From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>,
"Paul E . McKenney" <paulmck@kernel.org>,
Boqun Feng <boqun.feng@gmail.com>,
"H . Peter Anvin" <hpa@zytor.com>, Paul Turner <pjt@google.com>,
linux-api@vger.kernel.org, Christian Brauner <brauner@kernel.org>,
Florian Weimer <fw@deneb.enyo.de>,
David.Laight@ACULAB.COM, carlos@redhat.com,
Peter Oskolkov <posk@posk.io>,
Alexander Mikhalitsyn <alexander@mihalicyn.com>,
Chris Kennelly <ckennelly@google.com>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Subject: [PATCH 25/30] rseq: Extend struct rseq with per-memory-map NUMA-aware Concurrency ID
Date: Tue, 22 Nov 2022 15:39:27 -0500 [thread overview]
Message-ID: <20221122203932.231377-26-mathieu.desnoyers@efficios.com> (raw)
In-Reply-To: <20221122203932.231377-1-mathieu.desnoyers@efficios.com>
Expose a per-memory-map NUMA-aware concurrency ID to userspace. Each
concurrency ID stays associated with the same NUMA node except in case
of NUMA topology reconfiguration.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
---
include/uapi/linux/rseq.h | 9 +++++++++
kernel/rseq.c | 10 +++++++++-
2 files changed, 18 insertions(+), 1 deletion(-)
diff --git a/include/uapi/linux/rseq.h b/include/uapi/linux/rseq.h
index c233aae5eac9..5779249ed37f 100644
--- a/include/uapi/linux/rseq.h
+++ b/include/uapi/linux/rseq.h
@@ -148,6 +148,15 @@ struct rseq {
*/
__u32 mm_cid;
+ /*
+ * Restartable sequences mm_numa_cid field. Updated by the kernel.
+ * Read by user-space with single-copy atomicity semantics. This field
+ * should only be read by the thread which registered this data
+ * structure. Aligned on 32-bit. Contains the current thread's
+ * NUMA-aware concurrency ID (allocated uniquely within a memory map).
+ */
+ __u32 mm_numa_cid;
+
/*
* Flexible array member at end of structure, after last feature field.
*/
diff --git a/kernel/rseq.c b/kernel/rseq.c
index cb2512ab3256..58b09de0de47 100644
--- a/kernel/rseq.c
+++ b/kernel/rseq.c
@@ -91,14 +91,17 @@ static int rseq_update_cpu_node_id(struct task_struct *t)
u32 cpu_id = raw_smp_processor_id();
u32 node_id = cpu_to_node(cpu_id);
u32 mm_cid = task_mm_cid(t);
+ u32 mm_numa_cid = task_mm_numa_cid(t);
WARN_ON_ONCE((int) mm_cid < 0);
+ WARN_ON_ONCE((int) mm_numa_cid < 0);
if (!user_write_access_begin(rseq, t->rseq_len))
goto efault;
unsafe_put_user(cpu_id, &rseq->cpu_id_start, efault_end);
unsafe_put_user(cpu_id, &rseq->cpu_id, efault_end);
unsafe_put_user(node_id, &rseq->node_id, efault_end);
unsafe_put_user(mm_cid, &rseq->mm_cid, efault_end);
+ unsafe_put_user(mm_numa_cid, &rseq->mm_numa_cid, efault_end);
/*
* Additional feature fields added after ORIG_RSEQ_SIZE
* need to be conditionally updated only if
@@ -117,7 +120,7 @@ static int rseq_update_cpu_node_id(struct task_struct *t)
static int rseq_reset_rseq_cpu_node_id(struct task_struct *t)
{
u32 cpu_id_start = 0, cpu_id = RSEQ_CPU_ID_UNINITIALIZED, node_id = 0,
- mm_cid = 0;
+ mm_cid = 0, mm_numa_cid = 0;
/*
* Reset cpu_id_start to its initial state (0).
@@ -141,6 +144,11 @@ static int rseq_reset_rseq_cpu_node_id(struct task_struct *t)
*/
if (put_user(mm_cid, &t->rseq->mm_cid))
return -EFAULT;
+ /*
+ * Reset mm_numa_cid to its initial state (0).
