HACKS revision 1.2
1 # $NetBSD: HACKS,v 1.2 2003/03/28 21:15:20 he 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 specific knowledge of xterm escapes in curses code 31 cdate Wed, 11 Sep 2002 18:23:25 +0300 32 who jdc 33 port all 34 file lib/libcurses/setterm.c : 1.33 35 descr 36 Recognise that "\E[m" will turn off all attributes for xterm-like 37 terminals. Without this hack, we can't tell when turning off one 38 attribute affects another (which we check by comparing sequences). 39 kcah 40 41 hack gcc 2.95/vax cannot handle __builtin_ffs() 42 cdate Fri Aug 23 21:31:15 CEST 2002 43 who ragge 44 port vax 45 file sys/lib/libkern/libkern.h : 1.42 46 descr 47 __builtin_ffs() is ifdef'd away if __vax__. Matt Thomas has added 48 this feature to GCC 3 so it can be removed when switching. 49 kcah 50 51 hack gcc 2.95.3 -O2 bug 52 cdate Sun Jan 6 23:17:28 UTC 2002 53 who lukem 54 port alpha 55 file sbin/newfs/mkfs.c : (all?) 56 pr 15156 57 descr 58 newfs built with gcc -O2 dumps core part-way through. 59 using -O1 avoids the problem 60 kcah 61 62 hack binutil-2.11-sparc64-pltrela 63 mdate 14 Aug 2001 64 who eeh 65 port sparc64 66 file libexec/ld.elf_so/reloc.c: 1.41 : 597 : 619 67 descr 68 The first for PLT entries are reserved. There is some disagreement 69 whether they should have associated relocation entries. Both the 70 SPARC 32-bit and 64-bit ELF specifications say that they should have 71 relocation entries, but the 32-bit SPARC binutils do not generate 72 them, and now the 64-bit SPARC binutils have stopped generating them 73 too. 74 75 To provide binary compatibility, we will check the first entry, if it 76 is reserved it should not be of the type JMP_SLOT. If it is JMP_SLOT, 77 then the 4 reserved entries were not generated and our index is 4 78 entries too far, so we frob the rela pointer. 79 kcah 80 81 hack gcc-arm32-schedule 82 mdate 04 May 2000 83 who is 84 file lib/libc/stdlib/strtoull.c : 1.3 : 78 : 81 85 file lib/libc/stdlib/strtouq.c : 1.15 : 73 : 76 86 pr 9613 87 descr 88 strtoull() was returning a wrong result for small numbers with bit 31 89 set. This is a gcc/arm32 compiler bug in gcc < 2.95.2. Adding 90 (void)&acc; works around it. 91 kcah 92 93 hack gcc-sparc64-memorysize 94 mdate 17 Sep 2000 95 who mrg 96 port sparc64 97 file lib/libc/hash/sha1.c : 1.6 : 78 : 129 : 162 : 189 98 file sys/lib/libkern/sha1.c : 1.6 : 78 : 129 : 162 : 189 99 file usr.sbin/bind/lib/Makefile : 1.14 : 87 : 91 100 descr 101 the normal SHA1Transform() function causes the sparc64 compiler to 102 use excessive amounts of memory attempting to optimise this function. 103 using no optimisation allows the function to build. this hack 104 splits the mathematical functions inside SHA1Transform() into 4 105 separate functions, which avoids the optimiser explosion. 106 kcah 107 108 hack gcc-sparc64-double-arguments 109 mdate 18 Sep 2000 110 who mrg 111 port sparc64 112 file xsrc/xc/lib/fonts/Type1/type1.c : 1.2 : 1461 : 1507 : 1702 : 1709 113 descr 114 in passing 17 "double" arguments to a function, gcc incorrectly tries 115 to use a register that doesn't exist, causing an internal error. this 116 hack changes the way arguments are passed to the function, such that 117 a pointer to a "double" (really an array) is passed, and the old 118 arguments are assigned within the function. 