HACKS revision 1.3
1 # $NetBSD: HACKS,v 1.3 2003/04/06 10:12:20 jdc Exp $ 2 # 3 # This file is intended to document workarounds for currently unsolved 4 # (mostly) compiler bugs. 5 # 6 # Format: 7 # hack title 8 # cdate creation date 9 # mdate mod date 10 # who responsible developer 11 # port ... 12 # affected ports, space separated, if not "all" 13 # file affected file : revision : line from : line to 14 # affected files and revision and line numbers describing hack 15 # multiple lines if necessary. 16 # pr NNNN ... 17 # problem reports this hack works around, if known. Space 18 # separated. 19 # regress src/regress/directory/where/test/found 20 # regression test directories, if available. 21 # descr 22 # insert short informal description (multi-line). (Longer ones 23 # should be in the PR database. More formal descriptions might 24 # be in the regress tree. See above). 25 # kcah 26 # closing bracket. 27 # 28 # this is a comment. 29 30 hack gcc 2.95/vax cannot handle __builtin_ffs() 31 cdate Fri Aug 23 21:31:15 CEST 2002 32 who ragge 33 port vax 34 file sys/lib/libkern/libkern.h : 1.42 35 descr 36 __builtin_ffs() is ifdef'd away if __vax__. Matt Thomas has added 37 this feature to GCC 3 so it can be removed when switching. 38 kcah 39 40 hack gcc 2.95.3 -O2 bug 41 cdate Sun Jan 6 23:17:28 UTC 2002 42 who lukem 43 port alpha 44 file sbin/newfs/mkfs.c : (all?) 45 pr 15156 46 descr 47 newfs built with gcc -O2 dumps core part-way through. 48 using -O1 avoids the problem 49 kcah 50 51 hack binutil-2.11-sparc64-pltrela 52 mdate 14 Aug 2001 53 who eeh 54 port sparc64 55 file libexec/ld.elf_so/reloc.c: 1.41 : 597 : 619 56 descr 57 The first for PLT entries are reserved. There is some disagreement 58 whether they should have associated relocation entries. Both the 59 SPARC 32-bit and 64-bit ELF specifications say that they should have 60 relocation entries, but the 32-bit SPARC binutils do not generate 61 them, and now the 64-bit SPARC binutils have stopped generating them 62 too. 63 64 To provide binary compatibility, we will check the first entry, if it 65 is reserved it should not be of the type JMP_SLOT. If it is JMP_SLOT, 66 then the 4 reserved entries were not generated and our index is 4 67 entries too far, so we frob the rela pointer. 68 kcah 69 70 hack gcc-arm32-schedule 71 mdate 04 May 2000 72 who is 73 file lib/libc/stdlib/strtoull.c : 1.3 : 78 : 81 74 file lib/libc/stdlib/strtouq.c : 1.15 : 73 : 76 75 pr 9613 76 descr 77 strtoull() was returning a wrong result for small numbers with bit 31 78 set. This is a gcc/arm32 compiler bug in gcc < 2.95.2. Adding 79 (void)&acc; works around it. 80 kcah 81 82 hack gcc-sparc64-memorysize 83 mdate 17 Sep 2000 84 who mrg 85 port sparc64 86 file lib/libc/hash/sha1.c : 1.6 : 78 : 129 : 162 : 189 87 file sys/lib/libkern/sha1.c : 1.6 : 78 : 129 : 162 : 189 88 file usr.sbin/bind/lib/Makefile : 1.14 : 87 : 91 89 descr 90 the normal SHA1Transform() function causes the sparc64 compiler to 91 use excessive amounts of memory attempting to optimise this function. 92 using no optimisation allows the function to build. this hack 93 splits the mathematical functions inside SHA1Transform() into 4 94 separate functions, which avoids the optimiser explosion. 95 kcah 96 97 hack gcc-sparc64-double-arguments 98 mdate 18 Sep 2000 99 who mrg 100 port sparc64 101 file xsrc/xc/lib/fonts/Type1/type1.c : 1.2 : 1461 : 1507 : 1702 : 1709 102 descr 103 in passing 17 "double" arguments to a function, gcc incorrectly tries 104 to use a register that doesn't exist, causing an internal error. this 105 hack changes the way arguments are passed to the function, such that 106 a pointer to a "double" (really an array) is passed, and the old 107 arguments are assigned within the function. 