Comment #2 on issue 1155 by holisme: ARM segfaults with (Math.pow(2,32) -
1)*1 the second time
http://code.google.com/p/v8/issues/detail?id=1155
Before anything else, note that it does not happen with v8 2.5.9.15,
and that it is still happening with 3.0.12.22.
I have yet to check what happens on trunk.
It's compiling on a debian/sid box, with gcc 4.4.5, and those added
options :
-Wno-psabi, -fno-tree-sink, -O2
which gives, for example :
g++ -o obj/release/arm/assembler-arm.os -c -g -O2 -g -Wall -O2
-fno-strict-aliasing -Wall -Werror -W -Wno-unused-parameter
-Wnon-virtual-dtor -pedantic -Wno-psabi -fno-tree-sink -O2
-fomit-frame-pointer -fdata-sections -ffunction-sections -ansi -fno-rtti
-fno-exceptions -fvisibility=hidden -Wall -Werror -W -Wno-unused-parameter
-Wnon-virtual-dtor -pedantic -Wno-psabi -fno-tree-sink -O2
-fomit-frame-pointer -fdata-sections -ffunction-sections -ansi -fPIC
-DV8_TARGET_ARCH_ARM -DV8_SHARED -DENABLE_VMSTATE_TRACKING
-DENABLE_LOGGING_AND_PROFILING -DENABLE_DEBUGGER_SUPPORT -Isrc
src/arm/assembler-arm.cc
along with shell :
g++ -o obj/sample/shell/release/shell.o -c -g -O2 -g -Wall -O2
-fno-strict-aliasing -fno-rtti -fno-exceptions -O2 -Iinclude
samples/shell.cc
g++ -o shell obj/sample/shell/release/shell.o -L. -lv8-3.0.12.22 -lpthread
and the qemu machine :
Processor : ARM926EJ-S rev 5 (v5l)
BogoMIPS : 588.18
Features : swp half thumb fastmult vfp edsp java
CPU implementer : 0x41
CPU architecture: 5TEJ
CPU variant : 0x0
CPU part : 0x926
CPU revision : 5
Hardware : ARM-Versatile PB
Revision : 0000
Serial : 0000000000000000
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
