Julian Seward
2ee9e90486
Implement a GDB server in Valgrind. See #214909 .
...
(Philippe Waroquiers, philippe.waroquiers@skynet.be )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11727
2011-05-06 21:02:55 +00:00
Julian Seward
ed9d566463
mkLazyN: fix incorrect type decl for 'mergeTy64', as spotted by IBM's
...
BEAM checker.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11726
2011-05-04 09:09:31 +00:00
Julian Seward
21dc1d9d42
scalarts_limitations_fail_NORETURN: fix a format string error
...
spotted by IBM's BEAM checker. (Why didn't gcc or clang see this?)
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11725
2011-05-04 09:08:34 +00:00
Julian Seward
29c20eaeb0
setup_client_stack: use have_exename to consistently guard uses
...
of VG_(args_the_exename), thereby avoiding a potential segfault.
Spotted by IBM's BEAM checker.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11724
2011-05-04 09:07:38 +00:00
Julian Seward
1ccdd881a6
calling format_message: when passing frameNo == -1, also pass
...
tid == VG_INVALID_THREADID rather than an uninitialised ThreadId.
Also in format_message, improve precondition assertions for
frameNo and tid.
There's no error in the current code since if frameNo == -1 then
tid is unused, but it caused IBM's BEAM checker to complain.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11723
2011-05-04 09:06:17 +00:00
Julian Seward
a31f76b8fc
handle_maybe_load_notifier: assert when symbol == NULL rather than
...
segfaulting. Potential segfault was found by IBM's BEAM checker.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11722
2011-05-04 09:03:41 +00:00
Julian Seward
8911c2ffa5
VG_(env_unsetenv), VG_(env_clone): add assertions so as to cause
...
assertions instead of segfaults. Potential segfaults were detected by
IBM's BEAM checker.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11721
2011-05-04 09:01:58 +00:00
Julian Seward
555e3b2ccb
arm-linux: Set _start symbol alignment and type. Bug 266035 comment 1.
...
(Jeff Brown, jeffbrown@google.com )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11720
2011-05-03 14:24:11 +00:00
Bart Van Assche
b9d948061a
DRD: fix issues reported by the IBM static analysis tool.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11719
2011-04-30 07:27:41 +00:00
Julian Seward
3b0d8644cb
s390x : misc cleanups
...
- Forgotten cleanups in none/tests/s390x/Makefile.am
Partial fix for #271501 . (Florian Krohm, britzel@acm.org )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11718
2011-04-28 20:14:48 +00:00
Bart Van Assche
fbdd04d5a9
syswrap/Linux: trace ioctl() calls only once / do not report two-argument
...
ioctl() calls as an error. Patch provided by Mark Hills. Closes #272730 .
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11717
2011-04-28 18:36:49 +00:00
Julian Seward
5e6d4577de
Change the TT_FAST hash function for from "insn_address >> 2" to
...
"insn_address >> 1". The former is appropriate for ARM code, where
all insns are 4-sized and 4-aligned, but not for Thumb code, where the
minimum size and alignment is 2. The old scheme happened to work for
Thumb (indeed, any hash function would), but caused huge amounts of
conflict misses in the fast cache for some programs.
The change has been observed to reduce conflict misses by up to 100
times, and in some cases, improves performance significantly for Thumb
code. Performance of ARM code is unchanged or possibly a bit worse.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11716
2011-04-28 14:58:15 +00:00
Julian Seward
6045181f3c
Change the default (minimum) client malloc alignment from 8 to 16
...
on ppc32-linux. This is needed to make Altivec-using code work
correctly. Noticed when running ./auxprogs/gsl16test with gcc-4.6
with args -mcpu=970 -g -O3 -ftree-vectorize on Memcheck, in
which case a few of the tests failed because malloc() returns
8 byte aligned memory when it should return 16-aligned memory.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11715
2011-04-27 23:25:15 +00:00
Robert Walsh
7671cab5c8
Fix no-arg ioctls on Darwin.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11714
2011-04-27 19:04:31 +00:00
Julian Seward
482237674c
s390x: fpr - gpr transfer facility -- valgrind side fixes,
...
and test cases. Fixes #268619 .
(Florian Krohm, britzel@acm.org )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11713
2011-04-27 12:00:51 +00:00
Julian Seward
62c800bccb
Fix bogus asm constraints in jm-insns.c. Fixes #263919 .
...
(Maynard Johnson, maynardj@us.ibm.com )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11712
2011-04-27 11:33:44 +00:00
Julian Seward
63eda9ca5a
Add alternative expected output cases for more recent glibcs (eg,
...
2.12), which print a minus sign for NaNs. Fixes #262989 .
(Maynard Johnson, maynardj@us.ibm.com )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11711
2011-04-27 11:11:22 +00:00
Julian Seward
3a444148fd
Fix a bunch of printf format-string warnings when compiling this "-m64".
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11710
2011-04-27 08:09:37 +00:00
Julian Seward
d1ed036fc5
Fix a type mismatch that gcc-4.6 warns about.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11709
2011-04-26 10:23:54 +00:00
Julian Seward
46855c22e4
Remove some unused code that gcc-4.6 warns about.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11708
2011-04-26 10:23:33 +00:00
Julian Seward
f06f0f7185
Fix a couple of bogus asm constraints observed when trying to build
...
V with clang-2.9.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11707
2011-04-26 07:52:44 +00:00
Julian Seward
aa0d25d361
Fix a couple of warnings when building the regtests.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11706
2011-04-21 23:05:49 +00:00
Julian Seward
b1a118436a
Make reading of line number info from LLVM-2.9-generated Dwarf3 work.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11705
2011-04-21 08:55:51 +00:00
Julian Seward
9c2eb5a882
Change a bunch of < > style includes to " " style.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11704
2011-04-20 11:54:32 +00:00
Bart Van Assche
d077061870
Linux/syswrap: add support for USBDEVFS_RESET (patch contributed by Brad Hards <bradh@frogmouth.net>).
