2 Bugs reported after the release of 3.8.0
3 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5 This file lists bugs that were reported after 3.8.0 was released, both
6 in 3.8.0 and trunk, and that have not been closed. Bugs that have
7 been closed as "fixed" are listed in the top level NEWS file instead.
8 In other words they get listed here first and moved to NEWS if they
11 This file also tracks commits & fixes that should be pushed to the 3.8
16 (B38=2500) 2469 Fix a mixup ..
17 (B38=12930) 12873 n-i-bz: Use -s USR1 instead of -10 to send SIGUSR1 signal
18 (B38=12931) 12874 n-i-bz: shmat of a segment > 4Gb does not work
19 (B38=12933) 12878 n-i-bz: On Darwin, use __NR_poll_nocancel rather than __NR_poll
20 (B38=12934) 12879 + 12882 + 12883 (301281, OSX hang on "system")
21 (B38=12936) 12881 ptrace mips stuff
22 (B38=12938) 12885 (Iop_DPBtoBCD and Iop_BCDtoDPB fix)
23 (B38=2502) 2475 (289584: Handle a reg-reg encoding of MOVAPS)
24 (B38=2503) 2476 (305042: Implement 0F 7F encoding of movq reg-reg)
25 (B38=12940) 12887 (n-i-bz: libmpiwrap.c fixes)
26 (B38=2504) 2477 (305321: Re-enable 'prefetch m8' and 'prefetchw m8')
27 (B38=2505) 2478 (295808: Fix LZCNT and TZCNT properly)
28 (B38=2506) 2479 (305199: Implement QDADD and QDSUB)
29 (B38=2507) 2480 (304867: Implement MOVBE in 32 bit mode)
30 (B38=12941) 12891 (304980: Fix up range checking in ML_(addLineInfo))
31 (B38=12942) 12892 (305513: Fix skipping of block forms)
32 (B38=12943) 12893 (298281: Implement a wrapper for wcslen on Linux)
33 (B38=12944) 12894 (n-i-bz, exp-sgcheck memcpy)
34 (B38=12945) 12897/8 (305690: Handle non-zero sem_*wait() return values correctly)
35 (B38=2508) 2499 (305926: Invalid alignment checks for some AVX instructions)
36 (B38=12946) 12923 (304980: Guard against negative symbol sizes)
37 (B38=12948) 12947 (n-i-bz: volatile in stack check, re clang)
38 (B38=12984) 12964 (306310 3.8.0 release tarball missing some files)
39 (B38=2534) 2517/8 (n-i-bz: fix a couple of union tag-vs-field mixups)
40 (B38=2535) 2531 (306297: ARM: STM<c>.W <Rn>{!}, <registers> too strict)
41 (B38=2536) 2532 (306664: AMD64: Fix PCMPxSTRx variant $0x46)
42 (B38=12985) 12982 (306612: Intercept __GI_memmove)
45 (304035: ARM: uqsub16 shadd16 uhsub8 uhsub16)
47 -------- Bugs brought forward from 3_7_BUGSTATUS.txt --------
49 291310 FXSAVE instruction marks memory as undefined on amd64
50 x86 vs amd64 inconsistency?
51 **is fixed in 3.8.0, but was not marked as such
52 ** no it's not; v38BRANCH shows error, vTRUNK doesn't -- investigate
55 -------- Unfixed in 3.8.0
57 303741 sockaddr* parameter of recvfrom not handled correctly
58 (corruption not detected later on)
61 303877 valgrind doesn't support compressed debuginfo sections.
64 304259 support Darwin 12 (Mac OS 10.8, Mountain Lion)
67 304321 m_signals.c:2314 (sync_signalhandler_from_kernel):
68 Assertion 'tid != 0' failed
69 CLOSED as UNMAINTAINED.
71 304450 time logging tools ?
74 304675 make install-strip fails when cross-compiling
75 w/ one-liner patch; but needs checking (should be more general?
76 assumes auto* version?)
78 304676 regtest hangs on mac OS X (mchelp, coolo_sigaction, system)
79 probably related to 301281 et al. Still valid.
