HACKS revision 1.112
11.112Schristos# $NetBSD: HACKS,v 1.112 2009/01/11 18:25:17 christos 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.
851.25Smrgkcah
861.25Smrg
871.37Sjunyounghack	gcc-strict-aliasing
881.37Sjunyoungcdate	6 August 2004
891.37Sjunyoungwho	junyoung
901.37Sjunyoungpr	26516
911.37Sjunyoungfile	src/usr.sbin/racoon/racoon/Makefile : 1.21
921.37Sjunyoungdescr
931.37Sjunyoung	GCC issues a warning "dereferencing type-punned pointer will break
941.37Sjunyoung	strict-aliasing rules" when compiling
951.37Sjunyoung	crypto/dist/kame/racoon/crypto/openssl.c with -O[23s]. Note that this
961.38Sjunyoung	is due to use of type casting in a non-standard way in the code and
971.38Sjunyoung	not a compiler bug. Once a new release of KAME with a proper fix
981.38Sjunyoung	applied is imported this should be backed out.
991.37Sjunyoungkcah
1001.37Sjunyoung
1011.80Stsutsuihack	gcc4 wrong uninitialized variable
1021.58Smrgmdate	10 May 2006
1031.58Smrgwho	mrg
1041.58Smrgfile	bin/ksh/eval.c			: 1.6
1051.58Smrgfile	bin/sh/histedit.c		: 1.39
1061.58Smrgfile	bin/sh/parser.c			: 1.60
1071.58Smrgfile	crypto/dist/heimdal/kdc/524.c	: 1.10
1081.58Smrgfile	crypto/dist/ssh/sftp.c		: 1.20
1091.58Smrgfile	crypto/dist/ssh/ssh-keysign.c	: 1.11
1101.58Smrgfile	dist/ipf/lib/hostname.c		: 1.2
1111.58Smrgfile	dist/ipf/tools/ipmon.c		: 1.8
1121.72Smrgfile	dist/ntp/ntpd/ntp_request.c	: 1.4
1131.58Smrgfile	dist/ntp/ntpd/refclock_shm.c	: 1.4
1141.58Smrgfile	dist/ntp/sntp/timing.c		: 1.3
1151.58Smrgfile	dist/pppd/pppstats/pppstats.c	: 1.3
1161.58Smrgfile	dist/smbfs/lib/smb/rap.c	: 1.6
1171.58Smrgfile	dist/tcpdump/print-zephyr.c	: 1.5
1181.68Smrgfile	distrib/utils/sysinst/aout2elf.c : 1.12
1191.58Smrgfile	gnu/libexec/uucp/uucico/uucico.c : 1.6
1201.58Smrgfile	lib/libc/citrus/citrus_csmapper.c : 1.6
1211.58Smrgfile	lib/libc/citrus/citrus_pivot_factory.c : 1.5
1221.58Smrgfile	lib/libc/inet/inet_cidr_ntop.c	: 1.3
1231.58Smrgfile	lib/libc/inet/inet_ntop.c	: 1.3
1241.58Smrgfile	lib/libc/stdio/vfwprintf.c	: 1.8
1251.69Smrgfile	libexec/ld.elf_so/arch/m68k/mdreloc.c : 1.20
1261.76Smrgfile	libexec/ld.elf_so/arch/powerpc/ppc_reloc.c : 1.40
1271.74Smrgfile	libexec/ld.elf_so/arch/sh3/mdreloc.c : 1.22
1281.69Smrgfile	libexec/ld.elf_so/arch/sparc/mdreloc.c : 1.39
1291.58Smrgfile	libexec/ld.elf_so/arch/sparc64/mdreloc.c : 1.39
1301.74Smrgfile	libexec/ld.elf_so/arch/vax/mdreloc.c	: 1.21
1311.71Sdanfile	libexec/ld.elf_so/arch/x86_64/mdreloc.c	: 1.27
1321.58Smrgfile	sbin/fsck_ext2fs/dir.c		: 1.19
1331.58Smrgfile	sbin/routed/rtquery/rtquery.c	: 1.18
1341.73Sdanfile	sys/arch/amd64/amd64/pmap.c	: 1.26
1351.79Smrgfile	sys/arch/i386/pci/piixpcib.c	: 1.4
