HACKS revision 1.24
11.24Sragge# $NetBSD: HACKS,v 1.24 2003/11/08 17:21:15 ragge Exp $
21.1Slukem#
31.1Slukem# This file is intended to document workarounds for currently unsolved
41.1Slukem# (mostly) compiler bugs.
51.1Slukem#
61.1Slukem# Format:
71.1Slukem#   hack		title
81.1Slukem#   cdate		creation date
91.1Slukem#   mdate		mod date
101.1Slukem#   who			responsible developer
111.1Slukem#   port		...
121.1Slukem#  	  affected ports, space separated, if not "all"
131.1Slukem#   file		affected file : revision : line from : line to
141.1Slukem#  	  affected files and revision and line numbers describing hack
151.1Slukem#  	  multiple lines if necessary.
161.1Slukem#   pr			NNNN ...
171.1Slukem#  	  problem reports this hack works around, if known. Space
181.1Slukem#  	  separated.
191.1Slukem#   regress		src/regress/directory/where/test/found
201.1Slukem#  	  regression test directories, if available.
211.1Slukem#   descr
221.1Slukem#  	  insert short informal description (multi-line). (Longer ones
231.1Slukem#  	  should be in the PR database. More formal descriptions might
241.1Slukem#  	  be in the regress tree. See above).
251.1Slukem#   kcah
261.1Slukem#  	  closing bracket.
271.1Slukem#
281.1Slukem# this is a comment.
291.1Slukem
301.1Slukemhack	netstat ieee1394 address printing.
311.1Slukemmdate	14 Nov 2000
321.1Slukemwho	matt
331.1Slukemfile	lib/libc/net/getnameinfo.c	: 1.32 : 497 : 503
341.1Slukemdescr
351.1Slukem	Because the current implementation of IP over IEEE1394, the
361.1Slukem	fw device address contains more than just the IEEE1394 EUI-64.
371.1Slukem	So when printing out IEEE1394 addresses, ignore the extra stuff.
381.1Slukemkcah
391.1Slukem
401.1Slukemhack	xterm vs. libterm
411.1Slukemmdate	01 Aug 2000
421.1Slukemwho	jdc
431.1Slukemfile	xsrc/xc/programs/xterm/main.c	: 1.2 : 3609 : 3614
441.1Slukempr	10383
451.1Slukemdescr
461.1Slukem	In order to extend the termcap string over 1023 bytes, a ZZ entry was
471.1Slukem	introduced to point to a memory location containing the full entry.
481.1Slukem	Without this hack, xterm will export a termcap containing the ZZ
491.1Slukem	entry, which will then be ignored by libterm.  As xterm modifies the
501.1Slukem	exported termcap, this would cause those modifications to be ignored.
511.1Slukemkcah
521.1Slukem
531.1Slukemhack	wi-at-big-endian-bus
541.1Slukemcdate	15 Mar 2002
551.1Slukemwho	martin
561.1Slukemfile	dev/ic/wireg.h
571.1Slukemdescr	Add an option to access the underlying bus in big endian byte order
581.1Slukem	to work around deficiencies in bus_space_{read,write}_* macros.
591.1Slukem	Those don't allow the implementation of a proper pcmcia bus space
601.1Slukem	tag.
611.1Slukemkcah
621.1Slukem
631.3Sjdchack	specific knowledge of colours in curses code
641.3Sjdccdate	Sun Apr  6 11:05:24 BST 2003
651.3Sjdcwho	jdc
661.3Sjdcfile	lib/libcurses/color.c : r1.24
671.3Sjdcdescr
681.3Sjdc	Swap red/blue and yellow/cyan colours for COLOR_OTHER.
691.3Sjdc	Fix is to enhance libtermcap to understand terminfo-style % sequences.
701.3Sjdc	See also:
711.5Ssalo	    http://mail-index.NetBSD.org/tech-userlevel/2003/04/06/0000.html
721.4Sscwkcah
731.4Sscw
741.7Shehack	Compensation for differing types of LINUX_USRSTACK and USRSTACK
751.7Shecdate	21 Aug 2003
761.7Shewho	he
771.7Shefile	sys/miscfs/procfs/procfs_linux.c : 1.14
781.7Shedescr
791.7She	Not all ports have LINUX_USRSTACK and/or USRSTACK defined as
801.7She	literals/constants, but refer to variables of a type which is
811.7She	not "unsigned long", causing compilation of procfs_linux.c to
821.7She	fail with "makes integer from pointer without a cast".  This
831.7She	is observed on e.g. the sun3 port.  Ideally the "types" for
841.7She	symbols should be consistent across all ports.
