David Michael [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Lookup the startup server through /servers/startup
* sysdeps/mach/hurd/reboot.c: Include <hurd/paths.h>
(reboot): Lookup _SERVERS_STARTUP instead of calling proc_getmsgport to get a
port to the startup server.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name git-reboot-startup.diff
Samuel Thibault [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Make F_RDLCK/F_WRLCK atomic
lockf(LOCK_EX) would for instance drop any existing shared lock before taking
the exclusive lock. F_RDLCK/F_WRLCK need atomic changes, so introduce and use
__LOCK_ATOM
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
* misc/sys/file.h (__LOCK_ATOMIC): New macro.
* sysdeps/mach/hurd/fcntl.c (__libc_fcntl): Use __LOCK_ATOMIC along LOCK_SH and
LOCK_EX.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-WRLCK-upgrade.diff
2014-08-27 Samuel Thibault <samuel.thibault@ens-lyon.org>
Fix bind when umask is e.g. 0777.
* sysdeps/mach/hurd/bind.c (__bind): Pass mode 0666 to __dir_mkfile
instead of final mode, so that call __ifsock_getsockaddr can always
succeed, before calling __file_chmod to fix the mode according to umask,
before calling __dir_link to show the file.
Part of the original fix was committed, the other hasn't been yet, see Roland's
"Harumph" reply to
https://sourceware.org/ml/libc-alpha/2014-08/msg00408.html
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name submitted-bind_umask2.diff
Richard Braun [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Hurd: fix port leak in TLS
Depending on whether the thread is the main thread or not, the threading
library can have trouble determining whether the thread reference in the
TCB is valid. The simple solution is to let the threading library
initialize the TCB, and use a temporary reference when initializing TLS.
* sysdeps/mach/hurd/i386/tls.h (_hurd_tls_init): Use a temporary thread
reference.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name git-tls_thread_leak.diff
Richard Braun [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Hurd: make sigstates hold a reference on thread ports
This change is required in order to correctly release per-thread
resources. Directly reusing the threading library reference isn't
possible since the sigstate is also used early in the main thread,
before threading is initialized.
* hurd/hurd/signal.h (_hurd_self_sigstate): Drop thread reference after
calling _hurd_thread_sigstate.
(_hurd_critical_section_lock): Likewise.
* hurd/hurdsig.c (_hurd_thread_sigstate): Add a reference on the thread.
(_hurd_sigstate_delete): Drop thread reference.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-sigstate_thread_reference.diff
Thomas Schwinge [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
pipe2
2008-12-17 Thomas Schwinge <tschwinge@gnu.org>
pipe2 for GNU Hurd.
* sysdeps/mach/hurd/pipe2.c: New file, copy from pipe.c. Evolve it to
implement __pipe2.
* sysdeps/mach/hurd/pipe.c (__pipe): Reimplement using __pipe2.
* sysdeps/mach/hurd/kernel-features.h (__ASSUME_PIPE2): Define.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-pipe2.diff
Thomas Schwinge [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
socket_flags
2008-12-17 Thomas Schwinge <tschwinge@gnu.org>
SOCK_CLOEXEC and SOCK_NONBLOCK for socket on GNU Hurd.
* sysdeps/mach/hurd/socket.c (__socket): Handle SOCK_CLOEXEC and
SOCK_NONBLOCK.
* sysdeps/mach/hurd/kernel-features.h (__ASSUME_SOCK_CLOEXEC): Define.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-socket_flags.diff
To: libc-alpha@sourceware.org
Subject: [PATCH 2/2] Define _PATH_MOUNTED as "/etc/mtab"
Date: Thu, 15 Aug 2013 10:43:35 +0200
Change the definition of _PATH_MOUNTED to "/etc/mtab". This is the
value used on Linux.
The change is motivated by the fact that on Debian /etc/mtab is a
symbolic link to /proc/mounts. This patch adjusts the macro for
non-linux systems such as Hurd. Changing this using
sysdeps/mach/hurd/paths.h causes build problems because
/usr/include/hurd/paths.h is shadowed by this file. This change is
proposed in the hope that aligning the non-linux targets with the
glibc for Linux is perceived as a good thing while fixing this problem
on Debian/Hurd along the way.
