Gitiles
Code Review
Sign In
gerrit-public.fairphone.software
/
platform
/
external
/
valgrind
/
29dd9e62b241efb2c406ed658720e0dd2e68897d
/
coregrind
« Previous
2521d32
by njn
· 19 years ago
0e8c765
Enable the epoll system calls on amd64.
by tom
· 19 years ago
2d2b154
Remove dead function VG_(exit_single)().
by njn
· 19 years ago
2bc1012
A crucial, crucial change: update my email address.
by njn
· 19 years ago
a4469a7
Avoid compiler warnings.
by njn
· 19 years ago
502badb
by njn
· 19 years ago
4fbc86c
by njn
· 19 years ago
0ee4a95
Rework signal stack handling to avoid doing arithmetic on void pointers
by tom
· 19 years ago
be18a59
Only show invalid-CfiSI messages at verbosity level >= 1.
by sewardj
· 19 years ago
ad67ade
It helps if my changes actually compile...
by tom
· 19 years ago
cf941da
Allow for the red zone when generating new_mem_stack_signal and
by tom
· 19 years ago
a0e1656
Tidy up a bit.
by sewardj
· 19 years ago
f91889f
Add intercepts for unsigned long versions of the nothrow new and new[]
by tom
· 19 years ago
5bcde92
Fix stupid bug I introduced in r3556, which broke handling of RT signals. Sigh.
by sewardj
· 19 years ago
55022aa
* Clean up the DWARF2 CFI reader a bit
by sewardj
· 19 years ago
c4f204f
VG_(addCfiSI): ignore CfiSIs which state an address range outside of
by sewardj
· 19 years ago
7624567
Improve debug printing
by sewardj
· 19 years ago
f04fd62
Reset the address encoding to the default before parsing the
by tom
· 19 years ago
e0707a6
Handle DW_CFA_register.
by sewardj
· 19 years ago
cfe1046
Re-enable vex bb chasing. In fact this accidentally got reenabled in
by sewardj
· 19 years ago
520e349
rm debug printing
by sewardj
· 19 years ago
21c6d0f
Initial cleanups based on target-specific defines.
by sewardj
· 19 years ago
ee0bcbf
Move the replacement code for the amd64 vsyscalls into the trampoline page
by tom
· 19 years ago
bc7df20
Add macro definitions that make it possible to do cpu/os/both
by sewardj
· 19 years ago
3a1c7db
Sort the CFI summary table and do lookups in it using binary search.
by sewardj
· 19 years ago
bf60375
Create and use summary address range limits for the CFI records in
by sewardj
· 19 years ago
3a7cb97
CFI: if summarisation fails, don't add the result to our collection of
by sewardj
· 19 years ago
8eb4132
Fix various other CFI-frame-unwind bits and pieces.
by sewardj
· 19 years ago
325ec87
* handle DW_CFA_offset_extended_sf
by sewardj
· 19 years ago
2fd3890
Get DWARF CFI handling going on amd64 systems. This also required getting
by tom
· 19 years ago
ce058b0
New flag --trace-cfi=yes|no [no], which is for debugging the CFI reader.
by sewardj
· 19 years ago
3516553
Get the CFI-based frame-unwinder to the point where it works at least
by sewardj
· 19 years ago
5c638c2
Beginnings of a DWARF CFI-based frame-unwinder. Does not yet do
by sewardj
· 19 years ago
d485d69
Even more pissing with inline asm around to placate gcc-3.4 -fpic. We
by sewardj
· 19 years ago
d4d203b
local_sys_write_stderr: tell gcc what registers the asm trashes.
by sewardj
· 19 years ago
c612186
Make the debug log module work on amd64.
by tom
· 19 years ago
311400b
Rework the inline assembly implementations of write and getpid for x86 to
by tom
· 19 years ago
550f898
Add support for the fadvise system calls.
by tom
· 19 years ago
7efe7be
A bunch of redirections for SuSE 9.2 on amd64.
by sewardj
· 19 years ago
9be7c98
Use %lu for all length values in debug messages or you will sometimes
by tom
· 19 years ago
601371a
amd64 build fixes.
by sewardj
· 19 years ago
23c6a29
by sewardj
· 19 years ago
6ffee52
Use __FUNCTION__ correctly.
by sewardj
· 19 years ago
55f9d1a
by sewardj
· 19 years ago
8c61589
More debug printing
by sewardj
· 19 years ago
d85feff
More printing fine-tuning
by sewardj
· 19 years ago
1fbc1a5
Print debugging info at various places in the startup sequence.
by sewardj
· 19 years ago
a5ebfa9
Print a bit more prettily.
