malloc_printerr() on error detection "free(): invalid next size (fast)" ends up calling into: backtrace.c:init() dl-libc.c:do_dlopen() malloc.c:calloc() malloc.c:malloc_printerr() The malloc error reporting should only report the first error, not attempt to recusively report all error (we knew it was corrupted at the outer most point, so any further work inside malloc is also likely to find corruption). Full stack trace to follow. The main problem is the process does not abort() and die, it hangs around in: pthread_once.S:pthread_one() backtrace.c:__backtrace() I think due to recursive lock, this lock should trylock() on the 2nd time and abort() the process immediately. It does appear to deadlock itself.
(gdb) bt #0 pthread_once () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_once.S:95 #1 0x00007f8dfb540994 in __backtrace (array=<value optimized out>, size=64) at ../sysdeps/ia64/backtrace.c:85 #2 0x00007f8dfb4b280b in __libc_message (do_abort=2, fmt=0x7f8dfb599fc0 "*** glibc detected *** %s: %s: 0x%s ***\n") at ../sysdeps/unix/sysv/linux/libc_fatal.c:178 #3 0x00007f8dfb4b8126 in malloc_printerr (action=3, str=0x7f8dfb5980f9 "malloc(): memory corruption", ptr=<value optimized out>) at malloc.c:6311 #4 0x00007f8dfb4bbba4 in _int_malloc (av=0x7f8dfb7d0e80, bytes=<value optimized out>) at malloc.c:4411 #5 0x00007f8dfb4bc5e6 in __libc_calloc (n=<value optimized out>, elem_size=<value optimized out>) at malloc.c:4075 #6 0x00007f8dfda14d1f in _dl_new_object (realname=0x247de20 "/lib64/libgcc_s.so.1", libname=0x7f8dfb596e3e "libgcc_s.so.1", type=2, loader=0x0, mode=-1879048191, nsid=0) at dl-object.c:77 #7 0x00007f8dfda111ae in _dl_map_object_from_fd (name=0x7f8dfb596e3e "libgcc_s.so.1", fd=6, fbp=0x7fffd2c1ace0, realname=0x247de20 "/lib64/libgcc_s.so.1", loader=0x0, l_type=2, mode=-1879048191, stack_endp=0x7fffd2c1b028, nsid=0) at dl-load.c:975 #8 0x00007f8dfda1236a in _dl_map_object (loader=0x0, name=0x7f8dfb596e3e "libgcc_s.so.1", type=2, trace_mode=0, mode=<value optimized out>, nsid=<value optimized out>) at dl-load.c:2274 #9 0x00007f8dfda1ca34 in dl_open_worker (a=0x7fffd2c1b250) at dl-open.c:227 #10 0x00007f8dfda181a6 in _dl_catch_error (objname=0x7fffd2c1b2a0, errstring=0x7fffd2c1b298, mallocedp=0x7fffd2c1b2af, operate=0x7f8dfda1c910 <dl_open_worker>, args=0x7fffd2c1b250) at dl-error.c:178 #11 0x00007f8dfda1c4ea in _dl_open (file=0x7f8dfb596e3e "libgcc_s.so.1", mode=-2147483647, caller_dlopen=0x0, nsid=-2, argc=8, argv=<value optimized out>, env=0x7fffd2c30020) at dl-open.c:569 #12 0x00007f8dfb568340 in do_dlopen (ptr=<value optimized out>) at dl-libc.c:86 #13 0x00007f8dfda181a6 in _dl_catch_error (objname=0x7fffd2c1b460, errstring=0x7fffd2c1b458, mallocedp=0x7fffd2c1b46f, operate=0x7f8dfb568300 <do_dlopen>, args=0x7fffd2c1b440) at dl-error.c:178 #14 0x00007f8dfb568497 in dlerror_run (name=<value optimized out>, mode=<value optimized out>) at dl-libc.c:47 #15 __libc_dlopen_mode (name=<value optimized out>, mode=<value optimized out>) at dl-libc.c:160 #16 0x00007f8dfb540895 in init () at ../sysdeps/ia64/backtrace.c:41 #17 0x00007f8dfb7e1b23 in pthread_once () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_once.S:104 #18 0x00007f8dfb540994 in __backtrace (array=<value optimized out>, size=64) at ../sysdeps/ia64/backtrace.c:85 #19 0x00007f8dfb4b280b in __libc_message (do_abort=2, fmt=0x7f8dfb599fc0 "*** glibc detected *** %s: %s: 0x%s ***\n") at ../sysdeps/unix/sysv/linux/libc_fatal.c:178 #20 0x00007f8dfb4b8126 in malloc_printerr (action=3, str=0x7f8dfb59a2b8 "free(): invalid next size (fast)", ptr=<value optimized out>) at malloc.c:6311 #21 0x00007f8dfb4bac53 in _int_free (av=0x7f8dfb7d0e80, p=0x24d52c0, have_lock=0) at malloc.c:4811
See also bug#956
*** Bug 956 has been marked as a duplicate of this bug. ***
This is going to be difficult to fix and invasive. I will do my best to explain why. At the point of the failure we want to be able to print a backtrace. The only way to get a reliable backtrace is to use the unwinder provided by gcc via libgcc_s.so.1 (this may vary by machine). In order to get access to the unwinder we must dlopen that shared library. During the dlopen process we need to calloc enough structures to hookup the new shared library into the structures used by the dynamic linker. One resolution to this problem is to ensure that malloc has a fall-back allocation scheme that is robust against failure and then during the malloc_printerr we flip an internal bit and switch to the temporary reserve allocations. We could also create a new internal API for using the temporary allocations and then dlopen could use that in the event that we are crashing and need to dlopen one last library (the unwinder on demand). That would prevent other threads from consuming the reserve allocations after malloc_printerr is entered by another thread. This is a considerable amount of work and we aren't going to get to this issue until a core developer or someone with serious interest commits to fixing this. Therefore I'm moving this to SUSPENDED until we find the resources to fix the issue. This issue should remain open and new issues submited about this bug should be marked as duplicates of this issue.
On Wed, Nov 13, 2013 at 03:57:02AM +0000, carlos at redhat dot com wrote: > One resolution to this problem is to ensure that malloc has a fall-back > allocation scheme that is robust against failure and then during the > malloc_printerr we flip an internal bit and switch to the temporary reserve > allocations. We could also create a new internal API for using the temporary > allocations and then dlopen could use that in the event that we are crashing > and need to dlopen one last library (the unwinder on demand). That would > prevent other threads from consuming the reserve allocations after > malloc_printerr is entered by another thread. > > This is a considerable amount of work and we aren't going to get to this issue > until a core developer or someone with serious interest commits to fixing this. > Therefore I'm moving this to SUSPENDED until we find the resources to fix the > issue. > Why not reuse a singal-safe malloc for dlopen?
This fancy backtrace stuff is nice and all but... the process must die! Can't the pthread_once use a non-blocking lock ? Can the lock be a recursive type ? Can pthread_trylock() used in this non-critial path ? if already locked, and if possible to check if locked by our thread-id ? then we immediately abort the process (causing execution of the process to die, like it should). No backtrace is emitted, great! How do I stop this fancy backtrace stuff from working ? I want to setup an environment variable to turn it off as a workaround ? How do I make this fancy backtrace stuff work, by preloading the dlopen() stuff it might need, during initialization of malloc() ? I want to setup an environment variable for that too. There is no need to actually fix the bug, you are over thinking the issue. But this fancy stuff needs to be turned off or preloaded, before the process gets into an undefined state (due to memory bug).
Another idea, do not backtrace() every malloc() error, only the first one (the outer most one). But right now the process deadlocks itself, on what looks to be a non-recursive mutex trying to do fancy backtrace on every malloc() problem found. The process must die.
On Wed, Nov 13, 2013 at 01:00:06PM +0000, darryl.miles at darrylmiles dot org wrote: > How do I stop this fancy backtrace stuff from working ? I want to setup an > environment variable to turn it off as a workaround ? > > How do I make this fancy backtrace stuff work, by preloading the dlopen() stuff > it might need, during initialization of malloc() ? I want to setup an > environment variable for that too. > As a quick workaround you can add following code to your application/preload this. #include <execinfo.h> static void __attribute__ ((constructor)) init_backtrace() { void *bt[10]; backtrace (bt, 10); }
Carlos, this is yet another reason why dlopen'ing libgcc_s is simply the wrong thing to do, and libgcc_eh should be static-linked into libc. (The other big reason is the possibility of pthread_cancel aborting the program.) At one time in the distant past, it was necessary for there to only be one copy of this code (and its data) in the whole program; otherwise, exception propagation (or backtracing) across DSOs would not work reliably. But modern unwinding code uses dl_iterate_phdr and works fine even if multiple copies of the code are present in the program. Fixing this error in the way I describe will greatly simplify glibc and improve its reliability.
