From: "jroedel-l3A5Bk7waGM@public.gmane.org" <jroedel-l3A5Bk7waGM@public.gmane.org>
To: Shaohua Li <shli-b10kYP2dOMg@public.gmane.org>
Cc: "iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org"
<iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org>,
"Kernel-team-b10kYP2dOMg@public.gmane.org"
<Kernel-team-b10kYP2dOMg@public.gmane.org>,
"Woodhouse,
David" <david.woodhouse-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Subject: Re: [PATCH 2/4] iommu: add a bitmap based dma address allocator
Date: Wed, 25 Nov 2015 15:46:13 +0100 [thread overview]
Message-ID: <20151125144613.GA3763@suse.de> (raw)
In-Reply-To: <20151124220511.GA2106568-tb7CFzD8y5b7E6g3fPdp/g2O0Ztt9esIQQ4Iyu8u01E@public.gmane.org>
On Tue, Nov 24, 2015 at 02:05:12PM -0800, Shaohua Li wrote:
> The lib/iommu-common.c uses a bitmap and a lock. This implementation
> actually uses a percpu_ida which completely avoids locking. It would be
> possible to make lib/iommu-common.c use percpu_ida too if somebody wants
> to do it, but I think this shouldn't be a blocker for these patches
> giving it has huge performance gain.
It doesn't "completely avoids locking", the percpu_ida code uses a lock
internally too. Also, what is the memory and device address space
overhead per cpu?
Joerg
next prev parent reply other threads:[~2015-11-25 14:46 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-24 21:54 [PATCH 0/4] iommu: remove lock contention in iova allocation Shaohua Li
[not found] ` <cover.1448401089.git.shli-b10kYP2dOMg@public.gmane.org>
2015-11-24 21:54 ` [PATCH 1/4] iommu: alloc_iova returns a pfn Shaohua Li
2015-11-24 21:54 ` [PATCH 2/4] iommu: add a bitmap based dma address allocator Shaohua Li
[not found] ` <4c0804c85404be81acfe81fcd402f1af484e9d5f.1448401089.git.shli-b10kYP2dOMg@public.gmane.org>
2015-11-24 21:56 ` Woodhouse, David
[not found] ` <1448402209.44976.109.camel-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2015-11-24 22:05 ` Shaohua Li
[not found] ` <20151124220511.GA2106568-tb7CFzD8y5b7E6g3fPdp/g2O0Ztt9esIQQ4Iyu8u01E@public.gmane.org>
2015-11-25 14:46 ` jroedel-l3A5Bk7waGM [this message]
[not found] ` <20151125144613.GA3763-l3A5Bk7waGM@public.gmane.org>
2015-11-25 16:40 ` Woodhouse, David
[not found] ` <1448469599.44976.134.camel-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2015-11-25 16:43 ` Jens Axboe
[not found] ` <5655E545.1010509-b10kYP2dOMg@public.gmane.org>
2015-11-25 17:33 ` Shaohua Li
[not found] ` <20151125173320.GA92074-tb7CFzD8y5b7E6g3fPdp/g2O0Ztt9esIQQ4Iyu8u01E@public.gmane.org>
2015-11-25 17:38 ` Jens Axboe
[not found] ` <5655F1F8.9070203-b10kYP2dOMg@public.gmane.org>
2015-11-25 18:57 ` Shaohua Li
2015-11-24 21:54 ` [PATCH 3/4] iommu: enable bitmap allocation for intel iommu Shaohua Li
2015-11-24 21:54 ` [PATCH 4/4] iommu: free_iova doesn't need lock twice Shaohua Li
2015-12-07 17:52 ` [PATCH 0/4] iommu: remove lock contention in iova allocation Shaohua Li
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=20151125144613.GA3763@suse.de \
--to=jroedel-l3a5bk7wagm@public.gmane.org \
--cc=Kernel-team-b10kYP2dOMg@public.gmane.org \
--cc=david.woodhouse-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=shli-b10kYP2dOMg@public.gmane.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.