Massif: specify avg translation size at all, so as to avoid excessive
retranslations caused by the fact that the default value is far below
reality for Massif.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11494
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.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@11091
the changes to do with reading and using ELF and DWARF3 info.
This breaks all targets except amd64-linux and x86-linux.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@10982
This commit tidies up and rationalises what could be called the
"messaging" system -- that part of V to do with presenting output to
the user. In particular it brings significant improvements to XML
output.
Changes are:
* XML and normal text output now have separate file descriptors,
which solves longstanding problems for XML consumers caused by
the XML output getting polluted by unexpected non-XML output.
* This also means that we no longer have to hardwire all manner
of output settings (verbosity, etc) when XML is requested.
* The XML output format has been revised, cleaned up, and made
more suitable for use by error detecting tools in general
(various Memcheck-specific features have been removed). XML
output is enabled for Ptrcheck and Helgrind, and Memcheck is
updated to the new format.
* One side effect is that the behaviour of VG_(message) has been
made to be consistent with printf: it no longer automatically
adds a newline at the end of the output. This means multiple
calls to it can be used to build up a single line message; or a
single call can write a multi-line message. The ==pid==
preamble is automatically inserted at each newline.
* VG_(message)(Vg_UserMsg, ..args..) now has the abbreviated form
VG_(UMSG)(..args..); ditto VG_(DMSG) for Vg_DebugMsg and
VG_(EMSG) for Vg_DebugExtraMsg. A couple of other useful
printf derivatives have been added to pub_tool_libcprint.h,
most particularly VG_(vcbprintf).
* There's a small change in the core-tool interface to do with
error handling: VG_(needs_tool_errors) has a new method
void (*before_pp_Error)(Error* err) which, if non-NULL, is
called just before void (*pp_Error)(Error* err). This is to
give tools the chance to look at errors before any part of them
is printed, so they can print any XML preamble they like.
* coregrind/m_errormgr.c has been overhauled and cleaned up, and
is a bit simpler and more commented. In particular pp_Error
and VG_(maybe_record_error) are significantly changed.
The diff is huge, but mostly very boring. Most of the changes
are of the form
- VG_(message)(Vg_UserMsg, "this is a message %d", n);
+ VG_(message)(Vg_UserMsg, "this is a message %d\n", n);
Unfortunately as a result of this, it touches a large number
of source files.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@10465
DARWIN branch. A big ugly DARWIN/trunk sync commit, mostly to do with
changing the representation of SysRes and vki_sigset_t. Functionality of
the trunk shouldn't be changed by it.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@9876
relatively minor extensions to m_debuginfo, a major overhaul of
m_debuginfo/readdwarf3.c to get its space usage under control, and
changes throughout the system to enable heap-use profiling.
The majority of the merged changes were committed into
branches/PTRCHECK as the following revs: 8591 8595 8598 8599 8601 and
8161.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@8621
more cache friendly. This changes the mechanism from being a table of
pointers to (guest address, translated code pairs) to being a table of
pairs (guest address, pointer to translated code). The effect ranges
from zero up to about 20% performance improvement on memcheck, the
biggest effects being seen for programs which jump around a large
number of blocks of code and whose data set does not fit in L2.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@6582
isn't in effect. This should cause the dispatcher to segfault if it
should ever inadvertantly end up running the profiled dispatch loop
when it shouldn't. [The alternative is to run slowly for no apparent
reason.]
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@6310
translation cache. This ensures that the cache space is executable
which it isn't when it is allocated as a static variable in the data
segment, at least on my amd64 box.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@5525
branch hereby becomes inactive. This currently breaks everything
except x86; fixes for amd64/ppc32 to follow.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@5520
perf/tinycc.
- run_thread_for_a_while: just clear this thread's reservation when
starting, not all of them.
- use a different fast-cache hashing function for ppc32/64 than for
x86/amd64. This allows the former to use all the fast-cache entries
rather than just 1/4 of them.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@5441
add a bunch of code to detect what the cpu can do at startup by
catching SIGILLs. Shame PPC doesn't offer any sane mechanism for
finding out what instruction subsets the CPU is capable of (a la
x86/amd64 cpuid).
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@5108
the guest extents for the presented translation and also its original
un-redirected guest address. These changes are needed in particular
to make cachegrind's code cache management work properly.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4943
find and delete all translations intersecting with small address
ranges (8 k or less, currently). This makes it possible to simulate
ppc32 icbi instructions in reasonable time, and finally makes the
ppc32 port run at a usable speed.
The scheme is based around partitioning translations into equivalence
classes based on address ranges. For deletions whose range falls
within a single class, all translations intersecting it can be found
by inspecting just that class and one other. Given that there are 256
classes, this cuts the cost, relative to scanning the entire TC, by
approximately half that factor (viz, 128), assuming the translations
are distributed evenly over the classes.
