On Wed, Feb 24, 2016 at 01:43:32PM -0600, Bjorn Helgaas wrote:
> This is a revision of Jacek's v3 posting:
> http://lkml.kernel.org/r/1454152012-46337-1-git-send-email-jacek.lawrynow...@intel.com
>
> The changes from v3 are:
>
> - Split into smaller patches for reviewability
> - Move printk w
On Mon, Dec 28, 2015 at 8:17 AM, Adam Morrison wrote:
> From: Omer Peleg
>
> This patch introduces global and per-CPU caches of IOVAs, so that
> CPUs can usually allocate IOVAs without taking the rbtree spinlock
> to do so. The caching is based on magazines, as described in "Magazines
> and Vmem
Reviewed-by: Ben Serebrin
Looks good.
On Mon, Dec 28, 2015 at 8:17 AM, Adam Morrison wrote:
> From: Omer Peleg
>
> Make intel-iommu map/unmap/invalidate work with IOVA pfns instead of
> pointers to "struct iova". This avoids using the iova struct from the IOVA
> red-black tree and the resultin
Reviewed-by: Ben Serebrin
Looks good!
On Mon, Dec 28, 2015 at 8:17 AM, Adam Morrison wrote:
> From: Omer Peleg
>
> This patch avoids taking the device_domain_lock in iommu_flush_dev_iotlb()
> for domains with no dev iotlb devices.
>
> Signed-off-by: Omer Peleg
> [m...@cs.technion.ac.il: rebas
(From Adam's slightly newer patchset sent via email)
@@ -3542,6 +3543,7 @@ static void flush_unmaps(struct
deferred_flush_data *flush_data)
struct deferred_flush_entry *entry =
&flush_table->entries[j];
struct iova *iova = entry->iova;
+ unsigned long nrpages = entry->nrpages;
Is this alway
Reviewed-by: Ben Serebrin
Looks good.
On Mon, Dec 28, 2015 at 8:16 AM, Adam Morrison wrote:
> From: Omer Peleg
>
> Change flush_unmaps() to correctly pass iommu_flush_iotlb_psi()
> dma addresses. (Intel mm and dma have the same size for pages
> at the moment, but this usage improves consisten
Reviewed-by: Ben Serebrin
Commit message nit: s/defererred/deferred/
intel-iommu.c Line 3615: I'd suggest making the 10ms into a #defined constant.
On Mon, Dec 28, 2015 at 8:16 AM, Adam Morrison wrote:
> From: Omer Peleg
>
> The IOMMU's IOTLB invalidation is a costly process. When iommu mode
Reviewed-by: sereb...@google.com
Looks good!
On Mon, Dec 28, 2015 at 8:15 AM, Adam Morrison wrote:
> From: Omer Peleg
>
> Currently, deferred flushes' info is striped between several lists in
> the flush tables. Instead, move all information about a specific flush
> to a single entry in this ta
On Mon, Apr 11, 2016 at 10:59:48AM -0400, Mark Hounschell wrote:
> Yes this patch fixes my problem.
Awesome, thanks for testing. I'll send it upstream soon and then it
should hit the various stable-kernels.
Joerg
___
iommu mailing list
iommu@l
On 04/11/2016 09:54 AM, Joerg Roedel wrote:
Hi Mark,
On Fri, Mar 18, 2016 at 01:24:27PM -0400, Mark Hounschell wrote:
Should I submit a kernel bug report on this?
I just fixed a regression in v4.4. Can you please try the attached patch
and report if it fixes your problem too? The patch is aga
On Mon, Apr 11, 2016 at 10:14:46AM +0300, Dan Carpenter wrote:
> "devid" needs to be signed for the error handling to work.
>
> Fixes:b097d11a0fa3f ('iommu/amd: Manage iommu_group for ACPI HID devices')
> Signed-off-by: Dan Carpenter
Applied, thanks.
On Fri, Apr 08, 2016 at 10:17:03PM +0200, Tomasz Golinski wrote:
> >> > Can you
> >> > guys please test if the attached patch fixes the issues for you as it
> >> > did for me?
>
> Tested, works as expected. Thanks. Attaching dmesg from patched kernel
> in case you need it.
Thanks for verifying th
Hi Mark,
On Fri, Mar 18, 2016 at 01:24:27PM -0400, Mark Hounschell wrote:
> Should I submit a kernel bug report on this?
I just fixed a regression in v4.4. Can you please try the attached patch
and report if it fixes your problem too? The patch is against v4.6-rc2.
Thanks,
Joerg
>From
Am 11.04.2016 um 15:39 schrieb Oded Gabbay:
On Mon, Apr 11, 2016 at 4:28 PM, Christian König
wrote:
Am 09.04.2016 um 02:25 schrieb Luis R. Rodriguez:
On Tue, Mar 29, 2016 at 10:41 AM, Luis R. Rodriguez wrote:
We need to ensure amd iommu v2 initializes before
driver uses such as drivers/gpu/d
On Wed, Apr 06, 2016 at 07:59:31PM +0530, Sricharan R wrote:
> The driver currently works based on platform data. Remove this
> and add support for DT. A single master can have multiple ports
> connected to more than one iommu.
>
> master
> |
>
On Mon, Apr 11, 2016 at 4:28 PM, Christian König
wrote:
>
> Am 09.04.2016 um 02:25 schrieb Luis R. Rodriguez:
>>
>> On Tue, Mar 29, 2016 at 10:41 AM, Luis R. Rodriguez
>> wrote:
>>>
>>> We need to ensure amd iommu v2 initializes before
>>> driver uses such as drivers/gpu/drm/amd/amdkfd/kfd_modul
Am 09.04.2016 um 02:25 schrieb Luis R. Rodriguez:
On Tue, Mar 29, 2016 at 10:41 AM, Luis R. Rodriguez wrote:
We need to ensure amd iommu v2 initializes before
driver uses such as drivers/gpu/drm/amd/amdkfd/kfd_module.c,
to do this make its init routine a subsys_initcall() which
ensures its load
Stop wasting IOVA space by over-aligning scatterlist segments for a
theoretical worst-case segment boundary mask, and instead take the real
limits into account to merge consecutive segments wherever appropriate,
so our callers can benefit from getting back nicely simplified lists.
This also repres
On Fri, 2016-04-08 at 10:34 -0700, Doug Anderson wrote:
> Hi,
>
> On Fri, Apr 8, 2016 at 10:30 AM, Will Deacon wrote:
> >> > Am I barking up the wrong tree?
> >>
> >> I don't think min_order can be negative. Certainly we could enter the
> >> loop with order == 0 and min_order == 0, though.
> >
>
> From: Dan Carpenter [mailto:dan.carpen...@oracle.com]
> Sent: Monday, April 11, 2016 3:15 PM
> To: Joerg Roedel; Wan, Vincent
> Cc: iommu@lists.linux-foundation.org; linux-ker...@vger.kernel.org; kernel-
> janit...@vger.kernel.org
> Subject: [patch] iommu/amd: Signedness bug in acpihid_device_gro
"devid" needs to be signed for the error handling to work.
Fixes:b097d11a0fa3f ('iommu/amd: Manage iommu_group for ACPI HID devices')
Signed-off-by: Dan Carpenter
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index c430c10..12f7779 100644
--- a/drivers/iommu/amd_iommu.c
+++
21 matches
Mail list logo
