Re: [PATCH] percpu: Sync vmalloc mappings in pcpu_alloc() and free_percpu()

From: Steven Rostedt <rostedt@goodmis.org>
To: Joerg Roedel <jroedel@suse.de>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Ingo Molnar <mingo@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Peter Zijlstra <peterz@infradead.org>,
	Borislav Petkov <bp@alien8.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	Shile Zhang <shile.zhang@linux.alibaba.com>,
	Andy Lutomirski <luto@amacapital.net>,
	"Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Tzvetomir Stoyanov <tz.stoyanov@gmail.com>
Subject: Re: [PATCH] percpu: Sync vmalloc mappings in pcpu_alloc() and free_percpu()
Date: Mon, 4 May 2020 15:10:06 -0400	[thread overview]
Message-ID: <20200504151006.69d2a16c@gandalf.local.home> (raw)
In-Reply-To: <20200504183832.GL8135@suse.de>

On Mon, 4 May 2020 20:38:32 +0200
Joerg Roedel <jroedel@suse.de> wrote:

> Fair point, but this just shows how problematic it is to call something
> like vmalloc_sync_mappings() from a lower-level kernel API function.
> The obvious fix for this would be to make pgd_lock irq-safe, but this is
> getting more and more ridiculous.

Yeah, it's just getting more deeper into the rabbit hole.

> 
> I know you don't like to have a vmalloc_sync_mappings() call in the
> tracing code, but can you live with it until we get rid of this broken
> interface?

I'm fine with adding it to the tracing code (with that ridiculous
comment! ;-)

I'll even tag is as stable, but again, it's uncertain what commit that it
"fixes".

> 
> My plan for this is to use a small bitmap to track in the vmalloc and
> the (x86-)ioremap code at which levels of the page-tables the code made
> changes and combine that with an architecture-dependend mask to decide
> whether anything needs to be synced.
> 
> On x86-64 the sync would be necessary at most 64 times after boot, so I
> think this will only have a very small performance impact, even with
> VMAP_STACKS. And as a bonus it would also get rid of vmalloc faulting on
> x86, fixing the issue with tracing too.

Just because tracing get's a workaround, let's hopefully not get distracted
and forgot to add this. It looks like the best solution is to not need the
vmalloc_sync_mappings() call in the future. But, as we are about to release
trace-cmd with a feature that reliably triggers the bug, I can't wait for
the proper fix.

-- Steve