From: Julian Calaby <julian.calaby@gmail.com>
To: Boaz Harrosh <boaz@plexistor.com>, Christoph Hellwig <hch@lst.de>
Cc: "Linus Torvalds" <torvalds@linux-foundation.org>,
axboe@kernel.dk, linux-mips@linux-mips.org,
linux-ia64@vger.kernel.org, linux-nvdimm@ml01.01.org,
"David Howells" <dhowells@redhat.com>,
sparclinux <sparclinux@vger.kernel.org>,
"Hans-Christian Egtvedt" <egtvedt@samfundet.no>,
linux-arch@vger.kernel.org, linux-s390@vger.kernel.org,
x86@kernel.org, "David Woodhouse" <dwmw2@infradead.org>,
"Håvard Skinnemoen" <hskinnemoen@gmail.com>,
linux-xtensa@linux-xtensa.org, grundler@parisc-linux.org,
realmz6@gmail.com, alex.williamson@redhat.com,
linux-metag@vger.kernel.org, "Michal Simek" <monstr@monstr.eu>,
linux-parisc@vger.kernel.org, vgupta@synopsys.com,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
linux-alpha@vger.kernel.org,
"Linux Media Mailing List" <linux-media@vger.kernel.org>,
linuxppc-dev@lists.ozlabs.org
Subject: Re: RFC: prepare for struct scatterlist entries without page backing
Date: Thu, 13 Aug 2015 09:37:37 +1000 [thread overview]
Message-ID: <CAGRGNgUKkaPnyvn30DXyNpdiXQzS6J=1+mQ3ick8C8=bhx_RHA@mail.gmail.com> (raw)
In-Reply-To: <55CB3F47.3000902@plexistor.com>
Hi,
On Wed, Aug 12, 2015 at 10:42 PM, Boaz Harrosh <boaz@plexistor.com> wrote:
> On 08/12/2015 10:05 AM, Christoph Hellwig wrote:
>> It turns out most DMA mapping implementation can handle SGLs without
>> page structures with some fairly simple mechanical work. Most of it
>> is just about consistently using sg_phys. For implementations that
>> need to flush caches we need a new helper that skips these cache
>> flushes if a entry doesn't have a kernel virtual address.
>>
>> However the ccio (parisc) and sba_iommu (parisc & ia64) IOMMUs seem
>> to be operate mostly on virtual addresses. It's a fairly odd concept
>> that I don't fully grasp, so I'll need some help with those if we want
>> to bring this forward.
>>
>> Additional this series skips ARM entirely for now. The reason is
>> that most arm implementations of the .map_sg operation just iterate
>> over all entries and call ->map_page for it, which means we'd need
>> to convert those to a ->map_pfn similar to Dan's previous approach.
>>
>
[snip]
>
> It is a bit of work but is worth while, and accelerating tremendously
> lots of workloads. Not like this abomination which only branches
> things more and more, and making things fatter and slower.
As a random guy reading a big bunch of patches on code I know almost
nothing about, parts of this comment really resonated with me:
overall, we seem to be adding a lot of if statements to code that
appears to be in a hot path.
I.e. ~90% of this patch set seems to be just mechanically dropping
BUG_ON()s and converting open coded stuff to use accessor functions
(which should be macros or get inlined, right?) - and the remaining
bit is not flushing if we don't have a physical page somewhere.
Would it make sense to split this patch set into a few bits: one to
drop all the useless BUG_ON()s, one to convert all the open coded
stuff to accessor functions, then another to do the actual page-less
sg stuff?
