned char ch1 = p1 % ch ;
8345 For the modulus operation the variable ch will be promoted to unsigned int
8346 first then the modulus operation will be performed (this will lead to a
8347 call to support routine _moduint()), and the result will be casted to a
8349 If the code is changed to
8355 foobar(unsigned int p1, unsigned char ch)
8359 unsigned char ch1 = (unsigned char)p1 % ch ;
8370 It would substantially reduce the code generated (future versions of the
8371 compiler will be smart enough to detect such optimization oppurtunities).
8374 Notes on MCS51 memory layout
8377 The 8051 family of micro controller have a minimum of 128 bytes of internal
8378 memory which is structured as follows
8382 - Bytes 00-1F - 32 bytes to hold up to 4 banks of the registers R7 to R7
8385 - Bytes 20-2F - 16 bytes to hold 128 bit variables and
8387 - Bytes 30-7F - 60 bytes for general purpose use.
8391 Normally the SDCC compiler will only utilise the first bank of registers,
8392 but it is possible to specify that other banks of registers should be used
8393 in interrupt routines.
8394 By default, the compiler will place the stack after the last bank of used
8396 if the first 2 banks of registers are used, it will position the base of
8397 the internal stack at address 16 (0X10).
8398 This implies that as the stack grows, it will use up the remaining register
8399 banks, and the 16 bytes used by the 128 bit variables, and 60 bytes for
8400 general purpose use.
8403 By default, the compiler uses the 60 general purpose bytes to hold "near
8405 The compiler/optimiser may also declare some Local Variables in this area
8410 If any of the 128 bit variables are used, or near data is being used then
8411 care needs to be taken to ensure that the stack does not grow so much that
8412 it starts to over write either your bit variables or "near data".
8413 There is no runtime checking to prevent this from happening.
8416 The amount of stack being used is affected by the use of the "internal stack"
8417 to save registers before a subroutine call is made (---stack-auto will
8418 declare parameters and local variables on the stack) and the number of
8422 If you detect that the stack is over writing you data, then the following
8424 ---xstack will cause an external stack to be used for saving registers
8425 and (if ---stack-auto is being used) storing parameters and local variables.
8426 However this will produce more code which will be slower to execute.
8430 ---stack-loc will allow you specify the start of the stack, i.e.
8431 you could start it after any data in the general purpose area.
8432 However this may waste the memory not used by the register banks and if
8433 the size of the "near data" increases, it may creep into the bottom of
8437 ---stack-after-data, similar to the ---stack-loc, but it automatically places
8438 the stack after the end of the "near data".
8439 Again this could waste any spare register space.
8442 ---data-loc allows you to specify the start address of the near data.
8443 This could be used to move the "near data" further away from the stack
8444 giving it more room to grow.
8445 This will only work if no bit variables are being used and the stack can
8446 grow to use the bit variable space.
8454 If you find that the stack is over writing your bit variables or "near data"
8455 then the approach which best utilised the internal memory is to position
8456 the "near data" after the last bank of used registers or, if you use bit
8457 variables, after the last bit variable by using the ---data-loc, e.g.
8458 if two register banks are being used and no bit variables, ---data-loc
8459 16, and use the ---stack-after-data option.
8462 If bit variables are being used, another method would be to try and squeeze
8463 the data area in the unused register banks if it will fit, and start the
8464 stack after the last bit variable.
8467 Retargetting for other MCUs.
8470 The issues for retargetting the compiler are far too numerous to be covered
8472 What follows is a brief description of each of the seven phases of the
8473 compiler and its MCU dependency.
8476 Parsing the source and building the annotated parse tree.
8477 This phase is largely MCU independent (except for the language extensions).
8478 Syntax & semantic checks are also done in this phase, along with some initial
8479 optimizations like back patching labels and the pattern matching optimizations
8480 like bit-rotation etc.
8483 The second phase involves generating an intermediate code which can be easy
8484 manipulated during the later phases.
8485 This phase is entirely MCU independent.
8486 The intermediate code generation assumes the target machine has unlimited
8487 number of registers, and designates them with the name iTemp.
8488 The compiler can be made to dump a human readable form of the code generated
8489 by using the ---dumpraw option.
8492 This phase does the bulk of the standard optimizations and is also MCU independe
8494 This phase can be broken down into several sub-phases:
8498 Break down intermediate code (iCode) into basic blocks.
8500 Do control flow & data flow analysis on the basic blocks.
8502 Do local common subexpression elimination, then global subexpression elimination
8504 Dead code elimination
8508 If loop optimizations caused any changes then do 'global subexpression eliminati
8509 on' and 'dead code elimination' again.
