From: Julian Seward Date: Sun, 2 Nov 2008 23:46:32 +0000 (+0000) Subject: Add summaries for more than 60 bugs logged in bugzilla since 3.3.1 was X-Git-Tag: svn/VALGRIND_3_4_0~170 X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=79765330c3686f17970145ee3c7f85d47f541704;p=thirdparty%2Fvalgrind.git Add summaries for more than 60 bugs logged in bugzilla since 3.3.1 was released. Looking at this lot, you'd get the impression the system is so bug-riddled it's amazing it works at all. git-svn-id: svn://svn.valgrind.org/valgrind/trunk@8724 --- diff --git a/docs/internals/3_3_BUGSTATUS.txt b/docs/internals/3_3_BUGSTATUS.txt index cc556a801d..52e0a7d966 100644 --- a/docs/internals/3_3_BUGSTATUS.txt +++ b/docs/internals/3_3_BUGSTATUS.txt @@ -10,102 +10,201 @@ WF = waiting for feedback from bug reporter PRI: 33 = fix this for 3.3.1 Vfd = fix has been verified on 3.3.X branch +*** definitely fix +### merge fix to 3.3 branch +??? marginal +NNN won't fix +Vfd fix has been verified on 3.3.X branch +Fxd fixed ----------- Bugs reported after (in) 3.3.0 -------------------------- - -*** definitely fix -### merge fix to 3.3 branch -??? marginal -NNN won't fix -Vfd fix has been verified on 3.3.X branch - --- Last updated 9 May 08 - -get rid of pointless constraint !(--db-attach=yes && ---trace-children=yes) (users, 14 feb) - -Consider potential space/time effects of r7383 (see reply to -the commit message). - -Either back out r7353, or fix the resulting signedness problem -properly (see replies to the commit message). (now fixed ...) +---------- Bugs reported after (in) 3.3.1 -------------------------- re Keller vpath bug, don't use attached patch; use Ashley's one instead (on @dev ?) -*** -disInstr(ppc): unhandled instruction: 0x7D295F1E (users@, 29 Feb) -(0x7D295F1E == isel r9,r9,r11,28) - -64 bit tweak ("Solomon, Bernard", user@, 25 Mar, > 1 post the same day) -FIXED r8048. (don't merge) - -NNN disInstr(ppc): unhandled instruction: 0x7C1F42A6 primary 31(0x1F), secondary 678(0x2A6) "Gupta Richa", users@, 3 Apr 08 dis_proc_ctl(ppc)(mfspr,SPR)(0x11F) (0x7C1F42A6 == mfsprg r0,7) ---- start 27 March 08 --- -149878 add (proper) check for calloc integer overflow -*** (possibly with a viable patch now) - FIXME +BUG# STAT TRUNK WHAT + +n-i-bz Fixd 8048 Make return types for some of the requests 64-bit clean + + +92456 Fixd many Tracing the origin of uninitialised memory + +106497 Fixd ???? Valgrind does not demangles some C++template symbols +162222 Fixd ???? ==106497 + +126255 Wont pend vex x86->IR: 0xDF 0x75 (fbstp) (x87 BCD stores) + +140939 MED pend --track-fds reports leakage of stdout/in/err and + doesn't respect -q, w/ patch + +149878 LOW pend (proper) check for calloc integer overflow + (possibly with a viable patch now) + +151612 HIGH pend Suppression with "..." (w/ good patch) + +156404 Fixd 8712/3 Unable to start oocalc under memcheck on openSUSE + 10.3 (64-bit) + +156065 LOW pend VALGRIND_MEMPOOL_ALLOC vs. zero-size allocs + triggers false "overlapping" error + +158042 WF pend --db-attach at invalid free() gives broken stack + trace on x86_64 + Can't repro; reported queried + +158730 Wont pend ppc: illegal opcode crash on instruction fctiwz + Not sure this does have anything to do with fctiwz. + VEX generates code that doesn't run on e500 + +159285 WF pend unhandled syscall:25 + requested test case + +159452 ??? pend unhandled ioctl 0x8B01 on "valgrind