82 305170 linux: add FIOCLEX and FIONCLEX ioctls
83 SYSCALL. w/ patch, looks pretty harmless
85 n-i-bz exp-sgcheck asserts on gcc-4.6.2 generated Dwarf3
87 305957 m_debuginfo/d3basics.c:965 (vgModuleLocal_evaluate_GX):
88 Assertion 'aMin == (Addr)0' failed.
89 QUERIED -- no info so far -- close?
91 306004 Wishlist: port valgrind to sparc
94 -- Sat 1 Sept 2012 10pm
96 306171 Boost.Context appears to cause Valgrind to crash
97 Stack switching; should fix. (Conservative unwind?)
99 306235 unhandled syscall: 429 (freebsd)
100 Need to forward to FreeBSD port maintainers.
102 306299 Internal error when running a winelib application
105 306340 parse_var_DIE confused by DWARF from clang-3.1
106 Might be quite simple to fix; need to investigate.
109 306535 massif: force dump result every X minutes
112 306587 data cache line size is 128 and instruction cache line size is 32
113 PPC; obscure, needs investigatoin
115 306590 not support e500v2 spe 0x216 command
116 PPC SPE insns; with plausible looking patch? Need test cases, and
119 286864 strlen function redirection error
120 Gentoo glibc "miscompilation"; is this important?
122 -- Mon 17 Sept 2012, 10 am
124 valgrind.spec diff at https://bugzilla.mozilla.org/show_bug.cgi?id=750856#c20
125 r13008 to 382 ? check
126 change of leak-check-counting? 307465
127 gkw extra suppression?
128 gcp libnss3.so debuginfo read hang
130 306947 s390x: BFP rounding mode improvements
133 306949 s390x: BFP condition code improvements
136 211352 helgrind races in helgrind's own mythread_wrapper
139 307367 amd64->IR: 0xF2 0x48 0xF 0x5F 0x3E 0xF2 0x48 0xF
140 redundant-REX, should be easy to fix
142 307399 amd64->IR: 0xC4 0x62 0x55 0x2E 0x18 0x44 0x39 0x87
143 (vmaskmovps %ymm11,%ymm5,(%rax))
144 CONDITIONAL STORES -- probably fixed?
146 307415 3.8.1 build error - OSX Lion 10.7.4 duplicate symbol
147 _mach_make_memory_entry
150 307500 Valgrind docs have two <title> tags
151 QUERIED (unclear how to fix)
154 308027 Warning: negative range in .debug_loc section
155 QUERIED -- not sure this needs fixing or not
157 79362 Debug info is lost for .so files when they are dlclose'd
160 308158 running with --leack-check=full makes valgrind exit
161 FreeBSD -- although FreeBSD maintainers may not be aware
163 ------ Thu Oct 11 18:51:26 CEST 2012
165 197915 80-bit floats are not supported on x86 and x86-64
168 299316 Helgrind: hg_main.c:628 (map_threads_lookup): Assertion 'thr'
172 308582 massif: wrong stack usage at exit of threaded program
175 308668 pthread rwlock_rdlock doesn't understand modifications in
176 some cases (Helgrind)
177 Real; no action so far
179 308729 vex x86->IR: unhandled instruction bytes 0xf 0x5
180 HAS_PATCH, but NEEDINFO
182 243232 *Inconsistent Lock Orderings report with trylock (PBOS?)
183 Almost certainly real (+ probably a dup). No action so far.
185 237920 valgrind segfault on fork failure
186 Uninvestigated; filed by JJ
188 308816 valgrind GPF in signal handler context 3.8.1 (allignment
192 309100 s390x: Testcases for extended BFP
195 309430 Self hosting ppc64 encounters a vassert error on operand type
196 No action so far. Sounds a bit strange.
198 309554 Unhandled syscall remap_file_pages (216)
200 Possibly difficult to implement
202 309721 add client request returning current value of a statistic (e.g. Ir)
203 Wishlist. No action yet.
205 309783 Valgrind does not support separate source and build directories
206 Has patch, but needs further checking
207 Does not appear to be much enthusiasm for this
209 309827 *add support for nvidia (nvmem) ioctls
210 A bunch of Android hackery from VladV; needs pulling apart
213 309831 3.8.1: disInstr(arm): unhandled instruction: 0xECECA102
214 (stcl, gdb disassembles to stfp) .. IWMMXT
217 310188 need better diagnostic when mmap() fails with EINVAL
221 310692 [coregrind] Thread scheduling regression?
223 Severity: med Audience: low
225 311093 massif and memcheck fail to redirect malloc calls on
226 statically linked application
229 -- Wed Dec 5 23:41:20 CET 2012
231 311544 sys_fork() calls VG_(do_atfork_pre)() AFTER forking()
234 311649 exhausting fds can leak created files
235 Probably true; NEEDINFO; possible WONTFIX; QUERIED
237 311655 --log-file=FILE leads to apparent fd leak ("Open file
238 descriptor 3: $LOGFILE")
239 Probably true; should fix.
240 Severity: med Audience: low
242 311827 OS X 10.8.2 (eq_SyscallStatus): the 'impossible' happened.
