On s390x-linux, adds CFI based unwinding for %f0..%f7, since these are sometimes
used by gcc >= 8.0 to spill integer register values in leaf functions. Hence the
lack of unwinding them was causing unwind failures on this platform.
* sym name cache to cache the 'isText' characteristic of the cached sym_name
* implement the match anywhere also in the cache
Not handling matchAnywhere has bad performance effect on callgrind, as almost many
IPs were considered as a fn entry, while they were not.
* use the same convention to name the findText argument
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@16136
* Complain instead of asserting when negative line nr in inline info
* use a macro to factorise the complain once logic about lineno limits
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@15780
Line numbers should correctly reflect all instructions belonging to a source line,
regardless of is_stmt value. Previously only instructions covered by
'is_stmt = 1' were attributed to a source line.
Fixes BZ#356044
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@15741
This implements the interception of all globally public allocation
functions by default. It works by adding a flag to the spec to say the
interception only applies to global functions. Which is set for the
somalloc spec. The librarypath to match is set to "*" unless the user
overrides it. Then each DiSym keeps track of whether the symbol is local
or global. For a spec which has isGlobal set only isGlobal symbols will
match.
Note that because of padding to keep the addresses in DiSym aligned the
addition of the extra bool isGlobal doesn't actually grow the struct.
The comments explain how the struct could be made more compact on 32bit
systems, but this isn't as easy on 64bit systems. So I didn't try to do
that in this patch.
For ELF symbols keeping track of which are global is trivial. For pdb I
had to guess and made only the "Public" symbols global. I don't know
how/if macho keeps track of global symbols or not. For now I just mark
all of them local (which just means things work as previously on platforms
that use machos, no non-system symbols are matches by default for somalloc
unless the user explicitly tells which library name to match).
Included are two testcases for shared libraries (wrapmalloc) and staticly
linked (wrapmallocstatic) malloc/free overrides that depend on the new
default. One existing testcase (new_override) was adjusted to explicitly
not use the new somalloc default because it depends on a user defined
new implementation that has side-effects and should explicitly not be
intercepted.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@15726
Valgrind aspects, to match vex r3124.
See bug 339778 - Linux/TileGx platform support to Valgrind
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@15080
code. Previously functions exceeding that size were observed in the
field. Assert for 100x that amount.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14997
loop forever. In particular, it will do that when two symbols have
exactly the same address range but differ in their .isText attribute.
Fixes#342117.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14948
that once an element has been allocated and added to the pool it must
not be modified afterwards. See the documentation in pub_tool_deduppoolalloc.h
The rest of the patch is ripple.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14654
Tool files shall use tl_assert not vg_assert.
Fix code accordingly.
Adapted check_headers_and_includes to make sure the code
stays clean in that respect.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14542
- Document that the allocation function must ot return NULL.
- As a conequence of the previous requirement the various Create and AllocNode
functions cannot return NULL. Remove pointless asserts at call sites.
- Remove documentation of undefined function CreateWithCmp.
- Names of library functions (such as 'free') are reserved as a are names
beginning with underscores. Don't use those.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14531
(used for ppc64 platforms) #ifdef-ed and accessed by macros
that becomes NOP on non ppc64 platforms.
This decreases the debuginfo memory by about 2.5 Mb on a big 32 bit application.
Note : doing that, some questions were encountered in the way
tocptr and local_ep have (or do not have) to be copied/maintained
in storage.c canonicaliseSymtab
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14273
files in the dwarf3 reader.
Basically, the change consists in replacing in the DiInlLoc struct
const HChar* filename; /* caller source filename */
by
UInt fndn_ix; /* index in di->fndnpool of caller source
dirname/filename */
A similar change is done in DiVariable struct, as the
read_filename_Table code is shared between the inline info reader
and the varinfo reader.
Note however that outputting dirname in variable description
is not done. Unclear if that is desired or not.
