[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZZ_gssjTCyoWjjhP@tassilo>
Date: Thu, 11 Jan 2024 04:36:02 -0800
From: Andi Kleen <ak@...ux.intel.com>
To: Marco Elver <elver@...gle.com>
Cc: Oscar Salvador <osalvador@...e.de>, andrey.konovalov@...ux.dev,
Andrew Morton <akpm@...ux-foundation.org>,
Andrey Konovalov <andreyknvl@...il.com>,
Alexander Potapenko <glider@...gle.com>,
Dmitry Vyukov <dvyukov@...gle.com>,
Vlastimil Babka <vbabka@...e.cz>, kasan-dev@...glegroups.com,
Evgenii Stepanov <eugenis@...gle.com>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org,
Andrey Konovalov <andreyknvl@...gle.com>
Subject: Re: [PATCH v4 12/22] lib/stackdepot: use read/write lock
> stackdepot is severely limited in what kernel facilities it may use
> due to being used by such low level facilities as the allocator
> itself.
RCU can be done quite low level too (e.g. there is NMI safe RCU)
>
> I've been suggesting percpu-rwsem here, but looking at it in more
> detail that doesn't work because percpu-rwsem wants to sleep, but
> stackdepot must work in non-sleepable contexts. :-/
Yes something per CPU would work too I suppose. We used to have
big reader spinlocks for this.
-Andi
Powered by blists - more mailing lists