8512 This phase determines the live-ranges; by live range I mean those iTemp
8513 variables defined by the compiler that still survive after all the optimization
8515 Live range analysis is essential for register allocation, since these computati
8516 on determines which of these iTemps will be assigned to registers, and for
8520 Phase five is register allocation.
8521 There are two parts to this process.
8525 The first part I call 'register packing' (for lack of a better term).
8526 In this case several MCU specific expression folding is done to reduce
8531 The second part is more MCU independent and deals with allocating registers
8532 to the remaining live ranges.
8533 A lot of MCU specific code does creep into this phase because of the limited
8534 number of index registers available in the 8051.
8537 The Code generation phase is (unhappily), entirely MCU dependent and very
8538 little (if any at all) of this code can be reused for other MCU.
8539 However the scheme for allocating a homogenized assembler operand for each
8540 iCode operand may be reused.
8543 As mentioned in the optimization section the peep-hole optimizer is rule
8544 based system, which can reprogrammed for other MCUs.
8547 SDCDB - Source Level Debugger
8550 SDCC is distributed with a source level debugger.
8551 The debugger uses a command line interface, the command repertoire of the
8552 debugger has been kept as close to gdb (the GNU debugger) as possible.
8553 The configuration and build process is part of the standard compiler installati
8554 on, which also builds and installs the debugger in the target directory
8555 specified during configuration.
8556 The debugger allows you debug BOTH at the C source and at the ASM source
8560 Compiling for Debugging
8565 debug option must be specified for all files for which debug information
8567 The complier generates a .cdb file for each of these files.
8568 The linker updates the .cdb file with the address information.
8569 This .cdb is used by the debugger.
8572 How the Debugger Works
8575 When the ---debug option is specified the compiler generates extra symbol
8576 information some of which are put into the the assembler source and some
8577 are put into the .cdb file, the linker updates the .cdb file with the address
8578 information for the symbols.
8579 The debugger reads the symbolic information generated by the compiler &
8580 the address information generated by the linker.
8581 It uses the SIMULATOR (Daniel's S51) to execute the program, the program
8582 execution is controlled by the debugger.
8583 When a command is issued for the debugger, it translates it into appropriate
8584 commands for the simulator.
8587 Starting the Debugger
8590 The debugger can be started using the following command line.
8591 (Assume the file you are debugging has the file name foo).
8605 The debugger will look for the following files.
8608 foo.c - the source file.
8611 foo.cdb - the debugger symbol information file.
8614 foo.ihx - the intel hex format object file.
8617 Command Line Options.
8620 ---directory=<source file directory> this option can used to specify the
8621 directory search list.
8622 The debugger will look into the directory list specified for source, cdb
8624 The items in the directory list must be separated by ':', e.g.
8625 if the source files can be in the directories /home/src1 and /home/src2,
8626 the ---directory option should be ---directory=/home/src1:/home/src2.
8627 Note there can be no spaces in the option.
8631 -cd <directory> - change to the <directory>.
8634 -fullname - used by GUI front ends.
8637 -cpu <cpu-type> - this argument is passed to the simulator please see the
8638 simulator docs for details.
8641 -X <Clock frequency > this options is passed to the simulator please see
8642 the simulator docs for details.
8645 -s <serial port file> passed to simulator see the simulator docs for details.
8648 -S <serial in,out> passed to simulator see the simulator docs for details.
8654 As mention earlier the command interface for the debugger has been deliberately
8655 kept as close the GNU debugger gdb, as possible.
8656 This will help the integration with existing graphical user interfaces
8657 (like ddd, xxgdb or xemacs) existing for the GNU debugger.
8658 \layout Subsubsection
8660 break [line | file:line | function | file:function]
8663 Set breakpoint at specified line or function:
8672 sdcdb>break foo.c:100
8676 sdcdb>break foo.c:funcfoo
8677 \layout Subsubsection
8679 clear [line | file:line | function | file:function ]
8682 Clear breakpoint at specified line or function:
8691 sdcdb>clear foo.c:100
8695 sdcdb>clear foo.c:funcfoo
8696 \layout Subsubsection
8701 Continue program being debugged, after breakpoint.
8702 \layout Subsubsection
8707 Execute till the end of the current function.
8708 \layout Subsubsection
8713 Delete breakpoint number 'n'.
8714 If used without any option clear ALL user defined break points.
8715 \layout Subsubsection
8717 info [break | stack | frame | registers ]
8720 info break - list all breakpoints
8723 info stack - show the function call stack.
