All of lore.kernel.org
 help / color / mirror / Atom feed
From: Xunlei Pang <xpang@redhat.com>
To: Joerg Roedel <joro@8bytes.org>, Baoquan He <bhe@redhat.com>
Cc: Don Brace <don.brace@microsemi.com>,
	kexec@lists.infradead.org, Dave Young <dyoung@redhat.com>,
	xlpang@redhat.com, LKML <linux-kernel@vger.kernel.org>,
	iommu@lists.linux-foundation.org,
	Myron Stowe <myron.stowe@gmail.com>,
	Myron Stowe <myron.stowe@redhat.com>,
	David Woodhouse <dwmw2@infradead.org>
Subject: Re: [PATCH] iommu/vt-d: Flush old iotlb for kdump when the device gets context mapped
Date: Thu, 1 Dec 2016 10:15:45 +0800	[thread overview]
Message-ID: <583F87D1.6030402@redhat.com> (raw)
In-Reply-To: <20161130142642.GJ2078@8bytes.org>

On 11/30/2016 at 10:26 PM, Joerg Roedel wrote:
> On Wed, Nov 30, 2016 at 06:23:34PM +0800, Baoquan He wrote:
>> OK, talked with Xunlei. The old cache could be entry with present bit
>> set.
> -EPARSE
>
> Anyway, what I was trying to say is, that the IOMMU TLB is tagged with
> domain-ids, and that there is also a context-cache which maps device-ids
> to domain-ids.
>
> If we update the context entry then we need to flush only the context
> entry, as it will point to a new domain-id then and future IOTLB lookups
> in the IOMMU will be using the new domain-id and do not match the old
> entries.

Hi Joerg,

Thanks for the explanation, and we still need to flush context cache using old domain-id, right?
How about the following update?

index 3965e73..624eac9 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -2024,6 +2024,25 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
        if (context_present(context))
                goto out_unlock;
 
+       /*
+        * For kdump cases, old valid entries may be cached due to the
+        * in-flight DMA and copied pgtable, but there is no unmapping
+        * behaviour for them, thus we need an explicit cache flush for
+        * the newly-mapped device. For kdump, at this point, the device
+        * is supposed to finish reset at its driver probe stage, so no
+        * in-flight DMA will exist, and we don't need to worry anymore
+        * hereafter.
+        */
+       if (context_copied(context)) {
+               u16 did_old = context_domain_id(context);
+
+               if (did_old >= 0 && did_old < cap_ndoms(iommu->cap))
+                       iommu->flush.flush_context(iommu, did_old,
+                                                  (((u16)bus) << 8) | devfn,
+                                                  DMA_CCMD_MASK_NOBIT,
+                                                  DMA_CCMD_DEVICE_INVL);
+       }
+
        pgd = domain->pgd;


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

WARNING: multiple messages have this Message-ID (diff)
From: Xunlei Pang <xpang-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Joerg Roedel <joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>,
	Baoquan He <bhe-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Cc: Don Brace <don.brace-dzo6w/eZyo2tG0bUXCXiUA@public.gmane.org>,
	kexec-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
	Dave Young <dyoung-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	LKML <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
	Myron Stowe <myron.stowe-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Subject: Re: [PATCH] iommu/vt-d: Flush old iotlb for kdump when the device gets context mapped
Date: Thu, 1 Dec 2016 10:15:45 +0800	[thread overview]
Message-ID: <583F87D1.6030402@redhat.com> (raw)
In-Reply-To: <20161130142642.GJ2078-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>

On 11/30/2016 at 10:26 PM, Joerg Roedel wrote:
> On Wed, Nov 30, 2016 at 06:23:34PM +0800, Baoquan He wrote:
>> OK, talked with Xunlei. The old cache could be entry with present bit
>> set.
> -EPARSE
>
> Anyway, what I was trying to say is, that the IOMMU TLB is tagged with
> domain-ids, and that there is also a context-cache which maps device-ids
> to domain-ids.
>
> If we update the context entry then we need to flush only the context
> entry, as it will point to a new domain-id then and future IOTLB lookups
> in the IOMMU will be using the new domain-id and do not match the old
> entries.

Hi Joerg,

Thanks for the explanation, and we still need to flush context cache using old domain-id, right?
How about the following update?

index 3965e73..624eac9 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -2024,6 +2024,25 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
        if (context_present(context))
                goto out_unlock;
 
+       /*
+        * For kdump cases, old valid entries may be cached due to the
+        * in-flight DMA and copied pgtable, but there is no unmapping
+        * behaviour for them, thus we need an explicit cache flush for
+        * the newly-mapped device. For kdump, at this point, the device
+        * is supposed to finish reset at its driver probe stage, so no
+        * in-flight DMA will exist, and we don't need to worry anymore
+        * hereafter.
+        */
+       if (context_copied(context)) {
+               u16 did_old = context_domain_id(context);
+
+               if (did_old >= 0 && did_old < cap_ndoms(iommu->cap))
+                       iommu->flush.flush_context(iommu, did_old,
+                                                  (((u16)bus) << 8) | devfn,
+                                                  DMA_CCMD_MASK_NOBIT,
+                                                  DMA_CCMD_DEVICE_INVL);
+       }
+
        pgd = domain->pgd;