108 kcah 109 110 hack gcc-sparc64-gawk 111 mdate 28 Oct 2000 112 who mrg 113 port sparc64 114 file gnu/usr.bin/gawk/Makefile : 1.11 : 21 : 24 115 descr 116 gcc fails to build gawk properly, due to double floating point bugs. 117 this simple test program will fail to print ``ab'': 118 BEGIN { f = "abc"; print substr(f, 1, 2); exit } 119 the work around is simple: use -msoft-quad-float. 120 kcah 121 122 hack gcc-sparc64-libgcc 123 mdate 01 Dec 2001 124 who jmc 125 port sparc64 126 file gnu/lib/libgcc/Makefile : 1.29 : 24 : 27 127 descr 128 gcc fails to do double -> int conversions correctly without 129 soft-quad-float. Add it in there for the internal conversion 130 routines. This appears to be more than likely emulations bugs than 131 compiler bugs as the assembly looks correct on the surface. 132 133 kcah 134 135 hack gcc-sparc64-iostat-systat 136 mdate 04 Feb 2001 137 who mrg 138 port sparc64 139 file usr.sbin/iostat/Makefile : 1.17 : 7 : 10 140 file usr.bin/systat/Makefile : 1.23 : 6 : 9 141 descr 142 gcc fails to build iostat properly, due to double floating point bugs. 143 results will include "Inf" and other garbage. 144 the work around is simple: use -msoft-quad-float. 145 kcah 146 147 hack gcc-sparc64-nawk 148 mdate 25 Oct 2001 149 who martin 150 port sparc64 151 file usr.bin/awk/Makefile : 1.3 : 13 : 15 152 descr 153 Don't use long double on sparc64. 154 kcah 155 156 hack gcc-sparc64-ntp 157 mdate 12 Nov 2000 158 who martin 159 port sparc64 160 file usr.sbin/ntp/Makefile.inc : 1.5 : 17 : 19 161 descr 162 gcc fails to build ntpd properly, due to double floating point bugs. 163 Ntpd will fail to decode incoming packets; ntpdc -s will report 164 a delay, offset and dispersion of 0.0000 for all peers. 165 the work around is simple: use -msoft-quad-float. 166 kcah 167 168 hack gcc-sparc64-x_dhclient 169 mdate 27 Dec 2000 170 who mrg 171 port sparc64 172 file distrib/utils/x_dhclient/Makefile : 1.6 : 28 : 30 173 descr 174 gcc fails with an ICE while building dhclient's parse.c with -O2 175 and -Os. the work around is simple: don't use -Os. 176 kcah 177 178 hack lint-no-long-double 179 mdate 17 Aug 2001 180 who eeh 181 port sparc64 182 file usr.bin/xlint/lint1/param.h: 1.14 : 183 descr 184 Don't use long double on sparc64. 185 kcah 186 187 hack netstat ieee1394 address printing. 188 mdate 14 Nov 2000 189 who matt 190 file lib/libc/net/getnameinfo.c : 1.32 : 497 : 503 191 descr 192 Because the current implementation of IP over IEEE1394, the 193 fw device address contains more than just the IEEE1394 EUI-64. 194 So when printing out IEEE1394 addresses, ignore the extra stuff. 195 kcah 196 197 hack xterm vs. libterm 198 mdate 01 Aug 2000 199 who jdc 200 file xsrc/xc/programs/xterm/main.c : 1.2 : 3609 : 3614 201 pr 10383 202 descr 203 In order to extend the termcap string over 1023 bytes, a ZZ entry was 204 introduced to point to a memory location containing the full entry. 205 Without this hack, xterm will export a termcap containing the ZZ 206 entry, which will then be ignored by libterm. As xterm modifies the 207 exported termcap, this would cause those modifications to be ignored. 208 kcah 209 210 hack gcc-3-libtelnet 211 mdate 29 Jan 2002 212 who thorpej 213 file lib/libtelnet/Makefile : 1.14 : 14 : 17 214 descr 215 GCC 3.x issues a spurious warning when compiling a Duff's Device 216 with optimization, even though it compiles the file correctly. 