(In reply to Rich Felker from comment #9) > Carlos, this is yet another reason why dlopen'ing libgcc_s is simply the > wrong thing to do, and libgcc_eh should be static-linked into libc. (The > other big reason is the possibility of pthread_cancel aborting the program.) > At one time in the distant past, it was necessary for there to only be one > copy of this code (and its data) in the whole program; otherwise, exception > propagation (or backtracing) across DSOs would not work reliably. But modern > unwinding code uses dl_iterate_phdr and works fine even if multiple copies > of the code are present in the program. > > Fixing this error in the way I describe will greatly simplify glibc and > improve its reliability. That sounds like a good idea to me, I also agree that dlopening libgcc_s.so.1 always seemed like a terrible idea to me. We just need the resources to do the rewrite and fixup the linking to use libgcc_eh. I will leave this SUSPENDED until we find someone to clean this up.
On Wed, 13 Nov 2013, bugdal at aerifal dot cx wrote: > Carlos, this is yet another reason why dlopen'ing libgcc_s is simply the wrong > thing to do, and libgcc_eh should be static-linked into libc. (The other big Static-linking libgcc_eh into any glibc library is a bad idea because it complicates bootstrapping: it means glibc built with an initial bootstrap compiler (which was built without glibc headers available, implying full EH functionality is not present in libgcc) is not identical to glibc built with a compiler built using full shared glibc and headers. (It's *also* a bad idea because new compilers can start using new DWARF unwind opcodes that an old copy of the unwind code won't understand, causing problems using new programs with old glibc.) The answer for libpthread is for it to dlopen libgcc_s when loaded rather than at pthread_cancel time (or to be made to depend (DT_NEEDED) on libgcc_s in a way that doesn't require libgcc_s to be available when libpthread is built). The answer for other cases is to disable the backtracing by default as discussed in bug 12189 (possibly with an environment variable, not available in setuid programs, that can reenable it - in which case glibc would dlopen libgcc_s at startup).
On Wed, Nov 13, 2013 at 04:12:53PM +0000, joseph at codesourcery dot com wrote: > http://sourceware.org/bugzilla/show_bug.cgi?id=16159 > > --- Comment #11 from joseph at codesourcery dot com <joseph at codesourcery dot com> --- > On Wed, 13 Nov 2013, bugdal at aerifal dot cx wrote: > > > Carlos, this is yet another reason why dlopen'ing libgcc_s is simply the wrong > > thing to do, and libgcc_eh should be static-linked into libc. (The other big > > Static-linking libgcc_eh into any glibc library is a bad idea because it > complicates bootstrapping: it means glibc built with an initial bootstrap > compiler (which was built without glibc headers available, implying full > EH functionality is not present in libgcc) is not identical to glibc built > with a compiler built using full shared glibc and headers. (It's *also* a > bad idea because new compilers can start using new DWARF unwind opcodes > that an old copy of the unwind code won't understand, causing problems > using new programs with old glibc.) > Why did you jump from dlopening to static linking? Dynamic linking would work and if there is concern that user does not have one we could provide a stub implementation and function to test if we deal with stub or real one.
Joseph, the bootstrapping issue can presumably be fixed (and bootstrapping made easier) simply by providing a way to install headers without building glibc. This may even allow you to shave one or more steps off of the full bootstrap process. As for the issue of new DWARF opcodes, if they prevent older unwind code from being able to interpret the unwind information at all (rather than just failing to take advantage of the new features) that seems like a fundamental design bug elsewhere that should be reported. I'm not clear whether or not that's really the case. With that said, I find your alternate fix proposal acceptable. For the libpthread issue, I believe the DT_NEEDED could be generated at build time using a fake libgcc_s.so.1 in the glibc source tree. As for disabling backtrace by default, that's perfectly acceptable. Alternatively, glibc could always attempt to load libgcc_s.so.1 at startup and disable backtrace if it's not found.
Ondrej, is having glibc contain a DT_NEEDED entry for libgcc.so.1 really an option that's on the table? I think this would also interfere with bootstrapping issues Joseph and others may be concerned about, as well as hurting load-time performance.
On Wed, 13 Nov 2013, neleai at seznam dot cz wrote: > Why did you jump from dlopening to static linking? Dynamic linking would > work and if there is concern that user does not have one we could > provide a stub implementation and function to test if we deal with stub > or real one. I don't think default dlopening libgcc_s from libc at startup is desirable on performance grounds (most programs will never need it), whereas from libpthread it's likely to be less significant.
On Wed, 13 Nov 2013, bugdal at aerifal dot cx wrote: > Joseph, the bootstrapping issue can presumably be fixed (and bootstrapping made > easier) simply by providing a way to install headers without building glibc. There already is. But to install the correct set of headers (some generated at build time) you first need an appropriately configured compiler to configure glibc. That's the old three-compiler bootstrap process: first build a basic compiler, then install headers with it and crt*.o and build a dummy libc.so, then build a second compiler with shared libgcc, then build glibc, then build a third compiler. I changed things in glibc and GCC so that a two-compiler process suffices: the initial compiler built without headers can build glibc and the result is identical to what you get if you repeatedly alternate GCC and glibc builds. (Ideally you'd have a one-compiler process, where the second compiler build only builds/rebuilds GCC's runtime libraries where they depend on system headers or shared glibc, not GCC itself.)
Joseph, do you have a benchmark to measure libgcc overhead? I tried a following cat "int main() { return 42; }" > x.c gcc x.c -O3 -o nogcc gcc x.c -O3 -lgcc -o withgcc time for I in `seq 1 10000`; do ./nogcc; done time for I in `seq 1 10000`; do ./withgcc; done And I cannot distinguish these from noise. When I linked with -lpthread there was a noticable slowdown.
Ondrej, did you even check your results with readelf or ldd? -lgcc is a static library and is always linked, so of course it won't make any difference. You need to test with -lgcc_s (and double-check to make sure the dependency really got added). BTW, I'm not sure how well your test will do measuring exec time versus other overhead. If you'd like, I have a test I can post that execs itself and measures the actual time from just before the execve syscall to the start of main.
On Thu, Nov 14, 2013 at 03:54:30PM +0000, bugdal at aerifal dot cx wrote: > https://sourceware.org/bugzilla/show_bug.cgi?id=16159 > > --- Comment #18 from Rich Felker <bugdal at aerifal dot cx> --- > Ondrej, did you even check your results with readelf or ldd? -lgcc is a static > library and is always linked, so of course it won't make any difference. You > need to test with -lgcc_s (and double-check to make sure the dependency really > got added). > I asked for benchmark because of that, with a lgcc_s there is difference. plain real 0m3.039s user 0m0.195s sys 0m3.049s with lgcc_s real 0m3.141s user 0m0.169s sys 0m3.179s with lpthread real 0m3.282s user 0m0.182s sys 0m3.308s > BTW, I'm not sure how well your test will do measuring exec time versus other > overhead. If you'd like, I have a test I can post that execs itself and > measures the actual time from just before the execve syscall to the start of > main. > These also count as I wanted to show a relative performance impact. If this is taken into extreme we could improve performance by staticaly linking lm and lpthread Or using prelink.
On Thu, Nov 14, 2013 at 04:47:48PM +0000, neleai at seznam dot cz wrote: > These also count as I wanted to show a relative performance impact. If I agree this approach makes sense, but the relative performance impact could change when the program (possibly linked with libgcc_s) is invoked via posix_spawn or vfork+exec from a high-load server versus as part of an inefficient shell script where the shell may have a lot of additional syscall overhead on each command (this might also vary between shells; dash or busybox ash might perform very differently from bash). So while we may not care about the most extreme impact, I think it's important to consider how large the relative overhead is when the invocation conditions are a low-overhead, real-world scenario. > this is taken into extreme we could improve performance by staticaly linking lm > and lpthread Yes, of course -- actually, I would recommend merging all of the glibc .so's into libc.so, but I understand that the current situation with symbol versions greatly complicates this, and that there might be other issues. It would certainly improve load-time performance and memory overhead for small programs, though. But I think this is outside the scope of this bug report. The interest in looking at performance here is asking whether a proposed change would make performance noticably worse (a regression), not how we can best optimize startup performance.