1361.80Stsutsuifile	sys/arch/m68k/m68k/pmap_motorola.c	: 1.4
1371.58Smrgfile	sys/crypto/cast128/cast128.c	: 1.9
1381.58Smrgfile	sys/ddb/db_command.c		: 1.86
1391.68Smrgfile	sys/dev/cardbus/cardbus_map.c	: 1.21
1401.58Smrgfile	sys/dev/fss.c			: 1.25
1411.69Smrgfile	sys/dev/ic/igsfb.c		: 1.39
1421.68Smrgfile	sys/dev/ic/mb86950.c		: 1.5
1431.58Smrgfile	sys/dev/ic/midway.c		: 1.71
1441.68Smrgfile	sys/dev/kttcp.c			: 1.18
1451.68Smrgfile	sys/dev/rasops/rasops_bitops.h	: 1.9
1461.68Smrgfile	sys/dev/pci/cmpci.c		: 1.31
1471.58Smrgfile	sys/dev/pci/machfb.c		: 1.45
1481.68Smrgfile	sys/dev/usb/ohci.c		: 1.174
1491.68Smrgfile	sys/dev/usb/uhci.c		: 1.196
1501.68Smrgfile	sys/dev/rasops/rasops_bitops.h	: 1.9
1511.58Smrgfile	sys/dist/ipf/netinet/ip_nat.c	: 1.10
1521.58Smrgfile	sys/dist/ipf/netinet/ip_rpcb_pxy.c : 1.8
1531.58Smrgfile	sys/dist/pf/net/pf.c		: 1.22
1541.59Smrgfile	sys/fs/udf/udf_vnops.c		: 1.4
1551.58Smrgfile	sys/kern/kern_sig.c		: 1.219
1561.58Smrgfile	sys/kern/tty.c			: 1.181
1571.58Smrgfile	sys/net/bpf.c			: 1.116
1581.58Smrgfile	sys/net/zlib.c			: 1.26
1591.68Smrgfile	sys/netccitt/if_x25subr.c	: 1.37
1601.58Smrgfile	sys/netinet/in.c		: 1.107
1611.58Smrgfile	sys/netsmb/smb_smb.c		: 1.27
1621.58Smrgfile	sys/netsmb/smb_trantcp.c	: 1.22
1631.58Smrgfile	sys/nfs/nfs_serv.c		: 1.108
1641.58Smrgfile	sys/nfs/nfs_socket.c		: 1.129
1651.58Smrgfile	sys/nfs/nfs_syscalls.c		: 1.91
1661.58Smrgfile	sys/ufs/lfs/lfs_vfsops.c	: 1.207 [also (char *)]
1671.58Smrgfile	usr.bin/ftp/ftp.c		: 1.140
1681.58Smrgfile	usr.bin/find/function.c		: 1.54
1691.58Smrgfile	usr.bin/mail/tty.c		: 1.20
1701.100Stsutsuifile	usr.bin/msgc/msg_sys.def	: 1.33-1.34
1711.58Smrgfile	usr.bin/nl/nl.c			: 1.7
1721.58Smrgfile	usr.bin/systat/keyboard.c	: 1.23
1731.58Smrgfile	usr.bin/usbhidctl/usbhid.c	: 1.29
1741.58Smrgfile	usr.bin/vi/cl/cl_read.c		: 1.5
1751.58Smrgfile	usr.bin/vi/ex/ex_cscope.c	: 1.12
1761.58Smrgfile	usr.bin/vi/ex/ex_tag.c		: 1.19
1771.58Smrgfile	usr.bin/vi/vi/v_txt.c		: 1.15
1781.58Smrgfile	usr.sbin/altq/altqstat/qdisc_rio.c : 1.4
1791.58Smrgfile	usr.sbin/cron/do_command.c	: 1.19
1801.58Smrgfile	usr.sbin/timed/timed/slave.c	: 1.15
1811.58Smrgdescr
1821.58Smrg	GCC 4.1 gets many uninitialised variable warnings wrong.  We should
1831.58Smrg	really audit all the old hacks like this when older compilers are
1841.58Smrg	removed from the tree, as many are probably no longer required.
1851.65Schristos	The problem is that it does not recognize initialization via function
1861.65Schristos	call pointer. I.e.
1871.65Schristos		int p;
1881.65Schristos		foo(&p);
1891.65Schristos	does not mark p as initialized.
