Open/pending issues as of Jan 17

Open/pending issues as of Jan 17

[Rich Felker]

With a quick review of the mailing list since late November I found
the following possibly-open issues:

- resolver search domain support
- preinit_array
- mlock2 syscall wrapper
- mips siginfo_t fix
- avl tree: any changes still pending??
- adding pot file for translators
- unsetenv pointer eval after free
- printf big widths/precisions
- atomic.h overhaul
- dirent.h macros
- dlopen deadlock
- setvbuf

Also, not code issues, but the following possibly-open documentation
issues were found:

- documenting atomic.h primitives contracts
- documenting "known UB" in musl

There were a few other patches that were mostly trivial which I just
committed outright, and the big one, out-of-tree builds, that I
finally committed after some more testing.

Is there anything else I missed that's an open issue right now? Or
requests for prioritizing the above list? I'm going to try to take the
above (and anything else that comes up) and turn it into a plan for
what can reasonably go into the next release.

Rich

Re: Open/pending issues as of Jan 17

[Szabolcs Nagy]

* Rich Felker <dalias@libc.org> [2016-01-17 18:15:54 -0500]:
> With a quick review of the mailing list since late November I found
> the following possibly-open issues:
> 
> - resolver search domain support
> - preinit_array
> - mlock2 syscall wrapper
> - mips siginfo_t fix

i think this should have priority
(abi visible bug, should not wait for arch/bits cleanup)

> - avl tree: any changes still pending??

there are no known bugs in the current code, but
tsearch_small.c would be an improvement over it.
(and twalk,tfind should be separate tu)

> - adding pot file for translators
> - unsetenv pointer eval after free
> - printf big widths/precisions
> - atomic.h overhaul
> - dirent.h macros
> - dlopen deadlock
> - setvbuf

random stuff:
- linux 4.* uapi (includes mlock2, but a lot of other stuff too)
- symbol versioning (for at least gcc-6 libgcc_s.so)
- __cxa_thread_atexit_impl ? (i expect this to come up in real code)
- stdatomic.h
- string functions correctness
- regex bre extensions \+ \? \|
- regcomp increase tre_stack_new limit
- str{f,p}time %z %Z %k %l ?
- noreturn __assert_fail ?
- hidden visibility for internal functions
- compiler builtins in libc (sqrt, fabs, memcpy..)
- asm -> inline asm
- ld128 math

> Also, not code issues, but the following possibly-open documentation
> issues were found:
> 
> - documenting atomic.h primitives contracts
> - documenting "known UB" in musl
> 
> There were a few other patches that were mostly trivial which I just
> committed outright, and the big one, out-of-tree builds, that I
> finally committed after some more testing.
> 
> Is there anything else I missed that's an open issue right now? Or
> requests for prioritizing the above list? I'm going to try to take the
> above (and anything else that comes up) and turn it into a plan for
> what can reasonably go into the next release.
> 
> Rich

Re: Open/pending issues as of Jan 17

[Rich Felker]

On Mon, Jan 18, 2016 at 12:25:30PM +0100, Szabolcs Nagy wrote:
> random stuff:
> - linux 4.* uapi (includes mlock2, but a lot of other stuff too)

Done.

> - symbol versioning (for at least gcc-6 libgcc_s.so)

This is a fairly invasive change and needs testing and possibly
further discussion. It's big enough I'd like to allocate it attention
on the roadmap for a specific release cycle. It's tentatively a
secondary target for the next cycle but we could move it up to primary
if this is becoming a serious problem.

> - __cxa_thread_atexit_impl ? (i expect this to come up in real code)

It looks to me like gcc emulates it with a pthread key to get a dtor
and doesn't actually need help from the libc. If so I don't think it
makes sense to bother with doing a better implementation on the libc
side unless/until we can resolve the lazy allocation bug in their ABI.

> - stdatomic.h

Let's look at this later.

> - string functions correctness

I don't think there's much that can be done aside from adding the
may_alias attribute. But at least that small improvement can go on the
agenda for the next release cycle.

> - regex bre extensions \+ \? \|

Possibly going to apply this now unless you see a risk of breaking
anything else.

> - regcomp increase tre_stack_new limit

Sounds ok to apply now but I lost track of where the patch is.

> - str{f,p}time %z %Z %k %l ?
> - noreturn __assert_fail ?
> - hidden visibility for internal functions

Let's look at these after release.