(In reply to Darryl Miles from comment #6) > How do I stop this fancy backtrace stuff from working ? I want to setup an > environment variable to turn it off as a workaround ? According to: https://lists.gnu.org/archive/html/bug-gnulib/2013-11/msg00103.html setting MALLOC_CHECK_=2 in the environment is sufficient to prevent the error message attempts; but that sounds like something you set at program start rather than something we can do via setenv() at the time of reporting the first error (because setenv uses malloc). > There is no need to actually fix the bug, you are over thinking the issue. Yes, there IS a need to fix something. The link above points to a case of a user that is unhappy that their ./configure failed because the conftest program hung after tickling a malloc corruption bug in regex. Configure should never hang (thankfully, configure tests are one case where the MALLOC_CHECK_=2 trick may be sufficient - someone probing for known glibc bugs doesn't care about a bactrace, only about successful exit status).
Solution posted: https://sourceware.org/ml/libc-alpha/2015-02/msg00653.html
This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "GNU C Library master sources". The branch, master has been updated via fff94fa2245612191123a8015eac94eb04f001e2 (commit) via 99db95db37b4fd95986fadb263e4180b7381d10d (commit) via 920d70128baa41ce6ce3b1b4771fe912f8d1691a (commit) from 46f894d8c60afcc06056a376340df2f378694551 (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=fff94fa2245612191123a8015eac94eb04f001e2 commit fff94fa2245612191123a8015eac94eb04f001e2 Author: Siddhesh Poyarekar <siddhesh@redhat.com> Date: Tue May 19 06:40:37 2015 +0530 Avoid deadlock in malloc on backtrace (BZ #16159) When the malloc subsystem detects some kind of memory corruption, depending on the configuration it prints the error, a backtrace, a memory map and then aborts the process. In this process, the backtrace() call may result in a call to malloc, resulting in various kinds of problematic behavior. In one case, the malloc it calls may detect a corruption and call backtrace again, and a stack overflow may result due to the infinite recursion. In another case, the malloc it calls may deadlock on an arena lock with the malloc (or free, realloc, etc.) that detected the corruption. In yet another case, if the program is linked with pthreads, backtrace may do a pthread_once initialization, which deadlocks on itself. In all these cases, the program exit is not as intended. This is avoidable by marking the arena that malloc detected a corruption on, as unusable. The following patch does that. Features of this patch are as follows: - A flag is added to the mstate struct of the arena to indicate if the arena is corrupt. - The flag is checked whenever malloc functions try to get a lock on an arena. If the arena is unusable, a NULL is returned, causing the malloc to use mmap or try the next arena. - malloc_printerr sets the corrupt flag on the arena when it detects a corruption - free does not concern itself with the flag at all. It is not important since the backtrace workflow does not need free. A free in a parallel thread may cause another corruption, but that's not new - The flag check and set are not atomic and may race. This is fine since we don't care about contention during the flag check. We want to make sure that the malloc call in the backtrace does not trip on itself and all that action happens in the same thread and not across threads. I verified that the test case does not show any regressions due to this patch. I also ran the malloc benchmarks and found an insignificant difference in timings (< 2%). * malloc/Makefile (tests): New test case tst-malloc-backtrace. * malloc/arena.c (arena_lock): Check if arena is corrupt. (reused_arena): Find a non-corrupt arena. (heap_trim): Pass arena to unlink. * malloc/hooks.c (malloc_check_get_size): Pass arena to malloc_printerr. (top_check): Likewise. (free_check): Likewise. (realloc_check): Likewise. * malloc/malloc.c (malloc_printerr): Add arena argument. (unlink): Likewise. (munmap_chunk): Adjust. (ARENA_CORRUPTION_BIT): New macro. (arena_is_corrupt): Likewise. (set_arena_corrupt): Likewise. (sysmalloc): Use mmap if there are no usable arenas. (_int_malloc): Likewise. (__libc_malloc): Don't fail if arena_get returns NULL. (_mid_memalign): Likewise. (__libc_calloc): Likewise. (__libc_realloc): Adjust for additional argument to malloc_printerr. (_int_free): Likewise. (malloc_consolidate): Likewise. (_int_realloc): Likewise. (_int_memalign): Don't touch corrupt arenas. * malloc/tst-malloc-backtrace.c: New test case. https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=99db95db37b4fd95986fadb263e4180b7381d10d commit 99db95db37b4fd95986fadb263e4180b7381d10d Author: Siddhesh Poyarekar <siddhesh@redhat.com> Date: Tue May 19 06:36:29 2015 +0530 Succeed if make check does not report any errors The conditional that evaluates if there are any FAILed test cases currently always fails, since we ensure it fails if we find any unexpected results in tests.sum and it would obviously fail if it does not find failed results in tests.sum. This patch fixes this by simply inverting the result of the egrep, i.e. succeed if egrep fails (to find failed results) and fail if it succeeds. Tested with 'make subdirs=localedata check' and 'make subdirs=locale check' where all tests succeed and with 'make subdirs=elf check' where a couple of tests fail for me. * Makefile (summarize-tests): Fix return value on success. https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=920d70128baa41ce6ce3b1b4771fe912f8d1691a commit 920d70128baa41ce6ce3b1b4771fe912f8d1691a Author: Siddhesh Poyarekar <siddhesh@redhat.com> Date: Tue May 19 06:35:37 2015 +0530 Add envz_remove to the libc manual I was told that Ma Shimao submitted a patch to add envz_remove to the libc manual, but the patch could not be accepted since he does not have a copyright assignment in place. I have been woefully behind on libc-alpha recently and have not seen the patch or the discussion thread. I have also not read the man page for envz_remove, so Alexandre Oliva asked me if I could write this independently and post a patch. The patch below is the result of the same - I have written it based on the implementation in string/envz.c and Alex told me via email that the function is AS, AC and MT-safe like envz_strip. I assume Alex and Carlos cannot review this since they have been tainted by the original patch (I haven't even tried to look for a link to it since I don't want to be tainted) so someone else will have to review this. If there are no reviewers till the end of the week, I will commit this since I believe there is a chance that there are no other reviewers who haven't read that thread. * manual/string.texi (Envz Functions): Add envz_remove. ----------------------------------------------------------------------- Summary of changes: ChangeLog | 35 ++++ Makefile | 2 +- NEWS | 20 +- malloc/Makefile | 6 +- malloc/arena.c | 22 ++- malloc/hooks.c | 12 +- malloc/malloc.c | 173 ++++++++++++-------- .../tst-detach1.c => malloc/tst-malloc-backtrace.c | 49 +++--- manual/string.texi | 8 + 9 files changed, 213 insertions(+), 114 deletions(-) copy nptl/tst-detach1.c => malloc/tst-malloc-backtrace.c (57%)
Fixed in master.
