From: Leif Lindholm <leif.lindholm@linaro.org>
To: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: "Cohen, Eugene" <eugene@hp.com>,
edk2-devel-01 <edk2-devel@lists.01.org>,
Star Zeng <star.zeng@intel.com>
Subject: Re: [PATCH V3 4/4] MdeModulePkg EhciDxe: Use common buffer for AsyncInterruptTransfer
Date: Tue, 30 Oct 2018 12:50:06 +0000 [thread overview]
Message-ID: <20181030125006.4deveknlhrwehllb@bivouac.eciton.net> (raw)
In-Reply-To: <CAKv+Gu-U6rtrz11zsTH0+ea4X7WaDdf38ZcZG45Bsh_2cK+L=Q@mail.gmail.com>
On Tue, Oct 30, 2018 at 09:39:24AM -0300, Ard Biesheuvel wrote:
> (add back the list)
Oi! Go back on holiday!
> On 30 October 2018 at 09:07, Cohen, Eugene <eugene@hp.com> wrote:
> > Has this patch been tested on a system that does not have coherent DMA?
> >
> > It's not clear that this change would actually be faster on a system of that
> > type since using common buffers imply access to uncached memory. Depending
> > on the access patterns the uncached memory access could be more time
> > consuming than cache maintenance operations.
>
> I haven't had time to look at these patches yet.
>
> I agree with Eugene's concern: the directional DMA routines are much
> more performant on implementations with non-coherent DMA, and so
> common buffers should be avoided unless we are dealing with data
> structures that are truly shared between the CPU and the device.
>
> Since this is obviously not the case here, could we please have some
> numbers about the performance improvement we are talking about here?
> Would it be possible to improve the IOMMU handling code instead?
On an unrelated note to the concerns above:
Why has a fundamental change to the behaviour of one of the industry
standard drivers been pushed at the very end of the stable cycle?
Regards,
Leif
next prev parent reply other threads:[~2018-10-30 12:50 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-26 13:41 [PATCH V3 0/4] Remove unnecessary Map/Unmap in XhciDxe/EhciDxe for AsyncInterruptTransfer Star Zeng
2018-10-26 13:41 ` [PATCH V3 1/4] MdeModulePkg XhciDxe: Extract new XhciInsertAsyncIntTransfer function Star Zeng
2018-10-26 13:41 ` [PATCH V3 2/4] MdeModulePkg EhciDxe: Extract new EhciInsertAsyncIntTransfer function Star Zeng
2018-10-26 13:41 ` [PATCH V3 3/4] MdeModulePkg XhciDxe: Use common buffer for AsyncInterruptTransfer Star Zeng
2018-10-27 7:47 ` Wu, Hao A
2018-10-26 13:41 ` [PATCH V3 4/4] MdeModulePkg EhciDxe: " Star Zeng
[not found] ` <CS1PR8401MB1189428C2915107C583C0A42B4CC0@CS1PR8401MB1189.NAMPRD84.PROD.OUTLOOK.COM>
2018-10-30 12:39 ` Ard Biesheuvel
2018-10-30 12:50 ` Leif Lindholm [this message]
2018-10-31 4:38 ` Zeng, Star
2018-10-31 12:08 ` Leif Lindholm
2018-11-01 1:12 ` Zeng, Star
2018-11-01 10:32 ` Leif Lindholm
2018-11-06 9:49 ` Ard Biesheuvel
2018-11-06 14:37 ` Zeng, Star
2018-11-07 15:00 ` Zeng, Star
2018-11-07 15:14 ` Ard Biesheuvel
2018-10-28 13:35 ` [PATCH V3 0/4] Remove unnecessary Map/Unmap in XhciDxe/EhciDxe " Zeng, Star
2018-10-29 2:19 ` Ni, Ruiyu
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-list 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=20181030125006.4deveknlhrwehllb@bivouac.eciton.net \
--to=devel@edk2.groups.io \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox