Closed
Bug 1211977
Opened 9 years ago
Closed 9 years ago
Crash [@ used]
Categories
(Core :: JavaScript Engine, defect)
Tracking
()
RESOLVED
DUPLICATE
of bug 1209026
Tracking | Status | |
---|---|---|
firefox44 | --- | affected |
People
(Reporter: decoder, Unassigned)
References
(Blocks 1 open bug)
Details
(Keywords: assertion, regression, testcase, Whiteboard: [jsbugmon:ignore])
Crash Data
Attachments
(2 files)
(deleted),
application/octet-stream
|
Details | |
(deleted),
patch
|
bbouvier
:
review+
|
Details | Diff | Splinter Review |
The following testcase crashes on mozilla-central revision 9169f652fe5e (build with --enable-optimize --enable-posix-nspr-emulation --enable-valgrind --enable-gczeal --disable-tests --enable-debug, run with --fuzzing-safe --thread-count=2 --ion-check-range-analysis --ion-eager --baseline-eager --ion-extra-checks):
See attachment.
Backtrace:
Program terminated with signal SIGSEGV, Segmentation fault.
#0 used (this=0x0) at js/src/jit/Label.h:42
#1 js::jit::AssemblerX86Shared::bind (this=0x7f35fcaf1058, label=0x0) at js/src/jit/x86-shared/Assembler-x86-shared.h:932
#2 0x0000000000943975 in js::jit::CodeGenerator::getJumpLabelForBranch (this=this@entry=0x7f35fcaf1000, block=0x7f35fcdef9c8) at js/src/jit/CodeGenerator.cpp:664
#3 0x0000000000965462 in js::jit::CodeGenerator::visitTestVAndBranch (this=0x7f35fcaf1000, lir=0x7f35fc90d600) at js/src/jit/CodeGenerator.cpp:709
#4 0x00000000009a682e in js::jit::CodeGenerator::generateBody (this=this@entry=0x7f35fcaf1000) at js/src/jit/CodeGenerator.cpp:4149
#5 0x00000000009a6f6a in js::jit::CodeGenerator::generate (this=this@entry=0x7f35fcaf1000) at js/src/jit/CodeGenerator.cpp:7859
#6 0x00000000009a718e in js::jit::GenerateCode (mir=mir@entry=0x7f35fcddc1a8, lir=0x7f35fcdfa180) at js/src/jit/Ion.cpp:1954
#7 0x00000000009a7275 in js::jit::CompileBackEnd (mir=0x7f35fcddc1a8) at js/src/jit/Ion.cpp:1976
#8 0x0000000000675e1a in js::HelperThread::handleIonWorkload (this=this@entry=0x7f35fee2e000) at js/src/vm/HelperThreads.cpp:1180
#9 0x0000000000676fd9 in js::HelperThread::threadLoop (this=0x7f35fee2e000) at js/src/vm/HelperThreads.cpp:1482
#10 0x00000000006cd8b1 in nspr::Thread::ThreadRoutine (arg=0x7f35fee2c080) at js/src/vm/PosixNSPR.cpp:45
#11 0x00007f360007a182 in start_thread (arg=0x7f35fedff700) at pthread_create.c:312
#12 0x00007f35ff16a47d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
rax 0x0 0
rbx 0x0 0
rcx 0xa 10
rdx 0x400 1024
rsi 0xe073ca 14709706
rdi 0xa 10
rbp 0x7f35fedfea60 139869886081632
rsp 0x7f35fedfea10 139869886081552
r8 0x0 0
r9 0xe084d8 14714072
r10 0x30 48
r11 0x7f35fedfea10 139869886081552
r12 0x31 49
r13 0x31 49
r14 0x7f35fcdef9c8 139869852465608
r15 0x7f35fcaf1058 139869849325656
rip 0x5bbd42 <js::jit::AssemblerX86Shared::bind(js::jit::Label*)+82>
=> 0x5bbd42 <js::jit::AssemblerX86Shared::bind(js::jit::Label*)+82>: cmpb $0x0,0x3(%rbx)
0x5bbd46 <js::jit::AssemblerX86Shared::bind(js::jit::Label*)+86>: jns 0x5bbd50 <js::jit::AssemblerX86Shared::bind(js::jit::Label*)+96>
The testcase doesn't reproduce for me but h4writer seems to already have found the issue, so filing now.
Reporter | ||
Comment 1•9 years ago
|
||
Comment 2•9 years ago
|
||
Assignee: nobody → hv1989
Attachment #8670360 -
Flags: review?(benj)
Comment 3•9 years ago
|
||
Comment on attachment 8670360 [details] [diff] [review]
Propagate OOM
Review of attachment 8670360 [details] [diff] [review]:
-----------------------------------------------------------------
::: js/src/jit/CodeGenerator.cpp
@@ +659,4 @@
> // backedges, so emit inline code for the patchable jump. Heap allocating
> // the label allows it to be used by out of line blocks.
> Label* res = alloc().lifoAlloc()->new_<Label>();
> + masm.propagateOOM(res);
Can you make the coercion more explicit, please? !!res
Attachment #8670360 -
Flags: review?(benj) → review+
Comment 4•9 years ago
|
||
This looks like a duplicate of bug 1209026.
(I think with this patch we'll still crash though because we'll pass nullptr to masm.bind below.)
Updated•9 years ago
|
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → DUPLICATE
Updated•3 years ago
|
Assignee: hv1989 → nobody
You need to log in
before you can comment on or make changes to this bug.
Description
•