From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: libc-alpha@sourceware.org
Subject: [PATCH 02/11] nptl: Move createthread to pthread_create
Date: Wed, 26 May 2021 13:57:19 -0300 [thread overview]
Message-ID: <20210526165728.1772546-3-adhemerval.zanella@linaro.org> (raw)
In-Reply-To: <20210526165728.1772546-1-adhemerval.zanella@linaro.org>
The 'create_thread' function is moved to pthread_create.c. It removes
the START_THREAD_DEFN and START_THREAD_SELF macros and make the
lock usage more clear (no need to cross-reference multiple files).
No functional change.
---
nptl/createthread.c | 153 ------------------------------------------
nptl/pthread_create.c | 120 +++++++++++++++++++++++++++++++--
2 files changed, 113 insertions(+), 160 deletions(-)
delete mode 100644 nptl/createthread.c
diff --git a/nptl/createthread.c b/nptl/createthread.c
deleted file mode 100644
index bc3409b326..0000000000
--- a/nptl/createthread.c
+++ /dev/null
@@ -1,153 +0,0 @@
-/* Low-level thread creation for NPTL. Linux version.
- Copyright (C) 2002-2021 Free Software Foundation, Inc.
- This file is part of the GNU C Library.
- Contributed by Ulrich Drepper <drepper@redhat.com>, 2002.
-
- The GNU C Library is free software; you can redistribute it and/or
- modify it under the terms of the GNU Lesser General Public
- License as published by the Free Software Foundation; either
- version 2.1 of the License, or (at your option) any later version.
-
- The GNU C Library is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- Lesser General Public License for more details.
-
- You should have received a copy of the GNU Lesser General Public
- License along with the GNU C Library; if not, see
- <https://www.gnu.org/licenses/>. */
-
-#include <sched.h>
-#include <setjmp.h>
-#include <signal.h>
-#include <stdlib.h>
-#include <atomic.h>
-#include <ldsodefs.h>
-#include <tls.h>
-#include <stdint.h>
-
-#include <arch-fork.h>
-
-#ifdef __NR_clone2
-# define ARCH_CLONE __clone2
-#else
-# define ARCH_CLONE __clone
-#endif
-
-/* See the comments in pthread_create.c for the requirements for these
- two macros and the create_thread function. */
-
-#define START_THREAD_DEFN \
- static int __attribute__ ((noreturn)) start_thread (void *arg)
-#define START_THREAD_SELF arg
-
-/* pthread_create.c defines this using START_THREAD_DEFN
- We need a forward declaration here so we can take its address. */
-static int start_thread (void *arg) __attribute__ ((noreturn));
-
-static int
-create_thread (struct pthread *pd, const struct pthread_attr *attr,
- bool *stopped_start, STACK_VARIABLES_PARMS, bool *thread_ran)
-{
- /* Determine whether the newly created threads has to be started
- stopped since we have to set the scheduling parameters or set the
- affinity. */
- bool need_setaffinity = (attr != NULL && attr->extension != NULL
- && attr->extension->cpuset != 0);
- if (attr != NULL
- && (__glibc_unlikely (need_setaffinity)
- || __glibc_unlikely ((attr->flags & ATTR_FLAG_NOTINHERITSCHED) != 0)))
- *stopped_start = true;
-
- pd->stopped_start = *stopped_start;
- if (__glibc_unlikely (*stopped_start))
- /* See CONCURRENCY NOTES in nptl/pthread_creat.c. */
- lll_lock (pd->lock, LLL_PRIVATE);
-
- /* We rely heavily on various flags the CLONE function understands:
-
- CLONE_VM, CLONE_FS, CLONE_FILES
- These flags select semantics with shared address space and
- file descriptors according to what POSIX requires.
-
- CLONE_SIGHAND, CLONE_THREAD
- This flag selects the POSIX signal semantics and various
- other kinds of sharing (itimers, POSIX timers, etc.).
-
- CLONE_SETTLS
- The sixth parameter to CLONE determines the TLS area for the
- new thread.
-
- CLONE_PARENT_SETTID
- The kernels writes the thread ID of the newly created thread
- into the location pointed to by the fifth parameters to CLONE.
-
- Note that it would be semantically equivalent to use
- CLONE_CHILD_SETTID but it is be more expensive in the kernel.
-
- CLONE_CHILD_CLEARTID
- The kernels clears the thread ID of a thread that has called
- sys_exit() in the location pointed to by the seventh parameter
- to CLONE.