It should be trivially doable however.
Replacing filename by fndn_ix implies a bunch of semi-mechanical
changes.
The code to read the directory names is in the new function
static
XArray* read_dirname_xa (struct _DebugInfo* di, const HChar *compdir,
Cursor *c,
Bool td3 )
Note that readdwarf.c and readdwarf3.c have significant duplicated
logic. Would be nice to integrate these 2 dwarf readers in one
single reader. This function is directly inspired from
an equivalent piece of code in readdwarf.c.
Modified memcheck/tests/varinfo5.vgtest to test the dirname appears
in the inlined functions.
Impact on memory is neglectable (a few Kb on a big executable).
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14245
to add PPC64 LE support. The other two patches can be found in Bugzillas
334384 and 334836.
POWER PC, add the functional Little Endian support, patch 2
The IBM POWER processor now supports both Big Endian and Little Endian.
The ABI for Little Endian also changes. Specifically, the function
descriptor is not used, the stack size changed, accessing the TOC
changed. Functions now have a local and a global entry point. Register
r2 contains the TOC for local calls and register r12 contains the TOC
for global calls. This patch makes the functional changes to the
Valgrind tool. The patch makes the changes needed for the
none/tests/ppc32 and none/tests/ppc64 Makefile.am. A number of the
ppc specific tests have Endian dependencies that are not fixed in
this patch. They are fixed in the next patch.
Per Julian's comments renamed coregrind/m_dispatch/dispatch-ppc64-linux.S
to coregrind/m_dispatch/dispatch-ppc64be-linux.S Created new file for LE
coregrind/m_dispatch/dispatch-ppc64le-linux.S. The same was done for
coregrind/m_syswrap/syscall-ppc-linux.S.
Signed-off-by: Carl Love <carll@us.ibm.com>
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14239
to add PPC64 LE support. The other two patches can be found in Bugzillas
334834 and 334836. The commit does not have a VEX commit associated with it.
POWER PC, add initial Little Endian support
The IBM POWER processor now supports both Big Endian and Little Endian.
This patch renames the #defines with the name ppc64 to ppc64be for the BE
specific code. This patch adds the Little Endian #define ppc64le to the
Additionally, a few functions are renamed to remove BE from the name if the
function is used by BE and LE. Functions that are BE specific have BE put
in the name.
The goals of this patch is to make sure #defines, function names and
variables consistently use PPC64/ppc64 if it refers to BE and LE,
PPC64BE/ppc64be if it is specific to BE, PPC64LE/ppc64le if it is LE
specific. The patch does not break the code for PPC64 Big Endian.
The test files memcheck/tests/atomic_incs.c, tests/power_insn_available.c
and tests/power_insn_available.c are also updated to the new #define
definition for PPC64 BE.
Signed-off-by: Carl Love <carll@us.ibm.com>
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14238
On a big executable, the trunk needs:
dinfo: 134873088/71438336 max/curr mmap'd, 134607808/66717872 max/curr
With the patch, we have:
dinfo: 99065856/56836096 max/curr mmap'd, 97883776/51663656 max/curr
So, peak dinfo memory decreases by about 36Mb, and final by 15Mb.
(for info, valgrind 3.9.0 uses
dinfo: 158941184/109666304 max/curr mmap'd, 156775944/107590656 max/curr
So, compared to 3.9.0, dinfo peak decreases by about 40%, and the final
memory is divided by more than 2).
The memory decrease is obtained by:
* using a dedup pool to store filename/dirname pair for the loctab source/line
information.
As typically, there is not a lot of such pairs, typically a UShort is
good enough to identify a fn/dn pair in a dedup pool.
To avoid losing memory due to alignment, the fndn indexes are stored
in a "parallel" array to the DiLoc loctab array, with entries having
1, or 2 or 4 bytes according to the nr of fn/dn pairs in the dedup pool.
See priv_storage.h comments for details.