> - compiler builtins in libc (sqrt, fabs, memcpy..)
> - asm -> inline asm

A little bit of this was done (for arm) but I'd like to continue where
it makes sense.

> - ld128 math

Yes, that would be nice. Should we set a specific roadmap goal for
adding it or at least adding the most important stuff (like sqrtl)
that's currently badly broken.

Rich

Re: Open/pending issues as of Jan 17

[Christian Neukirchen]

Rich Felker <dalias@libc.org> writes:

> With a quick review of the mailing list since late November I found
> the following possibly-open issues:
>
> - resolver search domain support
> - preinit_array
> - mlock2 syscall wrapper
> - mips siginfo_t fix
> - avl tree: any changes still pending??
> - adding pot file for translators
> - unsetenv pointer eval after free
> - printf big widths/precisions
> - atomic.h overhaul
> - dirent.h macros
> - dlopen deadlock
> - setvbuf

So it wont be forgotten:
- adding GLOB_TILDE would simplify building quite some projects

-- 
Christian Neukirchen  <chneukirchen@gmail.com>  http://chneukirchen.org

Re: Open/pending issues as of Jan 17

[Rich Felker]

On Mon, Jan 18, 2016 at 01:12:31PM +0100, Christian Neukirchen wrote:
> Rich Felker <dalias@libc.org> writes:
> 
> > With a quick review of the mailing list since late November I found
> > the following possibly-open issues:
> >
> > - resolver search domain support
> > - preinit_array
> > - mlock2 syscall wrapper
> > - mips siginfo_t fix
> > - avl tree: any changes still pending??
> > - adding pot file for translators
> > - unsetenv pointer eval after free
> > - printf big widths/precisions
> > - atomic.h overhaul
> > - dirent.h macros
> > - dlopen deadlock
> > - setvbuf
> 
> So it wont be forgotten:
> - adding GLOB_TILDE would simplify building quite some projects

Thanks. I put this on the roadmap on the wiki, mainly as an aid to
remember it. I think we're going to want to shuffle and reprioritize a
lot after this release since there's a lot of stuff on there right now
and it's not clear to me that the current top items are actually in
the most demand.

Rich

Re: Open/pending issues as of Jan 17

[Natanael Copa]

On Thu, 28 Jan 2016 21:37:56 -0500
Rich Felker <dalias@libc.org> wrote:

> On Mon, Jan 18, 2016 at 01:12:31PM +0100, Christian Neukirchen wrote:
> > Rich Felker <dalias@libc.org> writes:
> >   
> > > With a quick review of the mailing list since late November I found
> > > the following possibly-open issues:
> > >
> > > - resolver search domain support
> > > - preinit_array
> > > - mlock2 syscall wrapper
> > > - mips siginfo_t fix
> > > - avl tree: any changes still pending??
> > > - adding pot file for translators
> > > - unsetenv pointer eval after free
> > > - printf big widths/precisions
> > > - atomic.h overhaul
> > > - dirent.h macros
> > > - dlopen deadlock
> > > - setvbuf  
> > 
> > So it wont be forgotten:
> > - adding GLOB_TILDE would simplify building quite some projects  
> 
> Thanks. I put this on the roadmap on the wiki, mainly as an aid to
> remember it. I think we're going to want to shuffle and reprioritize a
> lot after this release since there's a lot of stuff on there right now
> and it's not clear to me that the current top items are actually in
> the most demand.

An unrelated feature request: it would be nice to get netflix working
with the closed source libwidevinecdm.so and libwidevinecdmadapter.so 

ncopa-desktop:~$ ldd /usr/lib/chromium/libwidevinecdmadapter.so 
	ldd (0x723cd632b000)
	libwidevinecdm.so => /usr/lib/chromium/libwidevinecdm.so (0x723cd5ba2000)
	libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x723cd5852000)
	libpthread.so.0 => ldd (0x723cd632b000)
	libc.so.6 => ldd (0x723cd632b000)
	libm.so.6 => ldd (0x723cd632b000)
	libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x723cd563f000)
Error relocating /usr/lib/chromium/libwidevinecdm.so: __memcpy_chk: symbol not found
Error relocating /usr/lib/chromium/libwidevinecdm.so: __memset_chk: symbol not found
Error relocating /usr/lib/chromium/libwidevinecdm.so: __sprintf_chk: symbol not found


Adding those symbols for libc6 ABI compat would be a first step.

