From: Peter Xu <peterx@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>,
qemu-devel@nongnu.org, "Michael S . Tsirkin" <mst@redhat.com>,
Alex Williamson <alex.williamson@redhat.com>,
Jintack Lim <jintack@cs.columbia.edu>,
David Gibson <david@gibson.dropbear.id.au>,
Stefan Hajnoczi <stefanha@redhat.com>,
Fam Zheng <famz@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 03/10] intel-iommu: add iommu lock
Date: Thu, 3 May 2018 13:39:35 +0800 [thread overview]
Message-ID: <20180503053935.GA2378@xz-mi> (raw)
In-Reply-To: <54cee857-2492-95c8-e1c4-d6a6e2f6b552@redhat.com>
On Mon, Apr 30, 2018 at 09:20:42AM +0200, Paolo Bonzini wrote:
> On 28/04/2018 04:24, Peter Xu wrote:
> >>>> 2) Can we just reuse qemu BQL here?
> >>> I would prefer not. As I mentioned, at least I have spent too much
> >>> time on fighting BQL already. I really hope we can start to use
> >>> isolated locks when capable. BQL is always the worst choice to me.
> >> Just a thought, using BQL may greatly simplify the code actually (consider
> >> we don't plan to remove BQL now).
> > Frankly speaking I don't understand why using BQL may greatly simplify
> > the code... :( IMHO the lock here is really not a complicated one.
> >
> > Note that IMO BQL is mostly helpful when we really want something to
> > be run sequentially with some other things _already_ protected by BQL.
> > In this case, all the stuff is inside VT-d code itself (or other
> > IOMMUs), why bother taking the BQL to make our life harder?
> >
> > So, even if we want to provide a general lock for the translation
> > procedure, I would prefer we add a per AddressSpace lock but not BQL.
> > However still that will need some extra flag showing that whether we
> > need the protection of not. For example, we may need to expliclitly
> > turn that off for Power and s390. Would that really worth it?
> >
> > So my final preference is still current patch - we solve thread-safety
> > problems in VT-d and IOMMU code. Again, we really should make sure
> > all IOMMUs work with multithreads.
> >
>
> I agree. In particular, using BQL is _worse_ because it has very strict
> lock ordering requirements. Using fine-grained locks is greatly
> preferred as long as:
>
> 1) they are leaves in the lock ordering
>
> 2) they are not kept across calls to external callbacks (or there are no
> external callbacks involved)
Thanks Paolo for the input.
I'll temporarily keep this patch in my next post. We can further
discuss it there if we have better alternatives.
Regards,
--
Peter Xu
next prev parent reply other threads:[~2018-05-03 5:39 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-25 4:51 [Qemu-devel] [PATCH 00/10] intel-iommu: nested vIOMMU, cleanups, bug fixes Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 01/10] intel-iommu: send PSI always even if across PDEs Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 02/10] intel-iommu: remove IntelIOMMUNotifierNode Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 03/10] intel-iommu: add iommu lock Peter Xu
2018-04-25 16:26 ` Emilio G. Cota
2018-04-26 5:45 ` Peter Xu
2018-04-27 5:13 ` Jason Wang
2018-04-27 6:26 ` Peter Xu
2018-04-27 7:19 ` Tian, Kevin
2018-04-27 9:53 ` Peter Xu
2018-04-28 1:54 ` Tian, Kevin
2018-04-28 1:43 ` Jason Wang
2018-04-28 2:24 ` Peter Xu
2018-04-28 2:42 ` Jason Wang
2018-04-28 3:06 ` Peter Xu
2018-04-28 3:11 ` Jason Wang
2018-04-28 3:14 ` Peter Xu
2018-04-28 3:16 ` Jason Wang
2018-04-30 7:22 ` Paolo Bonzini
2018-04-30 7:20 ` Paolo Bonzini
2018-05-03 5:39 ` Peter Xu [this message]
2018-04-25 4:51 ` [Qemu-devel] [PATCH 04/10] intel-iommu: only do page walk for MAP notifiers Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 05/10] intel-iommu: introduce vtd_page_walk_info Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 06/10] intel-iommu: pass in address space when page walk Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 07/10] util: implement simple interval tree logic Peter Xu
2018-04-27 5:53 ` Jason Wang
2018-04-27 6:27 ` Peter Xu
2018-05-03 7:10 ` Peter Xu
2018-05-03 7:21 ` Jason Wang
2018-05-03 7:30 ` Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 08/10] intel-iommu: maintain per-device iova ranges Peter Xu
2018-04-27 6:07 ` Jason Wang
2018-04-27 6:34 ` Peter Xu
2018-04-27 7:02 ` Tian, Kevin
2018-04-27 7:28 ` Peter Xu
2018-04-27 7:44 ` Tian, Kevin
2018-04-27 9:55 ` Peter Xu
2018-04-27 11:40 ` Peter Xu
2018-04-27 23:37 ` Tian, Kevin
2018-05-03 6:04 ` Peter Xu
2018-05-03 7:20 ` Jason Wang
2018-05-03 7:28 ` Peter Xu
2018-05-03 7:43 ` Jason Wang
2018-05-03 7:53 ` Peter Xu
2018-05-03 9:22 ` Jason Wang
2018-05-03 9:53 ` Peter Xu
2018-05-03 12:01 ` Peter Xu
2018-04-28 1:49 ` Jason Wang
2018-04-25 4:51 ` [Qemu-devel] [PATCH 09/10] intel-iommu: don't unmap all for shadow page table Peter Xu
2018-04-25 4:51 ` [Qemu-devel] [PATCH 10/10] intel-iommu: remove notify_unmap for page walk Peter Xu
2018-04-25 5:05 ` [Qemu-devel] [PATCH 00/10] intel-iommu: nested vIOMMU, cleanups, bug fixes no-reply
2018-04-25 5:34 ` Peter Xu
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=20180503053935.GA2378@xz-mi \
--to=peterx@redhat.com \
--cc=alex.williamson@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=famz@redhat.com \
--cc=jasowang@redhat.com \
--cc=jintack@cs.columbia.edu \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
/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).