Home | History | Annotate | Download | only in internals
      1 
      2 ---------- Legend --------------------------------------------------
      3 
      4 n-i-bz  = not in bugzilla
      5 pending = is scheduled to be fixed (or at least considered) on this branch
      6 wontfix = will not fix on this branch
      7 many    = fix composed of many commits
      8 WF      = waiting for feedback from bug reporter
      9 
     10 PRI: 33  = fix this for 3.3.1
     11      Vfd = fix has been verified on 3.3.X branch
     12 
     13 ***  definitely fix
     14 ###  merge fix to 3.3 branch
     15 ???  marginal
     16 NNN  won't fix
     17 Vfd  fix has been verified on 3.3.X branch
     18 Fxd  fixed
     19 
     20 ---------- Bugs reported after (in) 3.3.1 --------------------------
     21 
     22 LOW/MED/HIGH describe impact of bug
     23 HIGH - major loss of functionality, no workaround
     24 MED  - loss of functionality, not critical or workaround possible
     25 LOW  - minor incorrectness, ought-to-fix etc
     26 
     27 re Keller vpath bug, don't use attached patch; use Ashley's one
     28 instead (on @dev ?)
     29 
     30 disInstr(ppc): unhandled instruction: 0x7C1F42A6
     31                  primary 31(0x1F), secondary 678(0x2A6)
     32  "Gupta Richa", users@, 3 Apr 08
     33  dis_proc_ctl(ppc)(mfspr,SPR)(0x11F)
     34  (0x7C1F42A6 == mfsprg  r0,7)
     35 
     36 
     37 BUG#    STAT  TRUNK  WHAT
     38 
     39 // not fixed in 3.4.0 -- triaged, but low prio
     40 
     41 126255  Wont  pend   vex x86->IR: 0xDF 0x75 (fbstp) (x87 BCD stores)
     42 
     43 140939  MED   pend   --track-fds reports leakage of stdout/in/err and
     44                      doesn't respect -q, w/ patch
     45 
     46 149878  LOW   pend   (proper) check for calloc integer overflow
     47                      (possibly with a viable patch now)
     48 
     49 156065  LOW   pend   VALGRIND_MEMPOOL_ALLOC vs. zero-size allocs
     50                      triggers false "overlapping" error
     51 
     52 158042  WF,LOW pend  --db-attach at invalid free() gives broken stack
     53                      trace on x86_64
     54                      Can't repro; reported queried
     55 
     56 158730  Wont  pend   ppc: illegal opcode crash on instruction fctiwz
     57                      Not sure this does have anything to do with fctiwz.
     58                      VEX generates code that doesn't run on e500
     59 
     60 159501  MED   pend   incorrect handling of ALSA ioctls
     61                      (has two patches, looks reasonable)
     62 
     63 159659  Wont  pend   scas does not use ES segment selector
     64 
     65 159720  Wont  pend   Opcode 0xFF 0x58 0xEB 0x5 unhandled on x86
     66                      (this is a "lcall" or callf -- far call)
     67                      (no test case)
     68 
     69 160595  Wont  pend   Why my program hang-up when work with valgrind ?
     70                      (no info so far, probably a threading bug in client)
     71 
     72 160872  MED   pend   race errors reported where libpthread.so
     73                      (might be a problem with pthread_setspecific)
     74 
     75 161374  WF,HIGH pend SYS_readdir syscall not implemented
     76                      will fix if reporter supplies a test case
     77 
     78 161609  LOW   pend   Results returned by VG_(get_StackTrace)() do not
     79                      match its documentation
     80 
     81 162020  MED   pend   Valgrinding an empty/zero-byte file crashes valgrind
     82 
     83 162229  LOW   pend   VALGRIND_DO_LEAK_CHECK emits false positive
     84 
     85 162456  ???   pend   valgrind-3.2.3 on ppc fails
     86                      problem undetermined (probably close)
     87 
     88 162630  Fixd  many   hg_main.c:926 (mk_SHVAL_fail): the 'impossible' happened.