851.11Slukemkcah
861.11Slukem
871.13Ssimonb
881.21Smycroftport	vax
891.21Smycroft
901.21Smycroft	hack	gcc 2.95/vax cannot handle __builtin_ffs()
911.21Smycroft	cdate	Fri Aug 23 21:31:15 CEST 2002
921.21Smycroft	who	ragge
931.21Smycroft	file	sys/lib/libkern/libkern.h : 1.42
941.21Smycroft	descr
951.21Smycroft		__builtin_ffs() is ifdef'd away if __vax__. Matt Thomas has
961.21Smycroft		added this feature to GCC 3 so it can be removed when switching.
971.21Smycroft	kcah
981.21Smycroft
991.21Smycroft	hack	gcc-vax-libbz2
1001.21Smycroft	mdate	27 Jun 2002
1011.21Smycroft	who	thorpej
1021.21Smycroft	file	lib/libbz2/Makefile
1031.21Smycroft	descr
1041.21Smycroft		libbz2 is mis-compiled with optimization with GCC 2.95.3
1051.24Sragge		on VAX.  -O0 works around this problem.
1061.24Sragge	kcah
1071.24Sragge
1081.24Sragge	hack	gawk
1091.24Sragge	mdate	8 Nov 2003
1101.24Sragge	who	ragge
1111.24Sragge	file	gnu/usr.bin/gawk/Makefile
1121.24Sragge	descr
1131.24Sragge		gawk is mis-compiled with optimization with GCC 2.95.3
1141.21Smycroft		on VAX.  -O0 works around this problem.
1151.21Smycroft	kcah
1161.21Smycroft
1171.21Smycroft	hack	vax Toolchain bug
1181.21Smycroft	cdate	28 Mar 2003
1191.21Smycroft	who	he
1201.21Smycroft	file	sys/arch/vax/vax/intvec.S : 1.5
1211.21Smycroft	descr
1221.21Smycroft		Workaround for PR toolchain/20924.  The assembler apparently
1231.21Smycroft		tries to range-check byte offsets when it doesn't have
1241.21Smycroft		sufficient information to make that decision.
1251.21Smycroft		Workaround uses `brw' instead of `brb' instruction.
1261.23She	kcah
1271.23She
1281.23She	hack	vax gcc 2.95.3 -Wuninitialized workarounds
1291.23She	cdate	6 Nov 2003
1301.23She	who	he
1311.23She	file	sys/dev/bi/if_ni.c : 1.22
1321.23She	file	sys/arch/vax/vax/ctu.c : 1.20
1331.23She	descr
1341.23She		Workaround for gcc 2.95.3 failing to detect that certain
1351.23She		variables will be initialized; so that this code compiles
1361.23She		with -Wuninitialized.  Should be reviewed and possibly
1371.23She		reverted when gcc 3.3.2 is ready for vax.
1381.21Smycroft	kcah
1391.21Smycroft
1401.21Smycroft
1411.21Smycroftport	sparc64
1421.21Smycroft
1431.21Smycroft	hack	binutil-2.11-sparc64-pltrela
1441.21Smycroft	mdate	14 Aug 2001
1451.21Smycroft	who	eeh
1461.21Smycroft	file	libexec/ld.elf_so/arch/sparc64/mdreloc.c
1471.21Smycroft	descr
1481.21Smycroft		The first for PLT entries are reserved.  There is some
1491.21Smycroft		disagreement whether they should have associated relocation
1501.21Smycroft		entries.  Both the SPARC 32-bit and 64-bit ELF specifications
1511.21Smycroft		say that they should have relocation entries, but the 32-bit
1521.21Smycroft		SPARC binutils do not generate them, and now the 64-bit SPARC
1531.21Smycroft		binutils have stopped generating them too.
1541.21Smycroft
1551.21Smycroft		To provide binary compatibility, we will check the first entry,
1561.21Smycroft		if it is reserved it should not be of the type JMP_SLOT.  If it
1571.21Smycroft		is JMP_SLOT, then the 4 reserved entries were not generated and
1581.21Smycroft		our index is 4 entries too far, so we frob the rela pointer.