119 kcah 120 121 hack gcc-sparc64-gawk 122 mdate 28 Oct 2000 123 who mrg 124 port sparc64 125 file gnu/usr.bin/gawk/Makefile : 1.11 : 21 : 24 126 descr 127 gcc fails to build gawk properly, due to double floating point bugs. 128 this simple test program will fail to print ``ab'': 129 BEGIN { f = "abc"; print substr(f, 1, 2); exit } 130 the work around is simple: use -msoft-quad-float. 131 kcah 132 133 hack gcc-sparc64-libgcc 134 mdate 01 Dec 2001 135 who jmc 136 port sparc64 137 file gnu/lib/libgcc/Makefile : 1.29 : 24 : 27 138 descr 139 gcc fails to do double -> int conversions correctly without 140 soft-quad-float. Add it in there for the internal conversion 141 routines. This appears to be more than likely emulations bugs than 142 compiler bugs as the assembly looks correct on the surface. 143 144 kcah 145 146 hack gcc-sparc64-iostat-systat 147 mdate 04 Feb 2001 148 who mrg 149 port sparc64 150 file usr.sbin/iostat/Makefile : 1.17 : 7 : 10 151 file usr.bin/systat/Makefile : 1.23 : 6 : 9 152 descr 153 gcc fails to build iostat properly, due to double floating point bugs. 154 results will include "Inf" and other garbage. 155 the work around is simple: use -msoft-quad-float. 156 kcah 157 158 hack gcc-sparc64-nawk 159 mdate 25 Oct 2001 160 who martin 161 port sparc64 162 file usr.bin/awk/Makefile : 1.3 : 13 : 15 163 descr 164 Don't use long double on sparc64. 165 kcah 166 167 hack gcc-sparc64-ntp 168 mdate 12 Nov 2000 169 who martin 170 port sparc64 171 file usr.sbin/ntp/Makefile.inc : 1.5 : 17 : 19 172 descr 173 gcc fails to build ntpd properly, due to double floating point bugs. 174 Ntpd will fail to decode incoming packets; ntpdc -s will report 175 a delay, offset and dispersion of 0.0000 for all peers. 176 the work around is simple: use -msoft-quad-float. 177 kcah 178 179 hack gcc-sparc64-x_dhclient 180 mdate 27 Dec 2000 181 who mrg 182 port sparc64 183 file distrib/utils/x_dhclient/Makefile : 1.6 : 28 : 30 184 descr 185 gcc fails with an ICE while building dhclient's parse.c with -O2 186 and -Os. the work around is simple: don't use -Os. 187 kcah 188 189 hack lint-no-long-double 190 mdate 17 Aug 2001 191 who eeh 192 port sparc64 193 file usr.bin/xlint/lint1/param.h: 1.14 : 194 descr 195 Don't use long double on sparc64. 196 kcah 197 198 hack netstat ieee1394 address printing. 199 mdate 14 Nov 2000 200 who matt 201 file lib/libc/net/getnameinfo.c : 1.32 : 497 : 503 202 descr 203 Because the current implementation of IP over IEEE1394, the 204 fw device address contains more than just the IEEE1394 EUI-64. 205 So when printing out IEEE1394 addresses, ignore the extra stuff. 206 kcah 207 208 hack xterm vs. libterm 209 mdate 01 Aug 2000 210 who jdc 211 file xsrc/xc/programs/xterm/main.c : 1.2 : 3609 : 3614 212 pr 10383 213 descr 214 In order to extend the termcap string over 1023 bytes, a ZZ entry was 215 introduced to point to a memory location containing the full entry. 216 Without this hack, xterm will export a termcap containing the ZZ 217 entry, which will then be ignored by libterm. As xterm modifies the 218 exported termcap, this would cause those modifications to be ignored. 219 kcah 220 221 hack gcc-3-libtelnet 222 mdate 29 Jan 2002 223 who thorpej 224 file lib/libtelnet/Makefile : 1.14 : 14 : 17 225 descr 226 GCC 3.x issues a spurious warning when compiling a Duff's Device 227 with optimization, even though it compiles the file correctly. 