243 OSX10.8 Maybe fixed in the trunk already. QUERIED
245 312989 ioctl handling needs to do POST handling on generic ioctls and
246 needs to handle BPF ioctls
247 OSX10.8 FIX 3.9 (commit)
249 313275 False positives for shared memory semaphores/mutexes/conditions
250 No action so far. Has testcase patches.
251 Potentially important to fix, but does not look easy.
253 313276 helgrind does not recognize shared memory initialized
254 semaphores, mutexes, conditions
255 == 313275, pretty much
256 Severity: med Audience: low
258 313875 cr_libinit.c:183 cri_init: sigaction() failed: Invalid argument
260 Severity: high Audience: low
262 ** [Bug 291310] FXSAVE instruction marks memory as undefined on amd64
263 Severity: med Audience: low
265 314236 OS X 10.8 testsuite build failure: test million.o
266 OSX108 No action so far.
267 Severity: low Audience: med
269 314365 enable VEX to run asm helpers that do callee register saving
270 In progress; not sure whether this is a good idea
272 -- Thu Feb 28 09:54:56 CET 2013
274 315936 UNKNOWN task message [id 3403, to mach_task_self(), reply
275 0x2603] when running fork()ing tests on Mac OS X
278 315952 tun/tap ioctls are not supported
279 QUERIED; leaving open for now
280 Severity: med Audience: low
282 316034 Building valgrind with gcc (4.7, trunk) fails on MacOS X 10.8
285 -- Mon Mar 4 11:36:17 CET 2013
287 294285 --partial-loads-ok does not work for 16-byte SSE loads
288 FIXED (multiple commits)
289 Requires regtest hookup on ppc
291 316382 Support for AMDs FMA4 instructions seem to be missing
293 Severity: high Audience: low
295 316528 Dump for cachegrind
298 316612 A -Wunused-value warning from Clang when including memcheck.h
299 Probably needs doc update
300 Severity: low Audience: med
302 316696 fluidanimate program of parsec 2.1 stuck (do not finish
303 for more than 5 hours)