The whole business is more complex and difficult than I would like.
A detailed comment will later be added.
Very thorough sanity checking has been added
(sanity_check_eclasses_in_sector). This is engaged at
--sanity-level=4 and above.
The TT hash function (HASH_TT) has been improved to reduce its
tendency to cluster TT entries in some circumstances. This has
allowed the TT maximum loading factor to be increased from 66% to 80%
and so the absolute size of the TC (in each sector) to be less than
2^16 entries. The latter change is important for the fast-deletion
changes.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4942
deal with any IR that happens to show up. This makes it work on ppc32
and should fix occasionally-reported bugs on x86/amd64 where it bombs
due to having to deal with multiple date references in a single
instruction.
The new scheme is based around the idea of a queue of memory events
which are outstanding, in the sense that no IR has yet been generated
to do the relevant helper calls. The presence of the queue --
currently 16 entries deep -- gives cachegrind more scope for combining
multiple memory references into a single helper function call. As a
result it runs 3%-5% faster than the previous version, on x86.
This commit also changes the type of the tool interface function
'tool_discard_basic_block_info' and clarifies its meaning. See
comments in include/pub_tool_tooliface.h.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4903
Changed some printf specifiers accordingly, plus some more that were
incorrect.
Also put commas in various output numbers, eg. the leak check stats.
This makes them much easier to read when they get big. One
exception is in XML number-only fields such as <leakedbytes>.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4874
- on 64-bit platforms, double the size of the supported address
space to 32G.
- Increase the size of the ExeContext table 6 times. Some very
large apps have been observed to have been doing a lot of
searching in the old 4999 entry table. This table may be
OSetified in the fullness of time.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4808
changes from r4341 through r4787 inclusive). That branch is now dead.
Please do not commit anything else to it.
For the most part the merge was not troublesome. The main areas of
uncertainty are:
- build system: I had to import by hand Makefile.core-AM_CPPFLAGS.am
and include it in a couple of places. Building etc seems to still
work, but I haven't tried building the documentation.
- syscall wrappers: Following analysis by Greg & Nick, a whole lot of
stuff was moved from -generic to -linux after the branch was created.
I think that is satisfactorily glued back together now.
- Regtests: although this appears to work, no .out files appear, which
is strange, and makes it hard to diagnose regtest failures. In
particular memcheck/tests/x86/scalar.stderr.exp remains in a
conflicted state.
- amd64 is broken (slightly), and ppc32 will be unbuildable. I'll
attend to the former shortly.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4789
via the use of self-checking translations. (Friendly platforms which
have icache-invalidation instructions we can observe, such as ppc32,
are already handled correctly.) This should finally fix the
longstanding problem of V incorrectly handling calls of statically
nested functions (a gcc extension), and more generally make it a lot
easier to use V to debug dynamic code generation systems.
Since self-checking is a large performance overhead, there is some
control via a command line flag:
--smc-support=none
Don't make any translations self-checking.
--smc-support=stack
Add checking code for translations taken from segments which
have the SF_GROWDOWN flag set -- stacks, basically.
This is the default. It should make gcc nested functions and
GNU Ada work correctly with no intervention from the user.
--smc-support=all
Make all translations self-checking. This is expensive and
you want to do this if you're debugging a JIT compiler or
some such.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4122
array at startup. This is used in m_transtab. However this info is
not yet fed to Vex, so it's still important to zero-out the auxv field
holding cache line size info passed to the client, so as to stop the
client's glibc using dcbz. This will be fixed.
Also get rid of a bunch more ppc32-specific vdso stuff in m_main that
doesn't need to be done. This now means ppc32-linux specifics in
m_main are only marginally intrusive than the x86-linux or amd64-linux
specifics in m_main.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4052
removes m_transtab's dependence on m_translate (breaking a circular
dependence) and m_debuginfo, hooray.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@4035
Plenty still to do, but simple programs like ls seem to run ok
Thanks, Paul, for having your ppc port of valgrind 2.4 to work from!
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3969
- Broke part of m_scheduler off into a new module m_threadstate. It
contains ThreadState, VG_(threads)[] and some basic operations on the
thread table. All simple stuff, the complex stuff stays in m_scheduler.
This avoids lots of circular dependencies between m_scheduler and other
modules.
- Managed to finally remove core.h and tool.h, double hurrah!
- Introduced pub_tool_basics.h and pub_core_basics.h, one of which is
include by every single C file.
- Lots of little cleanups and changes related to the above.
- I even did a small amount of documentation updating.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@3944