1901.58Smrgkcah
1911.58Smrg
1921.59Smrghack	gcc4 pointer sign and strict aliasing problems
1931.59Smrgmdate	10 May 2006
1941.59Smrgwho	mrg
1951.59Smrgfile	bin/ed/Makefile			: 1.33
1961.68Smrgfile	distrib/utils/sysinst/Makefile.inc : 1.44
1971.91Smartinfile	distrib/utils/x_dhclient/Makefile : 1.15
1981.60Smrgfile	games/bcd/Makefile		: 1.5
1991.60Smrgfile	games/dab/Makefile		: 1.5
2001.60Smrgfile	games/larn/Makefile		: 1.17
2011.60Smrgfile	games/pom/Makefile		: 1.5
2021.60Smrgfile	lib/libasn1/Makefile		: 1.26
2031.60Smrgfile	lib/libcrypt/Makefile		: 1.17
2041.60Smrgfile	lib/libgssapi/Makefile		: 1.16
2051.60Smrgfile	lib/libhdb/Makefile		: 1.20
2061.60Smrgfile	lib/libkadm5clnt/Makefile	: 1.21
2071.60Smrgfile	lib/libkadm5srv/Makefile	: 1.25
2081.60Smrgfile	lib/libkrb5/Makefile		: 1.35
2091.60Smrgfile	lib/libssh/Makefile		: 1.6
2101.60Smrgfile	lib/libtelnet/Makefile		: 1.26
2111.60Smrgfile	libexec/getty/Makefile		: 1.14
2121.60Smrgfile	libexec/kadmind/Makefile	: 1.19
2131.60Smrgfile	libexec/kpasswdd/Makefile	: 1.14
2141.60Smrgfile	sbin/atactl/Makefile		: 1.3
2151.60Smrgfile	sbin/cgdconfig/Makefile		: 1.7
2161.60Smrgfile	sbin/clri/Makefile		: 1.13
2171.60Smrgfile	sbin/dkctl/Makefile		: 1.4
2181.60Smrgfile	sbin/dump/Makefile		: 1.33
2191.60Smrgfile	sbin/fdisk/Makefile		: 1.35
2201.60Smrgfile	sbin/fsck_ext2fs/Makefile	: 1.11
2211.60Smrgfile	sbin/fsck_ffs/Makefile		: 1.29
2221.60Smrgfile	sbin/fsdb/Makefile		: 1.18
2231.60Smrgfile	sbin/mount_smbfs/Makefile	: 1.4
2241.60Smrgfile	sbin/newfs/Makefile		: 1.30
2251.60Smrgfile	sbin/newfs_sysvbfs/Makefile	: 1.2
2261.60Smrgfile	sbin/restore/Makefile		: 1.23
2271.60Smrgfile	sbin/veriexecctl/Makefile	: 1.11
2281.64Smrgfile	sys/lib/libsa/Makefile		: 1.59
2291.68Smrgfile	sys/arch/evbarm/adi_brh/brh_machdep.c : 1.24
2301.59Smrgfile	usr.bin/awk/Makefile		: 1.9
2311.64Smrgfile	usr.bin/crontab/Makefile	: 1.24
2321.60Smrgfile	usr.bin/ctags/Makefile		: 1.8
2331.61Smrgfile	usr.bin/gzip/Makefile		: 1.10
2341.60Smrgfile	usr.bin/ssh/sftp/Makefile	: 1.10
2351.60Smrgfile	usr.bin/ssh/ssh/Makefile	: 1.25
2361.59Smrgfile	usr.bin/vi/build/Makefile	: 1.26
2371.60Smrgfile	usr.bin/telnet/Makefile		: 1.40
2381.60Smrgfile	usr.bin/tn3270/tn3270/Makefile	: 1.36
2391.60Smrgfile	usr.bin/tr/Makefile		: 1.4
2401.59Smrgfile	usr.sbin/amd/amd/Makefile	: 1.27
2411.59Smrgfile	usr.sbin/amd/amq/Makefile	: 1.14
2421.59Smrgfile	usr.sbin/amd/libamu/Makefile	: 1.20
2431.59Smrgfile	usr.sbin/amd/pawd/Makefile	: 1.5
2441.59Smrgfile	usr.sbin/bind/Makefile.inc	: 1.22
2451.60Smrgfile	usr.sbin/bind/libdns/Makefile	: 1.3
2461.60Smrgfile	usr.sbin/bind/named/Makefile	: 1.17
2471.60Smrgfile	usr.sbin/bootp/bootptest/Makefile : 1.2
2481.60Smrgfile	usr.sbin/chrtbl/Makefile	: 1.6
2491.60Smrgfile	usr.sbin/cron/Makefile		: 1.12
2501.59Smrgfile	usr.sbin/dhcp/Makefile.inc	: 1.20
2511.59Smrgfile	usr.sbin/dumplfs/Makefile	: 1.11
2521.60Smrgfile	usr.sbin/hprop/Makefile		: 1.13
2531.60Smrgfile	usr.sbin/installboot/Makefile	: 1.35
2541.59Smrgfile	usr.sbin/ipf/ipftest/Makefile	: 1.32
2551.59Smrgfile	usr.sbin/isdn/isdnd/Makefile	: 1.6
2561.59Smrgfile	usr.sbin/isdn/isdnmonitor/Makefile : 1.3
