Make error messages at start-up more consistent. Every line of such
messages now begin with "valgrind: ", and they're more often printed before
the preamble. This required introducing a new message kind, Vg_FailMsg, and
functions VG_(fmsg) and VG_(fmsg_bad_option), and removing
VG_(err_bad_option).
Where we used to have horrible output like this:
[ocean:~/grind/ws2] vg5 --tool=massif --threshold=101 date
==31877== Massif, a heap profiler
==31877== Copyright (C) 2003-2010, and GNU GPL'd, by Nicholas Nethercote
==31877== Using Valgrind-3.6.0.SVN and LibVEX; rerun with -h for copyright info
==31877== Command: date
==31877==
==31877== --threshold must be between 0.0 and 100.0
valgrind: Bad option '--threshold'; aborting.
valgrind: Use --help for more information.
We now have nice output like this:
[ocean:~/grind/ws2] vg2 --tool=massif --threshold=101 date
valgrind: Bad option: --threshold=101
valgrind: --threshold must be between 0.0 and 100.0
valgrind: Use --help for more information or consult the user manual.
Convert x86-darwin to amd64-darwin at configure time, because Macs report as
32-bit even though almost all of them support 64-bit as well. A hack, but
one that will make just about everyone's lives easier.
Callgrind now uses Cachegrind's command line option to switch
on simulation: "--branch-sim=yes/no" for branch prediction,
and "--cache-sim=yes/no" for cache simulation (for more
consistency and to avoid confusion). However, the previously
used "--simulate-cache=yes/no" still is supported but deprecated.
Don't assert when trying to show a Darwin syscall number in an unknown
syscall class. This stops Valgrind asserting when doing
--trace-syscalls=yes and a sys_syscall is encountered, in 64-bit mode.
I don't know if this is a correct fix or whether it merely hides
symptoms of logic handling problems for sys_syscall on Darwin.
Then I had to fix up coregrind/link_tool_exe*.in -- those files had been
added independently on both the trunk and the branch, AFAICT. I just
overwrote the trunk versions with the branch versions.
Allow negative numbers in the cachegrind.out.* file. Also protect against
division-by-zero. Both are required for cg_diff (not yet committed) to work
properly.
Callgrind new feature: count global bus lock events "Ge"
To count global bus lock events, use "--collect-bus=yes".
For x86, this will count the number of executed instructions
with a lock prefix; for architectures with LL/SC, this will
count the number of executed SC instructions.
Callgrind: make globals for log_* helpers visible outside sim.c
At beginning of each BB, Callgrind inserts a call to setup_bbcc,
which (among a lot other things), sets global vars needed for
the log_* helpers called afterwards in this BB.
These globals, bb_base and cost_base, previously we static declared
and only visible in sim.c. Make them visible also in the rest of
callgrind to allow for log_* handlers outside sim.c.
Offsets of groups within event sets are now
stored in a offset table as part of the event set, and
can be accessed directly. Before, offsets were hardcoded
all over the place. This should make addition of further
event groups much simpler.
Julian Seward [Wed, 2 Jun 2010 00:31:34 +0000 (00:31 +0000)]
Add a new program (link_tool_exe.c) and use this to link the tool
executables. Gets rid of the linker script kludgery and uniformly
uses -Ttext=0x38000000 (or whatever) on Linux, so as to accomodate
both traditional ld and gold. Should fix #193413 although I have
been unable to test it. Using a whole new program seems like
overkill, but this is infrastructure to support static linking of
the tool executables on MacOS too.
Bart Van Assche [Thu, 13 May 2010 06:32:36 +0000 (06:32 +0000)]
Added support for glibc 2.12.
Note: many Helgrind and DRD regression tests still fail on Fedora 13 because
of differences in the call stacks of error reports compared to earlier
glibc/gcc combinations.
