This is rather on the obvious side.
While doing strace on an executable using libunwind, I noticed a
lot of:
msync(0, 1, MS_SYNC) = -1 ENOMEM (Cannot allocate memory)
Since we know that the first page isn't mapped (or at least doesn't
contain the data we are looking for), we can eliminate all such
msync calls.
Tested on Linux/x86_64 with no regressions.
Original code was accessing rs_cache memory without holding a lock
in some cases. If there was sufficient cache pressure, entry being
accessed may be overwritten by another thread, resulting in a data
race.
We now make a thread local copy of the data, before releasing the
lock. If we end up supporting UNW_CACHE_PER_THREAD properly
in the future, this memcpy should be unnecessary.
Greetings,
Attached patch gets rid of additional unnecessary branch (rs_get_cache
can not return NULL unless caching_policy is UNW_CACHE_NONE), gets rid of
goto's, and makes apply_reg_state (major CPU consumer) execute with cache
lock not held (before the patch, apply_reg_state was called with lock held
for newly-inserted entries, but not for found-in-cache entries).
Tested on Linux/x86_64 with no regressions.
Thanks,
--
Paul Pluzhnikov
Greetings,
Attached patch is rather on the obvious side:
- rs1 can't be NULL since it's assigned on previous line
- rs_new never returns NULL, and if it ever did, we'd crash on memcpy that
preceeds the NULL check.
Tested on Linux/x86_64 with no regressions.
Thanks,
--
Paul Pluzhnikov
Currently, libunwind allocates several PATH_MAX entries on stack, while
trying to find a binary via /proc/.../maps.
However stack space may be at premium (especially when sigaltstack is used),
and PATH_MAX on Linux is 4096, while SIGSTKSZ is only 8192 on x86.
Attached patch eliminates multiple PATH_MAX stack allocations, and simplifies
code in maps_next, at the cost of being unable to do anything if we can't
mmap one page. It appears to me that under such low-memory conditions,
libunwind will fail shortly elsewhere anyway.
This patch also disables more of debug_frame-handling code when
CONFIG_DEBUG_FRAME is undefined.
Tested on Linux/x86_64 with and without CONFIG_DEBUG_FRAME, no regressions.
The behavior on wait vs abort unwind depends on the locking primitive
chosen by the user. This makes the API consistent and independent of
the locking primitive.
Greetings,
We use libunwind just for stack traces (I suspect many others do as well).
The use pattern is:
GetStackTrace(void** result, int max_depth)
{
...
unw_getcontext(&uc);
unw_init_local(&cursor, &uc);
while (n < max_depth) {
if (unw_get_reg(&cursor, UNW_REG_IP, (unw_word_t *) &ip) < 0) {
break;
}
result[n++] = ip;
if (unw_step(&cursor) <= 0) {
break;
}
}
Given this usage, it is quite convenient for us to block signals (or
prevent signal handlers from re-entering libunwind by other means) at the
"top level", which makes most of the sigprocmask calls performed by
libunwind itself unneccessary.
The second patch in this series adds a configure option which removes most
of the sigprocmask calls.
Attached patch is a preliminary for it -- consolidating all of the
"sigprocmask; mutex_lock;" sequences into lock_acquire and "mutex_unlock;
sigprocmask;" sequences into lock_release.
Thanks,
--
Paul Pluzhnikov
commit 402d15b123d54a7669db7cf17a76dd315094e472
Author: Paul Pluzhnikov <ppluzhnikov@google.com>
Date: Mon Sep 21 10:18:28 2009 -0700
Replace "sigprocmask + mutext_lock" with a single lock_acquire.
Likewise, replace "mutext_unlock + sigprocmask" with lock_release.
This rule (no IP adjustment on ia64) may be correct for locating the right FDE.
Unfortunately the same adjusted/unadjusted return address is being used also by
__gxx_personality_v0() to locate the right call-site (the try {} block) for
unwinding. And this case is already sensitive for off-by-one PC values.
Unlike the FDE location where the function prologue + epilogue make it immune
against off-by-one PC calculations.
Therefore suggesting to unify it with non-ia64 case.
Signed-off-by: Jan Kratochvil <jan.kratochvil@redhat.com>
Provide a special implementation for ia64, because the unwind
information is such that an IP adjustment is not necessary before
looking up unwind info.