2571.59Smrgfile	usr.sbin/isdn/isdntel/Makefile	: 1.2
2581.59Smrgfile	usr.sbin/isdn/isdntrace/Makefile : 126
2591.61Smrgfile	usr.sbin/mopd/common/Makefile	: 1.10
2601.61Smrgfile	usr.sbin/mopd/mopd/Makefile	: 1.9
2611.61Smrgfile	usr.sbin/mopd/mopprobe/Makefile	: 1.7
2621.60Smrgfile	usr.sbin/makefs/Makefile	: 1.17
2631.60Smrgfile	usr.sbin/mscdlabel/Makefile	: 1.5
2641.60Smrgfile	usr.sbin/pkg_install/add/Makefile : 1.7
2651.60Smrgfile	usr.sbin/pkg_install/create/Makefile : 1.5
2661.60Smrgfile	usr.sbin/pkg_install/lib/Makefile : 1.28
2671.59Smrgfile	usr.sbin/ntp/ntpd/Makefile	: 1.10/1.11
2681.59Smrgfile	usr.sbin/ntp/ntptime/Makefile	: 1.4/1.5
2691.59Smrgfile	usr.sbin/pppd/Makefile.inc	: 1.3
2701.59Smrgfile	usr.sbin/pppd/pppd/Makefile	: 1.38
2711.60Smrgfile	usr.sbin/rarpd/Makefile		: 1.10
2721.59Smrgfile	usr.sbin/rbootd/Makefile	: 1.10
2731.60Smrgfile	usr.sbin/rpc.pcnfsd/Makefile	: 1.17
2741.59Smrgfile	usr.sbin/rtadvd/Makefile	: 1.9
2751.60Smrgfile	usr.sbin/tcpdump/Makefile	: 1.42
2761.59Smrgfile	usr.sbin/wiconfig/Makefile	: 1.3
2771.59Smrgdescr
2781.59Smrg	GCC 4.1 warns on pointer sign comparision/assignments and lots of
2791.59Smrg	code does not conform.  For now we use -Wno-pointer-sign and
2801.59Smrg	-fno-strict-aliasing.
2811.59Smrgkcah
2821.59Smrg
2831.105Sagchack	cross-building hack on Darwin
2841.105Sagcmdate	20 July 2008
2851.105Sagcwho	agc
2861.105Sagcfile	src/distrib/common/Makefile.mdset	: 1.33
2871.105Sagcdescr
2881.105Sagc	Darwin has problems with getopt() when used in mdsetimage -v,
2891.105Sagc	due to the difference between BSD and libiberty() getopt
2901.105Sagc	implementations, more fully described in
2911.105Sagc	http://mail-index.netbsd.org/current-users/2008/06/27/msg003242.html
2921.105Sagc	and the subsequent thread.  For just now, we just have an ugly
2931.105Sagc	workaround not to call mdsetimage with the -v argument on Darwin
2941.105Sagckcah
2951.105Sagc
2961.109Sveegohack	avoid rebuilding asn1 libraries
2971.109Sveegomdate	03 August 2008
2981.109Sveegowho	veego
2991.109Sveegofile	src/crypto/dist/heimdal/lib/asn1/gen.c	: 1.10
3001.109Sveegofile	src/lib/libasn1/Makefile	: 1.32
3011.109Sveegofile	src/lib/libhdb/Makefile	: 1.23
3021.109Sveegofile	src/lib/libgssapi/Makefile	: 1.20
3031.109Sveegofile	src/lib/libhx509/Makefile	: 1.3
3041.109Sveegopr	9702 39185
3051.109Sveegodescr
3061.109Sveego	asn1_compile does not check if generated header files do not have to
3071.109Sveego	be rebuild.
3081.109Sveego	Generate .hx files and copy it in the Makefiles if they changed.
3091.109Sveegokcah
3101.109Sveego
3111.111Stsutsuihack	avoid gcc4 -O1 bug in loop optimize
3121.111Stsutsuicdate	29 November 2008
3131.111Stsutsuiwho	tsutsui
3141.111Stsutsuifile	src/sbin/ifconfig/Makefile	: 1.39
3151.111Stsutsuipr	40036 40066
3161.111Stsutsuidescr
3171.111Stsutsui	gcc -O1 seems to generate wrong code in loop optimize so that
3181.111Stsutsui	ifconfig lo0 inet6 doesn't return error even if a kernel
3191.111Stsutsui	doesn't have IPv6 support. (see the above PRs for details)
3201.111Stsutsui	For workaround, specify -fno-loop-optimize on m68000,
3211.111Stsutsui	which uses -O1 by default.