* sysdeps/generic/paths.h (_PATH_MOUNTED): Change value to "/etc/mtab".
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name submitted-path_mounted.diff
Samuel Thibault [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Fix poll and select POSIX compliancy details about errors
This fixes the following:
- On error, poll must not return without polling, including EBADF, and instead
report POLLHUP/POLLERR/POLLNVAL
- Select must report EBADF if some set contains an invalid FD.
The idea is to move error management to after all select calls, in the
poll/select final treatment. The error is instead recorded in a new `error'
field, and a new SELECT_ERROR bit set.
Thanks Svante Signell for the initial version of the patch.
* hurd/hurdselect.c (SELECT_ERROR): New macro.
(_hurd_select):
- Add `error' field to `d' structures array.
- If a poll descriptor is bogus, set EBADF, but continue with a zero timeout.
- Go through the whole fd_set, not only until _hurd_dtablesize. Return EBADF
there is any bit set above _hurd_dtablesize.
- Do not request io_select on bogus descriptors (SELECT_ERROR).
- On io_select request error, record the error.
- On io_select bogus reply, use EIO error code.
- On io_select bogus or error reply, record the error.
- Do not destroy reply port for bogus FDs.
- On error, make poll set POLLHUP in the EPIPE case, POLLNVAL in the EBADF
case, or else POLLERR.
- On error, make select simulated readiness.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-poll_errors_fixes.diff
Richard Braun [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Fix timeout handling in _hurd_select
Rely on servers to implement timeouts, so that very short values (including
0) don't make mach_msg return before valid replies can be received. The
purpose of this scheme is to guarantee a full client-server round-trip,
whatever the timeout value.
This change depends on the new io_select_timeout RPC being implemented by
servers.
* hurd/Makefile (user-interfaces): Add io_reply and io_request.
* hurd/hurdselect.c: Include <sys/time.h>, <hurd/io_request.h> and <limits.h>.
(_hurd_select): Replace the call to __io_select with either __io_select_request
or __io_select_timeout_request, depending on the timeout. Count the number of
ready descriptors (replies for which at least one type bit is set). Implement
the timeout locally when there is no file descriptor.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-io_select_timeout.diff
Samuel Thibault [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Allow unknown root directory.
To be efficient, the remap translator simply returns ports from the underlying
filesystem, and thus the root directory found through browsing '..' is the
underlying root, not the remap root. This should not be a reason for getcwd to
fail.
* sysdeps/mach/hurd/getcwd.c (_hurd_canonicalize_directory_name_internal): Do
not remove the heading slash if we got an unknown root directory.
(__getcwd): Do not fail with EGRATUITOUS if we got an unknown root directory.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-remap_getcwd.diff
Samuel Thibault [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Workaround to add IPv6 support to getifaddrs
ifreq only contains sockaddr structures, which are not big enough for
IPv6 addresses. This takes another, ugly, approach, by parsing fsysopts
/servers/socket/2 options...
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-ifaddrs_v6.diff
Samuel Thibault [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
Fix sbrk in PIE binaries
In PIE mode, the program gets loaded at very low address, and thus _end is very
low, just before libraries, and thus initializing the brk to it does not make
sense, since there is no room left there. Hardcode the brk to 0x8000000 for now
as a workaround.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-pie-sbrk.diff
The preemptor sigcode doesn't match since the POSIX sigcode SI_TIMER is
used when SIGALRM is sent. In addition, The inline version of
hurd_preempt_signals doesn't update _hurdsig_preempted_set. For these
reasons, the preemptor would be skipped by post_signal.
The function attempts to optimize this case by performing one IPC system call
with the timeout included among the parameters, but in the absence of a reply,
it will call mach_msg again with the same timeout later, effectively doubling
the total timeout of the select/poll call.
Remove this optimization for the time being.
* hurd/hurdselect.c (_hurd_select): Always call __io_select with no timeout.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-single-select-timeout.diff
This dates when libpthread was moved to glibc and some features were added.
2.13-31/38/39 dates when the global signal disposition call was added.
TODO: _DEBIAN_ in versions however pose problem. Remove the _DEBIAN_ version
once packages are rebuilt against 2.21.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name libpthread_version.diff
2.13-19 dates when the global signal disposition call was added.