8726 info frame - show information about the current execution frame.
8729 info registers - show content of all registers.
8730 \layout Subsubsection
8735 Step program until it reaches a different source line.
8736 \layout Subsubsection
8741 Step program, proceeding through subroutine calls.
8742 \layout Subsubsection
8747 Start debugged program.
8748 \layout Subsubsection
8753 Print type information of the variable.
8754 \layout Subsubsection
8759 print value of variable.
8760 \layout Subsubsection
8765 load the given file name.
8766 Note this is an alternate method of loading file for debugging.
8767 \layout Subsubsection
8772 print information about current frame.
8773 \layout Subsubsection
8778 Toggle between C source & assembly source.
8779 \layout Subsubsection
8784 Send the string following '!' to the simulator, the simulator response is
8786 Note the debugger does not interpret the command being sent to the simulator,
8787 so if a command like 'go' is sent the debugger can loose its execution
8788 context and may display incorrect values.
8789 \layout Subsubsection
8796 My name is Bobby Brown"
8799 Interfacing with XEmacs.
8802 Two files (in emacs lisp) are provided for the interfacing with XEmacs,
8803 sdcdb.el and sdcdbsrc.el.
8804 These two files can be found in the $(prefix)/bin directory after the installat
8806 These files need to be loaded into XEmacs for the interface to work.
8807 This can be done at XEmacs startup time by inserting the following into
8808 your '.xemacs' file (which can be found in your HOME directory):
8814 (load-file sdcdbsrc.el)
8820 .xemacs is a lisp file so the () around the command is REQUIRED.
8821 The files can also be loaded dynamically while XEmacs is running, set the
8822 environment variable 'EMACSLOADPATH' to the installation bin directory
8823 (<installdir>/bin), then enter the following command ESC-x load-file sdcdbsrc.
8824 To start the interface enter the following command:
8838 You will prompted to enter the file name to be debugged.
8843 The command line options that are passed to the simulator directly are bound
8844 to default values in the file sdcdbsrc.el.
8845 The variables are listed below, these values maybe changed as required.
8848 sdcdbsrc-cpu-type '51
8851 sdcdbsrc-frequency '11059200
8857 The following is a list of key mapping for the debugger interface.
8865 ;; Current Listing ::
8882 binding\SpecialChar ~
8921 ------\SpecialChar ~
8961 sdcdb-next-from-src\SpecialChar ~
8987 sdcdb-back-from-src\SpecialChar ~
9013 sdcdb-cont-from-src\SpecialChar ~
9023 SDCDB continue command
9039 sdcdb-step-from-src\SpecialChar ~
9065 sdcdb-whatis-c-sexp\SpecialChar ~
9075 SDCDB ptypecommand for data at
9139 sdcdbsrc-delete\SpecialChar ~
9153 SDCDB Delete all breakpoints if no arg
9201 given or delete arg (C-u arg x)
9217 sdcdbsrc-frame\SpecialChar ~
9232 SDCDB Display current frame if no arg,
9281 given or display frame arg
9346 sdcdbsrc-goto-sdcdb\SpecialChar ~
9356 Goto the SDCDB output buffer
9372 sdcdb-print-c-sexp\SpecialChar ~
9383 SDCDB print command for data at
9447 sdcdbsrc-goto-sdcdb\SpecialChar ~
9457 Goto the SDCDB output buffer
9473 sdcdbsrc-mode\SpecialChar ~
9489 Toggles Sdcdbsrc mode (turns it off)
9493 ;; C-c C-f\SpecialChar ~
9501 sdcdb-finish-from-src\SpecialChar ~
9509 SDCDB finish command
9513 ;; C-x SPC\SpecialChar ~
9521 sdcdb-break\SpecialChar ~
9539 Set break for line with point
9541 ;; ESC t\SpecialChar ~
9551 sdcdbsrc-mode\SpecialChar ~
9567 Toggle Sdcdbsrc mode
9569 ;; ESC m\SpecialChar ~
9579 sdcdbsrc-srcmode\SpecialChar ~
9603 The Z80 and gbz80 port
9606 SDCC can target both the Zilog Z80 and the Nintendo Gameboy's Z80-like gbz80.
9607 The port is incomplete - long support is incomplete (mul, div and mod are
9608 unimplimented), and both float and bitfield support is missing.
