All of lore.kernel.org
 help / color / mirror / Atom feed
From: Catalin Marinas <catalin.marinas@arm.com>
To: Will Deacon <will@kernel.org>
Cc: linux-arm-kernel@lists.infradead.org, kernel-team@android.com,
	Mark Rutland <mark.rutland@arm.com>,
	Ard Biesheuvel <ardb@kernel.org>, Arnd Bergmann <arnd@arndb.de>,
	Vincent Whitchurch <vincent.whitchurch@axis.com>
Subject: Re: [PATCH] arm64: cache: Lower ARCH_DMA_MINALIGN to 64 (L1_CACHE_BYTES)
Date: Thu, 27 May 2021 14:11:20 +0100	[thread overview]
Message-ID: <20210527131120.GB8661@arm.com> (raw)
In-Reply-To: <20210527124356.22367-1-will@kernel.org>

On Thu, May 27, 2021 at 01:43:56PM +0100, Will Deacon wrote:
> Back in 97303480753e ("arm64: Increase the max granular size"),
> ARCH_DMA_MINALIGN was effectively increased to 128 bytes thanks to an
> increase in L1_CACHE_BYTES due to an unsubstantiated performance claim
> on the now obsolete ThunderX-1. Although this was reverted in
> d93277b9839b, ARCH_DMA_MINALIGN was kept at 128 bytes by ebc7e21e0fa2
> ("arm64: Increase ARCH_DMA_MINALIGN to 128").
> 
> During discussion of the original patch, it was reported that the change
> also prevented a warning during boot on (again, now obsolete) Qualcomm
> server hardware where the cache writeback granule was larger than 64
> bytes. The reason for this warning was because non-coherent DMA could
> lead to data corruption due to unexpected writeback from the CPU where a
> cacheline is shared with other allocations.
> 
> Since then, systems have appeared with larger cachelines still, and so
> commit 8f5c9037a55b ("arm64/mm: Correct the cache line size warning with
> non coherent device") reworked the warning so that it only appears on
> systems where non-coherent DMA is actually required and taints the
> kernel with TAINT_CPU_OUT_OF_SPEC. We are not aware of any systems, even
> including the aforementioned obsolete machines, which have a CWG larger
> than 64 bytes and require non-coherent DMA.
> 
> More recently, it has been reported that a ARCH_DMA_MINALIGN of 128
> bytes wastes considerable memory (~6% immediately after boot on one
> system).
> 
> Reduce ARCH_DMA_MINALIGN to 64 bytes and allow the warning/taint to
> indicate if there are machines that unknowingly rely on this.
> 
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Ard Biesheuvel <ardb@kernel.org>
> Cc: Arnd Bergmann <arnd@arndb.de>
> Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
> Link: https://lore.kernel.org/linux-arm-kernel/1442944788-17254-1-git-send-email-rric@kernel.org/
> Link: https://lore.kernel.org/linux-arm-kernel/CAOZdJXUiRMAguDV+HEJqPg57MyBNqEcTyaH+ya=U93NHb-pdJA@mail.gmail.com/
> Link: https://lore.kernel.org/linux-arm-kernel/20190614131141.4428-1-msys.mizuma@gmail.com/
> Link: https://lore.kernel.org/r/20210517074332.28280-1-vincent.whitchurch@axis.com
> Signed-off-by: Will Deacon <will@kernel.org>

Unless we hear from anyone that the warning is triggered, I think we
should change this in mainline.