TODO: _DEBIAN_ in versions however pose problem. Remove the _DEBIAN_ version
once packages are rebuilt against 2.21.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name local-hurdsig-global-dispositions-version.diff
Although they should not change the
default behaviors of signals for cthread programs, these patches add
new functions which can be used by libpthread to enable
POSIX-conforming behavior of signals on a per-thread basis.
YYYY-MM-DD Jeremie Koenig <jk@jk.fr.eu.org>
e407ae3 Hurd signals: implement global signal dispositions 38eb4b3 Hurd signals: provide a sigstate destructor 344dfd6 Hurd signals: fix sigwait() for global signals fb055f2 Hurd signals: fix global untraced signals.
YYYY-MM-DD Thomas Schwinge <thomas@codesourcery.com>
* sysdeps/mach/hurd/fork.c (__fork): In the child, reinitialize
the global sigstate's lock.
This is work in progress.
This cures an issue that would very rarely cause a deadlock in the child
in fork, tries to unlock ss' critical section lock at the end of fork.
This will typically (always?) be observed in /bin/sh, which is not
surprising as that is the foremost caller of fork.
To reproduce an intermediate state, add an endless loop if
_hurd_global_sigstate is locked after __proc_dostop (cast through
volatile); that is, while still being in the fork's parent process.
When that triggers (use the libtool testsuite), the signal thread has
already locked ss (which is _hurd_global_sigstate), and is stuck at
hurdsig.c:685 in post_signal, trying to lock _hurd_siglock (which the
main thread already has locked and keeps locked until after
__task_create). This is the case that ss->thread == MACH_PORT_NULL, that
is, a global signal. In the main thread, between __proc_dostop and
__task_create is the __thread_abort call on the signal thread which would
abort any current kernel operation (but leave ss locked). Later in fork,
in the parent, when _hurd_siglock is unlocked in fork, the parent's
signal thread can proceed and will unlock eventually the global sigstate.
In the client, _hurd_siglock will likewise be unlocked, but the global
sigstate never will be, as the client's signal thread has been configured
to restart execution from _hurd_msgport_receive. Thus, when the child
tries to unlock ss' critical section lock at the end of fork, it will
first lock the global sigstate, will spin trying to lock it, which can
never be successful, and we get our deadlock.
Options seem to be:
* Move the locking of _hurd_siglock earlier in post_signal -- but that
may generally impact performance, if this locking isn't generally
needed anyway?
On the other hand, would it actually make sense to wait here until we
are not any longer in a critical section (which is meant to disable
signal delivery anyway (but not for preempted signals?))?
* Clear the global sigstate in the fork's child with the rationale that
we're anyway restarting the signal thread from a clean state. This
has now been implemented.
Why has this problem not been observed before Jérémie's patches? (Or has
it? Perhaps even more rarely?) In _S_msg_sig_post, the signal is now
posted to a *global receiver thread*, whereas previously it was posted to
the *designated signal-receiving thread*. The latter one was in a
critical section in fork, so didn't try to handle the signal until after
leaving the critical section? (Not completely analyzed and verified.)
Another question is what the signal is that is being received
during/around the time __proc_dostop executes.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-hurdsig-global-dispositions.diff
These patches should not change the current
behavior, although they do fix a few minor bugs which were made
apparent in the process. They are unchanged from my previous post
earlier this month.
34f5960 _hurd_internal_post_signal: Split into more functions 420eec9 _hurd_internal_post_signal: Scope variables more restrictively 1f5accd _hurd_internal_post_signal: Split out inner functions 1bb0a1d Hurd signals: refactor check_pending_signals 1764465 Hurd signals: reindent 28473d2 Hurd signals: make sigsuspend POSIX-conformant. 26d091a Hurd signals: fix uninitialized value.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-hurdsig-fixes.diff
Thomas Schwinge [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
t/grantpt
/media/Stalin/tmp/glibc.hurd.gnu-2/libc_pic.os: In function `pts_name':
/media/data/home/thomas/tmp/source/glibc/git/glibc.hurd/login/../sysdeps/unix/grantpt.c:52: undefined reference to `__ptsname_internal'
collect2: ld returned 1 exit status
make[1]: *** [/media/Stalin/tmp/glibc.hurd.gnu-2/libc.so] Error 1
Svante Signell <svante.signell@gmail.com>
Samuel Thibault <samuel.thibault@ens-lyon.org>
* sysdeps/mach/hurd/sendmsg.c (__libc_sendmsg): On SCM_CREDS
control messages, record uids, pass a rendez-vous port in the
control message, and call __auth_user_authenticate_request to
make auth send credentials on that port. Do not wait for a
reply.