9609 Apart from that the code generated is correct.
9612 As always, the code is the authoritave reference - see z80/ralloc.c and z80/gen.c.
9613 The stack frame is similar to that generated by the IAR Z80 compiler.
9614 IX is used as the base pointer, HL is used as a temporary register, and
9615 BC and DE are available for holding varibles.
9616 IY is currently unusued.
9617 Return values are stored in HL.
9618 One bad side effect of using IX as the base pointer is that a functions
9619 stack frame is limited to 127 bytes - this will be fixed in a later version.
9625 SDCC has grown to be a large project.
9626 The compiler alone (without the preprocessor, assembler and linker) is
9627 about 40,000 lines of code (blank stripped).
9628 The open source nature of this project is a key to its continued growth
9630 You gain the benefit and support of many active software developers and
9632 Is SDCC perfect? No, that's why we need your help.
9633 The developers take pride in fixing reported bugs.
9634 You can help by reporting the bugs and helping other SDCC users.
9635 There are lots of ways to contribute, and we encourage you to take part
9636 in making SDCC a great software package.
9642 Send an email to the mailing list at 'user-sdcc@sdcc.sourceforge.net' or 'devel-sd
9643 cc@sdcc.sourceforge.net'.
9644 Bugs will be fixed ASAP.
9645 When reporting a bug, it is very useful to include a small test program
9646 which reproduces the problem.
9647 If you can isolate the problem by looking at the generated assembly code,
9648 this can be very helpful.
9649 Compiling your program with the ---dumpall option can sometimes be useful
9650 in locating optimization problems.
9656 The anatomy of the compiler
9661 This is an excerpt from an atricle published in Circuit Cellar MagaZine
9663 It's a little outdated (the compiler is much more efficient now and user/devell
9664 oper friendly), but pretty well exposes the guts of it all.
9670 The current version of SDCC can generate code for Intel 8051 and Z80 MCU.
9671 It is fairly easy to retarget for other 8-bit MCU.
9672 Here we take a look at some of the internals of the compiler.
9679 Parsing the input source file and creating an AST (Annotated Syntax Tree).
9680 This phase also involves propagating types (annotating each node of the
9681 parse tree with type information) and semantic analysis.
9682 There are some MCU specific parsing rules.
9683 For example the storage classes, the extended storage classes are MCU specific
9684 while there may be a xdata storage class for 8051 there is no such storage
9685 class for z80 or Atmel AVR.
9686 SDCC allows MCU specific storage class extensions, i.e.
9687 xdata will be treated as a storage class specifier when parsing 8051 C
9688 code but will be treated as a C identifier when parsing z80 or ATMEL AVR
9695 Intermediate code generation.
9696 In this phase the AST is broken down into three-operand form (iCode).
9697 These three operand forms are represented as doubly linked lists.
9698 ICode is the term given to the intermediate form generated by the compiler.
9699 ICode example section shows some examples of iCode generated for some simple
9706 Bulk of the target independent optimizations is performed in this phase.
9707 The optimizations include constant propagation, common sub-expression eliminati
9708 on, loop invariant code movement, strength reduction of loop induction variables
9709 and dead-code elimination.
9715 During intermediate code generation phase, the compiler assumes the target
9716 machine has infinite number of registers and generates a lot of temporary
9718 The live range computation determines the lifetime of each of these compiler-ge
9719 nerated temporaries.
9720 A picture speaks a thousand words.
9721 ICode example sections show the live range annotations for each of the
9723 It is important to note here, each iCode is assigned a number in the order
9724 of its execution in the function.
9725 The live ranges are computed in terms of these numbers.
9726 The from number is the number of the iCode which first defines the operand
9727 and the to number signifies the iCode which uses this operand last.
9733 The register allocation determines the type and number of registers needed
9735 In most MCUs only a few registers can be used for indirect addressing.
9736 In case of 8051 for example the registers R0 & R1 can be used to indirectly
9737 address the internal ram and DPTR to indirectly address the external ram.
9738 The compiler will try to allocate the appropriate register to pointer variables
9740 ICode example section shows the operands annotated with the registers assigned
9742 The compiler will try to keep operands in registers as much as possible;
9743 there are several schemes the compiler uses to do achieve this.
