From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Michael Jeanson <mjeanson@efficios.com>,
Jens Axboe <axboe@kernel.dk>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Peter Zijlstra <peterz@infradead.org>,
"Paul E. McKenney" <paulmck@kernel.org>,
x86@kernel.org, Sean Christopherson <seanjc@google.com>,
Wei Liu <wei.liu@kernel.org>
Subject: [patch V5 14/31] rseq: Cache CPU ID and MM CID values
Date: Wed, 22 Oct 2025 14:52:18 +0200 (CEST) [thread overview]
Message-ID: <20251022121942.277188265@linutronix.de> (raw)
In-Reply-To: 20251022121836.019469732@linutronix.de
In preparation for rewriting RSEQ exit to user space handling provide
storage to cache the CPU ID and MM CID values which were written to user
space. That prepares for a quick check, which avoids the update when
nothing changed.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
---
include/linux/rseq.h | 7 +++++--
include/linux/rseq_types.h | 21 +++++++++++++++++++++
include/trace/events/rseq.h | 4 ++--
kernel/rseq.c | 4 ++++
4 files changed, 32 insertions(+), 4 deletions(-)
--- a/include/linux/rseq.h
+++ b/include/linux/rseq.h
@@ -57,6 +57,7 @@ static inline void rseq_virt_userspace_e
static inline void rseq_reset(struct task_struct *t)
{
memset(&t->rseq, 0, sizeof(t->rseq));
+ t->rseq.ids.cpu_cid = ~0ULL;
}
static inline void rseq_execve(struct task_struct *t)
@@ -70,10 +71,12 @@ static inline void rseq_execve(struct ta
*/
static inline void rseq_fork(struct task_struct *t, u64 clone_flags)
{
- if (clone_flags & CLONE_VM)
+ if (clone_flags & CLONE_VM) {
rseq_reset(t);
- else
+ } else {
t->rseq = current->rseq;
+ t->rseq.ids.cpu_cid = ~0ULL;
+ }
}
#else /* CONFIG_RSEQ */
--- a/include/linux/rseq_types.h
+++ b/include/linux/rseq_types.h
@@ -31,17 +31,38 @@ struct rseq_event {
};
/**
+ * struct rseq_ids - Cache for ids, which need to be updated
+ * @cpu_cid: Compound of @cpu_id and @mm_cid to make the
+ * compiler emit a single compare on 64-bit
+ * @cpu_id: The CPU ID which was written last to user space
+ * @mm_cid: The MM CID which was written last to user space
+ *
+ * @cpu_id and @mm_cid are updated when the data is written to user space.
+ */
+struct rseq_ids {
+ union {
+ u64 cpu_cid;
+ struct {
+ u32 cpu_id;
+ u32 mm_cid;
+ };
+ };
+};
+
+/**
* struct rseq_data - Storage for all rseq related data
* @usrptr: Pointer to the registered user space RSEQ memory
* @len: Length of the RSEQ region
* @sig: Signature of critial section abort IPs
* @event: Storage for event management
+ * @ids: Storage for cached CPU ID and MM CID
*/
struct rseq_data {
struct rseq __user *usrptr;
u32 len;
u32 sig;
struct rseq_event event;
+ struct rseq_ids ids;
};
#else /* CONFIG_RSEQ */
--- a/include/trace/events/rseq.h
+++ b/include/trace/events/rseq.h
@@ -21,9 +21,9 @@ TRACE_EVENT(rseq_update,
),
TP_fast_assign(
- __entry->cpu_id = raw_smp_processor_id();
+ __entry->cpu_id = t->rseq.ids.cpu_id;
__entry->node_id = cpu_to_node(__entry->cpu_id);
- __entry->mm_cid = task_mm_cid(t);
+ __entry->mm_cid = t->rseq.ids.mm_cid;
),
TP_printk("cpu_id=%d node_id=%d mm_cid=%d", __entry->cpu_id,
--- a/kernel/rseq.c
+++ b/kernel/rseq.c
@@ -184,6 +184,10 @@ static int rseq_update_cpu_node_id(struc
rseq_unsafe_put_user(t, node_id, node_id, efault_end);
rseq_unsafe_put_user(t, mm_cid, mm_cid, efault_end);
+ /* Cache the user space values */
+ t->rseq.ids.cpu_id = cpu_id;
+ t->rseq.ids.mm_cid = mm_cid;
+
/*
* Additional feature fields added after ORIG_RSEQ_SIZE
* need to be conditionally updated only if
next prev parent reply other threads:[~2025-10-22 12:52 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-22 12:52 [patch V5 00/31] rseq: Optimize exit to user space Thomas Gleixner
