[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <010101747f12e499-1418e736-fe1b-449f-bc7c-4af35f855d13-000000@us-west-2.amazonses.com>
Date: Fri, 11 Sep 2020 21:29:38 +0000
From: Chris Goldsworthy <cgoldswo@...eaurora.org>
To: linux-kernel@...r.kernel.org
Cc: linux-kernel-owner@...r.kernel.org
Subject: Re: [PATCH v2] cma_alloc(), indefinitely retry allocations for -EBUSY
failures
On 2020-09-11 13:54, Chris Goldsworthy wrote:
> [PATCH v2] cma_alloc(), indefinitely retry allocations for -EBUSY
> failures
>
> On mobile devices, failure to allocate from a CMA area constitutes a
> functional failure. Sometimes during CMA allocations, we have observed
> that pages in a CMA area allocated through alloc_pages(), that we're
> trying
> to migrate away to make room for a CMA allocation, are temporarily
> pinned.
> This temporary pinning can occur when a process that owns the pinned
> page
> is being forked (the example is explained further in the commit text),
> or it is
> exiting. This patch addresses this issue by indefinitely retrying
> allocations
> that fail due to a return of -EBUSY.
>
> ** This change log was re-sent due to threading issues **
>
> Change log:
>
> v1: We were performing retries of the allocation a fixed number of
> times.
> Andrew Morton disliked this, as it didn't guarantee that the allocation
> would
> succeed.
>
> https://lkml.org/lkml/2020/8/5/1096
> https://lkml.org/lkml/2020/8/21/1490
>
> v2: To address this concern, we switched to retrying indefinitely, as
> opposed to
> doing to retrying the allocation a limited number of times.
>
> Chris Goldsworthy (1):
> mm: cma: indefinitely retry allocations in cma_alloc
>
> mm/cma.c | 25 +++++++++++++++++++++++--
> 1 file changed, 23 insertions(+), 2 deletions(-)
test
--
The Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora
Forum,
a Linux Foundation Collaborative Project
Powered by blists - more mailing lists