GCC Bugzilla – Attachment 4783 Details for
Bug 10746
[3.3 regression] [win32] garbage collection crash in GCJ
Home
|
New
|
Browse
|
Search
|
[?]
|
Reports
|
Help
|
New Account
|
Log In
Remember
[x]
|
Forgot Password
Login:
[x]
Another annotated debugging log
gdb_log2.txt (text/plain), 5.04 KB, created by
Hans Boehm
on 2003-09-19 01:16:14 UTC
(
hide
)
Description:
Another annotated debugging log
Filename:
MIME Type:
Creator:
Hans Boehm
Created:
2003-09-19 01:16:14 UTC
Size:
5.04 KB
patch
obsolete
>This is after GC_generic_push_regs fix, getenv enabled, debuggable GC. > >Saw several other kinds of symptoms, which were unfortunately hard to debug: > >1) String was corrupted at some point with a few garbage characters at beginning. >2) segfault (misinterpreted as NPE, and apparently caught!!) in >GC_gcj_malloc_ingore_off_page, at line 280. >3) bad branch. > >... >second thread 239000 ffegfb >first thread 723000 fffffffffdefgae >second thread 240000 ffee >first thread 724000 dddddggcged >second thread 241000 cccdec > >Program received signal SIGSEGV, Segmentation fault. >0x00410728 in java::lang::Object::getClass() () > at /home/hans/gcc/gcc/gcc/config/i386/cygwin.asm:29 > >(gdb) x/20i 0x410720 >0x410720 <_ZN4java4lang6Object8getClassEv>: push %ebp >0x410721 <_ZN4java4lang6Object8getClassEv+1>: mov %esp,%ebp >0x410723 <_ZN4java4lang6Object8getClassEv+3>: mov 0x8(%ebp),%eax >0x410726 <_ZN4java4lang6Object8getClassEv+6>: mov (%eax),%eax >0x410728 <_ZN4java4lang6Object8getClassEv+8>: mov (%eax),%eax >0x41072a <_ZN4java4lang6Object8getClassEv+10>: pop %ebp >0x41072b <_ZN4java4lang6Object8getClassEv+11>: ret >... >(gdb) p/x $esp >$3 = 0x22fce8 >(gdb) p/x $ebp >$4 = 0x22fce8 >(gdb) x/4wx $4 >0x22fce8: 0x0022fd08 0x0040adfc 0x0102bfd8 0x77f8f2eb >(gdb) x/10wx 0x0102bfd8 >0x102bfd8: 0x00000000 0x010a7a28 0x00000008 0x00000001 >0x102bfe8: 0x00000000 0x00000000 0x00000000 0x00000000 >0x102bff8: 0x00000000 0x00000000 >(gdb) x/20wx 0x0102b000 >0x102b000: 0x00000000 0x01040578 0x00000008 0x00000002 >0x102b010: 0x00000000 0x00000000 0x00598268 0x010405a0 >0x102b020: 0x00000008 0x00000001 0x0102b000 0x00000000 >0x102b030: 0x00598268 0x010405c8 0x00000008 0x00000001 >0x102b040: 0x00000000 0x00000000 0x00598268 0x010405f0 >[Same problem as before.] >(gdb) p GC_arrays._heap_sects >$11 = {{hs_start = 0xbd0000 "", hs_bytes = 65536}, {hs_start = 0xbf0000 "à²^", > hs_bytes = 65536}, {hs_start = 0xc00000 "", hs_bytes = 65536}, { > hs_start = 0x1020000 "", hs_bytes = 69632}, {hs_start = 0x1040000 "à²^", > hs_bytes = 90112}, {hs_start = 0x1060000 "h\202Y", hs_bytes = 122880}, { > hs_start = 0x1080000 "h\202Y", hs_bytes = 163840}, {hs_start = 0x0, > hs_bytes = 0} <repeats 377 times>} >(gdb) p 0x1020000 + 69632 >$12 = 16977920 >(gdb) p/x $12 >$13 = 0x1031000 >[ Block is well inside a heap