3221.111Stsutsuikcah
3231.111Stsutsui
3241.21Smycroftport	vax
3251.21Smycroft
3261.102Stsutsui	hack	gcc4/vax ICE
3271.102Stsutsui	cdate	
3281.102Stsutsui	who	tsutsui
3291.102Stsutsui	file	sys/arch/vax/conf/Makefile.vax
3301.102Stsutsui	descr
3311.102Stsutsui		GCC4 on vax gets ICE on compiling sys/ddb/db_command.c.
3321.102Stsutsui		-fno-tree-ter prevents it so add it to COPTS.
3331.102Stsutsui	kcah
3341.102Stsutsui
3351.81Smrg	hack	gcc4/vax compiler crash
3361.81Smrg	cdate	Fri Jun 30 22:39:12 PDT 2006
3371.81Smrg	who	mrg
3381.81Smrg	file	bin/csh/Makefile	: 1.27
3391.81Smrg	file	lib/i18n_module/UTF7/Makefile	: 1.2
3401.81Smrg	descr
3411.81Smrg		GCC4 on vax crashes.  -O0 stops it happening so far...
3421.81Smrg	kcah
3431.81Smrg
3441.52Sjmc	hack	gcc 2.95/vax doesn't like abort being used as function pointer
3451.52Sjmc	cdate	Tue Dec 13 05:54:50 GMT 2005
3461.52Sjmc	who	jmc
3471.52Sjmc	file	lib/libc/rpc/svc_vc.c 1.15
3481.52Sjmc	descr
3491.52Sjmc		Provide a local definition for abort which doesn't include
3501.52Sjmc		the noreturn attribute which trips up gcc 2.95 on vax.
3511.81Smrg	kcah
3521.52Sjmc	
3531.45Sjmc	hack	gcc 2.95/vax doesn't have stdbool.h and gettext needs it
3541.45Sjmc	cdate	Mon May  9 12:16:22 CDT 2005
3551.45Sjmc	who	jmc
3561.45Sjmc	file	gnu/usr.bin/gettext/Makefile.inc.prog 1.5
3571.45Sjmc		gnu/usr.bin/gettext/libgrep/Makefile 1.3
3581.45Sjmc		gnu/usr.bin/gettext/libnlspr/Makefile 1.5
3591.45Sjmc		gnu/usr.bin/gettext/libnlsut/Makefile 1.3   
3601.45Sjmc	descr
3611.45Sjmc		The latest gettext assumes a C99 environment or at least
3621.45Sjmc		configure checks to work around this. As we don't use configure
3631.45Sjmc		provide a stdbool.h for vax builds locally in directories from
3641.45Sjmc		the template provided by the gettext code.
3651.81Smrg	kcah
3661.45Sjmc	
3671.21Smycroft	hack	gcc 2.95/vax cannot handle __builtin_ffs()
3681.21Smycroft	cdate	Fri Aug 23 21:31:15 CEST 2002
3691.21Smycroft	who	ragge
3701.21Smycroft	file	sys/lib/libkern/libkern.h : 1.42
3711.21Smycroft	descr
3721.21Smycroft		__builtin_ffs() is ifdef'd away if __vax__. Matt Thomas has
3731.21Smycroft		added this feature to GCC 3 so it can be removed when switching.
3741.81Smrg	kcah
3751.21Smycroft
3761.102Stsutsui	hack	gcc 2.95/vax libbz2
3771.21Smycroft	mdate	27 Jun 2002
3781.21Smycroft	who	thorpej
3791.21Smycroft	file	lib/libbz2/Makefile
3801.21Smycroft	descr
3811.21Smycroft		libbz2 is mis-compiled with optimization with GCC 2.95.3
3821.24Sragge		on VAX.  -O0 works around this problem.
3831.24Sragge	kcah
3841.24Sragge
3851.21Smycroft	hack	vax Toolchain bug
3861.21Smycroft	cdate	28 Mar 2003
3871.21Smycroft	who	he
3881.21Smycroft	file	sys/arch/vax/vax/intvec.S : 1.5
3891.21Smycroft	descr
3901.21Smycroft		Workaround for PR toolchain/20924.  The assembler apparently
3911.21Smycroft		tries to range-check byte offsets when it doesn't have
3921.21Smycroft		sufficient information to make that decision.