9744 When the compiler runs out of registers the compiler will check to see
9745 if there are any live operands which is not used or defined in the current
9746 basic block being processed, if there are any found then it will push that
9747 operand and use the registers in this block, the operand will then be popped
9748 at the end of the basic block.
9752 There are other MCU specific considerations in this phase.
9753 Some MCUs have an accumulator; very short-lived operands could be assigned
9754 to the accumulator instead of general-purpose register.
9760 Figure II gives a table of iCode operations supported by the compiler.
9761 The code generation involves translating these operations into corresponding
9762 assembly code for the processor.
9763 This sounds overly simple but that is the essence of code generation.
9764 Some of the iCode operations are generated on a MCU specific manner for
9765 example, the z80 port does not use registers to pass parameters so the
9766 SEND and RECV iCode operations will not be generated, and it also does
9767 not support JUMPTABLES.
9774 <Where is Figure II ?>
9780 This section shows some details of iCode.
9781 The example C code does not do anything useful; it is used as an example
9782 to illustrate the intermediate code generated by the compiler.
9795 /* This function does nothing useful.
9802 for the purpose of explaining iCode */
9805 short function (data int *x)
9813 short i=10; /* dead initialization eliminated */
9818 short sum=10; /* dead initialization eliminated */
9831 while (*x) *x++ = *p++;
9845 /* compiler detects i,j to be induction variables */
9849 for (i = 0, j = 10 ; i < 10 ; i++, j---) {
9861 mul += i * 3; /* this multiplication remains */
9867 gint += j * 3;/* this multiplication changed to addition */
9884 In addition to the operands each iCode contains information about the filename
9885 and line it corresponds to in the source file.
9886 The first field in the listing should be interpreted as follows:
9891 Filename(linenumber: iCode Execution sequence number : ICode hash table
9892 key : loop depth of the iCode).
9897 Then follows the human readable form of the ICode operation.
9898 Each operand of this triplet form can be of three basic types a) compiler
9899 generated temporary b) user defined variable c) a constant value.
9900 Note that local variables and parameters are replaced by compiler generated
9902 Live ranges are computed only for temporaries (i.e.
9903 live ranges are not computed for global variables).
9904 Registers are allocated for temporaries only.
9905 Operands are formatted in the following manner:
9910 Operand Name [lr live-from : live-to ] { type information } [ registers
9916 As mentioned earlier the live ranges are computed in terms of the execution
9917 sequence number of the iCodes, for example
9919 the iTemp0 is live from (i.e.
9920 first defined in iCode with execution sequence number 3, and is last used
9921 in the iCode with sequence number 5).
9922 For induction variables such as iTemp21 the live range computation extends
9923 the lifetime from the start to the end of the loop.
9925 The register allocator used the live range information to allocate registers,
9926 the same registers may be used for different temporaries if their live
9927 ranges do not overlap, for example r0 is allocated to both iTemp6 and to
9928 iTemp17 since their live ranges do not overlap.
9929 In addition the allocator also takes into consideration the type and usage
9930 of a temporary, for example itemp6 is a pointer to near space and is used
9931 as to fetch data from (i.e.
9932 used in GET_VALUE_AT_ADDRESS) so it is allocated a pointer registers (r0).
9933 Some short lived temporaries are allocated to special registers which have
9934 meaning to the code generator e.g.
9935 iTemp13 is allocated to a pseudo register CC which tells the back end that
9936 the temporary is used only for a conditional jump the code generation makes
9937 use of this information to optimize a compare and jump ICode.
9939 There are several loop optimizations performed by the compiler.
9940 It can detect induction variables iTemp21(i) and iTemp23(j).
9941 Also note the compiler does selective strength reduction, i.e.
9942 the multiplication of an induction variable in line 18 (gint = j * 3) is
9943 changed to addition, a new temporary iTemp17 is allocated and assigned
9944 a initial value, a constant 3 is then added for each iteration of the loop.
9945 The compiler does not change the multiplication in line 17 however since
9946 the processor does support an 8 * 8 bit multiplication.
9948 Note the dead code elimination optimization eliminated the dead assignments
9949 in line 7 & 8 to I and sum respectively.
