docs/internals/3_0_BUGSTATUS.txt

------------------------------------------------------------------------
--- Bugs known prior to 3.0.0                                        ---
------------------------------------------------------------------------

x86 INT/INT3

Not started.  Seems low priority.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

87263 x86 segment stuff

Not started.  Seems low priority.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

88116 x86 enter variants assert

Not started.  Seems low priority.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

96542 x86 16-bit pop insns

Not started.  Seems low priority.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

109861 amd64 hangs at startup
110301 ditto

Will fix in 3.1.  Long delay seems to be caused by amd64-Gentoo kernel
not liking large mmap/munmap requests.

FIXED-TRUNK:    no (background hacking is in progress)
FIXED-30BRANCH: won't (3.1 fix only)

----------------------------------------------------------------

109313 x86 cmpxchg8b
110505 ditto

This ought to be fixed for 3.0.1.

FIXED-TRUNK:    done(1331, 4390 contains regtest 
                           + mistaken commit of this file)
FIXED-30BRANCH: done(1337)

----------------------------------------------------------------

109323 ppc32: dispatch.S uses Altivec insn, which doesn't work on POWER. 

Should fix for 3.1.  Any fix would be similar to that for 110274.

FIXED-TRUNK:    TODO
FIXED-30BRANCH: won't (3.1 fix only)

----------------------------------------------------------------

109345 ppc32 ptrace patch available    should be applied

Should fix for 3.1.  Any fix would be similar to that for 110274.

FIXED-TRUNK:    TODO
FIXED-30BRANCH: won't (3.1 fix only)

----------------------------------------------------------------

CrispinF x86 %eflags.ac problem

FIXED-TRUNK:    yes (1319/4334)
FIXED-30BRANCH: yes (1326, and 4334 was copied across as part of 4364)

----------------------------------------------------------------

Give more info about seginfo dropping.

FIXED-TRUNK:    yes (4425)
FIXED-30BRANCH: no

------------------------------------------------------------------------
--- Bugs reported after 3.0.0 shipped                                ---
------------------------------------------------------------------------

110102 dis_op2_E_G(amd64)

FIXED-TRUNK:    yes (1318)
FIXED-30BRANCH: yes (1325)

----------------------------------------------------------------

110183 tail of page with _end

Could be a problem for glibc developers.  Consider fixing.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110201 x86 FXTRACT

Could fix if important.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110202 x86 sys_waitpid(#286)

FIXED-TRUNK:    fixed(r4329)
FIXED-30BRANCH: fixed(r4332)

----------------------------------------------------------------

110203 clock_getres(,0)

FIXED-TRUNK:    fixed(r4328)
FIXED-30BRANCH: fixed(r4332)

----------------------------------------------------------------

110204 fmemopen false +ve

Seems low priority.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110205 sigcancel unwind fails

Tom is considering this.  It would be nice to fix it for 3.1 but
status currently unclear.

FIXED-TRUNK:    1320 - vex impl of sysenter
                4337 - minimal Valgrind-side; does not do anything
FIXED-30BRANCH: 

----------------------------------------------------------------

110207 mpn accuracy

Can't be easily fixed (x86 rounding/precision problem)
+ not convinced it's a big problem

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110208 execve fail wrong retval

FIXED-TRUNK:    yes (r4330)
FIXED-30BRANCH: yes (r4332)

----------------------------------------------------------------

110209 --show-emwarns misses some

Tom says: The math/test-fenv.c file in the glibc source is the code in
question and I can reproduce it with that code.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110240 x86 FP differences

really is the same as 110207; same comments apply

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110274 SSE1 now mandatory for x86

FIXED-TRUNK:    yes(1321/4339)
FIXED-30BRANCH: yes(1327/4374)

----------------------------------------------------------------

110388 amd64 0xDD 0xD1

FIXED-TRUNK:    yes(1322)
FIXED-30BRANCH: yes(1328)

----------------------------------------------------------------

110464 amd64 0xDC 0x1D FCOMP

FIXED-TRUNK:    yes(1323)
FIXED-30BRANCH: yes(1329)

----------------------------------------------------------------

110478 amd64 0xF 0xD PREFETCH

FIXED-TRUNK:    yes(1324)
FIXED-30BRANCH: yes(1330)

----------------------------------------------------------------

XML <unique> printing wrong

FIXED-TRUNK:    4355,4357,4358
FIXED-30BRANCH: 4585

----------------------------------------------------------------

Dirk r4359 (amd64 syscalls from trunk)