3931.21Smycroft		Workaround uses `brw' instead of `brb' instruction.
3941.23She	kcah
3951.23She
3961.23She	hack	vax gcc 2.95.3 -Wuninitialized workarounds
3971.23She	cdate	6 Nov 2003
3981.23She	who	he
3991.23She	file	sys/dev/bi/if_ni.c : 1.22
4001.23She	file	sys/arch/vax/vax/ctu.c : 1.20
4011.23She	descr
4021.23She		Workaround for gcc 2.95.3 failing to detect that certain
4031.23She		variables will be initialized; so that this code compiles
4041.23She		with -Wuninitialized.  Should be reviewed and possibly
4051.23She		reverted when gcc 3.3.2 is ready for vax.
4061.21Smycroft	kcah
4071.21Smycroft
4081.34Scl	hack	vax gcc 2.95.3 structure initialization
4091.34Scl	cdat	13 May 2004
4101.34Scl	who	cl
4111.34Scl	file	sys/miscfs/kernfs/kernfs_vnops.c : 1.104
4121.34Scl	descr
4131.34Scl		Workaround for gcc 2.95.3 failing to initialize structures
4141.34Scl		and/or unions inside structures using nested designators.
4151.34Scl		Should be reverted when gcc >=3.3.3 is ready for vax.
4161.34Scl	kcah
4171.21Smycroft
4181.35She	hack	vax gcc 2.95.3 needs -I. to build nslexer.c
4191.35She	cdat	6 Jun 2004
4201.36She	who	mhitch
4211.36She	file	lib/libc/arch/vax/Makefile.inc : 1.6
4221.35She	descr
4231.35She		It appears necessary to add -I. to CPPFLAGS when building
4241.35She		nslesxer.c.  This may be caused by gcc 2.95.3 being used.
4251.35She		Should be reverted when a newer gcc is ready for vax.
4261.35She	kcah
4271.35She
4281.43Stron	hack	vax gcc 2.95.3 doesn't understand "-std=c99"
4291.43Stron	cdat	5 April 2005
4301.43Stron	who	tron
4311.43Stron	file	src/share/mk/bsd.sys.mk
4321.43Stron	descr
4331.43Stron		We cannot use "-std=c99" for compiler warning level 4 and above
4341.43Stron		because gcc 2.95.3 doesn't support that option.
4351.43Stron	kcah
4361.43Stron
4371.54She	hack	declare boolean_t in two IPF user-mode programs
4381.54She	cdate	Tue Mar  7 19:19:20 CET 2006
4391.54She	who	he
4401.54She	file	dist/ipf/ipsend/iptests.c : 1.8
4411.54She		dist/ipf/ipsend/sock.c : 1.7
4421.54She	descr
4431.54She		The IPF user-mode programs ipsend and iptest first
4441.54She		include <sys/types.h> without _KERNEL defined, and
4451.54She		later include <sys/file.h> with _KERNEL defined.
4461.54She		This causes a build failure when building for vax,
4471.54She		since <sys/device.h> ends up being included without
4481.54She		bollean_t being defined by <sys/types.h>.
4491.54She		Build failure and further details documented in
4501.54She		PR#32907.
4511.54She	kcah
4521.54She
4531.104Sgmcgarry	hack	pcc 0.9.9 large string literals
4541.104Sgmcgarry	cdat	8 July 2008
4551.104Sgmcgarry	who	gmcgarry
4561.104Sgmcgarry	file	sys/conf/param.c : 1.58
4571.104Sgmcgarry	descr
4581.104Sgmcgarry		Workaround for pcc 0.9.9 not handling large string literals
4591.104Sgmcgarry		which causes kernels with 'options INCLUDE_CONFIG_FILE' to
4601.104Sgmcgarry		fail compilation.
4611.104Sgmcgarry		There is a proposal on the pcc mailing list to stuff config
4621.104Sgmcgarry		file in ELF section.