Julian Seward [Sun, 9 May 2010 22:30:43 +0000 (22:30 +0000)]
New flag --require-text-symbol=:sopatt:fnpatt, to be used to check
that specified shared objects contain specified symbols. Along with a
couple of regtests that unfortunately will fail on MacOSX.
Tom Hughes [Fri, 7 May 2010 15:32:04 +0000 (15:32 +0000)]
Add support for some EVIO ioctls. These are a bit abnormal because
the length part of the ioctl is not fixed so they are essentially
families of ioctls.
Based on patch from Peter Korsgaard. Fixes #235642.
Julian Seward [Wed, 5 May 2010 09:34:42 +0000 (09:34 +0000)]
Rollup changes for Helgrind:
* Add new client request VALGRIND_HG_CLEAN_MEMORY_HEAPBLOCK. This is
like VALGRIND_HG_CLEAN_MEMORY but doesn't take an address range.
Instead it takes a single argument which is supposed to be a pointer
to the start of, or anywhere within, a heap allocated block.
Helgrind then finds the block and paints it as belonging to the
calling thread. This is needed for correctly describing the
behaviour of threadsafe reference counting when applied to classes
involving inheritance of release methods or involving multiple
inheritance.
* Add statistics counters for all basic VTS operations (tick, join,
cmpLEQ, cmp_structural).
Julian Seward [Wed, 5 May 2010 09:23:41 +0000 (09:23 +0000)]
Add a method to get the contents of an XArray so we can index in
it really fast, or iterate over it. This is dangerous and breaks
the nice abstraction (sigh).
Tom Hughes [Thu, 29 Apr 2010 09:22:25 +0000 (09:22 +0000)]
Link the tls test against tls2.so as it references symbols in it
and Fedora 13 no longer allows so's to be required indirectly (via
tls.so in this case).
Tom Hughes [Thu, 29 Apr 2010 09:01:21 +0000 (09:01 +0000)]
Include valgrind headers before system headers to avoid problems
with the system headers #defining things which are used as names
of structure members in vki headers.
This is needed to allow valgrind to build on Fedora 13 and Rawhide.
Tom Hughes [Wed, 28 Apr 2010 08:09:30 +0000 (08:09 +0000)]
Add some basic DWARF4 support. Based on patch from Jakub Jelinek
but with support for VLIW architectures with multiple opcodes per
instruction removed. Fixes #233595.
search_all_symtabs: when mapping addresses to text symbols, consider
any symbol in the r-x mapped segment to be a valid candidate. This
relaxes the filtering criterion slightly, makes it consistent with
other is-it-text? checks. Some addresses which before didn't get
mapped to anything are now correctly mapped to "vtable for Foo"
symbols.
Change the method used in hg_intercepts.c to hide from the user, the
race between mythread_wrapper and the wrapper for pthread_create. The
previous scheme could lead to false race reports in obscure cases.
Julian Seward [Sun, 14 Mar 2010 15:09:27 +0000 (15:09 +0000)]
Improve performance of the fallback path when a translation is not
found in the fast-cache.
* reduce max loading of the per-sector TT hash tables from 80% to 65%.
This reduces the number of required probes by a factor of 3.
* when searching for a translation, don't visit the sectors in a fixed
order. Instead, use an MTF array in which the most popular sectors
(in terms of most likely to hold the translation we're looking for)
are visited first. This reduces the number of required probes by
another factor of 2.
These improvements have no effect on small programs, but improve
scalability on big apps. For an application comprising 300k
translations, runtime on Memcheck is reduced by 3% and on None by
about 20%. The average number of probes per fast-cache miss is
reduced from around 22 to less than 5.
Julian Seward [Thu, 11 Mar 2010 13:43:18 +0000 (13:43 +0000)]
If a race error is detected, check to see whether the raced-on address
is inside a heap block, and if so, print the allocation point of the
heap block. It's stupid not to do this considering that the
implementation already keeps track of all mallocs and frees.
projects / thirdparty / valgrind.git / log