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11702
2011-04-18 10:37:56 +00:00
Julian Seward
fce26577fc
Fix bogus .size directives which are now rejected by binutils 2.21.
...
Fixes #271043 . (Maynard Johnson, maynardj@us.ibm.com )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11700
2011-04-15 21:21:27 +00:00
Julian Seward
b0716eae17
Un-break configure check for VSX (a POWER7 extension). Fixes #271042 .
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11699
2011-04-15 21:16:00 +00:00
Julian Seward
438a573452
Mark this as executable. Followup for #267630 .
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11698
2011-04-15 12:33:14 +00:00
Julian Seward
8dd8315914
Add support for IBM Power ISA 2.06 -- stage 1. Valgrind-side changes
...
and test cases. Bug #267630 and followup fix #270794 .
(Maynard Johnson, maynardj@us.ibm.com )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11697
2011-04-15 11:57:05 +00:00
Julian Seward
aefafd9e19
Followup fix for r11693: add a file that was missed.
...
Related to #270115 . (Florian Krohm, britzel@acm.org )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11696
2011-04-13 16:05:35 +00:00
Julian Seward
7681f394c3
s390x: reconsider "long displacement" requirement -- hwcaps detection
...
changes. See #268620 . (Florian Krohm, britzel@acm.org )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11695
2011-04-13 15:40:10 +00:00
Julian Seward
1ddfe2cc21
Test cases for #270082 (s390x: Make sure to point the PSW address to
...
the next address on SIGILL)
(Christian Borntraeger <borntraeger@de.ibm.com>)
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11694
2011-04-13 15:15:52 +00:00
Julian Seward
7a3b89c927
s390x: rewrite some testcases. To make my testing and debugging life
...
simpler, I've rewritten those testcases from none/tests/s390x that
require binutils to recognize opcodes defined in the
extended-immediate and general-instruction-extension facilities. As a
side effect this change removes the special casing in
none/tests/s390x/Makefile.am and the configury bits to support it.
Fixes #270115 . (Florian Krohm, britzel@acm.org )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11693
2011-04-13 14:57:44 +00:00
Bart Van Assche
bd7dbb42d3
timerfd-syscall regression test: Allow even more scheduler variation.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11692
2011-04-12 10:28:31 +00:00
Julian Seward
7e54c5b2ca
Get rid of shadowing causing a perl warning. Minor fallout from
...
r11686.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11691
2011-04-11 22:14:03 +00:00
Julian Seward
945eccc8e4
Make MacOS builds work again following fixes for #259977
...
(revs 11687,8,9)
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11690
2011-04-11 22:08:06 +00:00
Julian Seward
e5a3e4f014
Add an alternative implementation of VG_MINIMAL_{SET,LONG}JMP
...
for ppc32-linux, that works for gcc >= 4.4. Related to #259977 .
(modified version of patch from Maynard Johnson <maynardj@us.ibm.com>)
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11689
2011-04-11 21:26:27 +00:00
Julian Seward
c2120cadde
Add an alternative implementation of VG_MINIMAL_{SET,LONG}JMP
...
for ppc32-linux, that works for gcc >= 4.4. Related to #259977 .
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11688
2011-04-11 18:36:34 +00:00
Julian Seward
b3827d6c33
Create new module m_libcsetjmp, which wraps up uses of
...
__builtin_setjmp and __builtin_longjmp so that they can be selectively
replaced, on a platform by platform basis. Does not change any
functionality. Related to #259977 .
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11687
2011-04-11 16:17:51 +00:00
Julian Seward
dca5c116bb
On OSX, post-process the tool executables to adjust the Mach-O headers
...
in certain circumstances. This works around a bug in the linker that
ships in Xcode 4.0.0 and 4.0.1 causing the 64-bit tool executables to
segfault at startup. Fixes #267997 .
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11686
2011-04-06 11:17:16 +00:00
Julian Seward
7970c3d273
Testcases for s390x conditional load and store. See #269209 .
...
(Christian Borntraeger, borntraeger@de.ibm.com )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11685
2011-04-05 13:20:53 +00:00
Julian Seward
bddd46f059
Add s390x-esque test for folding Iop_And16. From #268513 comment 3.
...
(Florian Krohm, britzel@acm.org )
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11684
2011-04-05 11:15:28 +00:00
Bart Van Assche
d411ca1dee
Made the line numbers in exp-ptrcheck/tests/partial.c and the regression test output match again.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11683
2011-04-04 10:15:33 +00:00
Bart Van Assche
8a3e9fdd7c
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11682
2011-04-03 17:42:19 +00:00
Bart Van Assche
954dc6c188
Also consider libpthread when testing for functions. Removed test for floor().
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11681
2011-04-03 16:46:01 +00:00
Bart Van Assche
f5001fcc32
Updated Subversion ignore list.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11680
2011-03-30 18:59:56 +00:00
Bart Van Assche
245dbc33b3
Generalized Subversion ignore lists.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11679
2011-03-30 18:58:40 +00:00
Julian Seward
857f34e035
Fix yet another gcc-4.6 assigned-to-but-not-used warning.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11677
2011-03-29 17:05:09 +00:00
Bart Van Assche
44a6867773
timerfd-syscall regression test: Allow even more scheduler variation.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11676
2011-03-29 10:56:32 +00:00
Julian Seward
b8a03bdc8e
Remove a bunch more warnings generated by gcc-4.6 about dead
...
assignments ("[-Wunused-but-set-variable]"), in the test suite.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11675
2011-03-28 21:19:00 +00:00