FIXED-TRUNK:    
FIXED-30BRANCH: done(4359)

----------------------------------------------------------------

Dirk r4360 (upd email addrs from trunk)

FIXED-TRUNK:    
FIXED-30BRANCH: done(4360)

----------------------------------------------------------------

110536/7/8/9/40 Valgrind crashes when tring to realloc memory

Uninvestigated.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110591 amd64: rdtsc not implemented properly

Under consideration. (Also afflicts x86)

FIXED-TRUNK:    1344 (x86), 1346 (amd64).
FIXED-30BRANCH: no

----------------------------------------------------------------

Nick r4384 (stub implementations of Addrcheck and Helgrind)

FIXED-TRUNK:    done(4384)
FIXED-30BRANCH: done(4397)

----------------------------------------------------------------

110652  AMD64 valgrind crashes on cwtd instruction

FIXED-TRUNK:    done(1333)
FIXED-30BRANCH: done(1335)

----------------------------------------------------------------

110653  AMD64 valgrind crashes on sarb $0x4,foo(%rip) instruction

FIXED-TRUNK:    done(1334)
FIXED-30BRANCH: done(1336)

----------------------------------------------------------------

110656  PATH=/usr/bin::/bin valgrind foobar stats ./fooba

FIXED-TRUNK:    done(4386)
FIXED-30BRANCH: done(4395)

----------------------------------------------------------------

110657  Small test fixes

(1) Filter out L3 cache warning messages causing problems
(2) Stop tests/mq failing on 2.4 kernels

I suppose it would be good to apply these.  They seem low risk.

FIXED-TRUNK:    4429
FIXED-30BRANCH: 4458

----------------------------------------------------------------

110669  valgrind attach to gdb and quitting gdb hangs valgrind

Not clear if this is really a Valgrind bug.

FIXED-TRUNK:    no
FIXED-30BRANCH: no

----------------------------------------------------------------

110671  vex x86->IR: unhandled instruction bytes: 0xF3 0xC3 (rep ret)

FIXED-TRUNK:    done(1332)
FIXED-30BRANCH: done(1338)

----------------------------------------------------------------

Nick (Cachegrind should not assert when it encounters a client
request.)

FIXED-TRUNK:    done(4391)
FIXED-30BRANCH: done(4393)

----------------------------------------------------------------

110685  amd64->IR: unhandled instruction bytes: 0xE1 0x56 (loope Jb)

FIXED-TRUNK:    1349
FIXED-30BRANCH: TODO

----------------------------------------------------------------

110830  configuring with --host fails to build 32 bit on 64 bit target

FIXED-TRUNK:    4442
FIXED-30BRANCH: 4459

----------------------------------------------------------------

110875  Assertion when execve fails

FIXED-TRUNK:    4435
FIXED-30BRANCH: 4457

----------------------------------------------------------------

Updates to Memcheck manual

FIXED-TRUNK:    4419, 4427, 4434
FIXED-30BRANCH: 4455

----------------------------------------------------------------

Fixed broken malloc_usable_size()

FIXED-TRUNK:    4439
FIXED-30BRANCH: 4453

----------------------------------------------------------------

110898  opteron instructions missing: btq sbbq btsq btrq bsfq

FIXED-TRUNK:    TODO
FIXED-30BRANCH: TODO

----------------------------------------------------------------

110954  x86->IR: unhandled instruction bytes: 0xE2 0xF6 (loop Jb)

FIXED-TRUNK:    1343
FIXED-30BRANCH: TODO

----------------------------------------------------------------

Make suppressions work for "???" lines in stacktraces.

FIXED-TRUNK:    4447
FIXED-30BRANCH: 4451

----------------------------------------------------------------

111006  bogus warnings from linuxthreads

FIXED-TRUNK:    4469, 4470
FIXED-30BRANCH: TODO

----------------------------------------------------------------

111090  segmentation fault in massif

FIXED-TRUNK:    TODO
FIXED-30BRANCH: TODO

----------------------------------------------------------------

111092  x86: dis_Grp2(Reg): unhandled case(x86) 

FIXED-TRUNK:    1341
FIXED-30BRANCH: TODO

----------------------------------------------------------------

111231  sctp_getladdrs() and sctp_getpaddrs() returns uninitialized
        memory

FIXED-TRUNK:    TODO
FIXED-30BRANCH: TODO

----------------------------------------------------------------

111102 (comment #4)   Fixed 64-bit unclean "silly arg" message

FIXED-TRUNK:    4476
FIXED-30BRANCH: TODO