This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "GNU C Library master sources". The annotated tag, glibc-2.22 has been created at be75ddf5e4dfab2aa4ceb2428cc146e7ea26a346 (tag) tagging 78bd7499af46d739ce94410eaeea006e874ca9e5 (commit) replaces glibc-2.21 tagged by Carlos O'Donell on Wed Aug 5 02:45:12 2015 -0400 - Log ----------------------------------------------------------------- The GNU C Library ================= The GNU C Library version 2.22 is now available. The GNU C Library is used as *the* C library in the GNU system and in GNU/Linux systems, as well as many other systems that use Linux as the kernel. The GNU C Library is primarily designed to be a portable and high performance C library. It follows all relevant standards including ISO C11 and POSIX.1-2008. It is also internationalized and has one of the most complete internationalization interfaces known. The GNU C Library webpage is at http://www.gnu.org/software/libc/ Packages for the 2.22 release may be downloaded from: http://ftpmirror.gnu.org/libc/ http://ftp.gnu.org/gnu/libc/ The mirror list is at http://www.gnu.org/order/ftp.html NEWS for version 2.22 ===================== * The following bugs are resolved with this release: 438, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152, 14094, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969, 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538, 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293, 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620, 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833, 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965, 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007, 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042, 18043, 18046, 18047, 18049, 18068, 18080, 18093, 18100, 18104, 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197, 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245, 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400, 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469, 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508, 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532, 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547, 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593, 18594, 18602, 18612, 18613, 18619, 18633, 18641, 18643, 18648, 18657, 18676, 18694, 18696. * Cache information can be queried via sysconf() function on s390 e.g. with _SC_LEVEL1_ICACHE_SIZE as argument. * A buffer overflow in gethostbyname_r and related functions performing DNS requests has been fixed. If the NSS functions were called with a misaligned buffer, the buffer length change due to pointer alignment was not taken into account. This could result in application crashes or, potentially arbitrary code execution, using crafted, but syntactically valid DNS responses. (CVE-2015-1781) * The time zone file parser has been made more robust against crafted time zone files, avoiding heap buffer overflows related to the processing of the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to large time zone data files. Overly long time zone specifiers in the TZ variable no longer result in stack overflows and crashes. * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors for LD and GD on x86 and x86-64, has been implemented. You will need binutils-2.24 or later to enable this optimization. * Character encoding and ctype tables were updated to Unicode 7.0.0, using new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red Hat). These updates cause user visible changes, such as the fix for bug 17998. * CVE-2014-8121 The NSS backends shared internal state between the getXXent and getXXbyYY NSS calls for the same database, causing a denial-of-service condition in some applications. * Added vector math library named libmvec with the following vectorized x86_64 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf, pow, powf. The library can be disabled with --disable-mathvec. Use of the functions is enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0. Shared library libmvec.so is linked in as needed when using -lm (no need to specify -lmvec explicitly for not static builds). Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information. * A new fmemopen implementation has been added with the goal of POSIX compliance. The new implementation fixes the following long-standing issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The old implementation is still present for use be by existing binaries. * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21 releases. It has been fixed to match 2.19 and older, but binaries built against 2.20 and 2.21 might need to be recompiled. See BZ#18694. * Port to Native Client running on ARMv7-A (--host=arm-nacl). Contributed by Roland McGrath (Google). Contributors ============ This release was made possible by the contributions of many people. The maintainers are grateful to everyone who has contributed changes or bug reports. These include: Adhemerval Zanella Alan Modra Alexandre Oliva Andreas Schwab Andrew Senkevich Andriy Rysin Arjun Shankar Aurelien Jarno Benno Schulenberg Brad Hubbard Carlos O'Donell Chris Metcalf Christian Schmidt Chung-Lin Tang Cong Wang Cyril Hrubis Daniel Marjamäki David S. Miller Dmitry V. Levin Eric Rannaud Evangelos Foutras Feng Gao Florian Weimer Gleb Fotengauer-Malinovskiy H.J. Lu Igor Zamyatin J William Piggott James Cowgill James Lemke John David Anglin Joseph Myers Kevin Easton Khem Raj Leonhard Holz Mark Wielaard Marko Myllynen Martin Galvan Martin Sebor Matthew Fortune Mel Gorman Mike Frysinger Miroslav Lichvar Nathan Lynch Ondřej Bílka Paul Eggert Paul Pluzhnikov Pavel Kopyl Pravin Satpute Rajalakshmi Srinivasaraghavan Rical Jasan Richard Henderson Roland McGrath Rüdiger Sonderfeld Samuel Thibault Siddhesh Poyarekar Stefan Liebler Steve Ellcey Szabolcs Nagy Torvald Riegel Tulio Magno Quites Machado Filho Vincent Bernat Wilco Dijkstra Yaakov Selkowitz Zack Weinberg -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAABAgAGBQJVwbEHAAoJECXvCkNsKkr//nwH/RbC+AmWbbrY7POeygVVxZVv 6ww/s4WOx3MJc0VNhQucelDCmRVRfKdoqtiex2bcysOiK2mv6K4efgYV7dkilT5O NhpjENGE2qCvRIeplmDdGDBTLwhxwcQoQXrFYtcayEXpeCHoJjSzY9PyeNWGvmLM eEah8kVPh6FsNf/YD28MXtChCpfoZf5IrVhXvn7+f2zPjUEy1PuHmo2kU9LzoCRu q3xtd8ICpVkAvFCoUnN7YEOITj3g9Qd+zGebfj8LpVL5zoQs9n2egSv+jIdNGQVI XuQ+oVXuMd9ho1p6LayZpTsY19jALxgk8ysnTzBofi+1Zkc8FTEB0fFdplDIwMg= =9uQ7 -----END PGP SIGNATURE----- Adhemerval Zanella (36): powerpc: multiarch Makefile cleanup for powerpc64 powerpc: Simplify bcopy default implementation powerpc: Remove POWER7 wordcopy ifunc powerpc: wordcopy/memmove cleanup for ppc64 powerpc: multiarch Makefile cleanup for powerpc32 powerpc: wordcopy/memmove cleanup for ppc32 powerpc: sysdeps/powerpc configure cleanup powerpc: drop R_PPC_REL16 check powerpc: Fix TABORT encoding for little endian powerpc: Fix memmove static build powerpc: Fix inline feraiseexcept, feclearexcept macros Update powerpc-fpu ULPs. powerpc: Fix incorrect results for pow when using FMA powerpc: Remove HAVE_ASM_GLOBAL_DOT_NAME define powerpc: Fix __wcschr static build libc-vdso.h place consolidation Fix non-portable echo usage in sysdeps/unix/make-syscalls.sh Add BZ #16704 as fixed Fix stdlib/tst-setcontext3 with dash [BZ#18418] i386: Remove six-argument specialized implementations Remove socket.S implementation Consolidate vDSO macros and usage Consolidate gettimeofday across aarch64/s390/tile Update powerpc-fpu libm-test-ulps. Fix ChangeLog entry x86: clock_gettime and timespec_get vDSO cleanup Use inline syscalls for non-cancellable versions nptl: Rewrite cancellation macros Consolidate sched_getcpu x86: Remove vsyscall usage libio: fmemopen rewrite to POSIX compliance libio: Update tst-fmemopen2.c libio: Update powerpc64le libc.abilist Avoid C++ tests when the C++ cannot be linked