-nc

> 
> Rich

Re: Open/pending issues as of Jan 17

[Rich Felker]

On Fri, Jan 29, 2016 at 08:54:49AM +0100, Natanael Copa wrote:
> On Thu, 28 Jan 2016 21:37:56 -0500
> Rich Felker <dalias@libc.org> wrote:
> 
> > On Mon, Jan 18, 2016 at 01:12:31PM +0100, Christian Neukirchen wrote:
> > > Rich Felker <dalias@libc.org> writes:
> > >   
> > > > With a quick review of the mailing list since late November I found
> > > > the following possibly-open issues:
> > > >
> > > > - resolver search domain support
> > > > - preinit_array
> > > > - mlock2 syscall wrapper
> > > > - mips siginfo_t fix
> > > > - avl tree: any changes still pending??
> > > > - adding pot file for translators
> > > > - unsetenv pointer eval after free
> > > > - printf big widths/precisions
> > > > - atomic.h overhaul
> > > > - dirent.h macros
> > > > - dlopen deadlock
> > > > - setvbuf  
> > > 
> > > So it wont be forgotten:
> > > - adding GLOB_TILDE would simplify building quite some projects  
> > 
> > Thanks. I put this on the roadmap on the wiki, mainly as an aid to
> > remember it. I think we're going to want to shuffle and reprioritize a
> > lot after this release since there's a lot of stuff on there right now
> > and it's not clear to me that the current top items are actually in
> > the most demand.
> 
> An unrelated feature request: it would be nice to get netflix working
> with the closed source libwidevinecdm.so and libwidevinecdmadapter.so 
> 
> ncopa-desktop:~$ ldd /usr/lib/chromium/libwidevinecdmadapter.so 
> 	ldd (0x723cd632b000)
> 	libwidevinecdm.so => /usr/lib/chromium/libwidevinecdm.so (0x723cd5ba2000)
> 	libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x723cd5852000)
> 	libpthread.so.0 => ldd (0x723cd632b000)
> 	libc.so.6 => ldd (0x723cd632b000)
> 	libm.so.6 => ldd (0x723cd632b000)
> 	libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x723cd563f000)
> Error relocating /usr/lib/chromium/libwidevinecdm.so: __memcpy_chk: symbol not found
> Error relocating /usr/lib/chromium/libwidevinecdm.so: __memset_chk: symbol not found
> Error relocating /usr/lib/chromium/libwidevinecdm.so: __sprintf_chk: symbol not found
> 
> 
> Adding those symbols for libc6 ABI compat would be a first step.

This is on the roadmap, and there's a pending patch on the list, but
it still needs some changes before it's ready to commit.

Are you aware of anything else blocking support for these libs? It
might be worth doing a symbol listing and checking it against
functions that might have glibc extensions we don't support or the
minor ABI mismatches (non-*64 funcs for 32-bit, or regexec regoff_t
for 64-bit).

Rich

Re: netflix support (WAS: Open/pending issues as of Jan 17)

[Natanael Copa]

On Fri, 29 Jan 2016 13:19:28 -0500
Rich Felker <dalias@libc.org> wrote:

> On Fri, Jan 29, 2016 at 08:54:49AM +0100, Natanael Copa wrote:
> > An unrelated feature request: it would be nice to get netflix working
> > with the closed source libwidevinecdm.so and libwidevinecdmadapter.so 
> > 
> > ncopa-desktop:~$ ldd /usr/lib/chromium/libwidevinecdmadapter.so 
> > 	ldd (0x723cd632b000)
> > 	libwidevinecdm.so => /usr/lib/chromium/libwidevinecdm.so (0x723cd5ba2000)
> > 	libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x723cd5852000)
> > 	libpthread.so.0 => ldd (0x723cd632b000)
> > 	libc.so.6 => ldd (0x723cd632b000)
> > 	libm.so.6 => ldd (0x723cd632b000)
> > 	libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x723cd563f000)
> > Error relocating /usr/lib/chromium/libwidevinecdm.so: __memcpy_chk: symbol not found
> > Error relocating /usr/lib/chromium/libwidevinecdm.so: __memset_chk: symbol not found
> > Error relocating /usr/lib/chromium/libwidevinecdm.so: __sprintf_chk: symbol not found
> > 
> > 
> > Adding those symbols for libc6 ABI compat would be a first step.  
> 
> This is on the roadmap, and there's a pending patch on the list, but
> it still needs some changes before it's ready to commit.
> 
> Are you aware of anything else blocking support for these libs?