217 GCC PR optimization/5230. kerberos.c contains a Duff's Device 218 in the checksum routine. Hack is to build libtelnet w/o optimization 219 if HAVE_GCC3 is set. 220 kcah 221 222 hack gcc-sh3-sed 223 mdate 23 Apr 2002 224 who thorpej 225 file usr.bin/sed/Makefile : 1.9 : 9 : 13 226 descr 227 The in-tree GCC 2.95.3-based compiler ICEs when building 228 with optimization for SuperH. Hack is to biuld with -O0. 229 kcah 230 231 hack gcc-unsigned-compare 232 cdate 09 Mar 2002 233 mdate 18 Mar 2002 234 who bjh21 235 port arm 236 file dist/bind/lib/nameser/ns_parse.c : 1.3 237 file dist/dhcp/minires/ns_parse.c : 1.3 238 file dist/dhcp/omapip/result.c : 1.2 239 file dist/dhcp/server/failover.c : 1.3 240 file gnu/dist/gawk/eval.c : 1.4 241 file gnu/dist/toolchain/bfd/bfd.c : 1.2 242 file gnu/dist/toolchain/bfd/format.c : 1.2 243 file gnu/dist/toolchain/gdb/target.c : 1.2 244 file sys/kern/vfs_subr.c : 1.172 245 descr When checking that a potentially-unsigned enum is >= 0, assign it 246 to an int first. This is necessary to avoid "comparison is always 247 true" warnings with -fshort-enums. Casting to an int really should 248 be enough, but turns out not to be. 249 kcah 250 251 hack wi-at-big-endian-bus 252 cdate 15 Mar 2002 253 who martin 254 file dev/ic/wireg.h 255 descr Add an option to access the underlying bus in big endian byte order 256 to work around deficiencies in bus_space_{read,write}_* macros. 257 Those don't allow the implementation of a proper pcmcia bus space 258 tag. 259 kcah 260 261 hack gcc 2.95.3 -O2 (-fgcse) bug 262 cdate Sun May 5 18:36:04 UTC 2002 263 who tsutsui 264 port macppc 265 file sys/arch/macppc/dev/adb_direct.c: 1.24 : 1895 : 1896 266 pr 16678 267 descr 268 gcc-2.95.3 does generates wrong code on optimization 269 by gcc -O2 (-fgcse), and adb_read_date_time() returns 270 wrong value on cuda system. 271 A null asm statement has been added to avoid this for workaround. 272 kcah 273 274 hack gcc-vax-libbz2 275 mdate 27 Jun 2002 276 who thorpej 277 port vax 278 file lib/libbz2/Makefile 279 descr 280 libbz2 is mis-compiled with optimization with GCC 2.95.3 281 on VAX. -O0 works around this problem. 282 kcah 283 284 hack egcs-pc532-ip6_mroute 285 cdate 09 Jul 2002 286 who simonb 287 port pc532 288 file sys/arch/pc532/conf/Makefile.pc532 : 1.70 289 file sys/arch/pc532/conf/files.pc532 : 1.47 290 descr 291 egcs 1.1.2 gets an "internal error--insn does not satisfy its 292 constraints" error compiling ip6_mroute.c with -O2 or greater. 293 -O1 works around this problem. 294 kcah 295 296 hack SuperH SH5 Toolchain Bugs 297 cdate 11 Jul 2002 298 who scw 299 port sh5 300 file usr.sbin/ndbootd/ndbootd.c : 1.6 301 file usr.sbin/traceroute/traceroute.c : 1.48 302 descr 303 The SuperH SH5 toolchain (2.97-sh5-010522) gets an internal 304 compiler error when assigning a bit-wise inverted value 305 under some circumstances. 306 Work around it by splitting the statement into two. 307 kcah 308 309 hack vax Toolchain bug 310 cdate 28 Mar 2003 311 who he 312 port vax 313 file sys/arch/vax/vax/intvec.S : 1.5 314 descr 315 Workaround for PR toolchain/20924. The assembler apparently 316 tries to range-check byte offsets when it doesn't have 317 sufficient information to make that decision. 318 Workaround uses `brw' instead of `brb' instruction. 319 kcah 320 321 hack specific knowledge of colours in curses code 322 cdate Sun Apr 6 11:05:24 BST 2003 323 who jdc 324 port all 325 file lib/libcurses/color.c : r1.24 326 descr 327 Swap red/blue and yellow/cyan colours for COLOR_OTHER. 328 Fix is to enhance libtermcap to understand terminfo-style % sequences. 329 See also: 330 http://mail-index.netbsd.org/tech-userlevel/2003/04/06/0000.html 331 kcah 332
Indexes created Wed Nov 05 22:09:55 GMT 2025