-
- The termination signal is chosen to be zero which means no signal
- is sent. */
- const int clone_flags = (CLONE_VM | CLONE_FS | CLONE_FILES | CLONE_SYSVSEM
- | CLONE_SIGHAND | CLONE_THREAD
- | CLONE_SETTLS | CLONE_PARENT_SETTID
- | CLONE_CHILD_CLEARTID
- | 0);
-
- TLS_DEFINE_INIT_TP (tp, pd);
-
- if (__glibc_unlikely (ARCH_CLONE (&start_thread, STACK_VARIABLES_ARGS,
- clone_flags, pd, &pd->tid, tp, &pd->tid)
- == -1))
- return errno;
-
- /* It's started now, so if we fail below, we'll have to cancel it
- and let it clean itself up. */
- *thread_ran = true;
-
- /* Now we have the possibility to set scheduling parameters etc. */
- if (attr != NULL)
- {
- int res;
-
- /* Set the affinity mask if necessary. */
- if (need_setaffinity)
- {
- assert (*stopped_start);
-
- res = INTERNAL_SYSCALL_CALL (sched_setaffinity, pd->tid,
- attr->extension->cpusetsize,
- attr->extension->cpuset);
-
- if (__glibc_unlikely (INTERNAL_SYSCALL_ERROR_P (res)))
- err_out:
- {
- /* The operation failed. We have to kill the thread.
- We let the normal cancellation mechanism do the work. */
-
- pid_t pid = __getpid ();
- INTERNAL_SYSCALL_CALL (tgkill, pid, pd->tid, SIGCANCEL);
-
- return INTERNAL_SYSCALL_ERRNO (res);
- }
- }
-
- /* Set the scheduling parameters. */
- if ((attr->flags & ATTR_FLAG_NOTINHERITSCHED) != 0)
- {
- assert (*stopped_start);
-
- res = INTERNAL_SYSCALL_CALL (sched_setscheduler, pd->tid,
- pd->schedpolicy, &pd->schedparam);
-
- if (__glibc_unlikely (INTERNAL_SYSCALL_ERROR_P (res)))
- goto err_out;
- }
- }
-
- return 0;
-}
diff --git a/nptl/pthread_create.c b/nptl/pthread_create.c
index 5680687efe..b53e3f30a0 100644
--- a/nptl/pthread_create.c
+++ b/nptl/pthread_create.c
@@ -215,9 +215,6 @@ late_init (void)
/* CREATE THREAD NOTES:
- createthread.c defines the create_thread function, and two macros:
- START_THREAD_DEFN and START_THREAD_SELF (see below).
-
create_thread must initialize PD->stopped_start. It should be true
if the STOPPED_START parameter is true, or if create_thread needs the
new thread to synchronize at startup for some other implementation
@@ -242,20 +239,129 @@ late_init (void)
so create_thread need not do that. On failure, *THREAD_RAN should
be set to true iff the thread actually started up and then got
canceled before calling user code (*PD->start_routine). */
+
+static int _Noreturn start_thread (void *arg);
+
static int create_thread (struct pthread *pd, const struct pthread_attr *attr,
bool *stopped_start, STACK_VARIABLES_PARMS,
- bool *thread_ran);
+ bool *thread_ran)
+{
+ /* Determine whether the newly created threads has to be started
+ stopped since we have to set the scheduling parameters or set the
+ affinity. */
+ bool need_setaffinity = (attr != NULL && attr->extension != NULL
+ && attr->extension->cpuset != 0);
+ if (attr != NULL
+ && (__glibc_unlikely (need_setaffinity)
+ || __glibc_unlikely ((attr->flags & ATTR_FLAG_NOTINHERITSCHED) != 0)))
+ *stopped_start = true;
+
+ pd->stopped_start = *stopped_start;
+ if (__glibc_unlikely (*stopped_start))
+ lll_lock (pd->lock, LLL_PRIVATE);
+
+ /* We rely heavily on various flags the CLONE function understands:
+
+ CLONE_VM, CLONE_FS, CLONE_FILES
+ These flags select semantics with shared address space and
+ file descriptors according to what POSIX requires.
+
+ CLONE_SIGHAND, CLONE_THREAD
+ This flag selects the POSIX signal semantics and various
+ other kinds of sharing (itimers, POSIX timers, etc.).
+
+ CLONE_SETTLS
+ The sixth parameter to CLONE determines the TLS area for the
+ new thread.
+
+ CLONE_PARENT_SETTID
+ The kernels writes the thread ID of the newly created thread
+ into the location pointed to by the fifth parameters to CLONE.
+
+ Note that it would be semantically equivalent to use
+ CLONE_CHILD_SETTID but it is be more expensive in the kernel.
+
+ CLONE_CHILD_CLEARTID
+ The kernels clears the thread ID of a thread that has called
+ sys_exit() in the location pointed to by the seventh parameter
+ to CLONE.
