From: David Vrabel <david.vrabel@citrix.com>
To: Thanos Makatos <thanos.makatos@citrix.com>,
"xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>
Cc: "boris.ostrovsky@oracle.com" <boris.ostrovsky@oracle.com>
Subject: Re: [PATCH] introduce grant copy for user land
Date: Tue, 11 Nov 2014 15:06:44 +0000 [thread overview]
Message-ID: <54622604.40807@citrix.com> (raw)
In-Reply-To: <2368A3FCF9F7214298E53C823B0A48EC0428962A@AMSPEX01CL02.citrite.net>
On 11/11/14 14:42, Thanos Makatos wrote:
>> On 11/11/14 12:27, Thanos Makatos wrote:
>>>> The arbitrary limitations in number of ops and page alignment should
>>>> be removed. I think both can be removed relatively easily by
>>>> consuming page aligned chunks of segments and doign the hypercall
>>>> when a batch of ops is filled.
>>>
>>> Wouldn't this lead to multiple calls to gnttab_batch_copy(), potentially
>> hurting performance?
>>
>> The incremental benefits of batching diminishes as the batch size increase.
>
> To overcome the page alignment limitation, struct gntdev_grant_copy_segment
> will have to contain a set of grant refs instead of just one grant ref,
> correct?
Oh, that would be awkward. Keeping the alignment requirement would be best.
>> We also don't want multi-page allocations in this driver, so the batch size
>> should be set accordingly.
>
> Why not?
Multi-page allocations are more likely to fail because of memory
fragmentation.
>> The interface should also support grant to grant copies.
>
> To support grant to grant copies, struct gntdev_grant_copy_segment will have
> to contain an additional set of grant refs, correct?
The hypervisor ABI uses a union of handle and grant ref. Perhaps
something similar for the kernel interface?
David
next prev parent reply other threads:[~2014-11-11 15:09 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-02 15:15 [PATCH] introduce grant copy for user land Thanos Makatos
2014-11-03 18:03 ` David Vrabel
2014-11-11 12:27 ` Thanos Makatos
2014-11-11 13:34 ` David Vrabel
2014-11-11 14:42 ` Thanos Makatos
2014-11-11 15:06 ` David Vrabel [this message]
2014-11-11 15:23 ` Thanos Makatos
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=54622604.40807@citrix.com \
--to=david.vrabel@citrix.com \
--cc=boris.ostrovsky@oracle.com \
--cc=thanos.makatos@citrix.com \
--cc=xen-devel@lists.xenproject.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).