Bad things happen if libunwind only provides parts of the ABI and
the rest come from libgcc.
Signed-off-by: Jan Kratochvil <jan.kratochvil@redhat.com>
getcontext in libc.
Also cleanup the namespace (check-name-space passes on x86_64 now).
Replace uses of offsets.h with ucontext_i.h.
Rename _x86_64_setcontext to _Ux86_64_setcontext.
TBD: Add CFI annotations for get/setcontext.
Signed-off-by: Paul Pluzhnikov <ppluzhnikov@google.com>
Signed-off-by: Arun Sharma <arun.sharma@google.com>
bad/missing unwind information, which could result in libunwind
dereferencing bad pointers. This mechanism is based on msync(2) system
call and significantly reduces the chances of a bad pointer
dereference in libunwind.
The original idea was to turn this mechanism on only when necessary
i.e. libunwind didn't find proper unwind information for a IP.
There are a couple of problems in the current implementation.
* The flag is global and is modified without locking
* The flag isn't reset when starting a new unwind
The attached patch makes ->validate a per-thread setting by moving it
into struct cursor from unw_local_addr_space and resets it to false
when starting a new unwind. As a result, cursor->as_arg points to the
cursor itself instead of the ucontext (for the local case).
This was found to reduce the number of msync() system calls from an
application using libunwind significantly.
Signed-off-by: Paul Pluzhnikov <ppluzhnikov@google.com>
Signed-off-by: Arun Sharma <arun.sharma@google.com>
* src/arm/unwind_i (arm_lock, arm_local_resume): Define.
* src/ptrace/_UPT_find_proc_info.c: Handle ARM like X86 etc.
* tests/flush-cache.S (flush_cache): Add (dummy) ARM-version.
ARM does need executable stack, even on Linux...
Signed-off-by: Anderson Lizardo <anderson.lizardo@indt.org.br>
Signed-off-by: Bruna Moreira <bruna.moreira@indt.org.br>
The current pattern is too restrictive and doesn't work well on
modern glibcs.
Signed-off-by: Paul Pluzhnikov <ppluzhnikov@google.com>
Signed-off-by: Arun Sharma <arun.sharma@google.com>
* Use explicit types for XMM registers
* Support full width (128 bits) access
Signed-off-by: Andrew Cagney <cagney@redhat.com>
Signed-off-by: Arun Sharma <aruns@google.com>
as "weak". Since the elf-support is in the library anyhow, this serves
no purpose and in fact causes problem because the weak reference alone
is not enough to pull in the ELF-code from an archive file, causing to
spurious failures of get_proc_name.
On some systems executable stacks are denied. Since libunwind and the
tests don't actually need executable stacks this patch marks all
assembly files as not needing it.
The original patch comes from frysk:
2007-04-05 Jan Kratochvil <jan.kratochvil@redhat.com>
* src/hppa/getcontext.S, src/hppa/setcontext.S, src/hppa/siglongjmp.S,
src/ia64/Ginstall_cursor.S, src/ia64/Linstall_cursor.S,
src/ia64/dyn_info_list.S, src/ia64/getcontext.S, src/ia64/longjmp.S,
src/ia64/setjmp.S, src/ia64/siglongjmp.S, src/ia64/sigsetjmp.S,
src/ppc64/longjmp.S, src/ppc64/siglongjmp.S, src/x86/longjmp.S,
src/x86/siglongjmp.S, src/x86_64/longjmp.S, src/x86_64/setcontext.S,
src/x86_64/siglongjmp.S: Stack should be non-executable, for SELinux.
I added a couple more markers for new files in current libunwind.
Before this patch you would get the following on selinux enabled
systems without allow_exec_stack: error while loading shared
libraries:
libunwind.so.7: cannot enable executable stack as shared object
requires: Permission denied
After the patch that error disappears and all test results are similar
to the results on systems without executable stack protection.
routine and add address-space argument. This is needed because on
PPC64, a the function-name symbol refers to a function descriptor
(unlike, for example, on ia64, where the @fptr() operator is needed to
refer to a function descriptor). Thus, in order to look up the name
of a function, we need to dereference the function descriptor. To
make matters more "interesting", the function descriptors are normally
resolved by the dynamic linker, so we can't get their values from the
ELF file. Instead, we have to read them from the running image, hence
the need for the address-space argument.
This is so that the source file gets distributed and _ucontext_i.h is
generated properly on the target machine.
