All of lore.kernel.org
 help / color / mirror / Atom feed
From: Xiaofeng Yuan <yuanxiaofeng@eswincomputing.com>
To: Paul Walmsley <pjw@kernel.org>,
	Palmer Dabbelt <palmer@dabbelt.com>,
	Albert Ou <aou@eecs.berkeley.edu>,
	Alexandre Ghiti <alex@ghiti.fr>
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
	linux-trace-kernel@vger.kernel.org, stable@vger.kernel.org
Subject: [PATCH v3] riscv: kprobes: free insn slot in arch_remove_kprobe() to avoid memory leak
Date: Fri, 20 Nov 2026 15:08:49 +0800	[thread overview]
Message-ID: <20261120070849.1239-1-yuanxiaofeng@eswincomputing.com> (raw)

When a kprobe is removed, arch_remove_kprobe() does not free the
instruction slot allocated by arch_prepare_kprobe(). This causes
a memory leak every time a kprobe is unregistered.

Other architectures such as arm64, x86, and powerpc all free the
slot in their arch_remove_kprobe() implementations. Add the
missing free_insn_slot() call for RISC-V.

Registered and unregistered a kprobe on kernel_clone in a loop
inside a QEMU (RISC-V virt) guest: before the fix the number of
kprobe_insn_page entries in /proc/kallsyms grew by one every 512
iterations (each insn-slot page holds 512 eight-byte slots), and
after the fix it stays at 1.

Fixes: c22b0bcb1dd0 ("riscv: Add kprobes supported")
Cc: stable@vger.kernel.org
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Xiaofeng Yuan <yuanxiaofeng@eswincomputing.com>
---
Changes in v3:
 - Add Fixes tag and Cc stable (Masami Hiramatsu).
 - Add Reviewed-by from Masami Hiramatsu. No code change.

 arch/riscv/kernel/probes/kprobes.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/riscv/kernel/probes/kprobes.c b/arch/riscv/kernel/probes/kprobes.c
index 9e2afabf9..a76eab297 100644
--- a/arch/riscv/kernel/probes/kprobes.c
+++ b/arch/riscv/kernel/probes/kprobes.c
@@ -128,6 +128,10 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
+	if (p->ainsn.api.insn) {
+		free_insn_slot(p->ainsn.api.insn, 0);
+		p->ainsn.api.insn = NULL;
+	}
 }
 
 static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
-- 
2.34.1


WARNING: multiple messages have this Message-ID (diff)
From: Xiaofeng Yuan <yuanxiaofeng@eswincomputing.com>
To: Paul Walmsley <pjw@kernel.org>,
	Palmer Dabbelt <palmer@dabbelt.com>,
	Albert Ou <aou@eecs.berkeley.edu>,
	Alexandre Ghiti <alex@ghiti.fr>
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
	linux-trace-kernel@vger.kernel.org, stable@vger.kernel.org
Subject: [PATCH v3] riscv: kprobes: free insn slot in arch_remove_kprobe() to avoid memory leak
Date: Fri, 20 Nov 2026 15:08:49 +0800	[thread overview]
Message-ID: <20261120070849.1239-1-yuanxiaofeng@eswincomputing.com> (raw)

When a kprobe is removed, arch_remove_kprobe() does not free the
instruction slot allocated by arch_prepare_kprobe(). This causes
a memory leak every time a kprobe is unregistered.

Other architectures such as arm64, x86, and powerpc all free the
slot in their arch_remove_kprobe() implementations. Add the
missing free_insn_slot() call for RISC-V.

Registered and unregistered a kprobe on kernel_clone in a loop
inside a QEMU (RISC-V virt) guest: before the fix the number of
kprobe_insn_page entries in /proc/kallsyms grew by one every 512
iterations (each insn-slot page holds 512 eight-byte slots), and
after the fix it stays at 1.

Fixes: c22b0bcb1dd0 ("riscv: Add kprobes supported")
Cc: stable@vger.kernel.org
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Xiaofeng Yuan <yuanxiaofeng@eswincomputing.com>
---
Changes in v3:
 - Add Fixes tag and Cc stable (Masami Hiramatsu).
 - Add Reviewed-by from Masami Hiramatsu. No code change.

 arch/riscv/kernel/probes/kprobes.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/riscv/kernel/probes/kprobes.c b/arch/riscv/kernel/probes/kprobes.c
index 9e2afabf9..a76eab297 100644
--- a/arch/riscv/kernel/probes/kprobes.c
+++ b/arch/riscv/kernel/probes/kprobes.c
@@ -128,6 +128,10 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
+	if (p->ainsn.api.insn) {
+		free_insn_slot(p->ainsn.api.insn, 0);
+		p->ainsn.api.insn = NULL;
+	}
 }
 
 static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
-- 
2.34.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

             reply	other threads:[~2026-11-20  7:09 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-20  7:08 Xiaofeng Yuan [this message]
2026-11-20  7:08 ` [PATCH v3] riscv: kprobes: free insn slot in arch_remove_kprobe() to avoid memory leak Xiaofeng Yuan

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=20261120070849.1239-1-yuanxiaofeng@eswincomputing.com \
    --to=yuanxiaofeng@eswincomputing.com \
    --cc=alex@ghiti.fr \
    --cc=aou@eecs.berkeley.edu \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mhiramat@kernel.org \
    --cc=palmer@dabbelt.com \
    --cc=pjw@kernel.org \
    --cc=stable@vger.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.