(there was a extensible WordArray local implementation in readdwarf.c.
As with this change, we use an xarray, the local implementation was
removed).
* the memory needed for --read-inline-info is slightly decreased (-2Mb)
by removing the (unused) dirname from the DiInlLoc struct.
Handling dirname for inlined function caller implies to rework
the dwarf3 parser read_filename_table common to the var and inlinfo parser.
Waiting for this to be done, the dirname component is removed from DiInlLoc.
* the stabs reader (readstabs.c) is broken since 3.9.0.
For this change, the code has been updated to make it compile with the new
DiLoc/FnDn dedup pool. As the code is completely broken, a vg_assert(0)
has been put at the begin of the stabs reader.
* the pdb reader (readpdb.c) has been trivially updated and should still work.
It has not been tested (how do we test this ?).
A follow-up patch will be done to avoid doing too many calls to
ML_(addFnDn) : instead of having one call per ML_(addLineInfo), one
should have a single call done when reading the filename table.
This has also be tested in an outer/inner setup, to verify no
memory leak/bugs.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14158
* Avoid printing the size of a null dedup pool
* Avoid warnings of 2 unused variables on some platforms
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14132
On a big executable, the trunk needs:
dinfo: 155844608/106737664 max/curr mmap'd 155572624/102276760 max/curr
With the patch, we have:
dinfo: 134873088/70389760 max/curr mmap'd 134607808/66717512 max/curr
So, peak dinfo memory decreases by 21Mb, and final by 36Mb.
The memory decrease is obtained by:
* using a dedup pool to store the machine dependent part (cfsi_m)
of the cfsi information as this information is highly duplicated.
For x86 and arm64, the duplication factor of cfsi machine dependent
part is very high (up to a factor 60).
For arm64, it is more like a factor 3.
A 'variable size' (1, 2 or 4 bytes) is automatically used to identify
the cfsi_m, if there is less than or more than 255/64K different cfsi_m.
* not storing explicitely the length of a range for which a cfsi_m
is to be used: in a large majority of the cases, ranges are
consecutive, and so the end of a range is just one byte before
the start of the next range.
So, we do not store the length of the ranges.
If there is a hole between 2 ranges, the hole is stored explicitely
as a range in which we have no cfsi_m information.
On x86 and amd64, we have quite some holes (something like one hole
every 7 cfsi). On arm64, we have very few holes (less than one hole
every 50 cfsi).
Even with the nr of holes on x86/amd64, it is more memory efficient
to store the holes rather than to store the length of each cfsi.
* Merging consecutive ranges that have the same cfsi_m info:
Many cfsi are "mergeable": there is no hole between 2 cfsi, and their
machine dependent part is identical
(I guess the unwind info needed by valgrind is subset of the full
unwind info, and so, the cfsi entries are not merged by the compiler,
but can be merged for simple unwind). Depending on the platform
(x86, amd64, arm64) and of the library/object file, we can have a
significant nr of mergeable entries.
The patch is not very small, but a lot is mechanical changes.
The patch has been compiled and tested on x86/amd64/ppc32/ppc64
(but ppc does not use cfsi so that just verifies it compiles).
It has been compiled on arm64, and "tested" by launching valgrind on
one executable.
It has not been compiled on s390 and mips.
With some luck, maybe it will compile on these platforms.
And if that uses the whole provision of luck for 2014, it might even work
on these platforms :).
If it does not compile, the fix should be straightforward.
Runtime problems might be more tricky (but arm64 "worked out of the box"
once x86/amd64 were ok).
This has also be tested in an outer/inner setup, to verify no memory leak/bugs.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14129
only decrease the size of a block, does not change the address,
does not need to alloc another block and copy the memory,
and (if big enough) makes the excess memory available for other
allocations.
VG_(arena_realloc_shrink) is then used for debuginfo storage.c
(replacing an allocation + copy).
