From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Paul Durrant <Paul.Durrant@citrix.com>,
Xen-devel <xen-devel@lists.xen.org>
Cc: Brian Woods <brian.woods@amd.com>, Wei Liu <wei.liu2@citrix.com>,
Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>,
Jan Beulich <JBeulich@suse.com>,
Roger Pau Monne <roger.pau@citrix.com>
Subject: Re: [PATCH 1/2] x86/SVM-IOMMU: Don't opencode memcpy() in queue_iommu_command()
Date: Mon, 24 Sep 2018 13:19:22 +0100 [thread overview]
Message-ID: <adca3d65-5c57-794a-b3e2-fc062b8d816a@citrix.com> (raw)
In-Reply-To: <83196ca8ba4740f891892a50f2aaeb9d@AMSPEX02CL03.citrite.net>
On 24/09/18 13:09, Paul Durrant wrote:
>>
>>>> @@ -35,12 +34,9 @@ static int queue_iommu_command(struct amd_iommu
>> *iommu,
>>>> u32 cmd[])
>>>> IOMMU_CMD_BUFFER_HEAD_OFFSET));
>>>> if ( head != tail )
>>>> {
>>>> - cmd_buffer = (u32 *)(iommu->cmd_buffer.buffer +
>>>> - (iommu->cmd_buffer.tail *
>>>> - IOMMU_CMD_BUFFER_ENTRY_SIZE));
>>>> -
>>>> - for ( i = 0; i < IOMMU_CMD_BUFFER_U32_PER_ENTRY; i++ )
>>>> - cmd_buffer[i] = cmd[i];
>>>> + memcpy(iommu->cmd_buffer.buffer +
>>>> + (iommu->cmd_buffer.tail * IOMMU_CMD_BUFFER_ENTRY_SIZE),
>>>> + cmd, IOMMU_CMD_BUFFER_ENTRY_SIZE);
>>> ...since the built-in memcpy may not guarantee to the copy in 4 byte
>> chunks in ascending order.
>>
>> "No functional change" != "The binary is identical".
>>
>> The functionality of queue_iommu_command() does not change, even if it's
>> code generation does. It is just copying bytes into a shared ring,
>> bounded later by updating the tail pointer.
> Yes, my point is that the ring is shared and so DMA by the h/w may race. This is clearly not a good situation but the fact that the code generation may change may have side effects.
This is writing past the tail pointer, and then updating the tail
pointer to match.
Hardware conforming to the spec won't read any of the data until it is
all ready.
~Andrew
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
next prev parent reply other threads:[~2018-09-24 12:19 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-24 10:55 [PATCH 1/2] x86/SVM-IOMMU: Don't opencode memcpy() in queue_iommu_command() Andrew Cooper
2018-09-24 10:55 ` [PATCH 2/2] x86/SVM-IOMMU: Drop get_field_from_byte() Andrew Cooper
2018-09-24 12:07 ` Paul Durrant
2018-09-24 12:17 ` Jan Beulich
2018-10-04 23:00 ` Woods, Brian
2018-10-05 8:22 ` Roger Pau Monné
2018-09-24 11:59 ` [PATCH 1/2] x86/SVM-IOMMU: Don't opencode memcpy() in queue_iommu_command() Paul Durrant
2018-09-24 12:05 ` Andrew Cooper
2018-09-24 12:09 ` Paul Durrant
2018-09-24 12:16 ` Jan Beulich
2018-09-24 12:18 ` Paul Durrant
2018-09-24 12:24 ` Andrew Cooper
2018-09-24 12:28 ` Paul Durrant
2018-09-24 12:32 ` Jan Beulich
2018-09-24 12:19 ` Andrew Cooper [this message]
2018-09-24 12:09 ` Jan Beulich
2018-09-24 12:14 ` Jan Beulich
2018-10-04 23:03 ` Woods, Brian
2018-10-05 8:22 ` Roger Pau Monné
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=adca3d65-5c57-794a-b3e2-fc062b8d816a@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=JBeulich@suse.com \
--cc=Paul.Durrant@citrix.com \
--cc=brian.woods@amd.com \
--cc=roger.pau@citrix.com \
--cc=suravee.suthikulpanit@amd.com \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xen.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.