I don't know.

> It might be worth doing a symbol listing and checking it against
> functions that might have glibc extensions we don't support or the
> minor ABI mismatches (non-*64 funcs for 32-bit, or regexec regoff_t
> for 64-bit).

I suppose strerror_r also belongs to the incompat function list.

Looks like there are a open64 and fopen64 atleast. But I suppose 64bit
should work.


ncopa-desktop:~$ nm -D /usr/lib/chromium/libwidevinecdm*.so

/usr/lib/chromium/libwidevinecdm.so:
000000000004e090 T CreateCdmInstance
000000000004e070 T DeinitializeCdmModule
000000000004e130 T GetCdmVersion
000000000004e060 T InitializeCdmModule_4
                 w _Jv_RegisterClasses
                 U _Unwind_Resume
                 U _ZNKSs4dataEv
                 U _ZNKSs4sizeEv
                 U _ZNKSs5c_strEv
                 U _ZNKSs5emptyEv
                 U _ZNKSs7_M_iendEv
                 U _ZNKSs7compareEPKc
                 U _ZNKSs7compareERKSs
                 U _ZNKSs8capacityEv
                 U _ZNKSs9_M_ibeginEv
                 U _ZNSaIcEC1Ev
                 U _ZNSaIcED1Ev
                 U _ZNSi4peekEv
                 U _ZNSi4readEPcl
                 U _ZNSi5seekgESt4fposI11__mbstate_tE
                 U _ZNSi5seekgElSt12_Ios_Seekdir
                 U _ZNSi5tellgEv
                 U _ZNSi7putbackEc
                 U _ZNSo5flushEv
                 U _ZNSo5writeEPKcl
                 U _ZNSs12_Alloc_hiderC1EPcRKSaIcE
                 U _ZNSs12_M_leak_hardEv
                 U _ZNSs12_S_empty_repEv
                 U _ZNSs14_M_replace_auxEmmmc
                 U _ZNSs4_Rep10_M_destroyERKSaIcE
                 U _ZNSs4_Rep10_M_refdataEv
                 U _ZNSs4_Rep11_S_terminalE
                 U _ZNSs4_Rep20_S_empty_rep_storageE
                 U _ZNSs4_Rep26_M_set_length_and_sharableEm
                 U _ZNSs4_Rep9_S_createEmmRKSaIcE
                 U _ZNSs5clearEv
                 U _ZNSs6appendEPKc
                 U _ZNSs6appendEPKcm
                 U _ZNSs6appendERKSs
                 U _ZNSs6appendEmc
                 U _ZNSs6assignEPKcm
                 U _ZNSs6assignERKSs
                 U _ZNSs6assignERKSsmm
                 U _ZNSs6insertEmRKSs
                 U _ZNSs6resizeEm
                 U _ZNSs6resizeEmc
                 U _ZNSs7replaceEmmPKcm
                 U _ZNSs7reserveEm
                 U _ZNSs9_M_mutateEmmm
                 U _ZNSs9push_backEc
                 U _ZNSsC1EOSs
                 U _ZNSsC1EPKcRKSaIcE
                 U _ZNSsC1EPKcmRKSaIcE
                 U _ZNSsC1ERKSs
                 U _ZNSsC1EmcRKSaIcE
                 U _ZNSsC1Ev
                 U _ZNSsD1Ev
                 U _ZNSsaSERKSs
                 U _ZNSsixEm
                 U _ZNSt13basic_filebufIcSt11char_traitsIcEE4openEPKcSt13_Ios_Openmode
                 U _ZNSt14basic_ifstreamIcSt11char_traitsIcEEC1Ev
                 U _ZNSt14basic_ofstreamIcSt11char_traitsIcEEC1Ev
                 U _ZNSt8__detail15_List_node_base11_M_transferEPS0_S1_
                 U _ZNSt8__detail15_List_node_base7_M_hookEPS0_
                 U _ZNSt8__detail15_List_node_base9_M_unhookEv
                 U _ZNSt8ios_base4InitC1Ev
                 U _ZNSt8ios_base4InitD1Ev
                 U _ZNSt9bad_allocD1Ev
                 U _ZNSt9basic_iosIcSt11char_traitsIcEE5clearESt12_Ios_Iostate
                 U _ZNSt9exceptionD2Ev
                 U _ZSt15set_new_handlerPFvvE
                 U _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l
                 U _ZSt17__throw_bad_allocv
                 U _ZSt18_Rb_tree_decrementPKSt18_Rb_tree_node_base
                 U _ZSt18_Rb_tree_decrementPSt18_Rb_tree_node_base
                 U _ZSt18_Rb_tree_incrementPKSt18_Rb_tree_node_base
                 U _ZSt18_Rb_tree_incrementPSt18_Rb_tree_node_base
                 U _ZSt18uncaught_exceptionv
                 U _ZSt19__throw_logic_errorPKc
                 U _ZSt20__throw_length_errorPKc
                 U _ZSt28_Rb_tree_rebalance_for_erasePSt18_Rb_tree_node_baseRS_
                 U _ZSt29_Rb_tree_insert_and_rebalancebPSt18_Rb_tree_node_baseS0_RS_
                 U _ZSt2wsIcSt11char_traitsIcEERSt13basic_istreamIT_T0_ES6_
                 U _ZSt9terminatev
                 U _ZTIPKc
                 U _ZTIPKh
                 U _ZTIPKi
                 U _ZTIPKw
                 U _ZTIPh
                 U _ZTISi
                 U _ZTISo
                 U _ZTISt9bad_alloc
                 U _ZTISt9exception
                 U _ZTIb
                 U _ZTIh
                 U _ZTIi
                 U _ZTIj
                 U _ZTVN10__cxxabiv116__enum_type_infoE
                 U _ZTVN10__cxxabiv117__class_type_infoE
                 U _ZTVN10__cxxabiv119__pointer_type_infoE
                 U _ZTVN10__cxxabiv120__si_class_type_infoE
                 U _ZTVN10__cxxabiv121__vmi_class_type_infoE
                 U _ZTVSt9bad_alloc
                 U _ZTVSt9exception
                 U _ZdaPv
