platforms, otherwise the address-masking operations to establish
alignment and primary-mappability are wrong on 64-bit platforms.
Also set the size of fast-mapped address space on 64-bit platforms to
16G.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3580
vex rev 1144.
* Observe that mkLazy2 generates IR which often turns into
long and slow code sequences in the back end, primarily because
PCast operations are expensive. Add a couple of special
cases which give noticably better performance when handling
FP-intensive code on x86.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3572
MC_(helperc_LOADV8) compiles down to just 12 instructions for the
fast-path, which is pretty darn good.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3547
account for the fact that on amd64 (really, on amd64-linux) the area
up to 128 bytes below the stack pointer is accessible. This meant
moving the definitions of VGA_STACK_REDZONE_SIZE to tool-visible
places.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3546
name the events, rather than just number them, which makes it a
lot easier to use
* Based on that, fill in some fast-path cases
{LOAD,STORE}V{4,2,1}. The assembly code looks about the same
length as it did before, on x86. Fast-path cases for the
stack have yet to be done.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3538
there is a better way to handle the 'pessimising cast' family of
operations in such a way that Vex's back-end instruction selectors can
generate better code than they do now, with less verbosity and general
confusingness in the insn selectors.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3536
32- and 64-bit targets, little- and big-endian. It does more or less
work on x86 as-is, although is unusably slow since I have knocked out
all the fast-path cases and am concentrating on getting the baseline
functionality correct. The fast cases will go back in in due course.
The fundamental idea is to retain the old 2-level indexing for speed,
even on a 64-bit target. Since that's clearly unviable on a 64-bit
target, the primary map handles only first N gigabytes of address
space (probably to be set to 16, 32 or 64G). Addresses above that are
handled slowly using an auxiliary primary map which explicitly lists
(base, &-of-secondary-map) pairs. The goal is to have the
address-space-manager try and put everything below the 16/32/64G
boundary, so we hit the fast cases almost all the time.
Performance of the 32-bit case should be unaffected since the fast map
will always cover at least the lowest 4G of address space.
There are many word-size and endianness cleanups.
Jeremy's distinguished-map space-compression scheme is retained, in
modified form, as it is simple and seems effective at reducing
Memcheck's space use.
Note this is all subject to rapid change.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3535
lot to one. This required two basic changes.
1. Tools are responsible for telling the tool about any functions they
provide that the tool may call. This includes basic functions like
TL_(instrument)(), functions that assist core services such as
TL_(pp_Error)(), and malloc-replacement-related functions like
TL_(malloc)().
2. Tools that replace malloc now specify the size of the heap block redzones
through an arg to the VG_(malloc_funcs)() function, rather than with a
variable VG_(vg_malloc_redzone_szB).
One consequence of these changes is that VG_(tool_init_dlsym)() no longer
needs to be generated by gen_toolint.pl.
There are a number of further improvements that could follow on from this one.
- Avoid the confusingly different definitions of the TL_() macro in the
core vs. for tools. Indeed, the functions provided by the tools now don't
need to use the TL_() macro at all, as they can have arbitrary names.
- Remove a lot of the auto-generated stuff in vg_toolint.c and vg_toolint.h
(indeed, it might be possible to not auto-generate these at all, which
would be nice).
- The handling of VgToolInterface is currently split across vg_needs.c and
vg_toolint.c, which isn't nice.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3487
two halves: stacktrace.c, which deals with getting, traversing and printing
stack traces; and execontext.c, which deals with storing stack traces
permanently in a way that avoids duplicates, and comparing them.
One nice outcome: previously we were often creating ExeContexts, which live
forever, even when they were only needed temporarily. Ie. this was a memory
leak, which has been removed.
As part of this, new headers have been created, carved off core.h and
tool.h. Lots of function names have changed, too.
In Massif, I also changed a lot of "eip" names to "ip" to make them less
x86-specific.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3429
sizes to the instrumentatation functions. Make most of the tools
abort if they are not the same; we can't handle that case yet.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3397
corecheck/tests and there is no point in duplicating it. Given that
it was intended to check the behaviour of our now-defunct pthread
implementation there's not much point in keeping it at all.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3378
user-specified block stuff, which confused me because I wasn't sure whether
to use the code already in SVN, or the code in CVS. Perhaps that code
doesn't need to be changed.
Unfortunately, Memcheck doesn't work entirely correctly -- I get some
spurious errors. Nonetheless I'm checking it in as a starting point.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3353
because the added VG_(find_root_memory)() is just a stub. And there's a
problem with overlap checking that I haven't worked out yet. Still it's a
start. The commit also brings Memcheck back into the build process,
although mc_main.c is entirely commented out at the moment.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3352
