linux-kernel - Re: [patch V3 21/29] tracing: Use percpu stack trace buffer more intelligently

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20190425132935.ae35l5oybby5ddgl@treble>
Date:   Thu, 25 Apr 2019 08:29:35 -0500
From:   Josh Poimboeuf <jpoimboe@...hat.com>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     LKML <linux-kernel@...r.kernel.org>, x86@...nel.org,
        Andy Lutomirski <luto@...nel.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Alexander Potapenko <glider@...gle.com>,
        Alexey Dobriyan <adobriyan@...il.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Christoph Lameter <cl@...ux.com>,
        Pekka Enberg <penberg@...nel.org>, linux-mm@...ck.org,
        David Rientjes <rientjes@...gle.com>,
        Catalin Marinas <catalin.marinas@....com>,
        Dmitry Vyukov <dvyukov@...gle.com>,
        Andrey Ryabinin <aryabinin@...tuozzo.com>,
        kasan-dev@...glegroups.com,
        Mike Rapoport <rppt@...ux.vnet.ibm.com>,
        Akinobu Mita <akinobu.mita@...il.com>,
        Christoph Hellwig <hch@....de>,
        iommu@...ts.linux-foundation.org,
        Robin Murphy <robin.murphy@....com>,
        Marek Szyprowski <m.szyprowski@...sung.com>,
        Johannes Thumshirn <jthumshirn@...e.de>,
        David Sterba <dsterba@...e.com>, Chris Mason <clm@...com>,
        Josef Bacik <josef@...icpanda.com>,
        linux-btrfs@...r.kernel.org, dm-devel@...hat.com,
        Mike Snitzer <snitzer@...hat.com>,
        Alasdair Kergon <agk@...hat.com>,
        Daniel Vetter <daniel@...ll.ch>,
        intel-gfx@...ts.freedesktop.org,
        Joonas Lahtinen <joonas.lahtinen@...ux.intel.com>,
        Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
        dri-devel@...ts.freedesktop.org, David Airlie <airlied@...ux.ie>,
        Jani Nikula <jani.nikula@...ux.intel.com>,
        Rodrigo Vivi <rodrigo.vivi@...el.com>,
        Tom Zanussi <tom.zanussi@...ux.intel.com>,
        Miroslav Benes <mbenes@...e.cz>, linux-arch@...r.kernel.org
Subject: Re: [patch V3 21/29] tracing: Use percpu stack trace buffer more
 intelligently

On Thu, Apr 25, 2019 at 11:45:14AM +0200, Thomas Gleixner wrote:
> @@ -2788,29 +2798,32 @@ static void __ftrace_trace_stack(struct
>  	 */
>  	preempt_disable_notrace();
>  
> -	use_stack = __this_cpu_inc_return(ftrace_stack_reserve);
> +	stackidx = __this_cpu_inc_return(ftrace_stack_reserve);
> +
> +	/* This should never happen. If it does, yell once and skip */
> +	if (WARN_ON_ONCE(stackidx >= FTRACE_KSTACK_NESTING))
> +		goto out;
> +
>  	/*
> -	 * We don't need any atomic variables, just a barrier.
> -	 * If an interrupt comes in, we don't care, because it would
> -	 * have exited and put the counter back to what we want.
> -	 * We just need a barrier to keep gcc from moving things
> -	 * around.
> +	 * The above __this_cpu_inc_return() is 'atomic' cpu local. An
> +	 * interrupt will either see the value pre increment or post
> +	 * increment. If the interrupt happens pre increment it will have
> +	 * restored the counter when it returns.  We just need a barrier to
> +	 * keep gcc from moving things around.
>  	 */
>  	barrier();
> -	if (use_stack == 1) {
> -		trace.entries		= this_cpu_ptr(ftrace_stack.calls);
> -		trace.max_entries	= FTRACE_STACK_MAX_ENTRIES;
> -
> -		if (regs)
> -			save_stack_trace_regs(regs, &trace);
> -		else
> -			save_stack_trace(&trace);
> -
> -		if (trace.nr_entries > size)
> -			size = trace.nr_entries;
> -	} else
> -		/* From now on, use_stack is a boolean */
> -		use_stack = 0;
> +
> +	fstack = this_cpu_ptr(ftrace_stacks.stacks) + (stackidx - 1);

nit: it would be slightly less surprising if stackidx were 0-based:

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index d3f6ec7eb729..4fc93004feab 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -2798,10 +2798,10 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer,
 	 */
 	preempt_disable_notrace();
 
-	stackidx = __this_cpu_inc_return(ftrace_stack_reserve);
+	stackidx = __this_cpu_inc_return(ftrace_stack_reserve) - 1;
 
 	/* This should never happen. If it does, yell once and skip */
-	if (WARN_ON_ONCE(stackidx >= FTRACE_KSTACK_NESTING))
+	if (WARN_ON_ONCE(stackidx > FTRACE_KSTACK_NESTING))
 		goto out;
 
 	/*
@@ -2813,7 +2813,7 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer,
 	 */
 	barrier();
 
-	fstack = this_cpu_ptr(ftrace_stacks.stacks) + (stackidx - 1);
+	fstack = this_cpu_ptr(ftrace_stacks.stacks) + stackidx;
 	trace.entries		= fstack->calls;
 	trace.max_entries	= FTRACE_KSTACK_ENTRIES;