From: John Snow <jsnow@redhat.com>
To: Stefan Hajnoczi <stefanha@redhat.com>
Cc: kwolf@redhat.com, pbonzini@redhat.com, qemu-devel@nongnu.org,
qemu-block@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 12/16] ahci: ncq migration
Date: Fri, 26 Jun 2015 12:46:07 -0400 [thread overview]
Message-ID: <558D81CF.3000208@redhat.com> (raw)
In-Reply-To: <20150626154823.GD31186@stefanha-thinkpad.redhat.com>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
On 06/26/2015 11:48 AM, Stefan Hajnoczi wrote:
> On Mon, Jun 22, 2015 at 08:21:11PM -0400, John Snow wrote:
>> @@ -1555,6 +1573,35 @@ static int ahci_state_post_load(void
>> *opaque, int version_id) return -1; }
>>
>> + for (j = 0; j < AHCI_MAX_CMDS; j++) { +
>> ncq_tfs = &ad->ncq_tfs[j]; + ncq_tfs->drive = ad; + +
>> if (ncq_tfs->used != ncq_tfs->halt) { + return
>> -1; + } + if (!ncq_tfs->halt) { +
>> continue; + } + if (!is_ncq(ncq_tfs->cmd))
>> { + return -1; + } + if
>> (ncq_tfs->slot != ncq_tfs->tag) { + return -1; +
>> } + if (ncq_tfs->slot > AHCI_MAX_CMDS) { +
>> return -1; + } + ncq_tfs->cmdh =
>> &((AHCICmdHdr *)ad->lst)[ncq_tfs->slot];
>
> Is there a guarantee that ->lst has been mapped? Maybe pr->cmd &
> PORT_CMD_START was 0.
>
> We need to check that the HBA is in a valid state for NCQ
> processing before attempting this.
>
Slipped my mind, there should be no way to have ncq_tfs->halt set
under normal circumstances except if the engine is started (and lst is
mapped.)
That said, stream corruption is always possible, so I'll add another
validation check.
An "is not null" check here should be sufficient, due to the "if not
halt" continue up above.
Thanks,
- --js
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=XQh5
-----END PGP SIGNATURE-----
next prev parent reply other threads:[~2015-06-26 16:46 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-23 0:20 [Qemu-devel] [PATCH 00/16] ahci: ncq cleanup, part 2 John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 01/16] ide: add limit to .prepare_buf() John Snow
2015-06-26 14:32 ` Stefan Hajnoczi
2015-06-26 18:16 ` John Snow
2015-06-29 13:34 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-06-29 18:52 ` John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 02/16] ahci: stash ncq command John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 03/16] ahci: assert is_ncq for process_ncq John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 04/16] ahci: refactor process_ncq_command John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 05/16] ahci: factor ncq_finish out of ncq_cb John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 06/16] ahci: record ncq failures John Snow
2015-06-26 15:35 ` Stefan Hajnoczi
2015-06-26 18:27 ` John Snow
2015-06-29 14:10 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-06-29 14:24 ` Stefan Hajnoczi
2015-06-29 15:42 ` John Snow
2015-06-29 15:47 ` John Snow
2015-06-30 13:56 ` Stefan Hajnoczi
2015-06-23 0:21 ` [Qemu-devel] [PATCH 07/16] ahci: kick NCQ queue John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 08/16] ahci: correct types in NCQTransferState John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 09/16] ahci: correct ncq sector count John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 10/16] qtest/ahci: halted NCQ test John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 11/16] ahci: add cmd header to ncq transfer state John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 12/16] ahci: ncq migration John Snow
2015-06-26 15:48 ` Stefan Hajnoczi
2015-06-26 16:46 ` John Snow [this message]
2015-06-29 14:25 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-06-23 0:21 ` [Qemu-devel] [PATCH 13/16] ahci: add get_cmd_header helper John Snow
2015-06-26 15:51 ` Stefan Hajnoczi
2015-06-26 18:32 ` John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 14/16] ahci: Do not map cmd_fis to generate response John Snow
2015-06-26 15:59 ` Stefan Hajnoczi
2015-06-26 17:31 ` John Snow
2015-06-29 14:51 ` Stefan Hajnoczi
2015-06-29 15:07 ` John Snow
2015-06-30 14:50 ` Stefan Hajnoczi
2015-06-23 0:21 ` [Qemu-devel] [PATCH 15/16] qtest/ahci: halted ncq migration test John Snow
2015-06-23 0:21 ` [Qemu-devel] [PATCH 16/16] ahci: fix sdb fis semantics John Snow
2015-06-26 16:11 ` Stefan Hajnoczi
2015-06-26 17:36 ` John Snow
2015-06-29 14:52 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-06-26 16:11 ` [Qemu-devel] [PATCH 00/16] ahci: ncq cleanup, part 2 Stefan Hajnoczi
2015-06-26 19:27 ` John Snow
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=558D81CF.3000208@redhat.com \
--to=jsnow@redhat.com \
--cc=kwolf@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--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).