Signed-off-by: Arun Sharma <arun.sharma@google.com>
When libunwind is linked with a C++ program that throws exceptions,
the exception that's thrown is passed in %rax. However, libc's
setcontext clears %rax, causing problems.
This patch implements a setcontext that doesn't clobber rax.
TBD: Add dwarf CFI annotations
Signed-off-by: Arun Sharma <arun.sharma@google.com>
Yes, it casts to signed, but then converts right back to
unsigned via the return type. Fixed thus.
Signed-off-by: Richard Henderson <rth@redhat.com>
Signed-off-by: Arun Sharma <arun.sharma@google.com>
CFA should be incremented by 16 in the hope that the previous frame
may have valid unwind info.
Also increase the default frame pointer recognition heuristic from 4k
to 16k.
Signed-off-by: Andrey Veskov <Andrey.Veskov@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@google.com>
Now that dwarf_find_save_locs() not just finds the save-locations but
also updates the cursor-state, document this fact (the function really
is misnamed now).
Be sure to clear pi->unwind_info when looking up the kernel table and
not needing the unwind-info. Otherwise, _UPT_put_unwind_info() may
erroneously call free() on the pointer, even though that space wasn't
malloc'd, leading to crashes..
(maps_next): After reading less than a full buffer of data, copy
remaining data to the top of the buffer so as to maintain the
invariant that all data between mi->buf and mi->buf_end is valid.
This avoids maps_next() parsing uninitialized data.
(get_initial_stack_pointers): Likewise.
(unw_init_local): Use new functions to make code more readable and add
Linux kernel support.
(Logical change 1.296)
Include "libunwind_i.h" instead of "internal.h" and "tdep.h".
Define rbs_cover_and_flush() only for the !UNW_REMOTE_ONLY case.
(ia64_is_big_endian): New macro.
(ia64_get_abi): Likewise.
(ia64_set_abi): Likewise.
(ia64_get_abi_marker): Likewise.
(Logical change 1.294)
Use ia64_get_abi_marker() instead of directly referencing c->last_abi_marker.
(linux_scratch_loc): When accessing an fp-reg, default to IA64_FPREG_LOC.
Handle more registers for the Linux interrupt-frame case.
(tdep_access_reg): Use ia64_get_abi() instead of c->as->abi.
Use ia64_is_big_endian() instead of c->as->big_endian.
(Logical change 1.294)
2005/05/17 14:14:23-07:00 hp.com!davidm
(linux_sigtramp): Add "prev_cfm_loc" argument. Use it to set c->ec_loc.
(linux_interrupt): Likewise.
(hpux_sigtramp): Likewise.
(update_frame_state): Copy old cfm_loc to prev_cfm_loc and pass it to
ABI-specific handlers. Initialize c->ec_loc to c->cfm_loc by default.
(Logical change 1.293)
Abstract ia64-specific code into bsp_match() routine.
Support any platform with at least 2 EH argument registers.
2004/12/06 01:33:16-08:00 mostang.com!davidm
Rename: src/longjmp.c -> src/setjmp/longjmp.c
(Logical change 1.290)
2004/11/17 02:43:39-08:00 mostang.com!davidm
(struct callback_data): New structure.
(linear_search): New function.
(callback): Convert to getting auxiliary info passed via a pointer to
a callback_data structure, rather than an unw_dyn_info_t
structure.
Keep track of the maximum load address in max_load_addr.
If an object doesn't have a binary search-table, fall back
on a linear search.
(dwarf_find_proc_info): Fill in callback-data structure before calling
dl_iterate_phdr(). Upon returning, check whether
cb_data.single_fde is set and, if so, return directly, without
searching the DWARF unwind-table.
(dwarf_search_unwind_table): Adjust for renaming of
dwarf_parse_fde() to dwarf_extract_proc_info_from_fde().
If IP doesn't fall into the IP-range of the FDE-entry found
by the search, return UNW_ENOINFO.
(Logical change 1.290)
2004/12/15 15:40:09-08:00 hp.com!davidm
Include "tdep.h" on ia64 (we need the cursor).
(bsp_match): Fix it so it compiles again.
(resume_restore_sigmask): Likewise.
2004/12/06 10:40:49-08:00 mostang.com!davidm
(Logical change 1.290)
2004/11/23 16:59:56-08:00 mostang.com!davidm
(_UPT_access_reg): If the register is out of range, set errno
to EINVAL.
