From: Joerg Roedel <joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
To: KarimAllah Ahmed <karahmed-ebkRAfMGSJGzQB+pC5nmwQ@public.gmane.org>
Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
David Woodhouse <dwmw-vV1OtcyAfmbQXOPxS62xeg@public.gmane.org>,
Anthony Liguori
<aliguori-vV1OtcyAfmbQT0dZR+AlfA@public.gmane.org>
Subject: Re: [PATCH] iommu: intel: Flush the IOTLB to get rid of the initial kdump mappings
Date: Wed, 17 May 2017 14:46:30 +0200 [thread overview]
Message-ID: <20170517124630.GC11068@8bytes.org> (raw)
In-Reply-To: <1494009599-20962-1-git-send-email-karahmed-ebkRAfMGSJGzQB+pC5nmwQ@public.gmane.org>
On Fri, May 05, 2017 at 11:39:59AM -0700, KarimAllah Ahmed wrote:
> Ever since commit 091d42e43d ("iommu/vt-d: Copy translation tables from
> old kernel")
Please put the commit-information into a Fixes: tag next time. I'll
added a Fixes tag to the commit this time.
> the kdump kernel copies the IOMMU context tables from the
> previous kernel. Each device mappings will be destroyed once the driver
> for the respective device takes over.
>
> This unfortunately breaks the workflow of mapping and unmapping a new
> context to the IOMMU. The mapping function assumes that either:
>
> 1) Unmapping did the proper IOMMU flushing and it only ever flush if the
> IOMMU unit supports caching invalid entries.
> 2) The system just booted and the initialization code took care of
> flushing all IOMMU caches.
>
> This assumption is not true for the kdump kernel since the context
> tables have been copied from the previous kernel and translations could
> have been cached ever since. So make sure to flush the IOTLB as well
> when we destroy these old copied mappings.
>
> Cc: Joerg Roedel <joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
> Cc: David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
> Cc: David Woodhouse <dwmw-vV1OtcyAfmbQXOPxS62xeg@public.gmane.org>
> Cc: Anthony Liguori <aliguori-vV1OtcyAfmbQT0dZR+AlfA@public.gmane.org>
> Signed-off-by: KarimAllah Ahmed <karahmed-ebkRAfMGSJGzQB+pC5nmwQ@public.gmane.org>
Applied to iommu/fixes, thanks.
WARNING: multiple messages have this Message-ID (diff)
From: Joerg Roedel <joro@8bytes.org>
To: KarimAllah Ahmed <karahmed@amazon.de>
Cc: iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org,
David Woodhouse <dwmw2@infradead.org>,
David Woodhouse <dwmw@amazon.co.uk>,
Anthony Liguori <aliguori@amazon.com>
Subject: Re: [PATCH] iommu: intel: Flush the IOTLB to get rid of the initial kdump mappings
Date: Wed, 17 May 2017 14:46:30 +0200 [thread overview]
Message-ID: <20170517124630.GC11068@8bytes.org> (raw)
In-Reply-To: <1494009599-20962-1-git-send-email-karahmed@amazon.de>
On Fri, May 05, 2017 at 11:39:59AM -0700, KarimAllah Ahmed wrote:
> Ever since commit 091d42e43d ("iommu/vt-d: Copy translation tables from
> old kernel")
Please put the commit-information into a Fixes: tag next time. I'll
added a Fixes tag to the commit this time.
> the kdump kernel copies the IOMMU context tables from the
> previous kernel. Each device mappings will be destroyed once the driver
> for the respective device takes over.
>
> This unfortunately breaks the workflow of mapping and unmapping a new
> context to the IOMMU. The mapping function assumes that either:
>
> 1) Unmapping did the proper IOMMU flushing and it only ever flush if the
> IOMMU unit supports caching invalid entries.
> 2) The system just booted and the initialization code took care of
> flushing all IOMMU caches.
>
> This assumption is not true for the kdump kernel since the context
> tables have been copied from the previous kernel and translations could
> have been cached ever since. So make sure to flush the IOTLB as well
> when we destroy these old copied mappings.
>
> Cc: Joerg Roedel <joro@8bytes.org>
> Cc: David Woodhouse <dwmw2@infradead.org>
> Cc: David Woodhouse <dwmw@amazon.co.uk>
> Cc: Anthony Liguori <aliguori@amazon.com>
> Signed-off-by: KarimAllah Ahmed <karahmed@amazon.de>
Applied to iommu/fixes, thanks.
next prev parent reply other threads:[~2017-05-17 12:46 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-05 18:39 [PATCH] iommu: intel: Flush the IOTLB to get rid of the initial kdump mappings KarimAllah Ahmed
2017-05-05 18:56 ` David Woodhouse
[not found] ` <1494009599-20962-1-git-send-email-karahmed-ebkRAfMGSJGzQB+pC5nmwQ@public.gmane.org>
2017-05-17 12:46 ` Joerg Roedel [this message]
2017-05-17 12:46 ` Joerg Roedel
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=20170517124630.GC11068@8bytes.org \
--to=joro-zlv9swrftaidnm+yrofe0a@public.gmane.org \
--cc=aliguori-vV1OtcyAfmbQT0dZR+AlfA@public.gmane.org \
--cc=dwmw-vV1OtcyAfmbQXOPxS62xeg@public.gmane.org \
--cc=dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=karahmed-ebkRAfMGSJGzQB+pC5nmwQ@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@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.