Also use it in the dedup pool, to recuperate the unused
memory of the last pool.
This also allows to re-increase the string pool size to the original
3.9.0 value of 64Kb. All this slightly decrease the peak and in use
memory of dinfo.
VG_(arena_realloc_shrink) will also be used to implement (in another patch)
a dedup pool which "numbers" the allocated elements.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14122
showing inlined function calls.
See 278972 valgrind stacktraces and suppression do not handle inlined function call debuginfo
Reading the inlined dwarf call info is activated using the new clo
--read-inline-info=yes
Default is currently no but an objective is to optimise the performance
and memory in order to possibly set it on by default.
(see below discussion about performances).
Basically, the patch provides the following pieces:
1. Implement a new dwarf3 reader that reads the inlined call info
2. Some performance improvements done for this new parser, and
on some common code between the new parser and the var info parser.
3. Use the parsed inlined info to produce stacktrace showing inlined calls
4. Use the parsed inlined info in the suppression matching and suppression generation
5. and of course, some reg tests
1. new dwarf3 reader:
---------------------
Two options were possible: add the reading of the inlined info
in the current var info dwarf reader, or add a 2nd reader.
The 2nd approach was preferred, for the following reasons:
The var info reader is slow, memory hungry and quite complex.
Having a separate parsing phase for the inlined information
is simpler/faster when just reading the inlined info.
Possibly, a single parser would be faster when using both
--read-var-info=yes and --read-inline-info=yes.
However, var-info being extremely memory/cpu hungry, it is unlikely
to be used often, and having a separate parsing for inlined info
does in any case make not much difference.
(--read-var-info=yes is also now less interesting thanks to commit
r13991, which provides a fast and low memory "reasonable" location
for an address).
The inlined info parser reads the dwarf info to make calls
to priv_storage.h ML_(addInlInfo).
2. performance optimisations
----------------------------
* the abbrev cache has been improved in revision r14035.
* The new parser skips the non interesting DIEs
(the var-info parser has no logic to skip uninteresting DIEs).
* Some other minor perf optimisation here and there.
In total now, on a big executable, 15 seconds CPU are needed to
create the inlined info (on my slow x86 pentium).
With regards to memory, the dinfo arena:
with inlined info: 172281856/121085952 max/curr mmap'd
without : 157892608/106721280 max/curr mmap'd,
So, basically, inlined information costs about 15Mb of memory for
my big executable (compared to first version of the patch, this is
already using less memory, thanks to the strpool deduppoolalloc.
The needed memory can probably be decreased somewhat more.
3. produce better stack traces
------------------------------
VG_(describe_IP) has a new argument InlIPCursor *iipc which allows
to describe inlined function calls by doing repetitive calls
to describe_IP. See pub_tool_debuginfo.h for a description.
4. suppression generation and matching
--------------------------------------
* suppression generation now also uses an InlIPCursor *iipc
to generate a line for each inlined fn call.
* suppression matching: to allow suppression matching to
match one IP to several function calls in a suppression entry,
the 'inputCompleter' object (that allows to lazily generate
function or object names for a stacktrace when matching
an error with a suppression) has been generalised a little bit
more to also lazily generate the input sequence.
VG_(generic_match) has been updated so as to be more generic
with respect to the input completer : when providing an
input completer, VG_(generic_match) does not need anymore
to produce/compute any input itself : this is all delegated
to the input completer.
5. various regtests
-------------------
to test stack traces with inlined calls, and suppressions
of (some of) these errors using inlined fn calls matching.
Work still to do:
-----------------
* improve parsing performance
* improve the memory overhead.
* handling the directory name for files of the inlined function calls is not yet done.
(probably implies to refactor some code)
* see if m_errormgr.c *offsets arrays cannot be managed via xarray
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14036
include/pub_tool_deduppoolalloc.h
coregrind/pub_core_deduppoolalloc.h
coregrind/m_deduppoolalloc.c
and uses it (currently only) for the strings in m_debuginfo/storage.c
The idea is that such ddup pool allocator will also be used for other
highly duplicated information (e.g. the DiCFSI information), where
significant gains can also be achieved.