+
+ The termination signal is chosen to be zero which means no signal
+ is sent. */
+ const int clone_flags = (CLONE_VM | CLONE_FS | CLONE_FILES | CLONE_SYSVSEM
+ | CLONE_SIGHAND | CLONE_THREAD
+ | CLONE_SETTLS | CLONE_PARENT_SETTID
+ | CLONE_CHILD_CLEARTID
+ | 0);
+
+ TLS_DEFINE_INIT_TP (tp, pd);
+
+#ifdef __NR_clone2
+# define ARCH_CLONE __clone2
+#else
+# define ARCH_CLONE __clone
+#endif
+ if (__glibc_unlikely (ARCH_CLONE (&start_thread, STACK_VARIABLES_ARGS,
+ clone_flags, pd, &pd->tid, tp, &pd->tid)
+ == -1))
+ return errno;
+
+ /* It's started now, so if we fail below, we'll have to cancel it
+ and let it clean itself up. */
+ *thread_ran = true;
-#include <createthread.c>
+ /* Now we have the possibility to set scheduling parameters etc. */
+ if (attr != NULL)
+ {
+ int res;
+
+ /* Set the affinity mask if necessary. */
+ if (need_setaffinity)
+ {
+ assert (*stopped_start);
+
+ res = INTERNAL_SYSCALL_CALL (sched_setaffinity, pd->tid,
+ attr->extension->cpusetsize,
+ attr->extension->cpuset);
+
+ if (__glibc_unlikely (INTERNAL_SYSCALL_ERROR_P (res)))
+ err_out:
+ {
+ /* The operation failed. We have to kill the thread.
+ We let the normal cancellation mechanism do the work. */
+
+ pid_t pid = __getpid ();
+ INTERNAL_SYSCALL_CALL (tgkill, pid, pd->tid, SIGCANCEL);
+
+ return INTERNAL_SYSCALL_ERRNO (res);
+ }
+ }
+
+ /* Set the scheduling parameters. */
+ if ((attr->flags & ATTR_FLAG_NOTINHERITSCHED) != 0)
+ {
+ assert (*stopped_start);
+
+ res = INTERNAL_SYSCALL_CALL (sched_setscheduler, pd->tid,
+ pd->schedpolicy, &pd->schedparam);
+
+ if (__glibc_unlikely (INTERNAL_SYSCALL_ERROR_P (res)))
+ goto err_out;
+ }
+ }
+
+ return 0;
+}
/* Local function to start thread and handle cleanup.
createthread.c defines the macro START_THREAD_DEFN to the
declaration that its create_thread function will refer to, and
START_THREAD_SELF to the expression to optimally deliver the new
thread's THREAD_SELF value. */
-START_THREAD_DEFN
+static int _Noreturn
+start_thread (void *arg)
{
- struct pthread *pd = START_THREAD_SELF;
+ struct pthread *pd = arg;
/* Initialize resolver state pointer. */
__resp = &pd->res;
--
2.30.2
next prev parent reply other threads:[~2021-05-26 16:57 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-26 16:57 [PATCH 00/11] nptl: pthread cancellation refactor Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 01/11] nptl: Move Linux createthread to nptl Adhemerval Zanella
2021-05-26 17:14 ` Florian Weimer
2021-05-26 16:57 ` Adhemerval Zanella [this message]
2021-05-26 17:16 ` [PATCH 02/11] nptl: Move createthread to pthread_create Florian Weimer
2021-05-26 17:29 ` Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 03/11] support: Add xpthread_attr_setaffinity_np wrapper Adhemerval Zanella
2021-05-26 17:17 ` Florian Weimer
2021-05-27 22:35 ` Joseph Myers
2021-05-28 1:08 ` Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 04/11] nptl: Add pthread_attr_setaffinity_np failure test Adhemerval Zanella
2021-05-26 17:21 ` Florian Weimer
2021-05-26 17:30 ` Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 05/11] nptl: Deallocate the thread stack on setup failure (BZ #19511) Adhemerval Zanella
2021-05-26 17:33 ` Florian Weimer
2021-05-26 17:51 ` Adhemerval Zanella
2021-05-26 17:58 ` Florian Weimer
2021-05-26 19:19 ` Adhemerval Zanella
2021-05-26 18:21 ` Andreas Schwab
2021-05-26 18:40 ` Adhemerval Zanella
2021-05-26 19:26 ` Adhemerval Zanella
2021-05-27 7:43 ` Florian Weimer
2021-05-26 16:57 ` [PATCH 06/11] nptl: Install cancellation handler on pthread_cancel Adhemerval Zanella
2021-05-26 17:38 ` Florian Weimer
2021-05-26 17:52 ` Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 07/11] nptl: Remove CANCELING_BITMASK Adhemerval Zanella
2021-05-26 18:02 ` Florian Weimer
2021-06-15 22:07 ` Florian Weimer
2021-06-15 23:33 ` Adhemerval Zanella
2021-06-16 12:46 ` Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 08/11] nptl: Move cancel state out of cancelhandling Adhemerval Zanella
2021-05-26 18:20 ` Florian Weimer
2021-05-27 16:40 ` Adhemerval Zanella
2021-05-27 16:48 ` Florian Weimer
2021-05-27 16:57 ` Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 09/11] nptl: Move cancel type " Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 10/11] nptl: Implement raise in terms of pthread_kill Adhemerval Zanella
2021-05-26 16:57 ` [PATCH 11/11] nptl: Use pthread_kill on pthread_cancel Adhemerval Zanella
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210526165728.1772546-3-adhemerval.zanella@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=libc-alpha@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).