iwconfig" + w/ plausible-looking patch + +159501 ??? pend incorrect handling of ALSA ioctls + (has two patches, looks reasonable) + +159659 Wont pend scas does not use ES segment selector + +159720 Wont pend Opcode 0xFF 0x58 0xEB 0x5 unhandled on x86 + (this is a "lcall" or callf -- far call) + (no test case) + +160595 Wont pend Why my program hang-up when work with valgrind ? + (no info so far, probably a threading bug in client) + +160872 ??? pend race errors reported where libpthread.so + (might be a problem with pthread_setspecific) + +160954 ??? pend ppc build of valgrind crashes with illegal instruction + (isel) + == 161251 + +160956 Fixd 7888 mallinfo implementation, w/ patch + +161374 WF pend SYS_readdir syscall not implemented + will fix if reporter supplies a test case -156065 VALGRIND_MEMPOOL_ALLOC vs. zero-size allocs triggers false -*** "overlapping" error - FIXME +161609 ??? pend Results returned by VG_(get_StackTrace)() do not + match its documentation -156404 Unable to start oocalc under memcheck on openSUSE 10.3 (64-bit) -??? openoffice on suse 10.3 readlink stack problem - POTENTIALLY SERIOUS (need to extend root stack before syscall) +162020 ??? pend Valgrinding an empty/zero-byte file crashes valgrind -158042 --db-attach at invalid free() gives broken stack trace on - x86_64 +162092 HIGH pend Valgrind fails to start gnome-system-monitor -158730 ppc: illegal opcode crash on instruction fctiwz -??? Not sure this does have anything to do with fctiwz. - VEX generates code that doesn't run on e500 +162229 LOW pend VALGRIND_DO_LEAK_CHECK emits false positive -159285 unhandled syscall:25 -??? (no test case provided) +162456 ??? pend valgrind-3.2.3 on ppc fails + problem undetermined (probably close) -159452 unhandled ioctl 0x8B01 on "valgrind iwconfig" -??? (queried user on documentation for this ioctl) +162630 Fixd many hg_main.c:926 (mk_SHVAL_fail): the 'impossible' happened. -159501 incorrect handling of ALSA ioctls -*** (has two patches, looks reasonable, checking for update) +162718 ??? pend x86: avoid segment selector 0 in sys_set_thread_area() + decide if this is a problem or not -159659 scas does not use ES segment selector -NNN +162819 ??? 8285/6 malloc_free_fill test doesn't pass on glibc2.8 x86 + need to look at this -159720 Opcode 0xFF 0x58 0xEB 0x5 unhandled on x86 -NNN (this is a "lcall" or callf -- far call) - (no test case) +163253 ??? pend canonicaliseSymtab forgot some fields in DiSym + is this correct? -160595 Why my program hang-up when work with valgrind ? -NNN (no info so far, probably a threading bug in client) +163794 Fixd 8228 assertion failure with "--track-origins=yes" -160872 race errors reported where libpthread.so -??? (might be a problem with pthread_setspecific) +163910 ??? pend Typo in Helgrind's Documentation +163933 Fixd 8229 sigcontext.err and .trapno must be set together -160954 ppc build of valgrind crashes with illegal instruction -*** 0x7D6B4F9E == isel r11,r11,r9,30 - == 161251 +163955 Fixd 8722 remove pointless constraint !