     89 
     90 162718  ???   pend   x86: avoid segment selector 0 in sys_set_thread_area()
     91                      decide if this is a problem or not
     92 
     93 163253  ???   pend   canonicaliseSymtab forgot some fields in DiSym
     94                      is this correct?
     95 
     96 163910  LOW   pend   Typo in Helgrind's Documentation
     97 
     98 164298  Wont  pend   Valgrind asserts on finitel with long double using
     99                      gcc-4.3 -- closed as wontfix
    100 
    101 164485  MED   pend   VG_N_SEGNAMES and VG_N_SEGMENTS are (still) too small
    102 
    103 165417  LOW   pend   helgrind/tests/hg05_race2: VG_(get_data_description)()
    104                      succeeds on x86 and amd64 but fails on ppc
    105 
    106 165468  LOW   pend   Inconsistent formatting in memcheck manual
    107 
    108 165598  ???   pend   fixed mmap() regions (via MAP_FIXED) cause various
    109                      strange failures
    110                      (I think this should be closed as invalid)
    111 
    112 166355  HIGH,WF  pend   unshare syscall not supported (amd64, #272)
    113 
    114 167212  ???   pend   Valgrind does not attach to GDB correctly, traceback
    115                      spits garbage
    116 
    117 167483  LOW   pend   __libc_freeres error valgrind with shared libraries with
    118                      multiple --rpath (w/ repro case)
    119 
    120 167640  HIGH,WF pend un-efficient memory management cause segmentation fault
    121                      w/ test case
    122                      Can't reproduce
    123 
    124 167676  ???   pend   Valgrind loads wrong threading library on Red Hat 9
    125 
    126 167700  HIGH,WF pend  vex x86->IR: unhandled instruction bytes: 0xD5 0x36 0x5B
    127                       0xC3 (aad)
    128 
    129 168538  ???   pend   Poor detection of stacks for clone can lead to
    130                      bad results, w/ patch
    131                      Looks plausible
    132 
    133 169539  WF    pend   mremap fails on specific address on 32-bit memcheck
    134                      w/ test case
    135                      (not clear if this is still valid)
    136 
    137 169562  MED   pend   memcheck warns on some unaligned bcopy()s
    138                      w/ test case
    139 
    140 169957  LOW   pend   Bad report about stack memory after long jump
    141                      from signal handler, w/ test case
    142 
    143 170510  MED   pend   Don't warn about ioctl of size 0 without
    144                      direction hint, w/ patch
    145 
    146 171627  ???   pend   Valgrind macros change program behavior
    147 
    148 175138  WF    pend   aspacem assertion failed: segment_is_sane at
    149                      m_aspacemgr/aspacemgr-linux.c:1412 (add_segment)
    150 
    151 FAQ.html and FAQ.html (Greg Czajknowski, @users, 21 Nov 08)
    152 
    153 Better return values from VG_(record_error) (kcc)
    154 
    155 %gs register not implemented in amd64 (test prog, 15 Dec)
    156 
    157 // not fixed in 3.4.0 -- not triaged
    158 
    159 175721 unexpected summarise_context messages
    160 175819 Support for ipv6 socket reporting with --track-fds
    161 175984 LDT is process-shared, not per-thread
    162 175985 missing modify_ldt syscall support on x86_64
    163 176641 The 'impossible' happened (dup of kernel bug)
    164 176926 memcheck floating point exception at valgrind startup with PPC 440EPX
    165 178092 Incomplete stack traces when using makecontext/swapcontext
