From: Peter Oskolkov <posk@google.com>
To: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
"Paul E . McKenney" <paulmck@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Boqun Feng <boqun.feng@gmail.com>,
linux-kernel@vger.kernel.org
Cc: Paul Turner <pjt@google.com>,
Chris Kennelly <ckennelly@google.com>,
Peter Oskolkov <posk@posk.io>, Peter Oskolkov <posk@google.com>
Subject: [PATCH 1/2 v2] rseq/membarrier: add MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU
Date: Thu, 6 Aug 2020 10:05:43 -0700 [thread overview]
Message-ID: <20200806170544.382140-1-posk@google.com> (raw)
This patchset is based on Google-internal RSEQ
work done by Paul Turner and Andrew Hunter.
When working with per-CPU RSEQ-based memory allocations,
it is sometimes important to make sure that a global
memory location is no longer accessed from RSEQ critical
sections. For example, there can be two per-CPU lists,
one is "active" and accessed per-CPU, while another one
is inactive and worked on asynchronously "off CPU" (e.g.
garbage collection is performed). Then at some point
the two lists are swapped, and a fast RCU-like mechanism
is required to make sure that the previously active
list is no longer accessed.
This patch introduces such a mechanism: in short,
membarrier() syscall issues an IPI to a CPU, restarting
a potentially active RSEQ critical section on the CPU.
v1->v2:
- removed the ability to IPI all CPUs in a single sycall;
- use task->mm rather than task->group_leader to identify
tasks belonging to the same process.
The second patch in the patchset adds a selftest
of this feature.
Signed-off-by: Peter Oskolkov <posk@google.com>
---
include/uapi/linux/membarrier.h | 5 ++++
kernel/sched/membarrier.c | 43 +++++++++++++++++++++++++++++++--
2 files changed, 46 insertions(+), 2 deletions(-)
diff --git a/include/uapi/linux/membarrier.h b/include/uapi/linux/membarrier.h
index 5891d7614c8c..ce4628ea17fa 100644
--- a/include/uapi/linux/membarrier.h
+++ b/include/uapi/linux/membarrier.h
@@ -114,6 +114,10 @@
* If this command is not implemented by an
* architecture, -EINVAL is returned.
* Returns 0 on success.
+ * @MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU:
+ * If a thread belonging to the current process
+ * is currently in an RSEQ critical section on the
+ * CPU identified by flags parameter, restart it.
* @MEMBARRIER_CMD_SHARED:
* Alias to MEMBARRIER_CMD_GLOBAL. Provided for
* header backward compatibility.
@@ -131,6 +135,7 @@ enum membarrier_cmd {
MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED = (1 << 4),
MEMBARRIER_CMD_PRIVATE_EXPEDITED_SYNC_CORE = (1 << 5),
MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_SYNC_CORE = (1 << 6),
+ MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU = (1 << 7),
/* Alias for header backward compatibility. */
MEMBARRIER_CMD_SHARED = MEMBARRIER_CMD_GLOBAL,
diff --git a/kernel/sched/membarrier.c b/kernel/sched/membarrier.c
index 168479a7d61b..1cdfde23696c 100644
--- a/kernel/sched/membarrier.c
+++ b/kernel/sched/membarrier.c
@@ -18,11 +18,19 @@
#define MEMBARRIER_PRIVATE_EXPEDITED_SYNC_CORE_BITMASK 0
#endif
+#ifdef CONFIG_RSEQ
+#define MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU_BITMASK \
+ MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU
+#else
+#define MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU_BITMASK 0
+#endif
+
#define MEMBARRIER_CMD_BITMASK \
(MEMBARRIER_CMD_GLOBAL | MEMBARRIER_CMD_GLOBAL_EXPEDITED \
| MEMBARRIER_CMD_REGISTER_GLOBAL_EXPEDITED \
| MEMBARRIER_CMD_PRIVATE_EXPEDITED \
| MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED \
+ | MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU_BITMASK \
| MEMBARRIER_PRIVATE_EXPEDITED_SYNC_CORE_BITMASK)
static void ipi_mb(void *info)
@@ -308,10 +316,39 @@ static int membarrier_register_private_expedited(int flags)
return 0;
}
+#ifdef CONFIG_RSEQ
+static void membarrier_rseq_ipi(void *arg)
+{
+ if (current->mm != arg) /* Not our process. */
+ return;
+ if (!current->rseq) /* RSEQ not set up for the current task/thread. */
+ return;
+
+ rseq_preempt(current);
+}
+#endif
+
+static int membarrier_private_restart_rseq_on_cpu(int cpu_id)
+{
+#ifdef CONFIG_RSEQ
+ /* syscalls are not allowed inside rseq critical sections. */
+ if (cpu_id == raw_smp_processor_id())
+ return 0;
+
+ return smp_call_function_single(cpu_id, membarrier_rseq_ipi,
+ current->mm, true);
+#else
+ return 0;
+#endif
+}
+
/**
* sys_membarrier - issue memory barriers on a set of threads
* @cmd: Takes command values defined in enum membarrier_cmd.
- * @flags: Currently needs to be 0. For future extensions.
+ * @flags: Currently needs to be 0 for all commands other than
+ * MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU: in the latter
+ * case it indicates the CPU on which to interrupt (= restart)
+ * the RSEQ critical section.
*
* If this system call is not implemented, -ENOSYS is returned. If the
* command specified does not exist, not available on the running
@@ -339,7 +376,7 @@ static int membarrier_register_private_expedited(int flags)
*/
SYSCALL_DEFINE2(membarrier, int, cmd, int, flags)
{
- if (unlikely(flags))
+ if (unlikely(flags && cmd != MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU))
return -EINVAL;
switch (cmd) {
case MEMBARRIER_CMD_QUERY:
@@ -369,6 +406,8 @@ SYSCALL_DEFINE2(membarrier, int, cmd, int, flags)
return membarrier_private_expedited(MEMBARRIER_FLAG_SYNC_CORE);
case MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_SYNC_CORE:
return membarrier_register_private_expedited(MEMBARRIER_FLAG_SYNC_CORE);
+ case MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU:
+ return membarrier_private_restart_rseq_on_cpu(flags);
default:
return -EINVAL;
}
--
2.28.0.163.g6104cc2f0b6-goog
next reply other threads:[~2020-08-06 17:08 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-06 17:05 Peter Oskolkov [this message]
2020-08-06 17:05 ` [PATCH 2/2 v2] rseq/selftests: test MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU Peter Oskolkov
2020-08-07 0:27 ` Boqun Feng
2020-08-07 12:54 ` Mathieu Desnoyers
2020-08-07 17:55 ` Peter Oskolkov
2020-08-07 18:25 ` Mathieu Desnoyers
2020-08-07 18:47 ` Peter Oskolkov
2020-08-07 20:29 ` Mathieu Desnoyers
2020-08-07 13:37 ` [PATCH 1/2 v2] rseq/membarrier: add MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU peterz
2020-08-07 17:50 ` Peter Oskolkov
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=20200806170544.382140-1-posk@google.com \
--to=posk@google.com \
--cc=boqun.feng@gmail.com \
--cc=ckennelly@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=posk@posk.io \
/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 NNTP newsgroup(s).