libio: Fix fmemopen 'w' mode with provided buffer Update powerpc-fpu libm-test-ulps. Alan Modra (5): Fix localplt test breakage with new readelf Remove HAVE_ASM_PPC_REL16 references powerpc64 configure message powerpc __tls_get_addr call optimization Harden powerpc64 elf_machine_fixup_plt Alexandre Oliva (6): Unicode 7.0.0 update; added generator scripts. Amendments to Unicode 7 update. BZ #15969: search locale archive again after alias expansion Fix constness error just introduced in findlocale. Avoid unsafe loc_name type casts with additional variable Fix DTV race, assert, DTV_SURPLUS Static TLS limit, and nptl_db garbage Andreas Schwab (16): Fix value of O_TMPFILE for architectures with non-default O_DIRECTORY (bug 17912) Filter out PTHREAD_MUTEX_NO_ELISION_NP bit in pthread_mutexattr_gettype (BZ #15790) Fix failure of elf/tst-audit2 when compiled with GCC-5 Fix read past end of pattern in fnmatch (bug 18032) Fix parallel build error Don't define __CORRECT_ISO_CPP_STRING_H_PROTO for non-GCC compilers (bug 17631) m68k: fix 64-bit arithmetic in atomic operations (bug 18128) aarch64: Increase MINSIGSTKSZ and SIGSTKSZ (bug 16850) Separate internal state between getXXent and getXXbyYY NSS calls (bug 18007) Simplify handling of nameserver configuration in resolver Record TTL also for DNS PTR queries (bug 18513) Fix buffer overflow for writes to memory buffer stream (bug 18549) Update NEWS m68k: update libm test ULPs Fix spurious conform test failures Properly terminate FDE in makecontext for ix86 (bug 18635) Andrew Senkevich (29): This is the beginning of series of patches with addition Refactoring of START for conditions in individual tests Last part of changes regarding to libm-test.inc: addition This patch adds infrastructure for addition of SIMD This is update for configure, build and install of vector math library. Localplt testing for vector math library and libmvec_hidden_* macro series. This patch adds detection of availability for AVX512F and AVX512DQ ISAs. Start of series of patches with x86_64 vector math functions. Addition of testing infrastructure for vector math functions. Vector cosf for x86_64. This patch adds vector cosf tests. More strict check of AVX512 support in assembler. Vector sin for x86_64 and tests. Vector sinf for x86_64 and tests. Vector log for x86_64 and tests. Vector logf for x86_64 and tests. Vector exp for x86_64 and tests. Vector expf for x86_64 and tests. Vector pow for x86_64 and tests. Vector powf for x86_64 and tests. Vector sincos for x86_64 and tests. Vector sincosf for x86_64 and tests. Fixed powerpc64 build. Combination of data tables for x86_64 vector functions sin, cos and sincos. Combination of data tables for x86_64 vector functions sinf, cosf and sincosf. More correct description of linking with vector math library. Fixed several libmvec bugs found during testing on KNL hardware. Added runtime check for AVX vector math tests. Prevent runtime fail of SSE vector math tests on non SSE4.1 machine. Andriy Rysin (1): Fix sorting order for Ukrainian locale (BZ 17293) Arjun Shankar (4): CVE-2015-1781: resolv/nss_dns/dns-host.c buffer overflow [BZ#18287] Ensure `wint_t' is defined before use in include/stdio.h Modify elf/tst-audit9.c to use test-skeleton.c Modify several tests to use test-skeleton.c Aurelien Jarno (1): Fix ldconfig segmentation fault with corrupted cache (Bug 18093). Benno Schulenberg (1): sprof: Make an error message identical to two others, and more accurate. Brad Hubbard (1): Use calloc to allocate xports (BZ #17542) Carlos O'Donell (15): Open development for 2.22. Fix missing ChangeLog attribution. NEWS: Fix spelling. Use alignment macros, pagesize and powerof2. hppa: Update libm-test-ulps. hppa: Fix feupdateenv and fesetexceptflag (Bug 18111). Enhance nscd's inotify support (Bug 14906). Bug 18125: Call exit after last linked context. Fail locale installation if localedef fails. Add sprintf benchmark. Fix ruserok scalability with large ~/.rhosts file. Add missing Advanced API (RFC3542) (1) defines. Regenerate libc.pot for 2.22 release. Updated translations for 2.22. Update version.h and include/features.h for 2.22 release Chris Metcalf (8): linux-generic: add a README tile: Enable PI_STATIC_AND_HIDDEN tile: use better variable naming in INLINE_SYSCALL math/test-fenvinline: avoid compiler warning tile: Regenerate ULPs. tst-leaks: raise timeout to 5 seconds tile: Fix BZ #18508 (makecontext yield infinite backtrace) tilepro: fix warnings in sysdeps/tile/tilepro/bits/atomic.h Christian Schmidt (1): Update currency_symbol in da_DK Chung-Lin Tang (5): Adjust timeouts for some tests, to accommodate slow processors, Fix order of arguments to rt_sigprocmask syscall when setting the signal mask Update Nios II ulps file. Add #include <string.h> to nptl/tst-join7mod.c to silence GCC warnings. Fixes extern protected data handling testcases elf/tst-protected1a Cong Wang (1): in.h: Coordinate in6_pktinfo and ip6_mtuinfo for kernel and glibc [BZ #15850] Cyril Hrubis (1): Set errno to ENOMEM on overflow in sbrk (bug 18592) Daniel Marjamäki (1): Add __nonnull attribute to wcscpy and wcsncpy [BZ#18265] David S. Miller (7): Update SPARC ulps. Rebuilt fresh sparc ULPS to get rid of removed tests. Convert sparc over to lowlevellock-futex.h Sparc memchr/memcmp/strncmp fixes from Il'ya Malakhov. Update sparc localplt.data Fix sparc build. Regenerate SPARC ULPs. Dmitry V. Levin (3): Prepare for restoration of .interp section in libpthread.so _res_hconf_reorder_addrs: fix typo in comment Fix potential hanging of gethostbyaddr_r/gethostbyname_r Eric Rannaud (1): linux: open and openat ignore 'mode' with O_TMPFILE in flags Evangelos Foutras (1): Fix __memcpy_chk on non-SSE2 CPUs Feng Gao (1): Use "|" instead of "+" when combine the _IO_LINE_BUF and _IO_UNBUFFERED flags Florian Weimer (32): NEWS: Also mention CVE-2015-1473 _nss_nis_initgroups_dyn: Return status instead of NSS_STATUS_SUCCESS vfprintf: Introduce THOUSANDS_SEP_T vfprintf: Introduce JUMP_TABLE_BASE_LABEL vfprintf: Define WORK_BUFFER_SIZE Avoid SIGFPE in wordexp [BZ #18100] pthread_setaffinity (Linux variant): Rewrite to use VLA instead of alloca Define libc_max_align_t for internal use Add struct scratch_buffer and its internal helper functions scratch_buffer_grow_preserve: Add missing #include <string.h> pldd: Use struct scratch_buffer instead of extend_alloca grp: Rewrite to use struct scratch_buffer instead of extend_alloca _nss_compat_initgroups_dyn: Use struct scratch_buffer instead of extend_alloca getnameinfo: Use struct scratch_buffer instead of extend_alloca nscd_getgr_r: Use struct scratch_buffer instead of extend_alloca scratch_buffer: Suppress truncation warning on 32-bit Do not build with -Winline Make time zone file parser more robust [BZ #17715] posix_fallocate, posix_fallocate64 stub: Do not set errno test-skeleton: Support temporary files without memory leaks [BZ#18333] CVE-2014-8121: Do not close NSS files database during iteration [BZ #18007] NEWS: BZ#18319 was fixed in commit ed159672eb3cd650a32b7e5cb4d5ec1fe0e63802 i386: Remove fallocate, fallocate64, posix_fallocate, posix_fallocate64 __ASSUME_FALLOCATE is always true on 32-bit architectures vfprintf: Move jump table definition and the macros out of function vfprintf: Introduce printf_positional function vfprintf: Remove label name switching for the jump table Avoid some aliasing violations in libio Fix indentation to match nesting in previous commit posix_fallocate: Emulation fixes and documentation [BZ #15661] Commit 7fe9e2e089f4990b7d18d0798f591ab276b15f2b fixes [BZ# 17322] pthread_key_create: Fix typo in comment Gleb Fotengauer-Malinovskiy (1): nptl: restore .interp section in libpthread.so H.J. Lu (23): Compile gcrt1.o with -fPIC Compile vismain with -fPIE and link with -pie Replace ELF_RTYPE_CLASS_NOCOPY with ELF_RTYPE_CLASS_COPY Replace __attribute__((visibility("protected"))) Preserve bound registers in _dl_runtime_resolve Add ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA to x86 Add a testcase for copy reloc against protected data Limit threads sharing L2 cache to 2 for SLM/KNL Check tzspec_len == 0 in __tzfile_read Remove a trailing `\' in make-syscalls.sh Don't issue an error if DT_PLTRELSZ is missing Make sure that calloc is called at least once Don't issue errors on GDB Python files Align TCB offset to the maximum alignment Support compilers defaulting to PIE Add a testcase for i386 LD_AUDIT Add and use sysdeps/i386/link-defines.sym Add la_symbind32 to x86-64 audit tests Improve bndmov encoding with zero displacement Replace %ld with %jd and cast to intmax_t Sort NEWS Add si_addr_bnd to _sigfault in x86 struct siginfo Extend local PLT reference check Igor Zamyatin (1): Preserve bound registers for pointer pass/return J William Piggott (1): [BZ #17969] James Cowgill (1): [BZ #17930] MIPS: Define SHM_NORESERVE. James Lemke (1): Fix for test "malloc_usable_size: expected 7 but got 11" John David Anglin (2): hppa: fix __O_SYNC to match the kernel hppa: Fix feholdexcpt and fesetenv (Bug 18110). Joseph Myers (162): soft-fp: Support floating-point extensions without quieting sNaNs. soft-fp: Refine FP_EX_DENORM handling for comparisons. soft-fp: Fix _FP_FMA when product is zero and third argument is finite (bug 17932). Remove sysdeps/mips soft-fp subdirectories. Fix sincos errno setting (bug 15467). Fix exp2 spurious underflows (bug 16560). Fix powerpc software sqrt (bug 17964). Fix powerpc software sqrtf (bug 17967). Fix dbl-64/wordsize-64 remquo (bug 17569). Fix MIPS __mips_isa_rev -Werror=undef build. Fix MIPS _COMPILING_NEWLIB -Werror=undef build. Fix MIPS _ABIO64 -Werror=undef build. Fix remquo spurious overflows (bug 17978). Fix sign of remquo zero remainder in round-downward mode (bug 17987). Refine documentation of libm exceptions goals. Fix posix_spawn getrlimit64 namespace (bug 17991). Fix search.h namespace (bug 17996). Fix atan / atan2 missing underflows (bug 15319). Fix scandir scandirat namespace (bug 17999). soft-fp: Adjust call to abort for kernel use. Fix x86/x86_64 scalb (qNaN, -Inf) (bug 16783). Fix ldbl-128ibm acoshl inaccuracy (bug 18019). Fix ldbl-128ibm asinhl inaccuracy (bug 18020). Fix ldbl-128ibm ilogbl near powers of 2 (bug 18029). Fix ldbl-128ibm logbl near powers of 2 (bug 18030). Fix asin missing underflows (bug 16351). Fix ldbl-128/ldbl-128ibm acosl inaccuracy (bug 18038, bug 18039). Avoid uninitialized warnings in Bessel functions. Avoid -Wno-write-strings for k_standard.c. Add comment to CSTR macro in k_standard.c. Fix ldbl-96, ldbl-128ibm atanhl inaccuracy (bug 18046, bug 18047). Correct __ASSUME_PRLIMIT64 for hppa/microblaze/sh (bug 17779). soft-fp: Condition sfp-machine.h include path on __KERNEL__. Fix /* in comment in previous commit. soft-fp: Support conditional zero-initialization in declarations. soft-fp: Use multiple-include guards. Add test for bug 18104. soft-fp: Add _FP_UNREACHABLE. soft-fp: Define and use _FP_STATIC_ASSERT. Make sem_timedwait use FUTEX_CLOCK_REALTIME (bug 18138). Note old commit as having resolved bug 11505. Add more tests of log2. Regenerate x86_64, x86 ulps from scratch. Add more tests of cosh, sinh. Add more tests of expm1. Add more tests of acos. Support six-argument syscalls from C for 32-bit x86, use generic lowlevellock-futex.h (bug 18138). Add more tests of asin. Remove unused macros from i386 lowlevellock.h. Add another test of asin. Add more tests of acosh, asinh and atanh. Fix dbl-64 atan in non-default rounding modes (bug 18197). Fix dbl-64 atan2 in non-default rounding modes (bug 18210, bug 18211). Add more tests of cabs. Add more tests of cbrt. Add more tests of atan. Add more tests of atanh. Add more tests of clog and clog10. Fix strtof decimal rounding close to half least subnormal (bug 18247). Fix ldbl-128 roundl for exponents in [31, 47] (bug 18346). Remove MIPS version of waitid.c. Add further tests of cosh and sinh. Add more tests of csqrt. Add more tests of erf, erfc. Add more tests of exp, exp10, exp2, expm1. Add more tests of log, log10, log1p, log2. Add more tests of lgamma. Add another test of pow. Add more tests of cos, sin, sincos. Add more tests of tan. Add more tests of tanh. Add more tests of tgamma. Add more tests of libm functions. Add further tests of libm functions. Add more tests of acosh, atanh, cos, csqrt, erfc, sin, sincos. Add more tests of csqrt, lgamma, log10, sinh. Fix mips16 __fpu_control static linking (bug 18397). Fix linknamespace test handling of architecture-specific st_other. Fix log1p missing underflows (bug 16339). Fix atanf spurious underflows (bug 18196). Fix erfcf spurious underflows (bug 18217). Fix lgammaf spurious underflows (bug 18220). Fix tanf spurious underflows (bug 18221). Fix atanhl missing underflows (bug 16352). Fix i386 atanhl spurious underflows (bug 18049). Fix ldbl-96 remquol (finite, Inf) (bug 18244). conformtest: clean up POSIX expectations for unistd.h. conformtest: correct POSIX expectations for locale.h. conformtest: use proper _POSIX_C_SOURCE value for POSIX. linknamespace: whitelist re_syntax_options. Fix sysdeps/ieee754/dbl-64/mpa.c for -Wuninitialized. Fix lgamma implementations for -Wuninitialized. Fix pathconf basename namespace (bug 18444). Restore _POSIX2_C_VERSION definition (bug 438). Fix ldbl-128 / ldbl-128ibm asinl for -Wuninitialized. Fix ldbl-128 / ldbl-128ibm erfcl for -Wuninitialized Fix ldbl-128 / ldbl-128ibm tanl for -Wuninitialized. Fix soft-fp fma for -Wuninitialized. Fix fnmatch towlower namespace (bug 18469). Use libc_hidden_proto / libc_hidden_def with __strnlen. Use better variable names in MIPS syscall macros. Fix fnmatch wmemchr namespace (bug 18468). Fix fnmatch strnlen namespace (bug 18470). Fix regex wctype namespace (bug 18495). Fix psignal, psiginfo declaration conditions (bug 18483). Fix regex wcrtomb namespace (bug 18496). Fix open_memstream namespace (bug 18498). Say "C++ tests" in comment on __open_memstream declaration. Fix pathconf statvfs namespace (bug 18507). Fix regcomp wcscoll, wcscmp namespace (bug 18497). Fix h_errno namespace (bug 18520). Fix ecvt_r, fcvt_r namespace (bug 18522). Fix aio_* pread namespace (bug 18519). Fix getlogin_r namespace (bug 18527). Fix grp.h endgrent, getgrent namespace (bug 18528). Fix netdb.h addrinfo namespace (bug 18529). Fix syslog fputs_unlocked namespace (bug 18530). Fix linknamespace expectations for in6addr_any, in6addr_loopback. Fix gethostbyaddr in6addr_any, in6addr_loopback namespace (bug 18532). Fix vsyslog namespace (bug 18533). Fix syslog dprintf namespace (bug 18534). Fix sem_* tdelete, tfind, tsearch, twalk namespace (bug 18536). Fix fmtmsg addseverity namespace (bug 18539). Fix getpass fflush_unlocked namespace (bug 18540). Fix swscanf vswscanf namespace (bug 18542). Fix mq_notify pthread_barrier_* namespace (bug 18544). Create hidden aliases for non-libc syscalls automatically. Fix mq_receive, mq_send mq_timed* namespace (bug 18545). Fix mq_notify socket, recv namespace (bug 18546). Fix ttyslot namespace (bug 18547). Fix nice getpriority, setpriority namespace (bug 18553). Remove ldbl-128ibm variants of complex math functions. Fix netinet/in.h MCAST_* namespace (bug 18558). Remove stray spurious-underflow markings from cexp test. Remove include/bits/ipc.h. Fix asinh missing underflows (bug 16350). conformtest: Support xfail markers on individual assertions. conformtest: Fix pselect expectations. Fix x86 / x86_64 expl, exp10l missing underflows (bug 16361). Correct ChangeLog syntax for conditional change within function. Fix x86_64 / x86 expm1l (-min_subnorm) result sign (bug 18569). Fix expm1 missing underflows (bug 16353). Fix exp2, exp2f spurious underflows (bug 18219). Fix csqrt spurious underflows (bug 18371). Fix math/Makefile dependency on