2025-10-22 12:52 ` [patch V5 01/31] rseq: Avoid pointless evaluation in __rseq_notify_resume() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 02/31] rseq: Condense the inline stubs Thomas Gleixner
2025-10-22 12:52 ` [patch V5 03/31] rseq: Move algorithm comment to top Thomas Gleixner
2025-10-22 12:52 ` [patch V5 04/31] rseq: Remove the ksig argument from rseq_handle_notify_resume() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 05/31] rseq: Simplify registration Thomas Gleixner
2025-10-22 12:52 ` [patch V5 06/31] rseq: Simplify the event notification Thomas Gleixner
2025-10-22 12:52 ` [patch V5 07/31] rseq, virt: Retrigger RSEQ after vcpu_run() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 08/31] rseq: Avoid CPU/MM CID updates when no event pending Thomas Gleixner
2025-10-22 12:52 ` [patch V5 09/31] rseq: Introduce struct rseq_data Thomas Gleixner
2025-10-22 12:52 ` [patch V5 10/31] entry: Cleanup header Thomas Gleixner
2025-10-22 12:52 ` [patch V5 11/31] entry: Remove syscall_enter_from_user_mode_prepare() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 12/31] entry: Inline irqentry_enter/exit_from/to_user_mode() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 13/31] sched: Move MM CID related functions to sched.h Thomas Gleixner
2025-10-22 12:52 ` Thomas Gleixner [this message]
2025-10-22 12:52 ` [patch V5 15/31] rseq: Record interrupt from user space Thomas Gleixner
2025-10-22 12:52 ` [patch V5 16/31] rseq: Provide tracepoint wrappers for inline code Thomas Gleixner
2025-10-22 12:52 ` [patch V5 17/31] rseq: Expose lightweight statistics in debugfs Thomas Gleixner
2025-10-22 12:52 ` [patch V5 18/31] rseq: Provide static branch for runtime debugging Thomas Gleixner
2025-10-22 12:52 ` [patch V5 19/31] rseq: Provide and use rseq_update_user_cs() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 20/31] rseq: Replace the original debug implementation Thomas Gleixner
2025-10-22 12:52 ` [patch V5 21/31] rseq: Make exit debugging static branch based Thomas Gleixner
2025-10-22 12:52 ` [patch V5 22/31] rseq: Use static branch for syscall exit debug when GENERIC_IRQ_ENTRY=y Thomas Gleixner
2025-10-22 12:52 ` [patch V5 23/31] rseq: Provide and use rseq_set_ids() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 24/31] rseq: Separate the signal delivery path Thomas Gleixner
2025-10-22 12:52 ` [patch V5 25/31] rseq: Rework the TIF_NOTIFY handler Thomas Gleixner
2025-10-22 13:31 ` Sean Christopherson
2025-10-22 12:52 ` [patch V5 26/31] rseq: Optimize event setting Thomas Gleixner
2025-10-22 12:52 ` [patch V5 27/31] rseq: Implement fast path for exit to user Thomas Gleixner
2025-10-23 8:51 ` Peter Zijlstra
2025-10-22 12:52 ` [patch V5 28/31] rseq: Switch to fast path processing on " Thomas Gleixner
2025-10-23 9:04 ` Peter Zijlstra
2025-10-22 12:52 ` [patch V5 29/31] entry: Split up exit_to_user_mode_prepare() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 30/31] rseq: Split up rseq_exit_to_user_mode() Thomas Gleixner
2025-10-22 12:52 ` [patch V5 31/31] rseq: Switch to TIF_RSEQ if supported Thomas Gleixner
2025-10-23 9:22 ` [patch V5 00/31] rseq: Optimize exit to user space Peter Zijlstra
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=20251022121942.277188265@linutronix.de \
--to=tglx@linutronix.de \
--cc=axboe@kernel.dk \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mjeanson@efficios.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=seanjc@google.com \
--cc=wei.liu@kernel.org \
--cc=x86@kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.