The dedup pool for strings also decreases significantly the memory
needed by the read inline information (patch still to be committed,
see bug 278972).
When testing with a big executable (tacot_process),
this reduces the size of the dinfo arena from
trunk: 158941184/109760512 max/curr mmap'd, 156775944/107882728 max/curr,
to
ddup: 157892608/106614784 max/curr mmap'd, 156362160/101414712 max/curr
(so 3Mb less mmap-ed once debug info is read, 1Mb less mmap-ed in peak,
6Mb less allocated once debug info is read).
This is all gained due to the string which changes from:
trunk: 17,434,704 in 266: di.storage.addStr.1
to
ddup: 10,966,608 in 750: di.storage.addStr.1
(6.5Mb less memory used by strings)
The gain in mmap-ed memory is smaller due to fragmentation.
Probably one could decrease the fragmentation by using bigger
size for the dedup pool, but then we would lose memory on the last
allocated pool (and for small libraries, we often do not use much
of a big pool block).
Solution might be to increase the pool size but have a "shrink_block"
operation. To be looked at in the future.
In terms of performance, startup of a big executable (on an old pentium)
is not influenced significantly (something like 0.1 seconds on 15 seconds
startup for a big executable, on a slow pentium).
The dedup pool uses a hash table. The hash function used currently
is the VG_(adler32) check sum. It is reported (and visible also here)
that this checksum is not a very good hash function (many collisions).
To have statistics about collisions, use --stats -v -v -v
As an example of the collisions, on the strings in debug info of memcheck tool on x86,
one obtain:
--4789-- dedupPA:di.storage.addStr.1 9983 allocs (8174 uniq) 11 pools (4820 bytes free in last pool)
--4789-- nr occurences of chains of len N, N-plicated keys, N-plicated elts
--4789-- N: 0 : nr chain 6975, nr keys 0, nr elts 0
--4789-- N: 1 : nr chain 3670, nr keys 6410, nr elts 8174
--4789-- N: 2 : nr chain 1070, nr keys 226, nr elts 0
--4789-- N: 3 : nr chain 304, nr keys 100, nr elts 0
--4789-- N: 4 : nr chain 104, nr keys 84, nr elts 0
--4789-- N: 5 : nr chain 72, nr keys 42, nr elts 0
--4789-- N: 6 : nr chain 44, nr keys 34, nr elts 0
--4789-- N: 7 : nr chain 18, nr keys 13, nr elts 0
--4789-- N: 8 : nr chain 17, nr keys 8, nr elts 0
--4789-- N: 9 : nr chain 4, nr keys 6, nr elts 0
--4789-- N:10 : nr chain 9, nr keys 4, nr elts 0
--4789-- N:11 : nr chain 1, nr keys 0, nr elts 0
--4789-- N:13 : nr chain 1, nr keys 1, nr elts 0
--4789-- total nr of unique chains: 12289, keys 6928, elts 8174
which shows that on 8174 different strings, we have only 6410 strings which have
a unique hash value. As other examples, N:13 line shows we have 13 strings
mapping to the same key. N:14 line shows we have 4 groups of 10 strings mapping to the
same key, etc.
So, adler32 is definitely a bad hash function.
Trials have been done with another hash function, giving a much lower
collision rate. So, a better (but still fast) hash function would probably
be beneficial. To be looked at ...
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@14029
Necessary changes to Valgrind to support MIPS64LE on Linux.
Minor cleanup/style changes embedded in the patch as well.
The change corresponds to r2687 in VEX.
Patch written by Dejan Jevtic and Petar Jovanovic.
More information about this issue:
https://bugs.kde.org/show_bug.cgi?id=313267
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@13292