WARNING: multiple messages have this Message-ID (diff)
From: Xunlei Pang <xpang@redhat.com>
To: Joerg Roedel <joro@8bytes.org>, Baoquan He <bhe@redhat.com>
Cc: xlpang@redhat.com, Don Brace <don.brace@microsemi.com>,
	Myron Stowe <myron.stowe@redhat.com>,
	kexec@lists.infradead.org, LKML <linux-kernel@vger.kernel.org>,
	iommu@lists.linux-foundation.org,
	Myron Stowe <myron.stowe@gmail.com>,
	Dave Young <dyoung@redhat.com>,
	David Woodhouse <dwmw2@infradead.org>
Subject: Re: [PATCH] iommu/vt-d: Flush old iotlb for kdump when the device gets context mapped
Date: Thu, 1 Dec 2016 10:15:45 +0800	[thread overview]
Message-ID: <583F87D1.6030402@redhat.com> (raw)
In-Reply-To: <20161130142642.GJ2078@8bytes.org>

On 11/30/2016 at 10:26 PM, Joerg Roedel wrote:
> On Wed, Nov 30, 2016 at 06:23:34PM +0800, Baoquan He wrote:
>> OK, talked with Xunlei. The old cache could be entry with present bit
>> set.
> -EPARSE
>
> Anyway, what I was trying to say is, that the IOMMU TLB is tagged with
> domain-ids, and that there is also a context-cache which maps device-ids
> to domain-ids.
>
> If we update the context entry then we need to flush only the context
> entry, as it will point to a new domain-id then and future IOTLB lookups
> in the IOMMU will be using the new domain-id and do not match the old
> entries.

Hi Joerg,

Thanks for the explanation, and we still need to flush context cache using old domain-id, right?
How about the following update?

index 3965e73..624eac9 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -2024,6 +2024,25 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
        if (context_present(context))
                goto out_unlock;
 
+       /*
+        * For kdump cases, old valid entries may be cached due to the
+        * in-flight DMA and copied pgtable, but there is no unmapping
+        * behaviour for them, thus we need an explicit cache flush for
+        * the newly-mapped device. For kdump, at this point, the device
+        * is supposed to finish reset at its driver probe stage, so no
+        * in-flight DMA will exist, and we don't need to worry anymore
+        * hereafter.
+        */
+       if (context_copied(context)) {
+               u16 did_old = context_domain_id(context);
+
+               if (did_old >= 0 && did_old < cap_ndoms(iommu->cap))
+                       iommu->flush.flush_context(iommu, did_old,
+                                                  (((u16)bus) << 8) | devfn,
+                                                  DMA_CCMD_MASK_NOBIT,
+                                                  DMA_CCMD_DEVICE_INVL);
+       }
+
        pgd = domain->pgd;

  reply	other threads:[~2016-12-01  2:14 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-16  9:02 [PATCH] iommu/vt-d: Flush old iotlb for kdump when the device gets context mapped Xunlei Pang
2016-11-16  9:02 ` Xunlei Pang
2016-11-16  9:02 ` Xunlei Pang
2016-11-16  9:13 ` Xunlei Pang
2016-11-16  9:13   ` Xunlei Pang
2016-11-16  9:13   ` Xunlei Pang
2016-11-16 14:58   ` Myron Stowe
2016-11-16 14:58     ` Myron Stowe
2016-11-16 14:58     ` Myron Stowe
2016-11-17  2:47     ` Xunlei Pang
2016-11-17  2:47       ` Xunlei Pang
2016-11-17  2:47       ` Xunlei Pang
     [not found]       ` <582D1A40.409-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-11-29 14:35         ` Joerg Roedel
2016-11-29 14:35           ` Joerg Roedel
2016-11-30  8:15           ` Xunlei Pang
2016-11-30  8:15             ` Xunlei Pang
2016-11-30  8:15             ` Xunlei Pang
2016-11-30  9:03             ` Baoquan He
2016-11-30  9:03               ` Baoquan He
2016-11-30  9:03               ` Baoquan He
2016-11-30  9:53               ` Baoquan He
2016-11-30  9:53                 ` Baoquan He
2016-11-30  9:53                 ` Baoquan He
2016-11-30 10:23                 ` Baoquan He
2016-11-30 10:23                   ` Baoquan He
2016-11-30 10:23                   ` Baoquan He
2016-11-30 14:26                   ` Joerg Roedel
2016-11-30 14:26                     ` Joerg Roedel
2016-12-01  2:15                     ` Xunlei Pang [this message]
2016-12-01  2:15                       ` Xunlei Pang
2016-12-01  2:15                       ` Xunlei Pang
     [not found]                       ` <583F87D1.6030402-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-12-01 10:33                         ` Joerg Roedel
2016-12-01 10:33                           ` Joerg Roedel
2016-12-01 11:44                           ` Xunlei Pang
2016-12-01 11:44                             ` Xunlei Pang
2016-12-01 11:44                             ` Xunlei Pang

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=583F87D1.6030402@redhat.com \
    --to=xpang@redhat.com \
    --cc=bhe@redhat.com \
    --cc=don.brace@microsemi.com \
    --cc=dwmw2@infradead.org \
    --cc=dyoung@redhat.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=joro@8bytes.org \
    --cc=kexec@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=myron.stowe@gmail.com \
    --cc=myron.stowe@redhat.com \
    --cc=xlpang@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 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.