+ */
+ if (put_user(mm_numa_cid, &t->rseq->mm_numa_cid))
+ return -EFAULT;
/*
* Additional feature fields added after ORIG_RSEQ_SIZE
* need to be conditionally reset only if
--
2.25.1
next prev parent reply other threads:[~2022-11-22 20:42 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-22 20:39 [PATCH 00/30] RSEQ node id and mm concurrency id extensions Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 01/30] selftests/rseq: Fix: Fail thread registration when CONFIG_RSEQ=n Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 02/30] rseq: Introduce feature size and alignment ELF auxiliary vector entries Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2023-01-04 18:44 ` [PATCH 02/30] " Nathan Chancellor
2023-01-04 19:00 ` Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 03/30] rseq: Introduce extensible rseq ABI Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 04/30] rseq: Extend struct rseq with numa node id Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 05/30] selftests/rseq: Use ELF auxiliary vector for extensible rseq Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2023-01-04 19:14 ` [PATCH 05/30] " Florian Weimer
2023-01-04 19:51 ` Mathieu Desnoyers
2023-01-05 16:19 ` Florian Weimer
2023-01-05 16:28 ` Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 06/30] selftests/rseq: Implement rseq numa node id field selftest Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 07/30] sched: Introduce per-memory-map concurrency ID Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 08/30] rseq: Extend struct rseq with " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 09/30] selftests/rseq: Remove RSEQ_SKIP_FASTPATH code Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 10/30] selftests/rseq: Implement rseq mm_cid field support Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 11/30] selftests/rseq: x86: Template memory ordering and percpu access mode Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 12/30] selftests/rseq: arm: " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 13/30] selftests/rseq: arm64: " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 14/30] selftests/rseq: mips: " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 15/30] selftests/rseq: ppc: " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 16/30] selftests/rseq: s390: " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 17/30] selftests/rseq: riscv: " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 18/30] selftests/rseq: Implement basic percpu ops mm_cid test Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 19/30] selftests/rseq: Implement parametrized " Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 20/30] selftests/rseq: parametrized test: Report/abort on negative concurrency ID Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 21/30] tracing/rseq: Add mm_cid field to rseq_update Mathieu Desnoyers
2022-12-27 12:13 ` [tip: sched/core] " tip-bot2 for Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 22/30] lib: Implement find_{first,next,nth}_notandnot_bit, find_first_andnot_bit Mathieu Desnoyers
2023-11-21 17:06 ` Yury Norov
2022-11-22 20:39 ` [PATCH 23/30] cpumask: Implement cpumask_{first,next}_{not,}andnot Mathieu Desnoyers
2023-11-21 17:13 ` Yury Norov
2022-11-22 20:39 ` [PATCH 24/30] sched: NUMA-aware per-memory-map concurrency ID Mathieu Desnoyers
2023-11-21 17:43 ` Yury Norov
2022-11-22 20:39 ` Mathieu Desnoyers [this message]
2022-11-22 20:39 ` [PATCH 26/30] selftests/rseq: x86: Implement rseq_load_u32_u32 Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 27/30] selftests/rseq: Implement mm_numa_cid accessors in headers Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 28/30] selftests/rseq: Implement numa node id vs mm_numa_cid invariant test Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 29/30] selftests/rseq: Implement mm_numa_cid tests Mathieu Desnoyers
2022-11-22 20:39 ` [PATCH 30/30] tracing/rseq: Add mm_numa_cid field to rseq_update Mathieu Desnoyers
2024-02-28 18:50 ` [PATCH 00/30] RSEQ node id and mm concurrency id extensions Marco Elver
2024-02-28 20:01 ` Mathieu Desnoyers
2024-02-29 9:31 ` Marco Elver
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=20221122203932.231377-26-mathieu.desnoyers@efficios.com \
--to=mathieu.desnoyers@efficios.com \
--cc=David.Laight@ACULAB.COM \
--cc=alexander@mihalicyn.com \
--cc=boqun.feng@gmail.com \
--cc=brauner@kernel.org \
--cc=carlos@redhat.com \
--cc=ckennelly@google.com \
--cc=fw@deneb.enyo.de \
--cc=hpa@zytor.com \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=posk@posk.io \
--cc=tglx@linutronix.de \
/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
all inboxes | Powered by JetHome®