304 No action so far (DRD specific)
306 Severity: high Audience: low
308 316874 amd64->IR: 0xC4 0xE2 0xF8 0xF2 0x85 0x30 0xFF 0xFF
309 Probably some AMD FMA insn?
310 Severity: high Audience: low
312 316903 memcheck/tests/origin5-bz2 FAIL with gcc-4.5+
314 Severity: med Audience: med
316 317078 _mm_madd_epi16, undefined * 0 should not give uninitialised result
317 True, but probably difficult to fix without perf hit
320 317127 Fedora18/x86_64 --sanity-level=3 : aspacem segment mismatch
321 Diagnosed (BTRFS specific); unclear what to do, if anything
323 317208 vex x86->IR: unhandled instruction bytes: 0xC8 0x0 0x0 0x0
324 "enter $0,$0" created by D language compiler
325 Severity: high Audience: low
327 317381 helgrind warns about xchg vs suppressed store
328 No action so far. Not sure there's an easy fix for this.
329 Severity: med Audience: low
331 317390 Patch for FBLD/FBSTP/FTST
332 Needs cleanup+test cases. Queried.
333 Severity: high Audience: low
335 317441 Add support for AMD's LWP (LightWeight Profiling) architecture
337 Contains plausible infrastructure patch; no insns so far tho
338 This is going to need some working through. Probably OK.
339 Not sure if can make 3.9.
340 Severity: high Audience: low
342 317698 parse_var_DIE: confused by: DW_TAG_compile_unit using
343 Intel 13.0 update 3 compiler
344 Reporter has queried Intel since this might be an ICC bug
346 Severity: med Audience: low
348 317893 massif terminates without any message
349 Probably just a memory limit thing. Close as a dup, but of what?
352 318030 addHRegUse takes a lot of CPU time; band-aid speedup
355 Needs review; would be nice to redo this with a more robust API
357 Thu Apr 11 18:16:04 CEST 2013
359 319237 [PATCH] Added support for ELF injection
362 319274 unhandled syscall: unix:410
365 319306 gdbserver tests call vgdb without path; patch within
366 Needs further discussion?
368 319307 Patches to let test suite run in parallel
370 Wishlist, effectively (has patch)
372 319393 bad rounding in cvtsi2ss instruction
373 Not started. Looks valid. (double rounding)
374 Severity: med Audience: low
376 319419 unhandled syscall: unix:369
379 319547 helgrind gives false positives on rwlocks
380 Needs investigation. Quite possibly a real bug.
381 Severity: med Audience: med
383 319896 vex amd64->IR: 0x8C 0xE0
384 Probably requires proper segment support, but need to check.
385 Severity: high Audience: low
387 320082 disInstr(thumb): unhandled instruction on instruction LDRB
389 LDRB; surely already supported? Queried.
391 Severity: high Audience: low
393 320100 Add nanosecond timer support
394 A bigger patch depends on it, but that has not appeared yet
395 Wishlist, + as yet incomplete
397 320546 valgrind exits with "Assertion 'in_rx' failed"
398 for /system/lib/libva-android.so on Android 4.0.4
399 Need more info (the .so); queried.
400 MAYBE FIX 3.9 (potentially serious)
402 320965 Unrecognised instruction __ieee754_pow_sse2
403 Not sure what this is. Queried.
404 Severity: high Audience: low (??)
406 321396 Add Support for Freescale e500v2 SPE Architecture.
407 Is metabug? -- no patch, no details
409 321400 Add Support for Freescale e5500 Architecture.
412 321401 Add Support for Freescale e6500 Architecture.
415 321681 printf false positive
416 Severity: med Audience: low
418 321836 valgrind cross compilation failure on Ubuntu machine
421 Tue Jul 2 12:15:07 CEST 2013
423 321888 *Unhandled instruction: LDRH (Thumb)
424 Has patch; plausible (needs checking)
427 321892 *Unhandled instruction: LDRSH (Thumb)
428 Has patch; plausible (needs checking)
431 321902 *disInstr(ARM): 0xECECA102
433 MAYBE 3.9 (query vgolubev)
435 322150 Add Support for Cavium Octeon Instructions
436 MIPS64; has patch; assigned to Petar
437 MAYBE 3.9 (query Petar)
439 322256 Provide more callstacks on memory pool related errors
440 Contains plausible patch; needs looking at
441 Severity: low Audience: low
443 322258 No thread notification is printed if new thread reused tid of
444 old thread that triggered last error
445 Contains plausible patch. Is bug fix. (should take)
448 322586 Unknown instruction (bextr) in ld-2.15.so with --march=native on AMD FX-8350
449 AMD specific. Uninvestigated.
450 Severity: high Audience: low (??)
452 322593 Power 8 code to implement Transactional support
453 Implements pass-to-host scheme; not intended to land.
454 Severity: low (right now) Audience: low
456 322621 False positives with pthread_barrier_wait / helgrind
457 Severity: med Audience: low
459 323020 more mysteries in valgrind thread stack handling
461 Severity: med Audience: low
463 323046 Implement ANNOTATE_THREAD_NAME() Hellgrind annotation
464 Request for implementation
465 QUERIED .. is 322254 good enough?
466 Severity: low Audience: med
468 323178 *Unhandled instruction: PLDW register (ARM)
469 Has patch; plausible (needs checking)
472 323179 *Unhandled instruction: PLDW immediate (ARM)
473 Has patch; plausible (needs checking)
476 323268 New calls for custom memory allocators
477 Obscure; needs investigation
478 Severity: low Audience: low
480 323431 vex amd64->IR: 0x8F 0xE8 0x78 0xA2 0xC1 0x40 0xC5 0xFB
482 Severity: high Audience: low (??)
483 FIX 3.9 if can do it at no risk and no CPUID hassle
486 323448 Check for memory overlap in sprintf()-like functions
487 Severity: low Wishlist
489 323464 VALGRIND_MEMPOOL_ALLOC problems with size >= 2^32
490 64-bit uncleanness; no action so far; QUERIED
491 Severity: med Audience: low
493 323777 *Documentation does not warn of sgcheck+darwin incompatibility
496 324050 Helgrind: SEGV because of unaligned stack when using movdqa
497 Needs to be looked at. Is this real, or some Gentoo-caused
499 Severity: high (hg crash) Audience: low
501 324078 Track and show in logs which thread allocated memory
502 Probably sync-event-loss in Helgrind
503 Severity: high (hg correctness bug?) Audience: low
505 324081 3.8.1 Valgrind periodically hangs in long-running application
508 324261 errorcount in xml output remains empty (no error pairs mentioned)
509 Hmm, looks like a real bug.
510 Severity: med Audience: low
512 324486 WARNING: unhandled syscall: 310
514 Hmm, tricky, sys_unshare.
515 Severity: high Audience: low
517 324525 valgrind can not detect any leak
518 PROBABLE INVALID; needs closing.
519 Severity: med Audience: low
521 324809 show nightly regression tests results (and later performance) in a
522 nice html page on valgrind.org
524 Severity: low Audience: low
526 324882 CPUID should depend on VEX capabilities, not on underlying CPU
527 Middle-term cleanup; not urgent
529 Thu Sep 19 10:34:49 CEST 2013