linux-kernel - Re: [PATCH RFCv3 10/23] uprobes/x86: Add support to emulate nop5 instruction

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Z_WFZT3rZtjts3u-@krava>
Date: Tue, 8 Apr 2025 22:21:57 +0200
From: Jiri Olsa <olsajiri@...il.com>
To: Jiri Olsa <olsajiri@...il.com>
Cc: Andrii Nakryiko <andrii.nakryiko@...il.com>,
	Oleg Nesterov <oleg@...hat.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Andrii Nakryiko <andrii@...nel.org>, bpf@...r.kernel.org,
	linux-kernel@...r.kernel.org, linux-trace-kernel@...r.kernel.org,
	x86@...nel.org, Song Liu <songliubraving@...com>,
	Yonghong Song <yhs@...com>,
	John Fastabend <john.fastabend@...il.com>,
	Hao Luo <haoluo@...gle.com>, Steven Rostedt <rostedt@...dmis.org>,
	Masami Hiramatsu <mhiramat@...nel.org>,
	Alan Maguire <alan.maguire@...cle.com>,
	David Laight <David.Laight@...lab.com>,
	Thomas Weißschuh <thomas@...ch.de>
Subject: Re: [PATCH RFCv3 10/23] uprobes/x86: Add support to emulate nop5
 instruction

On Mon, Apr 07, 2025 at 01:07:26PM +0200, Jiri Olsa wrote:
> On Fri, Apr 04, 2025 at 01:33:11PM -0700, Andrii Nakryiko wrote:
> > On Thu, Mar 20, 2025 at 4:43 AM Jiri Olsa <jolsa@...nel.org> wrote:
> > >
> > > Adding support to emulate nop5 as the original uprobe instruction.
> > >
> > > Signed-off-by: Jiri Olsa <jolsa@...nel.org>
> > > ---
> > >  arch/x86/kernel/uprobes.c | 16 ++++++++++++++++
> > >  1 file changed, 16 insertions(+)
> > >
> > 
> > This optimization is independent from the sys_uprobe, right? Maybe
> > send it as a stand-alone patch and let's land it sooner?
> 
> ok, will send it separately
> 
> > Also, how hard would it be to do the same for other nopX instructions?
> 
> will check, might be easy

we can't do all at the moment, nop1-nop8 are fine, but uprobe won't
attach on nop9/10/11 due unsupported prefix.. I guess insn decode
would need to be updated first

I'll send the nop5 emulation change, because of above and also I don't
see practical justification to emulate other nops

jirka


---
diff --git a/arch/x86/kernel/uprobes.c b/arch/x86/kernel/uprobes.c
index 9194695662b2..6616cc9866cc 100644
--- a/arch/x86/kernel/uprobes.c
+++ b/arch/x86/kernel/uprobes.c
@@ -608,6 +608,21 @@ static void riprel_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
 		*sr = utask->autask.saved_scratch_register;
 	}
 }
+
+static bool emulate_nop_insn(struct arch_uprobe *auprobe)
+{
+	unsigned int i;
+
+	/*
+	 * Uprobe is only allowed to be attached on nop1 through nop8. Further nop
+	 * instructions have unsupported prefix and uprobe fails to attach on them.
+	 */
+	for (i = 1; i < 9; i++) {
+		if (!memcmp(&auprobe->insn, x86_nops[i], i))
+			return true;
+	}
+	return false;
+}
 #else /* 32-bit: */
 /*
  * No RIP-relative addressing on 32-bit
@@ -621,6 +636,10 @@ static void riprel_pre_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
 static void riprel_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
 {
 }
+static bool emulate_nop_insn(struct arch_uprobe *auprobe)
+{
+	return false;
+}
 #endif /* CONFIG_X86_64 */
 
 struct uprobe_xol_ops {
@@ -840,6 +859,9 @@ static int branch_setup_xol_ops(struct arch_uprobe *auprobe, struct insn *insn)
 	insn_byte_t p;
 	int i;
 
+	if (emulate_nop_insn(auprobe))
+		goto setup;
+
 	switch (opc1) {
 	case 0xeb:	/* jmp 8 */
 	case 0xe9:	/* jmp 32 */