1591.21Smycroft	kcah
1601.21Smycroft
1611.21Smycroft
1621.21Smycroftport	arm
1631.21Smycroft
1641.21Smycroft	hack	gcc-unsigned-compare
1651.21Smycroft	cdate	09 Mar 2002
1661.21Smycroft	mdate	18 Mar 2002
1671.21Smycroft	who	bjh21
1681.21Smycroft	file	dist/bind/lib/nameser/ns_parse.c : 1.3
1691.21Smycroft	file	dist/dhcp/minires/ns_parse.c : 1.3
1701.21Smycroft	file	dist/dhcp/omapip/result.c : 1.2
1711.21Smycroft	file	dist/dhcp/server/failover.c : 1.3
1721.21Smycroft	file	gnu/dist/gawk/eval.c : 1.4
1731.21Smycroft	file	gnu/dist/toolchain/bfd/bfd.c : 1.2
1741.21Smycroft	file	gnu/dist/toolchain/bfd/format.c : 1.2
1751.21Smycroft	file	gnu/dist/toolchain/gdb/target.c : 1.2
1761.21Smycroft	file	sys/kern/vfs_subr.c : 1.172
1771.21Smycroft	descr	When checking that a potentially-unsigned enum is >= 0, assign
1781.21Smycroft		it to an int first.  This is necessary to avoid "comparison is
1791.21Smycroft		always true" warnings with -fshort-enums.  Casting to an int
1801.21Smycroft		really should be enough, but turns out not to be.
1811.21Smycroft	kcah
1821.21Smycroft
1831.21Smycroft	hack	gcc 3.3.x bug
1841.21Smycroft	cdate	Sun Oct 26 01:50:16 UTC 2003
1851.21Smycroft	who	lukem
1861.21Smycroft	file	usr.bin/newsyslog/newsyslog.c
1871.21Smycroft	pr	22986 23002 23217
1881.21Smycroft	descr
1891.21Smycroft		newsyslog fails with
1901.21Smycroft			newsyslog: config line 7: bad flags
1911.21Smycroft		solved when compiled with -fno-gcse-lm
1921.21Smycroft	kcah
1931.21Smycroft
1941.21Smycroft	hack	gcc 3.3.2/arm sys/kern/uipc_socket.c
1951.21Smycroft	cdate	Tue Oct 28 18:02:16 GMT 2003
1961.21Smycroft	who	skrll
1971.21Smycroft	file	sys/conf/arm/Makefile.arm : 1.19
1981.21Smycroft	descr
1991.21Smycroft		Performing a build.sh tools on a CATS with NFS mounted 
2001.21Smycroft		obj and src directories results in "panic: receive 1"
2011.21Smycroft		The hack is to compile sys/kern/uipc_socket.c with
2021.21Smycroft		-fno-strict-aliasing
2031.21Smycroft	kcah
2041.21Smycroft
2051.21Smycroft
2061.13Ssimonbport	pc532
2071.16Scl
2081.21Smycroft	hack	egcs-pc532-ip6_mroute
2091.21Smycroft	cdate	09 Jul 2002
2101.21Smycroft	who	simonb
2111.21Smycroft	file	sys/arch/pc532/conf/Makefile.pc532 : 1.70
2121.21Smycroft	file	sys/arch/pc532/conf/files.pc532 : 1.47
2131.21Smycroft	descr
2141.21Smycroft		egcs 1.1.2 gets an "internal error--insn does not satisfy its
2151.21Smycroft		constraints" error compiling ip6_mroute.c with -O2 or greater.
2161.21Smycroft		-O1 works around this problem.
2171.21Smycroft	kcah
2181.21Smycroft
2191.21Smycroft	hack	gcc-pc532-duffs_device
2201.21Smycroft	cdate	Mon Oct 27 07:23:05 UTC 2003
2211.21Smycroft	who	simonb
2221.21Smycroft	file	sys/arch/pc532/conf/Makefile.pc532 : 1.71
2231.21Smycroft	file	sys/arch/pc532/conf/files.pc532 : 1.52
2241.21Smycroft	pr	GCC PR optimization/5230
2251.21Smycroft	descr
2261.21Smycroft		gcc incorrectly gives an "unreachable code at beginning of
2271.21Smycroft		switch statement" for a Duff's device construct in
2281.21Smycroft		arch/pc532/dev/ncr.c.  There is no way to disable just this
2291.21Smycroft		warning, so -Wno-error is turned on for this file.