                 U _ZdlPv
                 U _Znam
                 U _Znwm
                 U __ctype_toupper_loc
                 U __cxa_allocate_exception
                 U __cxa_atexit
                 U __cxa_begin_catch
                 U __cxa_call_unexpected
                 U __cxa_end_catch
                 w __cxa_finalize
                 U __cxa_free_exception
                 U __cxa_get_exception_ptr
                 U __cxa_guard_acquire
                 U __cxa_guard_release
                 U __cxa_pure_virtual
                 U __cxa_rethrow
                 U __cxa_throw
                 U __dynamic_cast
                 U __errno_location
                 w __gmon_start__
                 U __gxx_personality_v0
                 U __memcpy_chk
                 U __memset_chk
                 U __sched_cpucount
                 U __sprintf_chk
                 U __stack_chk_fail
                 U __strdup
                 U __strtok_r
0000000000365cec T _fini
0000000000030bb8 T _init
                 U _setjmp
                 U abort
                 U accept
                 U acos
                 U atan
                 U atoi
                 U bind
                 U calloc
                 U ceil
                 U clock
                 U close
                 U connect
                 U cos
                 U exit
                 U exp
                 U exp2
                 U fclose
                 U fflush
                 U fgets
                 U floor
                 U floorf
                 U fopen
                 U fopen64
                 U fprintf
                 U fputc
                 U fputs
                 U fread
                 U free
                 U fseek
                 U ftell
                 U fwrite
                 U getenv
                 U geteuid
                 U gethostbyname
                 U gethostname
                 U getpeername
                 U getservbyname
                 U getsockname
                 U gettimeofday
                 U inet_addr
                 U ioctl
                 U isalpha
                 U ldexp
                 U listen
                 U localtime
                 U log
                 U log10
                 U longjmp
                 U malloc
                 U memalign
                 U memchr
                 U memcmp
                 U memcpy
                 U memmove
                 U memset
                 U open64
                 U pow
                 U printf
                 U pthread_attr_destroy
                 U pthread_attr_init
                 U pthread_attr_setschedpolicy
                 w pthread_cancel
                 U pthread_cond_broadcast
                 U pthread_cond_destroy
                 U pthread_cond_init
                 U pthread_cond_signal
                 U pthread_cond_timedwait
                 U pthread_cond_wait
                 U pthread_create
                 U pthread_getschedparam
                 U pthread_getspecific
                 U pthread_join
                 U pthread_key_create
                 U pthread_key_delete
                 U pthread_mutex_destroy
                 U pthread_mutex_init
                 U pthread_mutex_lock
                 U pthread_mutex_trylock
                 U pthread_mutex_unlock
                 U pthread_mutexattr_destroy
                 U pthread_mutexattr_init
                 U pthread_mutexattr_settype
                 U pthread_once
                 U pthread_setschedparam
                 U pthread_setspecific
                 U qsort
                 U rand
                 U read
                 U readlink
                 U realloc
                 U recv
                 U rint
                 U rintf
                 U round
                 U sched_get_priority_max
                 U sched_get_priority_min
                 U sched_getaffinity
                 U sched_yield
                 U select
                 U sem_destroy
                 U sem_init
                 U sem_post
                 U sem_wait
                 U send
                 U shutdown
                 U signal
                 U sin
                 U sleep
                 U snprintf
                 U socket
                 U sprintf
                 U sqrt
                 U stderr
                 U stdout
                 U strcasecmp
                 U strcat
                 U strcmp
                 U strcpy
                 U strerror
                 U strftime
                 U strlen
                 U strncat
                 U strncmp
                 U strncpy
                 U strrchr
                 U strstr
                 U strtol
                 U sysconf
                 U time
                 U times
                 U tolower
                 U toupper
                 U uname
                 U usleep
                 U vsnprintf
                 U wcstombs