Acked-by: Catalin Marinas <catalin.marinas@arm.com>

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2021-05-27 13:15 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-27 12:43 [PATCH] arm64: cache: Lower ARCH_DMA_MINALIGN to 64 (L1_CACHE_BYTES) Will Deacon
2021-05-27 13:11 ` Catalin Marinas [this message]
2021-05-27 13:19 ` Mark Rutland
2021-05-28  9:35   ` Arnd Bergmann
2021-06-01 10:14     ` Catalin Marinas
2021-05-31  5:38 ` Ard Biesheuvel
2021-06-01 18:21 ` Will Deacon
     [not found] ` <CGME20210602132541eucas1p17127696041c26c00d1d2f50bef9cfaf0@eucas1p1.samsung.com>
2021-06-02 13:25   ` Marek Szyprowski
2021-06-02 13:51     ` Mark Rutland
2021-06-02 14:09       ` Marek Szyprowski
2021-06-02 14:14         ` Arnd Bergmann
2021-06-02 14:28           ` Marek Szyprowski
2021-06-02 14:52             ` Arnd Bergmann
2021-06-07 12:17               ` Arnd Bergmann
2021-06-04 10:01         ` Mark Rutland
2021-06-07  9:58           ` Marek Szyprowski
2021-06-07 12:01             ` Mark Rutland
2021-06-07 13:08               ` Mark Rutland
2021-06-07 13:39                 ` Will Deacon
2021-06-07 13:39                   ` Will Deacon
2021-06-07 13:56                   ` Mark Rutland
2021-06-07 13:56                     ` Mark Rutland
2021-06-07 13:57                   ` Arnd Bergmann
2021-06-07 13:57                     ` Arnd Bergmann
2021-06-07 15:17                     ` Maxime Ripard
2021-06-07 15:17                       ` Maxime Ripard
2021-06-07 15:50                       ` Arnd Bergmann
2021-06-07 15:50                         ` Arnd Bergmann
2021-06-08  8:57                         ` Mark Rutland
2021-06-08  8:57                           ` Mark Rutland
2021-06-07 15:32                     ` Mark Rutland
2021-06-07 15:32                       ` Mark Rutland
2021-06-02 14:11       ` Arnd Bergmann
2021-06-02 14:15         ` Marek Szyprowski
2021-07-06  9:26 ` Yassine Oudjana
2021-07-06 10:26   ` Catalin Marinas
2021-07-06 10:26     ` Catalin Marinas
2021-07-06 13:29     ` Robin Murphy
2021-07-06 13:29       ` Robin Murphy
2021-07-06 13:33       ` Will Deacon
2021-07-06 13:33         ` Will Deacon
2021-07-06 13:44         ` Marc Zyngier
2021-07-06 13:44           ` Marc Zyngier
2021-07-06 14:21           ` Robin Murphy
2021-07-06 14:21             ` Robin Murphy
2021-07-06 14:30           ` Arnd Bergmann
2021-07-06 14:30             ` Arnd Bergmann
2021-07-06 14:46             ` Marc Zyngier
2021-07-06 14:46               ` Marc Zyngier
2021-07-06 15:43               ` Arnd Bergmann
2021-07-06 15:43                 ` Arnd Bergmann
2021-07-06 17:15                 ` Yassine Oudjana
2021-07-06 17:15                   ` Yassine Oudjana
2021-07-06 20:33                   ` Arnd Bergmann
2021-07-06 20:33                     ` Arnd Bergmann
2021-07-06 22:27                     ` Bjorn Andersson
2021-07-06 22:27                       ` Bjorn Andersson
2021-07-07  9:27                       ` Will Deacon
2021-07-07  9:27                         ` Will Deacon
2021-07-07  8:24                     ` Yassine Oudjana
2021-07-07  8:24                       ` Yassine Oudjana
2021-07-07  9:29                       ` Arnd Bergmann
2021-07-07  9:29                         ` Arnd Bergmann
2021-07-07 14:41                         ` Jeffrey Hugo
2021-07-07 14:41                           ` Jeffrey Hugo
2021-07-08 20:59                           ` Jeffrey Hugo
2021-07-08 20:59                             ` Jeffrey Hugo
2021-07-09  8:48                             ` Will Deacon
2021-07-09  8:48                               ` Will Deacon
2021-07-09 17:10                               ` Catalin Marinas
2021-07-09 17:10                                 ` Catalin Marinas
2021-07-06 16:20             ` Will Deacon
2021-07-06 16:20               ` Will Deacon

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210527131120.GB8661@arm.com \
    --to=catalin.marinas@arm.com \
    --cc=ardb@kernel.org \
    --cc=arnd@arndb.de \
    --cc=kernel-team@android.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=mark.rutland@arm.com \
    --cc=vincent.whitchurch@axis.com \
    --cc=will@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.