When encountering an error while accessing a register,
print the corresponding error string.
(Logical change 1.290)
a signal-trampoline, assume that it's a PLT stub.
If non-DWARF stepping fails to change IP and CFA, declare
it a bad frame.
2004/11/23 16:59:56-08:00 mostang.com!davidm
(unw_step): Also print IP as part of the function-trace.
2004/11/23 16:17:37-08:00 mostang.com!davidm
(unw_step): When dwarf_step() fails on a signal-frame, fill in all
the known locations because dwarf_step() fails on older
kernels which don't export the kernel vDSO even though every-
thing else may be providing proper DWARF unwind-info.
2004/10/25 17:43:57+02:00 homeip.net!davidm
Add Debug statement for return-value.
(Logical change 1.290)
2004/11/23 12:49:54-08:00 mostang.com!davidm
(unw_step): If dwarf_step() fails on a signal-frame, fill in the
save-locations for everything that gets saved in the
sigcontext structure.
2004/10/25 17:43:57+02:00 homeip.net!davidm
Add Debug statement for return-value.
(Logical change 1.290)
(parse_cie): Adjust for moving DWARF2 extra-info from
unw_dyn_dwarf_fde_info_t to dwarf_cie_info.
Add support for `S' augmentation ("special frames").
(dwarf_extract_proc_info_from_fde): Rename frm dwarf_parse_fde().
Convert to returning data via dwarf_cie_info rather than
unw_dyn-dwarf_fde_info_t. Allow 0-length FDE since those
can be used to indicate the end of the FDE-table. Return
the end of the FDE via *ADDRP. Ignore CIEs (we may pick them
up during linear searches through the FDE-table).
Fill in the unwind-info only if really needed. When needed,
return the info via a dynamically allocated dwarf_cie_info
structure. Add support for `S' augmentation (ABI/tag pairs).
(Logical change 1.290)
unwind-info, since DWARF expects us to do so (unlikey for ia64,
where we are guaranteed that the (return) IP points to the
correct unwind-info).
2004/11/17 02:43:39-08:00 mostang.com!davidm
(run_cfi_program): Switch over to using dwarf_cie_info instead of
unw_dyn_dwarf_fde_info_t.
(fetch_proc_info): Clear c->pi before filling it in.
(put_unwind_info): If we got non-dynamic unwind-info, free it
via a call to mempool_free().
(parse_fde): Switch over to using dwarf_cie_info instead of
unw_dyn_dwarf_fde_info_t.
(apply_reg_state): Drop accessees to c->cfa_is_sp.
2004/10/25 17:38:51+02:00 homeip.net!davidm
Fix typo: DW_CFA_CFA_expression -> DW_CFA_expression.
2004/10/21 11:15:44+02:00 homeip.net!davidm
Auto merged
2004/10/21 11:15:12+02:00 homeip.net!davidm
(run_cfi_program): Add missing newline in Debug statement.
(apply_reg_state): It was wrong to do a dwarf_get() on the CFA-location
returned by eval_location_expr(). Instead, we must make sure
we're dealing with a memory-location and, if so, just extract
the memory address as the CFA.
Also, update c->cfa only _after_ the other frame-state has
been updated. This seems to be necessary for correct parsing
of the sigtramp unwind info. Not sure yet this is 100% correct,
but it gets us through Gtest-resume-sig.
2004/10/19 23:24:56-07:00 mostang.com!davidm
(run_cfi_program): Fix typos in Debug statements.
(Logical change 1.290)
2005/04/07 12:19:38-07:00 hp.com!davidm
(ia64_make_proc_info): After looking up proc-info, cache it if
caching is enabled.
2005/02/20 21:42:55-08:00 mostang.com!davidm
Adjust for NELEMS to ARRAY_SIZE renaming.
(Logical change 1.290)
2004/12/02 18:37:47-08:00 hp.com!davidm
(get_list_addr): Even the "it's missing" version of this routine
must clear *countp to avoid spurious failures.
2004/11/03 11:45:24-08:00 hp.com!davidm
Auto merged
2004/10/21 16:06:28+02:00 homeip.net!davidm
(get_list_addr): Don't fail with -UNW_ENOINFO just because something
in /proc/maps isn't mappable.
(Logical change 1.290)