/usr/lib/chromium/libwidevinecdmadapter.so:
                 U CreateCdmInstance
                 U DeinitializeCdmModule
                 U InitializeCdmModule_4
000000000000be90 T PPP_GetInterface
000000000000be10 T PPP_InitializeModule
000000000000be70 T PPP_ShutdownModule
                 w _Jv_RegisterClasses
                 U _ZNKSs4findEcm
                 U _ZNSs12_M_leak_hardEv
                 U _ZNSs4_Rep10_M_destroyERKSaIcE
                 U _ZNSs4_Rep11_S_terminalE
                 U _ZNSs4_Rep20_S_empty_rep_storageE
                 U _ZNSs4swapERSs
                 U _ZNSs6appendEPKcm
                 U _ZNSs6appendERKSs
                 U _ZNSs6appendERKSsmm
                 U _ZNSs6appendEmc
                 U _ZNSs6assignEPKcm
                 U _ZNSs6assignERKSs
                 U _ZNSs6insertEmPKcm
                 U _ZNSs6resizeEmc
                 U _ZNSs7reserveEm
                 U _ZNSsC1EPKcRKSaIcE
                 U _ZNSsC1EPKcmRKSaIcE
                 U _ZNSsC1ERKSs
                 U _ZNSsC1ERKSsmm
0000000000005770 W _ZNSt6vectorIcSaIcEE13_M_assign_auxIPKhEEvT_S5_St20forward_iterator_tag
0000000000006260 W _ZNSt6vectorIcSaIcEE15_M_range_insertIN9__gnu_cxx17__normal_iteratorIPcS1_EEEEvS6_T_S7_St20forward_iterator_tag
0000000000005690 W _ZNSt6vectorIcSaIcEE17_M_default_appendEm
000000000000ee60 W _ZNSt6vectorIiSaIiEE13_M_insert_auxIJRKiEEEvN9__gnu_cxx17__normal_iteratorIPiS1_EEDpOT_
000000000000df90 W _ZNSt6vectorIiSaIiEE17_M_default_appendEm
000000000000cc70 W _ZNSt8_Rb_treeISsSt4pairIKSsPKvESt10_Select1stIS4_ESt4lessISsESaIS4_EE10_M_insert_IS4_EESt17_Rb_tree_iteratorIS4_EPKSt18_Rb_tree_node_baseSG_OT_
000000000000de70 W _ZNSt8_Rb_treeISsSt4pairIKSsPKvESt10_Select1stIS4_ESt4lessISsESaIS4_EE16_M_insert_uniqueIS4_EES0_ISt17_Rb_tree_iteratorIS4_EbEOT_
000000000000dc10 W _ZNSt8_Rb_treeISsSt4pairIKSsPKvESt10_Select1stIS4_ESt4lessISsESaIS4_EE17_M_insert_unique_IS4_EESt17_Rb_tree_iteratorIS4_ESt23_Rb_tree_const_iteratorIS4_EOT_
0000000000005b60 W _ZNSt8_Rb_treeISsSt4pairIKSsPKvESt10_Select1stIS4_ESt4lessISsESaIS4_EE4findERS1_
00000000000055d0 W _ZNSt8_Rb_treeISsSt4pairIKSsPKvESt10_Select1stIS4_ESt4lessISsESaIS4_EE8_M_eraseEPSt13_Rb_tree_nodeIS4_E
000000000000cc70 W _ZNSt8_Rb_treeISsSt4pairIKSsPvESt10_Select1stIS3_ESt4lessISsESaIS3_EE10_M_insert_IS3_EESt17_Rb_tree_iteratorIS3_EPKSt18_Rb_tree_node_baseSF_OT_
000000000000cd30 W _ZNSt8_Rb_treeISsSt4pairIKSsPvESt10_Select1stIS3_ESt4lessISsESaIS3_EE16_M_insert_uniqueIS3_EES0_ISt17_Rb_tree_iteratorIS3_EbEOT_
000000000000ca10 W _ZNSt8_Rb_treeISsSt4pairIKSsPvESt10_Select1stIS3_ESt4lessISsESaIS3_EE17_M_insert_unique_IS3_EESt17_Rb_tree_iteratorIS3_ESt23_Rb_tree_const_iteratorIS3_EOT_