* sysdeps/mach/hurd/recvmsg.c (contains_uid, contains_gid,
check_auth): New functions.
(__libc_recvmsg): On SCM_CREDS control messages, call check_auth
to check the passed credentials thanks to the answer from the
auth server.
* hurd/Makefile (user-interfaces): Add auth_request and
auth_reply.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-sendmsg-SCM_CREDS.diff
struct ortentry and struct ifalias req are actually not defined
struct arpreq is defined, but can not be passed to an ioctl on the Hurd.
so let's make packages not believe these are available.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name local-disable-ioctls.diff
Samuel Thibault [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
The critical section lock _can_ be held in these place.
At least since hurd_thread_cancel can be called by another thread and lock our
critical lock.
http://bugs.debian.org/46859
“
Thomas suggested that there is no need to take the critical section
lock. I believe that taking the critical section lock is necessary to
prevent the target thread from entering a signal handler. Roland will
look into the problem.
”
Taking the critical section lock makes these assertions bogus.
It happens that hurd_thread_cancel is only called from libports and inside
/hurd/term so this is rare in practice.
2006-08-05 Samuel Thibault <samuel.thibault@ens-lyon.org>
* hurd/thread-cancel.c (hurd_thread_cancel): Do not assert that
`&ss->critical_section_lock' is unlocked.
* sysdeps/mach/hurd/jmp-unwind.c (_longjmp_unwind): Likewise, and take
critical section lock before taking the sigstate lock.
* sysdeps/mach/hurd/spawni.c (__spawni): Likewise.
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-thread-cancel.diff
Thomas Schwinge [Mon, 16 Jul 2018 20:23:13 +0000 (21:23 +0100)]
getcontext, makecontext, setcontext, swapcontext
Adapted from the Linux x86 functions.
Not thoroughly tested, but manual testing as well as glibc tests look fine, and
manual -lpthread testing also looks fine (within the given bounds for a new
stack to be used with makecontext).
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name tg-context_functions.diff
# DP: Description: Enable ldconfig and such on hurd-i386
# DP: Author: Jeff Bailey <jbailey@nisa.net>
# DP: Related Bugs: #309489
# DP: Upstream status: Not submitted
# DP: Status Details: Upstream disagrees with this patch, but I'm
# DP: putting it in so that we have expected Debian behaviour on the
# DP: Hurd. We should review this when the ELF standard supports runpath.
# DP: Date: 08 Apr 2003
Gbp-Pq: Topic hurd-i386
Gbp-Pq: Name local-enable-ldconfig.diff
Increase the maximal overall growth of the compilation unit caused
by inlining for dl-reloc.c on hppa. This remove some warnings and
strangely it reduces the size of the final binary.
Gbp-Pq: Topic hppa
Gbp-Pq: Name local-inlining.diff
For backward compatibility with armhf binaries built with the
old linker SONAME, we need to fake out the linker to believe
the new is the old, until such a point as everything is rebuilt
Gbp-Pq: Topic arm
Gbp-Pq: Name local-soname-hack.diff
The alpha assembly version of various string functions do not work
correctly on some corner cases, and thus doesn't pass the testsuite.
This patch removes them.
Gbp-Pq: Topic alpha
Gbp-Pq: Name local-string-functions.diff
older versions of glibc would build dl-sysdep as shared-only and dl-support as
static-only. alpha hooks in a cache variable via dl-auxv.h. newer versions of
glibc build dl-sysdep as both shared and static which means we now have symbol
duplication for static builds with dl-sysdep and dl-support. since dl-sysdep
is both shared/static, there is no point in hooking dl-support anymore, so we
can punt it.