libm-test.stmp for libmvec tests. Fix spurious "inexact" exceptions from __kernel_standard_l (bug 18245, bug 18583). Fix sin, sincos missing underflows (bug 16526, bug 16538). Fix ldbl-128 expl missing underflows (bug 18586). Fix csin, csinh overflow in directed rounding modes (bug 18593). Move csin, csinh tests to auto-libm-test-in. Fix cexp, ccos, ccosh, csin, csinh spurious underflows (bug 18594). Refactor libm tests. Use round-to-nearest internally in jn, test with ALL_RM_TEST (bug 18602). Update headers for Linux 4.0, 4.1 definitions. Fix j1, jn missing underflows (bug 16559). Fix ldbl-128 j1l spurious underflows (bug 18612). Improve tgamma accuracy (bug 18613). Regenerate MIPS libm-test-ulps. Regenerate ARM libm-test-ulps. Regenerate powerpc-nofpu libm-test-ulps. Fix ldbl-128 expm1l (-min_subnorm) result sign (bug 18619). Mark bug 2981 (elf/tst-audit* fail on MIPS) as fixed. Kevin Easton (1): Reduce lock contention in __tz_convert() [BZ #16145] (partial fix) Khem Raj (2): Reflect renaming of bh_IN and tu_IN in SUPPORTED file [BZ #17475] locale: Do not define lang_ab for tcy_IN and bhb_IN Leonhard Holz (6): Remove unused definitions Improve strcoll with strdiff. Split locale generation snippet into a separate file Add strcoll benchmark remove now unused idxnow in strcoll remove unnecessary memset in strcoll Mark Wielaard (2): elf.h SHF_EXCLUDE signed int 31 bit shift triggers undefined behaviour. elf.h: Add section compression constants and structures. Marko Myllynen (4): Fix bo_CN and bo_IN. Fix monetary.h comment Remove unused PREDEFINED_CLASSES code locale: Remove obsolete repertoire map references Martin Galvan (2): NPTL: swap comments for THREAD_SETMEM and THREAD_SETMEM_NC for i386 and x86_64 NPTL: Remove duplicate definition of PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP Martin Sebor (4): powerpc: setcontext.S uses power6 mtfsf when not supported [BZ #18116] Attempting to install glibc configured with --prefix=/usr into The C++ 2011 std::call_once function is specified to allow The patch committed to fix bug #18435 caused regressions on aarch64 Matthew Fortune (2): ia64: remove fixed page size macros and others [BZ #17792] Add support for DT_MIPS_RLD_MAP_REL. Mel Gorman (2): malloc: Consistently apply trim_threshold to all heaps [BZ #17195] malloc: Do not corrupt the top of a threaded heap if top chunk is MINSIZE [BZ #18502] Mike Frysinger (27): ia64: drop custom getpagesize hppa: fix build failure with RTLD_PRIVATE_ERRNO add changelog for previous commit alloca: fix buf interaction manual: drop strerror C89 compatibility note hppa: update __O_SYNC fix with [BZ #18068] pwd.h: add __nonnull markings [BZ #18641] nscd: drop selinux/flask.h include tst-tzset: raise timeout to 5 seconds hppa/ia64: _dl_symbol_address: add PLT bypass for rtld hppa/ia64: _dl_unmap: make it hidden sparc: fix sigaction for 32bit builds [BZ #18694] ia64: siginfo.h: delete siginfo name ia64: sifaction.h: change sa_flags to an int ia64: stat.h: rename pad0 to __glibc_reserved0 ia64: msg.h: fix msg_qnum/msg_qbytes types ia64: sigaction.h: fix sa_flags ordering conform/linknamespace: whitelist matherrf/matherrl pwd.h: revert __nonnull markings on putpwent [BZ #18641] ia64: clean up old kernel headers cruft ia64: atomic.h: fix atomic_exchange_and_add 64bit handling ia64: drop __tls_get_addr from expected ld.so plt usage hppa: rewrite INLINE_SYSCALL hppa: fix sysdep.h header setup hppa: sigaction.h: change sa_flags to an int hppa: fix pthreadtypes.h namespace failures hppa: add bz entry for pthreadtypes.h fix Miroslav Lichvar (1): Update timex.h for ADJ_SETOFFSET. Nathan Lynch (1): ARM: VDSO support Ondřej Bílka (2): Handle mblen return code when n is zero. Use strspn/strcspn/strpbrk ifunc in internal calls. Paul Eggert (6): Add ersatz _Static_assert on older C hosts * manual/time.texi (TZ Variable): glibc no longer comes with tzdata. * stdlib/setenv.c (__add_to_environ): * stdlib/setenv.c (__add_to_environ): Revert previous change. Better fix for setenv (..., NULL, ...) Remove obsolete aliases that broke 'locale -a' Paul Pluzhnikov (13): Cleanup: add missing #include's Fix BZ #17269 -- _IO_wstr_overflow integer overflow Fix BZ #17916 - fopen unbounded stack usage for ccs= modes Fix minor formatting violation. Fix BZ 18036 buffer overflow (read past end of buffer) in internal_fnmatch Fix BZ #18043: buffer-overflow (read past the end) in wordexp/parse_dollars/parse_param Cleanup: in preparation for fixing BZ #16734, fix memory leaks exposed by Refactor wordexp-test.c such that words always ends at the edge of Fix off-by-one which caused BZ #18042 and add a test for it. Mention BZ #18042 in NEWS. Fix BZ #18043 (c4): buffer-overflow (read past the end) in wordexp/parse_dollars/parse_param Minor refactoring: Fix BZ #18043 comment # 19: don't call undefined setenv(..., NULL, 1). Pavel Kopyl (1): Add forced deletion support to _dl_close_worker Pravin Satpute (1): Correcting language code for Bhili and Tulu locales (bug 17475) Rajalakshmi Srinivasaraghavan (2): powerpc: POWER7 strncpy optimization for unaligned string powerpc: strstr optimization Rical Jasan (1): manual: complete example in error message documentation Richard Henderson (5): alpha: Unconditionally include dl-sysdep.h in sysdep.h alpha: Update libm-test-ulps math/test-fenvinline: Cast fe_exc to unsigned int before printing alpha: Update libm-test-ulps soft-fp: Fix alpha kernel build problem Roland McGrath (97): Clean up sysdep-dl-routines variable. Exclude rpcent functions and NSS backends for rpc, key when excluding sunrpc. x86: Clean up __vdso_clock_gettime variable. Clean up NPTL fork to be compat-only. Clean up NPTL longjmp to be compat-only. Clean up NPTL system to be compat-only. Clean up nptl/tst-join5 use of nanosleep. Fix nptl/tst-kill5 not to presume SIGRTMAX exists. Fix dirent/tst-fdopendir not to presume O_NOATIME exists. Fix libio/tst-atime not to presume ST_NOATIME exists. Move tst-getlogin to login/ subdirectory. Do not use SA_NOCLDWAIT in tst-pselect. Conditionalize some tests' use of SA_SIGINFO. Use signal rather than sigaction in nptl/tst-cleanup2. NPTL: Build tests using clone directly only for Linux. Don't set unused field in rt/tst-timer2. Conditionalize use of SIGRTMIN in nptl/tst-locale1.c. NPTL: Conditionalize some sanity tests for SIGCANCEL/SIGSETXID. ARM: Add missing sfi_breg in LDR_GLOBAL macro. Clean up math/test-snan. Pointless update in README. Another pointless update in README. Support after-link variable to run a final step on binaries. Use -Werror=undef for assembly code. NPTL: Initializer for .init_array-only configurations. Add placeholder c++-types.data and *.abilist files. Don't crash in iconv setup when getcwd fails. Convert tst-iconv5 to use test-skeleton. Convert tst-iconv3 to use test-skeleton. Convert dlfcn/tststatic2 to use test-skeleton. Deglobalize internal variables in timer_routines.c. Avoid C++ tests when the C++ cannot be linked. Avoid more C++ tests. Conditionalize some test code for SIGRTMIN, SA_SIGINFO. Split rpcent tests out of tst-netdb. Define ETH_ALEN in generic <netinet/if_ether.h>. Avoid re-exec-self in bug-setlocale1. ChangeLog format Document test-wrapper-env-only in INSTALL. Harmonize posix/regcomp.c with gnulib: comment formatting Let tests result in UNSUPPORTED; use that for unbuildable C++ cases ARM: Rewrite sysdeps/arm/tls-macros.h ARM: Fix memcpy & memmove for [ARM_ALWAYS_BX] Minor cleanups in libio/iofdopen.c Convert dlfcn/tststatic to use test-skeleton. Make test-skeleton.c grok TEST_DIRECT magic environment variable. Let non-add-on preconfigure scripts set libc_config_ok. Omit libc-modules.h for all .v.i files. Add arm-nacl port. Fuller check for invalid NSID in _dl_open. Avoid confusing compiler with dynamically impossible statically invalid dereference in _dl_close_worker. ARM: Define PI_STATIC_AND_HIDDEN. NaCl: Make __suseconds_t be long int rather than int32_t. NaCl: Fix symbol names for euidaccess. NaCl: Change clock_t to long int. NaCl: Fix elf_loader file name in nacl-test-wrapper.sh BZ#18383: Add test case for large alignment in TLS blocks. NaCl: Implement gethostname. NaCl: Provide non-default values for uname. Add a test case for scandir. Break __scandir_cancel_handler out into its own file. Refactor scandir/scandirat to use common tail. Nit fixes in last change. NaCl: Make fdopendir skip fcntl check. Refactor opendir. BZ#18434: Fix sem_post EOVERFLOW check for [!