0000000000005b60 W _ZNSt8_Rb_treeISsSt4pairIKSsPvESt10_Select1stIS3_ESt4lessISsESaIS3_EE4findERS1_
00000000000055d0 W _ZNSt8_Rb_treeISsSt4pairIKSsPvESt10_Select1stIS3_ESt4lessISsESaIS3_EE8_M_eraseEPSt13_Rb_tree_nodeIS3_E
0000000000005ed0 W _ZNSt8_Rb_treeISsSt4pairIKSsbESt10_Select1stIS2_ESt4lessISsESaIS2_EE10_M_insert_IS2_EESt17_Rb_tree_iteratorIS2_EPKSt18_Rb_tree_node_baseSE_OT_
0000000000005f90 W _ZNSt8_Rb_treeISsSt4pairIKSsbESt10_Select1stIS2_ESt4lessISsESaIS2_EE16_M_insert_uniqueIS2_EES0_ISt17_Rb_tree_iteratorIS2_EbEOT_
0000000000005c70 W _ZNSt8_Rb_treeISsSt4pairIKSsbESt10_Select1stIS2_ESt4lessISsESaIS2_EE17_M_insert_unique_IS2_EESt17_Rb_tree_iteratorIS2_ESt23_Rb_tree_const_iteratorIS2_EOT_
0000000000005b60 W _ZNSt8_Rb_treeISsSt4pairIKSsbESt10_Select1stIS2_ESt4lessISsESaIS2_EE4findERS1_
00000000000055d0 W _ZNSt8_Rb_treeISsSt4pairIKSsbESt10_Select1stIS2_ESt4lessISsESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E
                 U _ZNSt8ios_base4InitC1Ev
                 U _ZNSt8ios_base4InitD1Ev
                 U _ZSt17__throw_bad_allocv
                 U _ZSt18_Rb_tree_decrementPKSt18_Rb_tree_node_base
                 U _ZSt18_Rb_tree_decrementPSt18_Rb_tree_node_base
                 U _ZSt18_Rb_tree_incrementPKSt18_Rb_tree_node_base
                 U _ZSt20__throw_length_errorPKc
                 U _ZSt20__throw_out_of_rangePKc
                 U _ZSt28_Rb_tree_rebalance_for_erasePSt18_Rb_tree_node_baseRS_
                 U _ZSt29_Rb_tree_insert_and_rebalancebPSt18_Rb_tree_node_baseS0_RS_
                 U _ZSt4cout
                 U _ZdaPv
                 U _ZdlPv
                 U _Znam
                 U _Znwm
0000000000014038 A __bss_start
                 U __cxa_atexit
                 w __cxa_finalize
                 U __cxa_pure_virtual
                 w __gmon_start__
                 U __stack_chk_fail
0000000000014038 A _edata
00000000000141f9 A _end
0000000000010f78 T _fini
0000000000003860 T _init
                 U close
                 U memcmp
                 U memcpy
                 U memmove
                 U memset
                 U pthread_attr_init
                 U pthread_attr_setstacksize
                 w pthread_cancel
                 U pthread_create
                 U pthread_join
                 U pthread_mutex_destroy
                 U pthread_mutex_init
                 U pthread_mutex_lock
                 U pthread_mutex_unlock
                 U snprintf
                 U strcmp
                 U strlen
ncopa-desktop:~$ 