by sewardj
· 19 years ago
1cf558c
by sewardj
· 19 years ago
5efa55f
Relax the requirement that a pub_tool_<modulename>.h file must exist
by sewardj
· 19 years ago
8a7fdd2
Make amd64-linux build again after m_sigframe hackery.
by sewardj
· 19 years ago
985fabb
Create a new module, "sigframe", responsible for creating/destroying
by sewardj
· 19 years ago
267100d
Rename the first three modules as per naming scheme.
by sewardj
· 19 years ago
cc9549a
Add statement-of-intent re top level module structure.
by sewardj
· 19 years ago
663a1bd
Add initialisation-order sanity checks.
by sewardj
· 19 years ago
fa8202f
Update comment about stack management, and remove some unused functions.
by sewardj
· 19 years ago
a752970
A leftover from the days of our own libpthread; now irrelevant.
by sewardj
· 19 years ago
9b3328d
Add intercepts for operator new(unsigned long) and operator
by sewardj
· 19 years ago
045a405
Allow memcheck to take account of VGA_STACK_REDZONE_SIZE -- that is,
by sewardj
· 19 years ago
9ea0901
Add another redirect that we need. This has no effect at present
by sewardj
· 19 years ago
d2b1711
by njn
· 19 years ago
50ae1a7
by njn
· 19 years ago
703eec5
Reword error messages in a not-quite-so-terse way.
by sewardj
· 19 years ago
5d2ca93
Even more syscalls.
by sewardj
· 19 years ago
f8bb9e7
Get rid of the --sloppy-malloc= flag and the functionality it
by sewardj
· 19 years ago
33ef39c
Removed bogus (UInt) casts from the amd64 signal frame builder and some
by tom
· 19 years ago
990e90c
by njn
· 19 years ago
2e1da50
Removed references to client_stack_base which no longer exists.
by tom
· 19 years ago
90eecea
64-bit cleanness -- use UWord instead of UInt.
by njn
· 19 years ago
50ba34e
by njn
· 19 years ago
fc14e45
Deal with apparently non-cfront compliant new[]/delete[] manglings
by sewardj
· 19 years ago
10e1cfb
gen_intercepts.pl doesn't exist any more.
by njn
· 19 years ago
97724e5
In vg_memory.c, allow the stack-change threshold to be specified by a
by sewardj
· 19 years ago
074edc6
More amd64 system calls - emacs will now run under valgrind.
by tom
· 19 years ago
9ee81f5
by sewardj
· 19 years ago
c657eb6
Remove vg_instrument.c from the makefile.
by tom
· 19 years ago
8b75b41
Get rid of apparently-redundant file.
by sewardj
· 19 years ago
cb2b13a
Assert that cpuid is available - this is just done to force the
by tom
· 19 years ago
02d73d9
Get VG_(has_cpuid) and VG_(cpuid) working on amd64.
by tom
· 19 years ago
748a131
Add VGP_(setup_redirects) to the platform specific layers and use
by tom
· 19 years ago
f26d068
Don't truncate symbol values in ELF symbol tables.
by tom
· 19 years ago
686d4ca
Even more amd64 system calls.
by tom
· 19 years ago
9749a50
Yet more amd64 system calls.
by tom
· 19 years ago
7286be5
More amd64 system calls.
by tom
· 19 years ago
d26f242
Rework the vsyscall redirections to work in pie code - the old form
by tom
· 19 years ago
a8d1ec6
Move the gettid system call to the linux specific section.
by tom
· 19 years ago
fb154a5
Increase maximum translation size. This can happen when translating
by sewardj
· 19 years ago
60df38c
Implement the gettid system call which seems to have got lost in the merge.
by tom
· 19 years ago
cf81d55
by njn
· 19 years ago
8a97c6d
by njn
· 19 years ago
dd53b7a
Make the dispatcher code position independent so that PIE mode works.
by tom
· 19 years ago
23f6cae
Comment-only change
by sewardj
· 19 years ago
0312f51
Completely get rid of VG_(instr_ptr_offset).
by sewardj
· 19 years ago
f89f8fa
Get rid of the use of VG_(instr_ptr_offset) since we know what that is
by sewardj
· 19 years ago
3c35375
rm unused function
by sewardj
· 19 years ago
6f917fd
Get thew DWARF reading going on 64 bit machines.
by tom
· 19 years ago
c03c4fd
Discard the sys_socketcall wrapper (there is no socketcall system
by tom
· 19 years ago
de2ec26
Fixed ROUNDDN to avoid truncating the result when the first argument
by tom
· 19 years ago
5e82ade
Build the arch and platform libraries as PIE code when appropriate.
by tom
· 19 years ago
b9c3d46
Revive VG_(cpuid) which had been commented out during the merge and
by tom
· 19 years ago
Next »