Nicholas Nethercote
e2c5384124
De-document now-defunct command line options.
...
MERGED FROM CVS HEAD
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3271
2005-03-11 04:35:00 +00:00
Nicholas Nethercote
c3187ad891
Added the Quick Start Guide, in XML. Not entirely happy with the formatting
...
(it should be all one page), but it's a start.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3270
2005-03-11 04:33:29 +00:00
Julian Seward
d80694b265
Summarise the breakage caused by the Big Merge (hence the current
...
situation).
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3269
2005-03-11 03:07:23 +00:00
Julian Seward
ae8df0379b
Fix PRE wrapper for sys_exit.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3268
2005-03-11 02:47:32 +00:00
Nicholas Nethercote
0331bb5f9e
Fix broken xml tags.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3267
2005-03-11 02:43:17 +00:00
Julian Seward
d5b545da24
These files are no longer needed. Bye!
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3266
2005-03-11 00:22:10 +00:00
Julian Seward
0356d27ca6
Merge in changes from the 2.4.0 line. This basically brings in the
...
overhaul of the thread support. Many things are now probably broken,
but at least with --tool=none, simple and not-so-simple threaded and
non-thread programs work.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3265
2005-03-10 23:59:00 +00:00
Julian Seward
7c542ccd39
Add new files resulting from merging in the 2.4.0 line. Many of these
...
seem to be simply duplication of the x86 instruction set tests into
the addrcheck and helgrind trees. I'm not sure what this duplication
achieves.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3264
2005-03-10 23:23:45 +00:00
Nicholas Nethercote
306f41edc7
Test commit (whitespace-only change).
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3263
2005-03-10 22:20:11 +00:00
Julian Seward
ef0743de0a
Test commit for new repo.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3262
2005-03-10 13:28:41 +00:00
Julian Seward
2e00d9e22a
Fix enough syscall stuff so that simple X applications run (xfontsel,
...
xedit).
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3261
2005-03-02 16:01:23 +00:00
Julian Seward
fb5a19f31f
A place to record notes on how to clean up the structure. This is
...
stuff that I discover as I go along.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3260
2005-03-02 14:22:49 +00:00
Julian Seward
243713971c
Minimal, horrible hacks needed to make vsyscalls work on amd64.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3259
2005-03-02 14:06:08 +00:00
Julian Seward
3fe41b4fd9
Fix struct ume_auxv so that auxc walking on 64-bit platforms works.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3258
2005-03-02 14:04:19 +00:00
Julian Seward
435d820b6d
Clean up handling of socket system calls on amd64-linux, so that the
...
code for them is shared between amd64-linux and x86-linux.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3257
2005-03-01 19:00:30 +00:00
Julian Seward
a978be5bcb
Factor out all the socket-related syscall handlers into their own
...
functions, so the same handlers can be used on amd64-linux as
x86-linux without tons of code duplications.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3256
2005-03-01 16:45:23 +00:00
Julian Seward
d2445e3d81
Fix more RES-is-not-set assertions, and move a couple of comments to
...
their correct homes.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3255
2005-02-28 20:50:29 +00:00
Julian Seward
006c9e84c3
When handling syscalls, don't try to figure out if the pre-handler set
...
the syscall result by inspecting RES after the pre-handler has run.
Instead, give each thread a syscall_result_set Bool, and make
SET_RESULT set it. Inspect that Bool.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3254
2005-02-28 17:27:04 +00:00
Julian Seward
be6b72b2d3
Various cleanups to revised low-level memory manager.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3253
2005-02-28 14:39:21 +00:00
Julian Seward
2cfb6bc855
On AMD64 it's valid to access up to 128 bytes below %rsp. Or to be
...
more accurate, on the amd64-linux ABI that is allowable. Anyway,
parameterise the signal handler so it isn't confused by such accesses.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3252
2005-02-25 02:45:39 +00:00
Julian Seward
4bfc4e090b
Rearrange debug printing a bit.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3251
2005-02-23 11:36:56 +00:00
Julian Seward
a84f1bd656
More amd64 system calls.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3250
2005-02-19 22:45:15 +00:00
Julian Seward
f62998dc1b
Clean up startup grunge a bit.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3249
2005-02-18 09:39:05 +00:00
Julian Seward
4b34a53a19
Make ELF symbol reading work on both 32 and 64 bit platforms.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3248
2005-02-18 09:38:08 +00:00
Julian Seward
f122d86dc0
Further cleanups to low-level memory management. It's still a
...