(--db-attach=yes && + --trace-children=yes) -161374 SYS_readdir syscall not implemented -??? readdir is pretty prehistoric +164298 Wont pend Valgrind asserts on finitel with long double using + gcc-4.3 -- closed as wontfix -161609 Results returned by VG_(get_StackTrace)() do not match -*** its documentation +164476 Fixd 8263,6 Missing kernel module loading system calls +**** But need to move non-generic code to Linux section ---- end 8 May 08 --- +164485 ??? pend VG_N_SEGNAMES and VG_N_SEGMENTS are (still) too small ---- start 28 May 08 --- +164669 HIGH pend SVN regression: mmap() drops posix file locks ---- end 28 May 08 --- +165417 ??? pend helgrind/tests/hg05_race2: VG_(get_data_description)() + succeeds on x86 and amd64 but fails on ppc + +165468 ??? pend Inconsistent formatting in memcheck manual + +165598 ??? pend fixed mmap() regions (via MAP_FIXED) cause various + strange failures + (I think this should be closed as invalid) + +166355 ??? pend unshare syscall not supported (amd64, #272) + +166581 Fixd 8704 Callgrind output corruption when program forks + +167212 ??? pend Valgrind does not attach to GDB correctly, traceback + spits garbage + +167288 ??? pend Patch file for missing system calls on Cell BE + +167483 ??? pend __libc_freeres error valgrind with shared libraries with + multiple --rpath (w/ repro case) + +167640 HIGH pend un-efficient memory management cause segmentation fault + w/ test case + +167676 ??? pend Valgrind loads wrong threading library on Red Hat 9 + +167700 ??? pend vex x86->IR: unhandled instruction bytes: 0xD5 0x36 0x5B + 0xC3 (aad) + +168538 ??? pend Poor detection of stacks for clone can lead to + bad results, w/ patch + Looks plausible + +168943 ??? pend unsupported scas instruction pentium + +169539 WF pend mremap fails on specific address on 32-bit memcheck + w/ test case + (not clear if this is still valid) + +169562 ??? pend memcheck warns on some unaligned bcopy()s + w/ test case + +169957 ??? pend Bad report about stack memory after long jump + from signal handler, w/ test case + +170510 ??? pend Don't warn about ioctl of size 0 without + direction hint, w/ patch + +171627 ??? pend Valgrind macros change program behavior + +171645 ??? WF Unrecognised instruction (MOVSD, non-binutils + encoding) (patch proposed, waiting for feedback) + +172417 Fixd vx1867 x86->IR: 0x82 ... (dun, vx1867) + +172563 Fixd vx???? amd64->IR: 0xD9 0xF5 - fprem1 + +173099 ??? pend .lds linker script generation error + w/ plausible patch + +173177 Fixd 8720 [x86_64] WARNING: unhandled syscall: 125/126/179 + (capget/capset/quotactl) + + + +---------- Bugs fixed in 3.3.1 ------------------------------------- + +///////////////////////////////////////////////////////////////// +// +// Changes in 3.3.1 only below here r1816 (x86 CondNS after SUBB) r1817 (amd64 CondNS after SUBB) @@ -133,7 +232,6 @@ r7350 r8080 Mgd n-i-bz intercept memcpy in 64-bit ld.so's r8047 r8148 Mgd n-i-bz Fix wrappers for sys_{futimesat,utimensat} -many wontfix Vfd 92456 Tracing the origin of uninitialised memory vx1836/7 vx1845 Vfd 126389 vex x86->IR: 0xF 0xAE (FXRSTOR) == 158525 @@ -167,8 +265,6 @@ vx1848 vx1849 Vfd 158744 vex amd64->IR: 0xF0 0x41 0xF 0xC0 (xaddb) r7917,8051 OR r7942,r7983,r8051 r8148 Mgd 160907 Support for a couple of recent Linux syscalls -r7888 pending 33 160956 PATCH: mallinfo implementation - r7917 r8148 Mgd 161285 Patch -- support for eventfd() syscall vx1826 vx1842 Vfd 161378 illegal opcode in debug libm (FUCOMPP) @@ -193,6 +289,4 @@ various r8148 Mgd 162663 signalfd_wrapper fails on 64bit linux r8154 r8155 Mgd n-i-bz Massif/Cachegrind output corruption when programs fork -r8704 pending 166581 Callgrind output corruption when program forks - vx1853 vx1854 33 n-i-bz regalloc: don't incorrectly omit spill stores