2301.21Smycroft	kcah
2311.21Smycroft
2321.21Smycroft
2331.21Smycroftport	sh3
2341.21Smycroft
2351.21Smycroft	hack	gcc-sh3-sed
2361.21Smycroft	mdate	23 Apr 2002
2371.21Smycroft	who	thorpej
2381.21Smycroft	file	usr.bin/sed/Makefile	: 1.9 : 9 : 13
2391.21Smycroft	descr
2401.21Smycroft		The in-tree GCC 2.95.3-based compiler ICEs when building
2411.21Smycroft		with optimization for SuperH.  Hack is to biuld with -O0.
2421.21Smycroft	kcah
2431.21Smycroft
2441.21Smycroft
2451.21Smycroftport	sh5
2461.21Smycroft
2471.21Smycroft	hack	SuperH SH5 Toolchain Bugs
2481.21Smycroft	cdate	11 Jul 2002
2491.21Smycroft	who	scw
2501.21Smycroft	file	usr.sbin/ndbootd/ndbootd.c : 1.6
2511.21Smycroft	file	usr.sbin/traceroute/traceroute.c : 1.48
2521.21Smycroft	descr
2531.21Smycroft		The SuperH SH5 toolchain (2.97-sh5-010522) gets an internal
2541.21Smycroft		compiler error when assigning a bit-wise inverted value
2551.21Smycroft		under some circumstances.
2561.21Smycroft		Work around it by splitting the statement into two.
2571.21Smycroft	kcah
2581.21Smycroft
2591.21Smycroft	hack	SuperH SH5 Toolchain Bugs
2601.21Smycroft	cdate	17 May 2003
2611.21Smycroft	who	scw
2621.21Smycroft	file	lib/libpthread/pthread_lock.c : 1.7
2631.21Smycroft	descr
2641.21Smycroft		The SuperH SH5 toolchain generates incorrect PIC code when
2651.21Smycroft		faced with a symbol which is declared extern, but has local
2661.21Smycroft		scope due to being defined within an asm() statement (without
2671.21Smycroft		being declared .globl in that statement). Work around it by
2681.21Smycroft		adding the .globl.
2691.21Smycroft	kcah
2701.17Smrg
2711.17Smrg
2721.17Smrgport	sun2
2731.20Sskrll
2741.21Smycroft	hack	gcc 3.3.2/mdsetimage
2751.21Smycroft	cdate	Tue Oct 28 18:43:05 EST 2003
2761.21Smycroft	who	fredette
2771.21Smycroft	file	gnu/usr.sbin/mdsetimage/Makefile : 1.15
2781.21Smycroft	descr
2791.21Smycroft		mdsetimage.c causes cc1 to barf for as-yet unknown reasons.
2801.21Smycroft		this hack ues -O0 to avoid the problem.
2811.21Smycroft	kcah
2821.21Smycroft
2831.21Smycroft	hack	gcc 3.3.2/installboot/ffs.c
2841.21Smycroft	cdate	Tue Oct 28 18:43:05 EST 2003
2851.21Smycroft	who	fredette
2861.21Smycroft	file	usr.sbin/installboot/Makefile : 1.26
2871.21Smycroft	descr
2881.21Smycroft		ffs.c causes cc1 to barf for as-yet unknown reasons.  this
2891.21Smycroft		hack ues -O0 to avoid the problem.
2901.22Smrg	kcah
2911.22Smrg
2921.22Smrg
2931.22Smrgport	sun2, m68k
2941.22Smrg
2951.22Smrg	hack	gcc 3.3.2/gcc/ra-colorize.c
2961.22Smrg	cdate	Wed Nov  5 11:03:13 EST 2003
2971.22Smrg	who	mrg
2981.22Smrg	file	gnu/usr.bin/gcc/backend/Makefile : 1.11
2991.22Smrg	descr
3001.22Smrg		ra-colorize.c causes a whole spew of unreferenced .L* local
3011.22Smrg		symbols.  using -O0 (or even -O2 reportedly) avoids the problem.
3021.21Smycroft	kcah
303