    166 
    167 // 22 Dec 2008
    168 
    169 
    170 ---------- Bugs fixed in 3.4.0 -------------------------------------
    171 
    172 n-i-bz  Fixd  8048   Make return types for some of the requests 64-bit clean
    173 
    174 n-i-bz  Fixd  8760   glibc 2.9 support
    175 
    176 n-i-bz  Fixd  8800   ignore unsafe .valgrindrc's (CVE-2008-4865)
    177 
    178 n-i-bz  Fixd   8812  MPI_Init(0,0) is valid but libmpiwrap.c segfaults
    179 
    180 n-i-bz  Fixd  ????   Building in an env without gdb gives bogus gdb attach
    181 
    182 92456   Fixd  many   Tracing the origin of uninitialised memory
    183 
    184 106497  Fixd  ????   Valgrind does not demangles some C++template symbols
    185 162222  Fixd  ????   ==106497
    186 
    187 151612  Fixd  8725-7 Suppression with "..." (w/ good patch)
    188 
    189 156404  Fixd  8712/3 Unable to start oocalc under memcheck on openSUSE
    190                      10.3 (64-bit)
    191 159285  Fixd  8736   unhandled syscall:25 (stime, on x86-linux)
    192 
    193 159452  Fixd  8734   unhandled ioctl 0x8B01 on "valgrind iwconfig"
    194                      w/ plausible-looking patch
    195 160954  Fixd  1870,1  ppc build of valgrind crashes with illegal instruction
    196                      (isel)
    197                      == 161251
    198 
    199 160956  Fixd  7888   mallinfo implementation, w/ patch
    200 
    201 162092  Fixd  8729,32  Valgrind fails to start gnome-system-monitor
    202 
    203 162819  ???   8285/6 malloc_free_fill test doesn't pass on glibc2.8 x86
    204                      need to look at this
    205 163794  Fixd  8228   assertion failure with "--track-origins=yes"
    206 
    207 163933  Fixd  8229   sigcontext.err and .trapno must be set together
    208 
    209 163955  Fixd  8722   remove pointless constraint !(--db-attach=yes &&
    210                      --trace-children=yes)
    211 
    212 164476  Fixd  8263,6 Missing kernel module loading system calls
    213 
    214 164669  Fixd  r8747  SVN regression: mmap() drops posix file locks
    215 
    216 166581  Fixd  8704   Callgrind output corruption when program forks
    217 
    218 167288  Fixd  8731   Patch file for missing system calls on Cell BE
    219 
    220 168943  Fixd  vx1872 unsupported scas instruction pentium
    221 
    222 171645  Fixd  vx1869 Unrecognised instruction (MOVSD, non-binutils
    223                      encoding)
    224 
    225 172417  Fixd  vx1867 x86->IR: 0x82 ...
    226 
    227 172563  Fixd  vx???? amd64->IR: 0xD9 0xF5  -  fprem1
    228 
    229 173099  Fixd  8758   .lds linker script generation error
    230 
    231 173177  Fixd  8720   [x86_64] WARNING: unhandled syscall: 125/126/179
    232                      (capget/capset/quotactl)
    233 
    234 173751  Fixd  vx1876 amd64->IR: 0x48 0xF 0x6F 0x45
    235                      (even more redundant prefixes)
    236 
    237 174532  WF    DUP    amd64->IR: 0x48 0xF 0xED 0x0 
    238               == 173751
    239 
    240 174908  Fixd  8774   --log-file value not expanded correctly for core file
    241 
    242 175044  Fixd  8769   Add lookup_dcookie for amd64
    243 
    244 175150  Fixd  vx1873 x86->IR: 0xF2 0xF 0x11 0xC1 (movss xmm1, xmm0)
    245                      non-binutils encoding
    246 
    247 
    248 ---------- Bugs fixed in 3.3.1 -------------------------------------
    249 
    250 /////////////////////////////////////////////////////////////////
    251 //
    252 // Changes in 3.3.1 only below here
    253 
    254 r1816 (x86 CondNS after SUBB)
