#5003: test 4038 (ghci) fails on OS X 64 bit.
-------------------------------+--------------------------------------------
Reporter: gwright | Owner: gwright
Type: bug | Status: new
Priority: normal | Component: GHCi
Version: 7.0.2 | Keywords:
Testcase: | Blockedby:
Os: MacOS X | Blocking:
Architecture: x86_64 (amd64) | Failure: Runtime crash
-------------------------------+--------------------------------------------
Comment(by altaic):
Compiling and running your modified test on OS X (ghc-stage2 built with
`-Rghc-timing -O0 -DDEBUG -debug`) fails similarly:
{{{
[...]
first: 501
second: 501
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_PROTECTION_FAILURE at address: 0x00007fff5f3fe8a0
0x000000010031ac72 in stg_ap_p_info ()
}}}
Interestingly, the failure occurs more quickly in ghci:
{{{
[...]
first: 908
second: 908
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x00000001037ff720
[Switching to process 27370 thread 0x2003]
0x0000000101b7331a in stg_ap_p_info ()
}}}
Note that all the failures in `stg_ap_p_info` happen at the same
instruction:
{{{
0x0000000101b7331a <stg_ap_p_info+890>: mov %rax,0x40(%rsp)
}}}
----
Using ghc-stage2 built with `BuildFlavour = perf`, the modified test fails
somewhat differently.
ghci:
{{{
[...]
first: 296
second: 296
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_PROTECTION_FAILURE at address: 0x0000000103700fd0
[Switching to process 5685 thread 0x2003]
0x0000000101aad47e in stg_takeMVarzh ()
}}}
But similarly, the instruction is the first access of `rsp`:
{{{
0x0000000101aad47e <stg_takeMVarzh+6>: mov %rax,0x40(%rsp)
}}}
----
And using ghc-stage2 built with `BuildFlavour = devel2`...
ghci:
{{{
first: 1000
second: 1000
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0000000000000000
[Switching to process 21612 thread 0x2103]
0x0000000000000000 in ?? ()
(gdb) bt
#0 0x0000000000000000 in ?? ()
#1 0x00000001037de4dd in ?? ()
#2 0x0000000104a1d9c8 in ?? ()
Previous frame inner to this frame (gdb could not unwind past this frame)
(gdb) frame 1
#1 0x00000001037de4dd in ?? ()
(gdb) disas $rip $rip+1
Dump of assembler code from 0x1037de4dd to 0x1037de4de:
0x00000001037de4dd: mov 0x50(%rsp),%rcx
End of assembler dump.
(gdb) p $rsp
$1 = (void *) 0x10397ce60
}}}
----
Finally, using ghc built with `-Rghc-timing -O0 -DDEBUG`, the original
test fails as originally reported (note: callq accesses `rsp`).
ghci:
{{{
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_PROTECTION_FAILURE at address: 0x0000000103900f88
[Switching to process 21633 thread 0x2103]
0x0000000101b5197b in stg_makeStablePtrzh ()
(gdb) disas $rip $rip+1
Dump of assembler code from 0x101b5197b to 0x101b5197c:
0x0000000101b5197b <stg_makeStablePtrzh+51>: callq 0x101b34f80
<getStablePtr>
End of assembler dump.
(gdb) p $rsp
$1 = (void *) 0x103900f90
}}}
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5003#comment:6>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