228 GCC PR optimization/5230. kerberos.c contains a Duff's Device 229 in the checksum routine. Hack is to build libtelnet w/o optimization 230 if HAVE_GCC3 is set. 231 kcah 232 233 hack gcc-sh3-sed 234 mdate 23 Apr 2002 235 who thorpej 236 file usr.bin/sed/Makefile : 1.9 : 9 : 13 237 descr 238 The in-tree GCC 2.95.3-based compiler ICEs when building 239 with optimization for SuperH. Hack is to biuld with -O0. 240 kcah 241 242 hack gcc-unsigned-compare 243 cdate 09 Mar 2002 244 mdate 18 Mar 2002 245 who bjh21 246 port arm 247 file dist/bind/lib/nameser/ns_parse.c : 1.3 248 file dist/dhcp/minires/ns_parse.c : 1.3 249 file dist/dhcp/omapip/result.c : 1.2 250 file dist/dhcp/server/failover.c : 1.3 251 file gnu/dist/gawk/eval.c : 1.4 252 file gnu/dist/toolchain/bfd/bfd.c : 1.2 253 file gnu/dist/toolchain/bfd/format.c : 1.2 254 file gnu/dist/toolchain/gdb/target.c : 1.2 255 file sys/kern/vfs_subr.c : 1.172 256 descr When checking that a potentially-unsigned enum is >= 0, assign it 257 to an int first. This is necessary to avoid "comparison is always 258 true" warnings with -fshort-enums. Casting to an int really should 259 be enough, but turns out not to be. 260 kcah 261 262 hack wi-at-big-endian-bus 263 cdate 15 Mar 2002 264 who martin 265 file dev/ic/wireg.h 266 descr Add an option to access the underlying bus in big endian byte order 267 to work around deficiencies in bus_space_{read,write}_* macros. 268 Those don't allow the implementation of a proper pcmcia bus space 269 tag. 270 kcah 271 272 hack gcc 2.95.3 -O2 (-fgcse) bug 273 cdate Sun May 5 18:36:04 UTC 2002 274 who tsutsui 275 port macppc 276 file sys/arch/macppc/dev/adb_direct.c: 1.24 : 1895 : 1896 277 pr 16678 278 descr 279 gcc-2.95.3 does generates wrong code on optimization 280 by gcc -O2 (-fgcse), and adb_read_date_time() returns 281 wrong value on cuda system. 282 A null asm statement has been added to avoid this for workaround. 283 kcah 284 285 hack gcc-vax-libbz2 286 mdate 27 Jun 2002 287 who thorpej 288 port vax 289 file lib/libbz2/Makefile 290 descr 291 libbz2 is mis-compiled with optimization with GCC 2.95.3 292 on VAX. -O0 works around this problem. 293 kcah 294 295 hack egcs-pc532-ip6_mroute 296 cdate 09 Jul 2002 297 who simonb 298 port pc532 299 file sys/arch/pc532/conf/Makefile.pc532 : 1.70 300 file sys/arch/pc532/conf/files.pc532 : 1.47 301 descr 302 egcs 1.1.2 gets an "internal error--insn does not satisfy its 303 constraints" error compiling ip6_mroute.c with -O2 or greater. 304 -O1 works around this problem. 305 kcah 306 307 hack SuperH SH5 Toolchain Bugs 308 cdate 11 Jul 2002 309 who scw 310 port sh5 311 file usr.sbin/ndbootd/ndbootd.c : 1.6 312 file usr.sbin/traceroute/traceroute.c : 1.48 313 descr 314 The SuperH SH5 toolchain (2.97-sh5-010522) gets an internal 315 compiler error when assigning a bit-wise inverted value 316 under some circumstances. 317 Work around it by splitting the statement into two. 318 kcah 319 320 hack vax Toolchain bug 321 cdate 28 Mar 2003 322 who he 323 port vax 324 file sys/arch/vax/vax/intvec.S : 1.5 325 descr 326 Workaround for PR toolchain/20924. The assembler apparently 327 tries to range-check byte offsets when it doesn't have 328 sufficient information to make that decision. 329 Workaround uses `brw' instead of `brb' instruction. 330 kcah 331
Indexes created Wed Nov 05 22:09:55 GMT 2025