4631.104Sgmcgarry	kcah
4641.104Sgmcgarry
4651.107Smrg	hack	xorg warnings
4661.107Smrg	cdat	30 July 2008
4671.107Smrg	who	mrg
4681.108Smrg	file	external/mit/xorg/lib/libSM/Makefile : 1.2
4691.107Smrg		external/mit/xorg/lib/libX11/Makefile.libx11 : 1.2
4701.107Smrg		external/mit/xorg/lib/libXext/Makefile : 1.2
4711.107Smrg		external/mit/xorg/lib/libXfont/Makefile : 1.2
4721.107Smrg	descr
4731.107Smrg		Disable several warnings or use -Wno-error across Xorg sources
4741.107Smrg		while we get them working
4751.107Smrg	kcah
4761.107Smrg
4771.112Schristos	hack	32 bit time leftovers
4781.112Schristos	cdat	11 January 2009
4791.112Schristos	who	christos
4801.112Schristos	file	lib/libc/time/localtime.c : 1.41
4811.112Schristos		lib/libc/time/zic.c : 1.23
4821.112Schristos	descr
4831.112Schristos		The timezone compiled files still contain 32 bit time_t
4841.112Schristos		quantities. I did not want to version the files because
4851.112Schristos		the ``parser'' is too ugly for words. What needs to be
4861.112Schristos		done, is to rewrite the parser from scratch also to avoid
4871.112Schristos		potential core-dumps from parsing invalid files.
4881.112Schristos	kcah
4891.112Schristos
4901.112Schristos	hack	32 bit time leftovers
4911.112Schristos	cdat	11 January 2009
4921.112Schristos	who	christos
4931.112Schristos	file	various
4941.112Schristos	descr
4951.112Schristos		Many filesystem on-disk formats have 332 bit times.
4961.112Schristos	kcah
4971.112Schristos
4981.21Smycroftport	arm
4991.21Smycroft
5001.21Smycroft	hack	gcc-unsigned-compare
5011.21Smycroft	cdate	09 Mar 2002
5021.21Smycroft	mdate	18 Mar 2002
5031.21Smycroft	who	bjh21
5041.21Smycroft	file	dist/bind/lib/nameser/ns_parse.c : 1.3
5051.21Smycroft	file	dist/dhcp/minires/ns_parse.c : 1.3
5061.21Smycroft	file	dist/dhcp/omapip/result.c : 1.2
5071.21Smycroft	file	dist/dhcp/server/failover.c : 1.3
5081.21Smycroft	file	gnu/dist/toolchain/bfd/bfd.c : 1.2
5091.21Smycroft	file	gnu/dist/toolchain/bfd/format.c : 1.2
5101.21Smycroft	file	gnu/dist/toolchain/gdb/target.c : 1.2
5111.21Smycroft	file	sys/kern/vfs_subr.c : 1.172
5121.21Smycroft	descr	When checking that a potentially-unsigned enum is >= 0, assign
5131.21Smycroft		it to an int first.  This is necessary to avoid "comparison is
5141.21Smycroft		always true" warnings with -fshort-enums.  Casting to an int
5151.21Smycroft		really should be enough, but turns out not to be.
5161.21Smycroft	kcah
5171.21Smycroft
5181.21Smycroft
5191.21Smycroftport	sh3
5201.21Smycroft
5211.74Smrg	hack	gcc4-sh3-bz2
5221.103Stsutsui	cdate	Sun May 21 03:34:57 UTC 2006
5231.103Stsutsui	mdate	Fri May 16 13:13:00 UTC 2008
5241.103Stsutsui	who	mrg, tsutsui
5251.74Smrg	file	lib/libbz2/Makefile	: 1.10
5261.74Smrg	descr
5271.74Smrg		The in-tree GCC 4.1-based compiler generated too-far
5281.103Stsutsui		pc-relative addresses.  Hack is to build with
5291.103Stsutsui		-fno-loop-optimize.
5301.74Smrg	kcah
5311.74Smrg
5321.21Smycroft
5331.96Smrgport	m68000
5341.79Smrg
5351.96Smrg	hack	gcc4-m68000
5361.102Stsutsui	cdate	Fri Feb  8 10:29:37 PST 2008
5371.102Stsutsui	mdate	Sun May  4 15:37:19 UTC 2008
5381.99Stsutsui	who	mrg, tsutsui
5391.96Smrg	file	rescue/Makefile	: 1.21
5401.99Stsutsui	file	sbin/dump_lfs/Makefile	: 1.9
5411.101Stsutsui	file	sbin/fsck_ffs/Makefile	: 1.35
5421.101Stsutsui	file	sbin/fsdb/Makefile	: 1.22
5431.101Stsutsui	file	share/mk/sys.mk	: 1.96
5441.99Stsutsui	file	usr.sbin/ndbootd/Makefile	: 1.5
5451.95Smrg	descr
5461.101Stsutsui		Several internal compiler errors with gcc -O1
5471.101Stsutsui		around 64bit integer arithmetic.