    255 r1817 (amd64 CondNS after SUBB)
    256 r1833 (both, CondNZ after SUBW)
    257 merged; 1847
    258 
    259 r7795 (handle r-x mapped .bss sections - merge to trunk)
    260 
    261 
    262 TRUNK   33BRANCH  PRI    BUG#  WHAT
    263 
    264 r8058   r8059     Vfd  n-i-bz  Massif segfaults at exit
    265 
    266 vx1832  vx1839    Vfd  n-i-bz  Memcheck asserts on Altivec code
    267 
    268 r7299   r8081     Mgd  n-i-bz  fix sizeof bug in Helgrind
    269 r7322   r8073     Mgd  n-i-bz  check fd on sys_llseek
    270 r7323   r8082     Mgd  n-i-bz  update syscall lists to kernel 2.6.23.1
    271 r7324   r8076     Mgd  n-i-bz  support sys_sync_file_range
    272 
    273 r7334   r8078     Mgd  n-i-bz  handle sys_sysinfo, sys_getresuid,
    274                                sys_getresgid on ppc64-linux
    275 
    276 r7350   r8080     Mgd  n-i-bz  intercept memcpy in 64-bit ld.so's
    277 
    278 r8047   r8148     Mgd  n-i-bz  Fix wrappers for sys_{futimesat,utimensat}
    279 
    280 
    281 vx1836/7 vx1845   Vfd  126389  vex x86->IR: 0xF 0xAE (FXRSTOR)
    282                        == 158525
    283 
    284 vx1838  vx1846    Vfd  152818  vex x86->IR: 0xF3 0xAC (repz lodsb) 
    285 
    286 vx1834  vx1843    Vfd  153196  vex x86->IR: 0xF2 0xA6 (repnz cmpsb) 
    287 
    288 vx1807  vx1840    Mgd  155011  vex x86->IR: 0xCF (iret)
    289 
    290 r7347   r8085     Mgd  155091  Warning [...] unhandled DW_OP_ opcode 0x23
    291                        ==156960
    292 
    293 vx1808/9/10,r7384/5/7/9
    294         vx1841,r8056
    295                   Vfd  155528  support Core2/SSSE3 insns on x86/amd64
    296 
    297 r7355   r7356     Mgd  155929  ms_print fails on massif outputs 
    298                                containing long lines
    299 
    300 r7399   r8083     Mgd  157665  valgrind fails on shmdt(0) after shmat to 0
    301 
    302 vx1835  vx1844    Vfd  157748  support x86 PUSHFW/POPFW
    303 
    304 r7454   r8084     Mgd  158212  hg: handle pthread_rwlock_try{rd,wr}lock.
    305 
    306 r8049   r8086     Vfd  158425  sys_poll incorrectly emulated when RES==0
    307 
    308 vx1848  vx1849    Vfd  158744  vex amd64->IR: 0xF0 0x41 0xF 0xC0 (xaddb)
    309 
    310 r7917,8051 OR r7942,r7983,r8051
    311         r8148     Mgd  160907  Support for a couple of recent Linux syscalls
    312 
    313 r7917   r8148     Mgd  161285  Patch -- support for eventfd() syscall
    314 
    315 vx1826  vx1842    Vfd  161378  illegal opcode in debug libm (FUCOMPP)
    316                        == 160136
    317 
    318 r8065/8 r8069     Mgd  161487  number of suppressions files is limited to 10
    319 
    320 r7361   r7362     Mgd  162386  ms_print typo in milliseconds time unit for
    321                                massif
    322 
    323 r7895   r8063     Mgd  161036  exp-drd: client allocated memory was never freed
    324 
    325 r8092   r8143     Mgd  n-i-bz  libmpiwrap.c: add a wrapper for MPI_Waitany
    326 
    327 r8140   r8145     Mgd  n-i-bz  helgrind support for glibc-2.8
    328 
    329 r8144   r8146     Vfd  n-i-bz  partial fix for mc_leakcheck.c:698 assert:
    330                                'lc_shadows[i]->data + lc_shadows[i] ...
    331 
    332 various r8148     Mgd  162663  signalfd_wrapper fails on 64bit linux
    333 
    334 r8154   r8155     Mgd  n-i-bz  Massif/Cachegrind output corruption when
    335                                programs fork
    336 
    337 vx1853  vx1854     33  n-i-bz  regalloc: don't incorrectly omit spill stores
    338