Gbp-Pq: Topic alpha
Gbp-Pq: Name submitted-dl-support.diff
# All lines beginning with `# DP:' are a description of the patch.
# DP: Description: Rewrite collation rules to include iso14651_t1
# DP: es_ES: BZ664, by Pablo Saratxaga
# DP: da_DK fi_FI nb_NO: BZ672, by Denis Barbier
# DP: ar_SA cs_CZ et_EE hr_HR lt_LT pl_PL sl_SI tr_TR: not
# DP: submitted yet.
# DP: Related bugs: BZ664 BZ672
# DP: Dpatch author: Denis Barbier
# DP: Patch author: Denis Barbier, Pablo Saratxaga
# DP: Upstream status: BZ664 BZ672
# DP: Date: 2006-01-18
2008-08-08 Aurelien Jarno <aurel32@debian.org>
Loosy update for glibc 2.8.
2007-05-25 Pierre Habouzit <madcoder@debian.org>
Loosy update for glibc 2.6.
2005-01-16 Denis Barbier <barbier@linuxfr.org>
[BZ #672]
Replace current collation rules by including iso14651_t1 and adding
extra rules if needed. There should be no noticeable changes in
sorted text. only ligatures and ignoreable characters have modified
weights.
* locales/da_DK: Likewise.
Gbp-Pq: Topic localedata
Gbp-Pq: Name tailor-iso14651_t1.diff
# All lines beginning with `# DP:' are a description of the patch.
# DP: Description: Improve en_DK and add en_DK.ISO-8859-15 to SUPPORTED
# DP: This locale is only useful to provide ISO8601 date formats.
# DP: Related bugs: #323159
# DP: Dpatch author: Denis Barbier
# DP: Patch author: Jakob Bohm <jbj@image.dk>
# DP: Upstream status: not submitted
# DP: Date: 2006-01-16
Gbp-Pq: Topic localedata
Gbp-Pq: Name locale-en_DK.diff
# All lines beginning with `# DP:' are a description of the patch.
# DP: Description: Various fixes for all French locales
# DP: Related bugs: #248377 #351786 #345481
# DP: Dpatch author: Denis Barbier <barbier@debian.org>
# DP: Patch author: Denis Barbier
# DP: Upstream status: not submitted
# DP: Date: 2006-01-10
Gbp-Pq: Topic localedata
Gbp-Pq: Name locales-fr.diff
# All lines beginning with `# DP:' are a description of the patch.
# DP: Description: Fix d_t_fmt and date_fmt in fo_FO
# DP: Related bugs: #307194
# DP: Dpatch author: Denis Barbier
# DP: Patch author: Jacob Sparre Andersen <sparre@nbi.dk>
# DP: Upstream status: not submitted
# DP: Date: 2006-01-10
Gbp-Pq: Topic localedata
Gbp-Pq: Name fo_FO-date_fmt.diff
# DP: Description: Please add Kurdish locale
# DP: Related bugs: BZ870
# DP: Dpatch author: Denis Barbier <barbier@debian.org>
# DP: Patch author: Kader DILSIZ, Pablo Saratxaga
# DP: Upstream status: Version 0.1 submitted in the original bugreport
# DP: has been committed, this new version has been sent just after.
# DP: Date: 2006-01-06
Gbp-Pq: Topic localedata
Gbp-Pq: Name locale-ku_TR.diff
# All lines beginning with `# DP:' are a description of the patch.
# DP: Description: Add several locales to localedata.SUPPORTED
# DP: - ru_RU.CP1251 was requested in #225516
# DP: - uz_UZ.UTF-8 was clearly missing, uz_UZ had no UTF-8 variant
# DP: - da_DK.ISO-8859-15, en_GB.ISO-8859-15, en_US.ISO-8859-15, no_NO.UTF-8,
# DP: no_NO and sv_SE.ISO-8859-15 are found in fedora-branch CVS branch.
# DP: Related bugs: #225516 (ru_RU.CP1251)
# DP: Dpatch author: Denis Barbier
# DP: Upstream status: not submitted
# DP: Date: 2006-01-10
Gbp-Pq: Topic localedata
Gbp-Pq: Name supported.diff
projects / glibc.git / log