Re: [PATCH] kprobes: retry pending optprobe after freeing blocker
From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Date: 2025-10-29 04:02:51
Also in:
lkml
On Mon, 27 Oct 2025 21:05:35 +0800 hongao [off-list ref] wrote:
The freeing_list cleanup now retries optimizing any sibling probe that was deferred while this aggregator was being torn down. Track the pending address in struct optimized_kprobe so __disarm_kprobe() can defer the retry until kprobe_optimizer() finishes disarming.
Thanks Hongao! I have some comment below. Please make it simpler.
quoted hunk ↗ jump to hunk
Signed-off-by: hongao <redacted>diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 8c4f3bb24..33d65b008 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h@@ -338,6 +338,7 @@ DEFINE_INSN_CACHE_OPS(insn); struct optimized_kprobe { struct kprobe kp; struct list_head list; /* list for optimizing queue */ + kprobe_opcode_t *pending_reopt_addr; /* addr that should trigger re-optimization */
You may need "bool reopt_unblocked_probes".
quoted hunk ↗ jump to hunk
struct arch_optimized_insn optinsn; };diff --git a/kernel/kprobes.c b/kernel/kprobes.c index da59c68df..0976ab57d 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c@@ -514,6 +514,7 @@ static LIST_HEAD(freeing_list); static void kprobe_optimizer(struct work_struct *work); static DECLARE_DELAYED_WORK(optimizing_work, kprobe_optimizer); +static void optimize_kprobe(struct kprobe *p); #define OPTIMIZE_DELAY 5 /*@@ -591,6 +592,20 @@ static void do_free_cleaned_kprobes(void) */ continue; } + if (op->pending_reopt_addr) { + struct kprobe *blocked;
unblocked?
+ + /* + * The aggregator was holding back another probe while it sat on the + * unoptimizing/freeing lists. Now that the aggregator has been fully + * reverted we can safely retry the optimization of that sibling. + */ + + blocked = get_optimized_kprobe(op->pending_reopt_addr);
You can use op->kp.addr.
quoted hunk ↗ jump to hunk
+ if (unlikely(blocked)) + optimize_kprobe(blocked); + } + free_aggr_kprobe(&op->kp); } }@@ -1009,13 +1024,13 @@ static void __disarm_kprobe(struct kprobe *p, bool reopt) _p = get_optimized_kprobe(p->addr); if (unlikely(_p) && reopt) optimize_kprobe(_p); + } else if (reopt && kprobe_aggrprobe(p)) {
Here, the @p is queued. This means @p is an optprobe (and aggrprobe).
+ struct optimized_kprobe *op = + container_of(p, struct optimized_kprobe, kp); + + /* Defer the re-optimization until the worker finishes disarming. */ + op->pending_reopt_addr = p->addr;
You should save the @reopt flag instead of addr. Thank you,
} - /* - * TODO: Since unoptimization and real disarming will be done by - * the worker thread, we can not check whether another probe are - * unoptimized because of this probe here. It should be re-optimized - * by the worker thread. - */ } #else /* !CONFIG_OPTPROBES */ -- 2.47.2
-- Masami Hiramatsu (Google) [off-list ref]