__HAVE_64B_ATOMICS]. BZ#18434: Mark fixed in NEWS. Move usleep.c using nanosleep to sysdeps/posix. NaCl: Set tid field to a unique value. Fix nptl-init.c use of INTERNAL_SYSCALL_DECL. Split timed-wait functions out of nptl/lowlevellock.c. NaCl: Add NaCl-specific __lll_timedlock_wait. NaCl: Fix thinko in last change. NaCl: Fix lll_futex_timed_wait timeout calculation. NaCl: Make thread exit wake pthread_join. Fix setenv.c diagnostic pragma to be compatible with GCC 4.6 BZ#18383: Another test case, with TLS refs and defs in separate TUs. NaCl: Implement nacl_interface_ext_supply entry point. Line-wrap some log entries. Print more information in tst-getcpu failure case. NaCl: Fix glob.c build after getlogin_r -> __getlogin_r. Use unsigned types for counters in AIO code. Use unsigned types for counters in getaddrinfo_a code. NPTL: Use unsigned type for setxid_futex. Install a dummy <rpc/netdb.h> when not building sunrpc/. Fix some places to use $(LN_S) makefile variable. BZ#18383: Conditionalize test-xfail-tst-tlsalign{,-static} on ARM assembler bug. PLT avoidance for _exit in rtld. Provide __libc_fatal for rtld. NaCl: Make pthread_condattr_setclock reject CLOCK_MONOTONIC. Factor file identity rules out of generic rtld code. Add abilist files and NEWS item for arm-nacl port. NaCl: Use only nacl_irt_dev_filename, never nacl_irt_filename. NaCl: Fix missing getdtablesize symbol. Add SIGWINCH to generic <bits/signum.h>. Make sysdeps/posix bring in login subdir. NaCl: Remove bogus O_SHLOCK, O_EXLOCK definitions. Rüdiger Sonderfeld (1): Document tv_sec is of type time_t: Samuel Thibault (28): hurd: fix build with pthread aio hurd: fix f?chflags prototypes, declare them and their flags hurd: allow poll() array bigger than FD_SETSIZE hurd: map nice levels 1-to-1 with Mach prio levels hurdselect: Let select get interrupted by signals hurd: fix sigstate locking hurdselect: remove dead code. hurd: support mmap with PROT_NONE hurd: add basic types for ioctls hurd: fix compilation of signal.h in C++ hurd: fix compilation of signal.h in C++ hurd: Ignore bytes beyond sockaddr length for AF_UNIX hurd: fix tls.h build hurd: Fix abi-tag, following ba90e05 Fix time/getdate.c build. add hurd/hurdsocket.h file missing from a5eb23d hurd: fix unwind-resume.c build hurd: fix unwind-resume.c build Add fixed bug numbers to NEWS Revert "hurd: Fix abi-tag, following ba90e05" Fix aio_error thread-safety. hurd: Make libc able to call pthread stubs Add missing dependency Fix warnings Fix visibility of EXTPROC macro Add more exception to local headers list mach: fix typo hurd: permit to use mlock from non-root process Siddhesh Poyarekar (25): Consolidate arena_lookup and arena_lock into a single arena_get Skip logging for DNSSEC responses [BZ 14841] Fix up NEWS merge goof-up Update NEWS Minor changelog fixup Add *.pyc to .gitignore Add envz_remove to the libc manual Succeed if make check does not report any errors Avoid deadlock in malloc on backtrace (BZ #16159) Fix typo in safety annotations in envz_remove Fix monetary.h comment New module to import and process benchmark output benchtest: script to compare two benchmarks Avoid boolean coercion in tst-tls-atexit test case Remove unnecessary mutex locks from tst-tls-atexit test case Whitespace fix in tst-tls-atexit.c Fix up ChangeLog Fix up typo in tst-tls-atexit Set NODELETE flag when opening already open objects with RTLD_NODELETE Whitespace fixup in cxa_thread_atexit_impl.c Add comment to clarify how the test can fail Remove Linuxism from tst-tls-atexit Also use l_tls_dtor_count to decide on object unload (BZ #18657) Mention dl_load_lock by name in the comments Use IE model for static variables in libc.so, libpthread.so and rtld Stefan Liebler (18): S390: Build failure due to nptl/pt-longjmp.c changes. s390: Use generic lowlevellock-futex.h S/390: Regenerate ULPs S/390: Fix setcontext/swapcontext which are not restoring sigmask. Update tst_mbrlen/tst_mbrtowc for mblen change Set errno for log1p on pole/domain error. Use correct signedness in wcsncmp S/390: Get cache information via sysconf S/390: Regenerate ULPs Adjust tst-strfmon1 after da_DK locale change. S/390: Regenerate ULPs Fix timezone tests run in parallel. Fix benchtests build failure after 'add benchmark for strcoll' S390: Fix sem.h conformance test failures. S390: Regenerate ULPs. S390: Fix "backtrace() returns infinitely deep stack frames with makecontext()" [BZ #18508]. S390: Regenerate ULPs i686: Mark stdlib/tst-makecontext as XFAIL. Steve Ellcey (6): 2015-02-13 Steve Ellcey <sellcey@imgtec.com> 2015-02-17 Steve Ellcey <sellcey@imgtec.com> 2015-02-17 Steve Ellcey <sellcey@imgtec.com> 2015-02-18 Steve Ellcey <sellcey@imgtec.com> * inet/rcmd.c (rresvport_af): Change ss to anonymous union * resolv/res_hconf.c (_res_hconf_reorder_addrs): Use a union to Szabolcs Nagy (11): [AArch64] Fix the big endian loader name. [AArch64] Fix inline asm clobber list in tls-macros.h struct stat is not posix conform [BZ 18034][AArch64] Lazy TLSDESC relocation data race fix [AArch64] Fix cfi_adjust_cfa_offset usage in dl-tlsdesc.S Regenerate aarch64 libm-test-ulps [AArch64] make setcontext etc functions consistent with the kernel [AArch64][BZ 18400] fix elf_prpsinfo in procfs.h [AArch64][BZ 18648] change greg_t definition in ucontext.h [AArch64][BZ #17711] Fix extern protected data handling [ARM][BZ #17711] Fix extern protected data handling Torvald Riegel (11): Make error checking effective in nptl/tst-cond25.c. ia64: Remove custom lowlevellock.h Fix lost wake-up when pthread_rwlock_timedrwlock times out. Fix missing wake-ups in pthread_rwlock_rdlock. Fix atomic_full_barrier on x86 and x86_64. Clean up BUSY_WAIT_NOP and atomic_delay. Remove documentation of lowlevellock systemtap probes. Do not create invalid pointers in C code of string functions. Add and use new glibc-internal futex API. Clean up semaphore EINTR handling after Linux futex docs clarification. hppa: Remove custom lowlevellock.h. Tulio Magno Quites Machado Filho (2): BZ #18116: Mark fixed in NEWS. Avoid outputting to TTY after an expected memory corruption in testcase Vincent Bernat (1): time: ensure failing strptime() tests are reported correctly Wilco Dijkstra (14): Rather than using a C implementation of memset, directly call memset, which Rather than using a C implementation of memmove, directly call memmove, which Use __copysign rather than copysign. 2015-05-06 Szabolcs Nagy <szabolcs.nagy@arm.com> Remove various ABS macros and replace uses with fabs (or in one case abs) Add missing math_private includes. 2015-05-28 Wilco Dijkstra <wdijkstr@arm.com> 2015-06-02 Szabolcs Nagy <szabolcs.nagy@arm.com> This patch renames all uses of __isinf*, __isnan*, __finite* and __signbit* to use standard C99 macros. This has no effect on generated code. Replace finite with isfinite. Remove unused file sysdeps/ieee754/support.c Inline __ieee754_sqrt and __ieee754_sqrtf. Also add external definitions. Optimize the strlen implementation by using a page cross check and a fast check Add AArch64 versions of math_opt_barrier and math_force_eval that avoid going via memory. Yaakov Selkowitz (1): manual: fix XPG basename prototype Zack Weinberg (1): Deprecate the use of regexp.h -----------------------------------------------------------------------