9956 Sample.c (5:1:0:0) _entry($9) :
9961 Sample.c(5:2:1:0) proc _function [lr0:0]{function short}
9966 Sample.c(11:3:2:0) iTemp0 [lr3:5]{_near * int}[r2] = recv
9971 Sample.c(11:4:53:0) preHeaderLbl0($11) :
9976 Sample.c(11:5:55:0) iTemp6 [lr5:16]{_near * int}[r0] := iTemp0 [lr3:5]{_near
9982 Sample.c(11:6:5:1) _whilecontinue_0($1) :
9987 Sample.c(11:7:7:1) iTemp4 [lr7:8]{int}[r2 r3] = @[iTemp6 [lr5:16]{_near *
9993 Sample.c(11:8:8:1) if iTemp4 [lr7:8]{int}[r2 r3] == 0 goto _whilebreak_0($3)
9998 Sample.c(11:9:14:1) iTemp7 [lr9:13]{_far * int}[DPTR] := _p [lr0:0]{_far
10004 Sample.c(11:10:15:1) _p [lr0:0]{_far * int} = _p [lr0:0]{_far * int} + 0x2
10010 Sample.c(11:13:18:1) iTemp10 [lr13:14]{int}[r2 r3] = @[iTemp7 [lr9:13]{_far
10016 Sample.c(11:14:19:1) *(iTemp6 [lr5:16]{_near * int}[r0]) := iTemp10 [lr13:14]{int
10022 Sample.c(11:15:12:1) iTemp6 [lr5:16]{_near * int}[r0] = iTemp6 [lr5:16]{_near
10023 * int}[r0] + 0x2 {short}
10028 Sample.c(11:16:20:1) goto _whilecontinue_0($1)
10033 Sample.c(11:17:21:0)_whilebreak_0($3) :
10038 Sample.c(12:18:22:0) iTemp2 [lr18:40]{short}[r2] := 0x0 {short}
10043 Sample.c(13:19:23:0) iTemp11 [lr19:40]{short}[r3] := 0x0 {short}
10048 Sample.c(15:20:54:0)preHeaderLbl1($13) :
10053 Sample.c(15:21:56:0) iTemp21 [lr21:38]{short}[r4] := 0x0 {short}
10058 Sample.c(15:22:57:0) iTemp23 [lr22:38]{int}[r5 r6] := 0xa {int}
10063 Sample.c(15:23:58:0) iTemp17 [lr23:38]{int}[r7 r0] := 0x1e {int}
10068 Sample.c(15:24:26:1)_forcond_0($4) :
10073 Sample.c(15:25:27:1) iTemp13 [lr25:26]{char}[CC] = iTemp21 [lr21:38]{short}[r4]
10079 Sample.c(15:26:28:1) if iTemp13 [lr25:26]{char}[CC] == 0 goto _forbreak_0($7)
10084 Sample.c(16:27:31:1) iTemp2 [lr18:40]{short}[r2] = iTemp2 [lr18:40]{short}[r2]
10085 + ITemp21 [lr21:38]{short}[r4]
10090 Sample.c(17:29:33:1) iTemp15 [lr29:30]{short}[r1] = iTemp21 [lr21:38]{short}[r4]
10096 Sample.c(17:30:34:1) iTemp11 [lr19:40]{short}[r3] = iTemp11 [lr19:40]{short}[r3]
10097 + iTemp15 [lr29:30]{short}[r1]
10102 Sample.c(18:32:36:1:1) iTemp17 [lr23:38]{int}[r7 r0]= iTemp17 [lr23:38]{int}[r7
10108 Sample.c(18:33:37:1) _gint [lr0:0]{int} = _gint [lr0:0]{int} + iTemp17 [lr23:38]{
10114 Sample.c(15:36:42:1) iTemp21 [lr21:38]{short}[r4] = iTemp21 [lr21:38]{short}[r4]
10120 Sample.c(15:37:45:1) iTemp23 [lr22:38]{int}[r5 r6]= iTemp23 [lr22:38]{int}[r5
10126 Sample.c(19:38:47:1) goto _forcond_0($4)
10131 Sample.c(19:39:48:0)_forbreak_0($7) :
10136 Sample.c(20:40:49:0) iTemp24 [lr40:41]{short}[DPTR] = iTemp2 [lr18:40]{short}[r2]
10137 + ITemp11 [lr19:40]{short}[r3]
10142 Sample.c(20:41:50:0) ret iTemp24 [lr40:41]{short}
10147 Sample.c(20:42:51:0)_return($8) :
10152 Sample.c(20:43:52:0) eproc _function [lr0:0]{ ia0 re0 rm0}{function short}
10158 Finally the code generated for this function:
10199 ; ----------------------------------------------
10204 ; function function
10209 ; ----------------------------------------------