Thanks,
--
Julian Calaby
Email: julian.calaby@gmail.com
Profile: http://www.google.com/profiles/julian.calaby/
next prev parent reply other threads:[~2015-08-12 23:37 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-08-12 7:05 RFC: prepare for struct scatterlist entries without page backing Christoph Hellwig
2015-08-12 7:05 ` [PATCH 01/31] scatterlist: add sg_pfn and sg_has_page helpers Christoph Hellwig
2015-08-12 7:05 ` [PATCH 02/31] scatterlist: use sg_phys() Christoph Hellwig
2015-08-12 7:05 ` [PATCH 03/31] dma-debug: handle page-less SG entries Christoph Hellwig
2015-08-12 7:05 ` [PATCH 04/31] x86/pci-nommu: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 05/31] x86/pci-calgary: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 06/31] alpha/pci-noop: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 07/31] alpha/pci_iommu: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 08/31] c6x: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 09/31] ia64/pci_dma: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 10/31] powerpc/iommu: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 11/31] sparc/iommu: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 12/31] mn10300: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 13/31] sparc/ldc: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 14/31] sparc32/io-unit: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 15/31] sparc32/iommu: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 16/31] s390: " Christoph Hellwig
2015-08-12 11:51 ` Sebastian Ott
2015-08-12 7:05 ` [PATCH 17/31] ia64/sba_iommu: remove sba_sg_address Christoph Hellwig
2015-08-12 7:05 ` [PATCH 18/31] nios2: handle page-less SG entries Christoph Hellwig
2015-08-12 7:05 ` [PATCH 19/31] arc: " Christoph Hellwig
2015-08-12 10:28 ` Vineet Gupta
2015-08-12 7:05 ` [PATCH 20/31] avr32: " Christoph Hellwig
2015-08-12 9:36 ` Hans-Christian Egtvedt
2015-08-12 7:05 ` [PATCH 21/31] blackfin: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 22/31] metag: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 23/31] sh: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 24/31] xtensa: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 25/31] frv: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 26/31] openrisc: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 27/31] mips: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 28/31] powerpc: " Christoph Hellwig
2015-08-12 7:05 ` [PATCH 29/31] parisc: " Christoph Hellwig
2015-08-12 16:01 ` Linus Torvalds
2015-08-13 14:31 ` Christoph Hellwig
2015-08-14 3:30 ` Dan Williams
2015-08-14 3:59 ` James Bottomley
2015-08-14 4:11 ` David Miller
2015-08-14 16:17 ` Dan Williams
2015-08-12 7:05 ` [PATCH 30/31] intel-iommu: " Christoph Hellwig
2015-08-12 9:51 ` David Woodhouse
2015-08-12 7:05 ` [PATCH 31/31] dma-mapping-common: skip kmemleak checks for " Christoph Hellwig
2015-08-12 16:05 ` Linus Torvalds
2015-08-13 14:33 ` Christoph Hellwig
2015-08-12 16:26 ` Catalin Marinas
2015-08-12 12:42 ` RFC: prepare for struct scatterlist entries without page backing Boaz Harrosh
2015-08-12 23:37 ` Julian Calaby [this message]
2015-08-13 14:35 ` Christoph Hellwig
2015-08-13 23:40 ` Julian Calaby
2015-08-13 14:40 ` Christoph Hellwig
2015-08-13 15:42 ` Boaz Harrosh
2015-08-12 17:00 ` James Bottomley
2015-08-12 17:56 ` Grant Grundler
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='CAGRGNgUKkaPnyvn30DXyNpdiXQzS6J=1+mQ3ick8C8=bhx_RHA@mail.gmail.com' \
--to=julian.calaby@gmail.com \
--cc=alex.williamson@redhat.com \
--cc=axboe@kernel.dk \
--cc=boaz@plexistor.com \
--cc=dhowells@redhat.com \
--cc=dwmw2@infradead.org \
--cc=egtvedt@samfundet.no \
--cc=grundler@parisc-linux.org \
--cc=hch@lst.de \
--cc=hskinnemoen@gmail.com \
--cc=linux-alpha@vger.kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-ia64@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-metag@vger.kernel.org \
--cc=linux-mips@linux-mips.org \
--cc=linux-nvdimm@ml01.01.org \
--cc=linux-parisc@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=linux-xtensa@linux-xtensa.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=monstr@monstr.eu \
--cc=realmz6@gmail.com \
--cc=sparclinux@vger.kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=vgupta@synopsys.com \
--cc=x86@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).