From: Julian Seward Date: Sun, 11 May 2008 08:57:01 +0000 (+0000) Subject: Update. X-Git-Tag: svn/VALGRIND_3_4_0~597 X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=04859bfbbad2eec403dd6ae812c523d82a90fb1a;p=thirdparty%2Fvalgrind.git Update. git-svn-id: svn://svn.valgrind.org/valgrind/trunk@8042 --- diff --git a/docs/internals/3_3_BUGSTATUS.txt b/docs/internals/3_3_BUGSTATUS.txt index f3a02a6aca..5a04cd944c 100644 --- a/docs/internals/3_3_BUGSTATUS.txt +++ b/docs/internals/3_3_BUGSTATUS.txt @@ -13,6 +13,13 @@ PRI: 33 = fix this for 3.3.1 ---------- Bugs reported after (in) 3.3.0 -------------------------- +*** definitely fix +### merge fix to 3.3 branch +??? marginal +NNN won't fix + +-- Last updated 9 May 08 + tidy up static int linux_kernel_2_6_22(void) (syswrap-linux.c) check r7322 is merged to branch @@ -27,14 +34,9 @@ get rid of pointless constraint !(--db-attach=yes && utimensat syswrap bug (Kegel, users, 24 Feb) -r7454: Intercept pthread_rwlock_try{rd,wr}lock. Fixes #158212. openoffice on suse 10.3 readlink stack problem -Core2/ssse3 support in the trunk: vx1808, vx1809, vx1810, - r7384, r7385, r7387, r7389. -(in essence #155528) - Consider potential space/time effects of r7383 (see reply to the commit message). @@ -44,116 +46,144 @@ properly (see replies to the commit message). (now fixed ...) re Keller vpath bug, don't use attached patch; use Ashley's one instead (on @dev ?) ---- start 8 May 08 --- - disInstr(ppc): unhandled instruction: 0x7D295F1E (users@, 29 Feb) +*** 64 bit tweak ("Solomon, Bernard", user@, 25 Mar, > 1 post the same day) disInstr(ppc): unhandled instruction: 0x7C1F42A6 primary 31(0x1F), secondary 678(0x2A6) "Gupta Richa", users@, 3 Apr 08 -92456: Tracing the origin of uninitialised memory - -149878 add (proper) check for calloc integer overflow - (possibly with a viable patch now) - -160595 Why my program hang-up when work with valgrind ? - (no info so far) +--- start 27 March 08 --- -160872 race errors reported where libpthread.so - (no response from reporter) +92456: Tracing the origin of uninitialised memory + FIXED -160907 Support for a couple of recent Linux syscalls, syscall wrapper - bugfixes (doesn't compile on ppc32/64-linux) +126389 (fxrstr) vex x86->IR: unhandled instruction bytes: +### 0xF 0xAE 0x4D 0xF0 + == 158525 Valgrind doesn't recognise fxrstor instruction + FIXED trunk r1836, (test) r1837 + MERGE -160954 ppc build of vagrind crashes with illegal instruction +149878 add (proper) check for calloc integer overflow +*** (possibly with a viable patch now) + FIXME -160956 PATCH: mallinfo implementation +152818 (repz lodsb) vex x86->IR: unhandled instruction bytes: +*** 0xF3 0xAC 0xFC 0x9C -161199 Valgrind restarts my application without my app crashing, The - legend at the end of the file indicates that it may be a - valgind issue (probably bogus, should close as invalid) +153196 (repnz cmpsb) vex x86->IR: unhandled instruction bytes: 0xF2 +### 0xA6 0x89 0x45 + FIXED trunk r1834 -161251 ppc Power-QUICC-III (8555, e500 core) not supported: isel - instruction +155528 valgrind cupinfo reports a CPU, which does not reflect + valgrind's instruction set. + Core2/ssse3 support in the trunk: + vx1808, vx1809, vx1810, + r7384, r7385, r7387, r7389. + (in essence #155528) + FIXED, MERGE -161285 Patch -- support for eventfd() syscall +156065 VALGRIND_MEMPOOL_ALLOC vs. zero-size allocs triggers false +*** "overlapping" error + FIXME -161374 SYS_readdir syscall not implemented +156404 Unable to start oocalc under memcheck on openSUSE 10.3 (64-bit) +??? POTENTIALLY SERIOUS (need to extend root stack before syscall) -161378 illegal opcode in debug libm +157665 Valgrind fails on shmdt(0) after shmat to 0 in X int10 code +### FIXED r7399, MERGE -161487 number of suppressions files is limited to 10 +157748 valgrind seems to not recognize what would be sse2 +### instructions + FIXED vx1835, MERGE -161609 Results returned by VG_(get_StackTrace)() do not match - its documentation +158004 "Assertion 'szB == 32 || szB == 128' failed" with valgrind +*** cross-compiled for ppc32 + Easy to fix? -161628 exp-drd/tests/sigalrm triggers an assertion failure in - Valgrind's core on ppc64. +158042 --db-attach at invalid free() gives broken stack trace on + x86_64 ---- end 8 May 08 --- +158212 Locks taken by pthread_rwlock_tryrdlock and +### pthread_rwlock_trywrlock not recognised by helgrind + r7454: Intercept pthread_rwlock_try{rd,wr}lock. Fixes #158212. + MERGE ---- start 27 March 08 --- +158425 sys_poll incorrectly emulated when RES==0 +*** PATCH + test case provided -156065 VALGRIND_MEMPOOL_ALLOC vs. zero-size allocs triggers false -"overlapping" error +158730 ppc: illegal opcode crash on instruction fctiwz +??? Not sure this does have anything to do with fctiwz. + Queried reporter. -156404 Unable to start oocalc under memcheck on openSUSE 10.3 (64-bit) +158744 Valgrind fails to handle interlocked instructions on extended +*** registers (r8, etc) on amd64 (vex amd64->IR: unhandled + instruction bytes: 0xF0 0x41 0xF 0xC0 0x87 0xB8) + Patch posted; waiting for response. -157665 Valgrind fails on shmdt(0) after shmat to 0 in X int10 code +159285 unhandled syscall:25 +??? (no test case provided) -126389 (fxrstr) vex x86->IR: unhandled instruction bytes: 0xF 0xAE -0x4D 0xF0 -158525 Valgrind doesn't recognise fxrstor instruction +159452 unhandled ioctl 0x8B01 on "valgrind iwconfig" +??? (queried user on documentation for this ioctl) +159501 incorrect handling of ALSA ioctls +*** (has two patches, looks reasonable, checking for update) -157748 New: valgrind seems to not recognize what would be sse2 -instructions +159659 scas does not use ES segment selector +NNN -158004 "Assertion 'szB == 32 || szB == 128' failed" with valgrind -cross-compiled for ppc32 +159720 Opcode 0xFF 0x58 0xEB 0x5 unhandled on x86 +NNN (this is a "lcall" or callf -- far call) + (no test case) -158042 --db-attach at invalid free() gives broken stack trace on -x86_64 +160595 Why my program hang-up when work with valgrind ? +NNN (no info so far, probably a threading bug in client) -149878 add (proper) check for calloc integer overflow +160872 race errors reported where libpthread.so +??? (might be a problem with pthread_setspecific) -158212 Locks taken by pthread_rwlock_tryrdlock and -pthread_rwlock_trywrlock not recognised by helgrind +160907 Support for a couple of recent Linux syscalls, syscall wrapper +*** bugfixes (doesn't compile on ppc32/64-linux) -158425 sys_poll incorrectly emulated when RES==0 +160954 ppc build of valgrind crashes with illegal instruction +??? e500 specific instruction -158730 ppc: illegal opcode crash on instruction fctiwz +160956 PATCH: mallinfo implementation +??? fixed. merge? difficult; the patch is in so many commits -158744 Valgrind fails to handle interlocked instructions on extended -registers (r8, etc) on amd64 (vex amd64->IR: unhandled instruction -bytes: 0xF0 0x41 0xF 0xC0 0x87 0xB8) +161251 ppc Power-QUICC-III (8555, e500 core) not supported: isel +??? instruction (it's "just" a cmov) -153196 (repnz cmpsb) vex x86->IR: unhandled instruction bytes: 0xF2 -0xA6 0x89 0x45 +161285 Patch -- support for eventfd() syscall +??? Allegedly fixed, but what rev? -159285 unhandled syscall:25 +161374 SYS_readdir syscall not implemented +??? readdir is pretty prehistoric -159452 unhandled ioctl 0x8B01 on "valgrind iwconfig" +161378 illegal opcode in debug libm + == 160136 +### FIXED vx1826 -159501 incorrect handling of ALSA ioctls +161487 number of suppressions files is limited to 10 +??? Patch exists -159659 scas does not use ES segment selector +161609 Results returned by VG_(get_StackTrace)() do not match +*** its documentation -159720 Opcode 0xFF 0x58 0xEB 0x5 unhandled on x86 +--- end 8 May 08 --- ---- end 27 March 08 --- -r1816 (x86 CondNS after SUBB) -r1817 (amd64 CondNS after SUBB) -r1833 (both, CondNZ after SUBW) +### r1816 (x86 CondNS after SUBB) +### r1817 (amd64 CondNS after SUBB) +### r1833 (both, CondNZ after SUBW) -!!ppc32-linux memcheck craps out on some Altivec code! -(none/tests/ppc32/jm-insns -a -v) +*** !!ppc32-linux memcheck craps out on some Altivec code! + (none/tests/ppc32/jm-insns -a -v) -!!massif segfaults on ooo on amd64 +*** !!massif segfaults on ooo on amd64 r7795 (handle r-x mapped .bss sections - merge to trunk)