10219 ; iTemp0 [lr3:5]{_near * int}[r2] = recv
10231 ; iTemp6 [lr5:16]{_near * int}[r0] := iTemp0 [lr3:5]{_near * int}[r2]
10243 ;_whilecontinue_0($1) :
10253 ; iTemp4 [lr7:8]{int}[r2 r3] = @[iTemp6 [lr5:16]{_near * int}[r0]]
10258 ; if iTemp4 [lr7:8]{int}[r2 r3] == 0 goto _whilebreak_0($3)
10317 ; iTemp7 [lr9:13]{_far * int}[DPTR] := _p [lr0:0]{_far * int}
10336 ; _p [lr0:0]{_far * int} = _p [lr0:0]{_far * int} + 0x2 {short}
10383 ; iTemp10 [lr13:14]{int}[r2 r3] = @[iTemp7 [lr9:13]{_far * int}[DPTR]]
10423 ; *(iTemp6 [lr5:16]{_near * int}[r0]) := iTemp10 [lr13:14]{int}[r2 r3]
10449 ; iTemp6 [lr5:16]{_near * int}[r0] =
10454 ; iTemp6 [lr5:16]{_near * int}[r0] +
10471 ; goto _whilecontinue_0($1)
10483 ; _whilebreak_0($3) :
10493 ; iTemp2 [lr18:40]{short}[r2] := 0x0 {short}
10505 ; iTemp11 [lr19:40]{short}[r3] := 0x0 {short}
10517 ; iTemp21 [lr21:38]{short}[r4] := 0x0 {short}
10529 ; iTemp23 [lr22:38]{int}[r5 r6] := 0xa {int}
10548 ; iTemp17 [lr23:38]{int}[r7 r0] := 0x1e {int}
10577 ; iTemp13 [lr25:26]{char}[CC] = iTemp21 [lr21:38]{short}[r4] < 0xa {short}
10582 ; if iTemp13 [lr25:26]{char}[CC] == 0 goto _forbreak_0($7)
10627 ; iTemp2 [lr18:40]{short}[r2] = iTemp2 [lr18:40]{short}[r2] +
10632 ; iTemp21 [lr21:38]{short}[r4]
10658 ; iTemp15 [lr29:30]{short}[r1] = iTemp21 [lr21:38]{short}[r4] * 0x3 {short}
10691 ; iTemp11 [lr19:40]{short}[r3] = iTemp11 [lr19:40]{short}[r3] +
10696 ; iTemp15 [lr29:30]{short}[r1]
10715 ; iTemp17 [lr23:38]{int}[r7 r0]= iTemp17 [lr23:38]{int}[r7 r0]- 0x3 {short}
10762 ; _gint [lr0:0]{int} = _gint [lr0:0]{int} + iTemp17 [lr23:38]{int}[r7 r0]
10809 ; iTemp21 [lr21:38]{short}[r4] = iTemp21 [lr21:38]{short}[r4] + 0x1 {short}
10821 ; iTemp23 [lr22:38]{int}[r5 r6]= iTemp23 [lr22:38]{int}[r5 r6]- 0x1 {short}
10835 cjne r5,#0xff,00104$
10847 ; goto _forcond_0($4)
10859 ; _forbreak_0($7) :
10869 ; ret iTemp24 [lr40:41]{short}
10912 A few words about basic block successors, predecessors and dominators
10915 Successors are basic blocks that might execute after this basic block.
10917 Predecessors are basic blocks that might execute before reaching this basic
10920 Dominators are basic blocks that WILL execute before reaching this basic
10946 a) succList of [BB2] = [BB4], of [BB3] = [BB4], of [BB1] = [BB2,BB3]
10949 b) predList of [BB2] = [BB1], of [BB3] = [BB1], of [BB4] = [BB2,BB3]
10952 c) domVect of [BB4] = BB1 ...
10953 here we are not sure if BB2 or BB3 was executed but we are SURE that BB1
10961 \begin_inset LatexCommand \url{http://sdcc.sourceforge.net#Who}
10971 Thanks to all the other volunteer developers who have helped with coding,
10972 testing, web-page creation, distribution sets, etc.
10973 You know who you are :-)
10980 This document was initially written by Sandeep Dutta
10983 All product names mentioned herein may be trademarks of their respective
10989 \begin_inset LatexCommand \printindex{}