Re: Open/pending issues as of Jan 17

[Kurt H Maier]

On Fri, Jan 29, 2016 at 01:19:28PM -0500, Rich Felker wrote:
> On Fri, Jan 29, 2016 at 08:54:49AM +0100, Natanael Copa wrote:
> > 
> > An unrelated feature request: it would be nice to get netflix working
> > with the closed source libwidevinecdm.so and libwidevinecdmadapter.so 
> > 
> > ncopa-desktop:~$ ldd /usr/lib/chromium/libwidevinecdmadapter.so 
> > 	ldd (0x723cd632b000)
> > 	libwidevinecdm.so => /usr/lib/chromium/libwidevinecdm.so (0x723cd5ba2000)
> > 	libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x723cd5852000)
> > 	libpthread.so.0 => ldd (0x723cd632b000)
> > 	libc.so.6 => ldd (0x723cd632b000)
> > 	libm.so.6 => ldd (0x723cd632b000)
> > 	libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x723cd563f000)
> > Error relocating /usr/lib/chromium/libwidevinecdm.so: __memcpy_chk: symbol not found
> > Error relocating /usr/lib/chromium/libwidevinecdm.so: __memset_chk: symbol not found
> > Error relocating /usr/lib/chromium/libwidevinecdm.so: __sprintf_chk: symbol not found
> > 
> > 
> > Adding those symbols for libc6 ABI compat would be a first step.
> 
> This is on the roadmap, and there's a pending patch on the list, but
> it still needs some changes before it's ready to commit.
> 
> Are you aware of anything else blocking support for these libs? It
> might be worth doing a symbol listing and checking it against
> functions that might have glibc extensions we don't support or the
> minor ABI mismatches (non-*64 funcs for 32-bit, or regexec regoff_t
> for 64-bit).

I haven't been following the status of this closely -- if the mentioned
patch was committed, can someone point me to it?  Is it still worth
dumping a symbol table for this binary?

khm

Re: Open/pending issues as of Jan 17

[Bobby Bingham]

On Sun, Jan 17, 2016 at 06:15:54PM -0500, Rich Felker wrote:
> There were a few other patches that were mostly trivial which I just
> committed outright, and the big one, out-of-tree builds, that I
> finally committed after some more testing.
>

There's still the pthread_{try,timed}join_np patch outstanding.

Re: Open/pending issues as of Jan 17

[Khem Raj]

On Mon, Jan 18, 2016 at 5:14 PM, Bobby Bingham <koorogi@koorogi.info> wrote:
> On Sun, Jan 17, 2016 at 06:15:54PM -0500, Rich Felker wrote:
>> There were a few other patches that were mostly trivial which I just
>> committed outright, and the big one, out-of-tree builds, that I
>> finally committed after some more testing.
>>
>
> There's still the pthread_{try,timed}join_np patch outstanding.

+1

Re: Open/pending issues as of Jan 17

[Rich Felker]

On Sun, Jan 17, 2016 at 06:15:54PM -0500, Rich Felker wrote:
> With a quick review of the mailing list since late November I found
> the following possibly-open issues:
> 
> - resolver search domain support

Done.

> - preinit_array
> - mlock2 syscall wrapper

Left pending; I'll look at these after release.

> - mips siginfo_t fix

Done.

> - avl tree: any changes still pending??

No bugs just possible improvements; I'll revisit after release.

> - adding pot file for translators
> - unsetenv pointer eval after free
> - printf big widths/precisions
> - atomic.h overhaul

Done.

> - dirent.h macros

Unsure. Any opinions on this? Perhaps defer to after release.

> - dlopen deadlock
> - setvbuf

Revisit after release.

Rich