5481.99Stsutsui		This hack uses -O1 and adds some -fno-tree-foo options
5491.99Stsutsui		to avoid the problem.
5501.101Stsutsui		This might be related with GCC Bugzilla Bug 32424.
5511.95Smrg	kcah
5521.95Smrg
5531.22Smrg
5541.31Smartinport	sparc64
5551.31Smartin
5561.31Smartin	hack	disable optimzations for uvm_bio.c on 32 bit kernels
5571.31Smartin	cdate	Sun Mar 21 14:14:04 MET 2004
5581.31Smartin	who	martin
5591.32Smartin	file	sys/arch/sparc64/conf/Makefile.sparc64 : 1.53-1.54
5601.32Smartin	file	sys/arch/sparc64/conf/files.sparc64 : 1.93
5611.31Smartin	descr
5621.31Smartin		The sparc compiler miscompiles uvm_bio.c when using
5631.31Smartin		any optimization. This results in ubc_release panics.
5641.31Smartin	kcah
5651.42Schristos
5661.51Ssimonbport	mips
5671.51Ssimonb
5681.51Ssimonb	hack	mips-shared-linker-load-address
5691.51Ssimonb	cdate	Fri Oct  7 08:33:10 UTC 2005
5701.51Ssimonb	who	simonb
5711.51Ssimonb	file	src/sys/kern/exec_elf32.c : 1.107
5721.51Ssimonb	descr
5731.51Ssimonb		With COMPAT_16 or previous enabled (which enables
5741.51Ssimonb		ELF_INTERP_NON_RELOCATABLE) a recent ld.elf_so will
5751.51Ssimonb		load and run at address 0.  The check to fix this in
5761.51Ssimonb		rev 1.107 only checks the first psection of the ELF
5771.51Ssimonb		executable, which may not be loadable.  A more correct
5781.51Ssimonb		fix is to check the first loadable psection instead of
5791.51Ssimonb		just the first psection.
5801.51Ssimonb	kcah
5811.54She
5821.85Smartin	hack	mips-duplicate-ras-end-label
5831.85Smartin	cdate	Sat Sep  2 23:29:42 2006
5841.85Smartin	who	martin
5851.92Schs	file	src/regress/sys/kern/ras/ras3/Makefile : 1.3
5861.85Smartin	descr
5871.85Smartin		Add -fno-reorder-blocks to CFLAGS to avoid duplicate
5881.85Smartin		labels by duplicated __asm output from RAS_END()
5891.85Smartin		macro.
5901.85Smartin	kcah
5911.85Smartin
5921.106Slukem	hack	mips-mcount-assembler-warning
5931.106Slukem	cdate	Tue Jul 29 14:16:52 UTC 2008
5941.106Slukem	who	lukem
5951.106Slukem	file	src/lib/libc/gmon/Makefile.inc : 1.8
5961.106Slukem	descr
5971.106Slukem		Workaround for PR port-mips/39192.
5981.106Slukem		common/lib/libc/gmon/mcount.c generates a (fatal)
5991.106Slukem		assembler warning on MIPS:
6001.106Slukem			Warning: No .cprestore pseudo-op used in PIC code
6011.106Slukem		Add COPTS.count.c+=-Wa,--no-warn to avoid -Wa,--fatal-warnings
6021.106Slukem	kcah
6031.106Slukem
6041.82Schristosport	i386
6051.82Schristos
6061.84Sdrochner	hack	use volatile intermediate variable to enforce rounding
6071.84Sdrochner	cdate	Tue Aug  1 22:15:55 MEST 2006
6081.84Sdrochner	who	drochner
6091.84Sdrochner	file	src/lib/libm/src/lrintf.c : 1.4
6101.84Sdrochner	file	src/lib/libm/src/s_rintf.c : 1.8
6111.84Sdrochner	descr
6121.84Sdrochner		gcc-4 does subsequent operations on "float" values within
6131.84Sdrochner		the i387 FPU without rounding the intermediate results
6141.84Sdrochner	kcah
6151.110Smacallan
6161.110Smacallanport	powerpc
6171.110Smacallan
6181.110Smacallan	hack	avoid using __builtin_return_address(0) because it fails in
6191.110Smacallan		Xorg's module loader
6201.110Smacallan	cdate	Sat Sep 27 03:52:05 UTC 2008
6211.110Smacallan	who	macallan
6221.110Smacallan	file	src/libexec/ld.elf_so/rtld.c : 1.121
6231.110Smacallan	descr
6241.110Smacallan		workaround for PR port-macppc/37812
6251.110Smacallan	kcah
6261.110Smacallan
627