conceptual mess and needs a redesign, but this is a start. Most stuff
now works again.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3247
2005-02-18 08:28:32 +00:00
Julian Seward
a5787a4558
Fix enough syscalls to get through 'hello world' on amd64.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3246
2005-02-17 09:30:21 +00:00
Julian Seward
07be6896f6
Fix comment.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3245
2005-02-17 09:29:33 +00:00
Julian Seward
353c6ca6c5
Turn off tons of debug printing.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3244
2005-02-16 01:58:51 +00:00
Julian Seward
e57df07ce8
Duh; mmap doesn't work if you don't use the post-mmap wrapper too.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3243
2005-02-16 01:58:01 +00:00
Julian Seward
245173117c
Major rewrite of the machinery which keeps track of segments:
...
* remove initialisation-time circularities by not using the
skiplist mechanism and therefore not using any dynamically
allocated memory in support of this mechanism
* Add comments about how it works (it is pretty opaque) so as
to help future maintainers/bug-fixers
It only just works and many things are still broken. That should
improve rapidly however.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3242
2005-02-16 01:31:37 +00:00
Julian Seward
d2897b0a2b
Remove more debug printing.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3241
2005-02-13 11:05:31 +00:00
Julian Seward
d033437190
Remove misc debug printing.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3240
2005-02-13 11:02:12 +00:00
Julian Seward
67a7a983cf
Change signature of thread_syscall a bit.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3239
2005-02-13 10:57:01 +00:00
Julian Seward
3e6c170f75
Debugging hacks (switchback routine -- unused)
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3238
2005-02-13 10:56:22 +00:00
Julian Seward
2e98ffac01
Debug printing at startup.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3237
2005-02-13 10:55:47 +00:00
Julian Seward
54e34d76ee
Debug hacks (currently unused)
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3236
2005-02-13 10:53:51 +00:00
Julian Seward
a804b287ad
Fix bug causing __NR_read on amd64-linux to always be considered to have
...
failed -- because __NR_read == 0.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3235
2005-02-13 10:49:58 +00:00
Julian Seward
60454f22a5
amd64-linux syscall support.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3234
2005-02-13 10:47:24 +00:00
Julian Seward
768ce51a26
Make amd64 signal returns work.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3233
2005-02-08 15:00:48 +00:00
Julian Seward
8d9a222274
Track renaming of Vex's Iop_*128* primops to Iop_*V128*.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3232
2005-02-07 23:50:18 +00:00
Julian Seward
bb0c8347a1
Make --wait-for-gdb sort-of usable on amd64. I don't know why this
...
works on x86 -- and I can't make any analogous version work on amd64.
Hence return to a delay loop for the time being.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3231
2005-02-06 14:10:40 +00:00
Julian Seward
f224865e9e
Fix enough stuff so it will start up and run for a few bbs on amd64,
...
before dying.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3230
2005-02-05 18:27:14 +00:00
Julian Seward
a576e88663
Fix amd64 assembly for presumably newer binutils.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3229
2005-02-05 14:59:25 +00:00
Julian Seward
056ed053a8
Increase the number of top-bbs in a profile to a useful number.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3228
2005-01-24 10:42:46 +00:00
Julian Seward
24777514e0
AMD64 build fixes.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3227
2005-01-19 14:05:09 +00:00
Julian Seward
efcf1e61ec
Complete rewrite (I think this is the 4th incarnation) of translation
...
table/cache management. Two main changes. (1) Translation areas are
recorded using VexGuestExtents, so that Vex is now properly supported
and code deletion works correctly. (2) Low overhead BB profiling,
enabled by the --profile-flags=<XXXXXXXX> option. This finds the top
N bbs at exit and shows them, so as to give a basis from which to do
performane tuning.
To support this, the way tt/tc work is changed. It is still a
sectored arrangement, but now each sector has its own hash table.
This simplifies a lot of things.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3226
2005-01-19 11:55:34 +00:00
Julian Seward
d19f789514
Track VEX API changes introduced in Vex rev 718.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3225
2005-01-17 18:35:30 +00:00
Julian Seward
990997c313
Handle Add64/Sub64 (naively).
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3224
2005-01-15 20:43:58 +00:00
Julian Seward
9091a78381
Add support for 64-bit SIMD primops.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3223
2005-01-13 19:17:06 +00:00
Julian Seward
7956de0e96
Re-enable regression tests for PSLLDQ.
...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3222
2005-01-11 15:02:34 +00:00