section. ] >(gdb) info threads > 3 thread 1772.0x6e8 0x00405a9f in gnu::gcj::runtime::StringBuffer::append(jav >a::lang::String*) () at /home/hans/gcc/gcc/gcc/config/i386/cygwin.asm:29 > 2 thread 1772.0x634 0x77f88e63 in ?? () >* 1 thread 1772.0x678 0x00410728 in java::lang::Object::getClass() () > at /home/hans/gcc/gcc/gcc/config/i386/cygwin.asm:29 >(gdb) p &thread_table >$16 = (struct <unknown> (*)[]) 0x607620 >(gdb) p sizeof(struct thread_entry) >$17 = 736 >(gdb) p (struct thread_entry *)0x607620 >$18 = (struct thread_entry *) 0x607620 >(gdb) p *$18 >$19 = {in_use = 1, id = 1588, handle = 0xa4, stack = 0xe20000, context = { > ContextFlags = 65539, Dr0 = 4271566032, Dr1 = 4271566032, Dr2 = 642, > Dr3 = 2151861871, Dr6 = 3184217444, Dr7 = 0, FloatSave = { > ControlWord = 3188259364, StatusWord = 4272195008, TagWord = 1, > ErrorOffset = 2250942056, ErrorSelector = 3221241968, DataOffset = 0, > DataSelector = 1, > RegisterArea = "\001\000\000\000`ú\b_\200ì\232_\000¤/\202+\tC\200Xú\b_à¤/\ >202\227\nC\200\000\000\000\000\000\000\000\000L¥/\202\001\205¤_\000\000\000\000L >¥/\202F\002\000\000W¡B\200\000\000\000\000Xú\b_A\205¤_@KE½", > Cr0NpxState = 2152614501}, SegGs = 3188259416, SegFs = 0, SegEs = 0, > SegDs = 2147888912, Edi = 2, Esi = 2012778072, Ebx = 2, Edx = 0, > Ecx = 1024, Eax = 0, Ebp = 14810604, Eip = 2012778083, SegCs = 27, > EFlags = 582, Esp = 14810528, SegSs = 35, > ExtendedRegisters = '\000' <repeats 511 times>}, suspended = 0} >(gdb) p $18)[1] >Junk after end of expression. >(gdb) p $18[1] >$21 = {in_use = 1, id = 1768, handle = 0xb0, stack = 0x1020000, context = { > ContextFlags = 65539, Dr0 = 4277974792, Dr1 = 2152259297, Dr2 = 2032639, > Dr3 = 3188259628, Dr6 = 1, Dr7 = 2173227072, FloatSave = { > ControlWord = 3188260020, StatusWord = 3802093920, TagWord = 176, > ErrorOffset = 4272194984, ErrorSelector = 4272194984, > DataOffset = 3188260020, DataSelector = 2152061872, > RegisterArea = "D\b@\200ÿÿÿÿdú\b_hHM\200\b·ü_Üú\b_"\205¤_E\235Xÿ\000\000\0 >00\000fIO\200A\205¤_oZÆâ\001\000\000\000\bû\b_OOD\200"\205¤_Hû\b_º\201\006\200\0 >34\000\000\000º\201\006\200", Cr0NpxState = 28}, SegGs = 48, > SegFs = 2152099173, SegEs = 4285003546, SegDs = 48, Edi = 24, > Esi = 6261248, Ebx = 22, Edx = 0, Ecx = 9, Eax = 10, Ebp = 16907588, > Eip = 2012777415, SegCs = 27, EFlags = 514, Esp = 16907536, SegSs = 35, > ExtendedRegisters = '\000' <repeats 511 times>}, suspended = 0} >[ Further thread_table entries appear to be zero. In_use bits are 0.] >(gdb) p/x 1588 >$25 = 0x634 >(gdb) p/x 1768 >$26 = 0x6e8 >
You cannot view the attachment while viewing its details because your browser does not support IFRAMEs.
View the attachment on a separate page
.
View Attachment As Raw
Actions:
View
Attachments on
bug 10746
:
4528
|
4716
|
4732
|
4733
|
4764
|
4765
| 4783 |
4888