Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 17 Jun 2022 16:52:06 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Stephane Eranian <eranian@google.com>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Eric Dumazet <edumazet@google.com>, jpoimboe@redhat.com,
        Steven Rostedt <rostedt@goodmis.org>
Subject: Re: [RFC] pr_warn_once() issue in x86 MSR extable code
Message-ID: <YqyVFsbviKjVGGZ9@worktop.programming.kicks-ass.net>
References: <CABPqkBRVF9qmxKFgmjZpzN3tx=U+_8udECMLHs7BrtzfPwmuhQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CABPqkBRVF9qmxKFgmjZpzN3tx=U+_8udECMLHs7BrtzfPwmuhQ@mail.gmail.com>
Precedence: bulk

On Fri, Jun 17, 2022 at 02:08:52PM +0300, Stephane Eranian wrote:
> Hi,
> 
> Some changes to the way invalid MSR accesses are reported by the kernel is
> causing some problems with messages printed on the console.
> 
> We have seen several cases of ex_handler_msr() printing invalid MSR
> accesses once but
> the callstack multiple times causing confusion on the console.
> 
> The last time the exception MSR code was modified (5.16) by PeterZ was:
> 
>   d52a7344bdfa x86/msr: Remove .fixup usage:
> 
>   if (!safe && wrmsr &&  pr_warn_once("unchecked MSR access error: ..."))
>                show_stack_regs(regs);
> 
> Note that this code pattern was also present, though in a different
> form, before this commit.
> 
> The problem here is that another earlier commit (5.13):
> 
> a358f40600b3 once: implement DO_ONCE_LITE for non-fast-path "do once"
> functionality
> 
> Modifies all the pr_*_once() calls to always return true claiming that
> no caller is ever
> checking the return value of the functions.
> 
> This is why we are seeing the callstack printed without the associated
> printk() msg.
> 
> I believe that having the pr_*_once() functions return true the first
> time they are called
> is useful especially when extra information, such as callstack, must
> be printed to help
> track the origin of the problem.
> 
> The exception handling code seems to be the only place where the
> return value is checked
> for pr_warn_once(). A minimal change would be to create another
> version of that function
> that calls DO_ONCE() instead of DO_ONCE_LITE(), e.g., pr_warn_once_return().
> 
> I can post a patch to that effect if we all agree on the approach.
> 
> Thanks.

How about something like this?

diff --git a/arch/x86/mm/extable.c b/arch/x86/mm/extable.c
index dba2197c05c3..331310c29349 100644
--- a/arch/x86/mm/extable.c
+++ b/arch/x86/mm/extable.c
@@ -94,16 +94,18 @@ static bool ex_handler_copy(const struct exception_table_entry *fixup,
 static bool ex_handler_msr(const struct exception_table_entry *fixup,
 			   struct pt_regs *regs, bool wrmsr, bool safe, int reg)
 {
-	if (!safe && wrmsr &&
-	    pr_warn_once("unchecked MSR access error: WRMSR to 0x%x (tried to write 0x%08x%08x) at rIP: 0x%lx (%pS)\n",
-			 (unsigned int)regs->cx, (unsigned int)regs->dx,
-			 (unsigned int)regs->ax,  regs->ip, (void *)regs->ip))
+	if (__ONCE_LITE_IF(!safe && wrmsr)) {
+		pr_warn("unchecked MSR access error: WRMSR to 0x%x (tried to write 0x%08x%08x) at rIP: 0x%lx (%pS)\n",
+			(unsigned int)regs->cx, (unsigned int)regs->dx,
+			(unsigned int)regs->ax,  regs->ip, (void *)regs->ip);
 		show_stack_regs(regs);
+	}
 
-	if (!safe && !wrmsr &&
-	    pr_warn_once("unchecked MSR access error: RDMSR from 0x%x at rIP: 0x%lx (%pS)\n",
-			 (unsigned int)regs->cx, regs->ip, (void *)regs->ip))
+	if (__ONCE_LITE_IF(!safe && !wrmsr)) {
+		pr_warn("unchecked MSR access error: RDMSR from 0x%x at rIP: 0x%lx (%pS)\n",
+			(unsigned int)regs->cx, regs->ip, (void *)regs->ip);
 		show_stack_regs(regs);
+	}
 
 	if (!wrmsr) {
 		/* Pretend that the read succeeded and returned 0. */
diff --git a/include/linux/once_lite.h b/include/linux/once_lite.h
index 861e606b820f..63c3bbcef694 100644
--- a/include/linux/once_lite.h
+++ b/include/linux/once_lite.h
@@ -9,15 +9,27 @@
  */
 #define DO_ONCE_LITE(func, ...)						\
 	DO_ONCE_LITE_IF(true, func, ##__VA_ARGS__)
-#define DO_ONCE_LITE_IF(condition, func, ...)				\
+
+#define __ONCE_LITE_IF(condition)					\
 	({								\
 		static bool __section(".data.once") __already_done;	\
-		bool __ret_do_once = !!(condition);			\
+		bool __ret_cond = !!(condition);			\
+		bool __ret_once = false;				\
 									\
 		if (unlikely(__ret_do_once && !__already_done)) {	\
 			__already_done = true;				\
-			func(__VA_ARGS__);				\
+			__ret_once = true;				\
 		}							\
+		unlikely(__ret_once);					\
+	})
+
+#define DO_ONCE_LITE_IF(condition, func, ...)				\
+	({								\
+		bool __ret_do_once = !!(condition);			\
+									\
+		if (__ONCE_LITE_IF(__ret_do_once))			\
+			func(__VA_ARGS__);				\
+									\
 		unlikely(__ret_do_once);				\
 	})