Hi,
We encounter a issue with the patch: our platform is ARM64, and we run
DPDK with smmu disable on VM (without iommu=smmuv3 etc),
so we use noiommu mode with enable_unsafe_noiommu_mode=1 to passthrough
the device to VM with following steps (those steps are on VM) :
insmod vfio.ko
From: Xiang Chen
Barry Song will not focus on this area, and Xiang Chen will continue his
work to maintain this module.
Signed-off-by: Xiang Chen
---
MAINTAINERS | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index ea3e6c914384..48335022b0e4
From: Xiang Chen
The series ("Optimizing iommu_[map/unmap] performance") improve the
iommu_[map/unmap] performance. Based on the series, implement [map/unmap]_pages
callbacks for ARM SMMUV3.
Use tool dma_map_benchmark to test the latency of map/unmap, and it promotes
much on it. The test result
From: Xiang Chen
Implement the map_pages() callback for ARM SMMUV3 driver to allow calls
from iommu_map to map multiple pages of the same size in one call.
Also remove the map() callback for the ARM SMMUV3 driver as it will no
longer be used.
Signed-off-by: Xiang Chen
---
From: Xiang Chen
Implement the unmap_pages() callback for ARM SMMUV3 driver to allow calls
from iommu_unmap to unmap multiple pages of the same size in one call.
Also remove the unmap() callback for the ARM SMMUV3 driver as it will
no longer be used.
Signed-off-by: Xiang Chen
---
在 2021/7/29 18:59, Robin Murphy 写道:
On 2021-07-29 03:55, chenxiang (M) wrote:
Hi Robin,
在 2021/7/28 23:58, Robin Murphy 写道:
Hi all,
Here's v2 where things start to look more realistic, hence the expanded
CC list. The patches are now based on the current iommu/core branch to
take John's
Hi Robin,
在 2021/7/28 23:58, Robin Murphy 写道:
Hi all,
Here's v2 where things start to look more realistic, hence the expanded
CC list. The patches are now based on the current iommu/core branch to
take John's iommu_set_dma_strict() cleanup into account.
The series remiains in two (or
在 2021/7/15 9:23, Lu Baolu 写道:
On 7/14/21 10:24 PM, Georgi Djakov wrote:
On 16.06.21 16:38, Georgi Djakov wrote:
When unmapping a buffer from an IOMMU domain, the IOMMU framework
unmaps
the buffer at a granule of the largest page size that is supported by
the IOMMU hardware and fits within
Hi Robin,
在 2021/6/7 19:23, Robin Murphy 写道:
On 2021-06-07 03:42, chenxiang wrote:
From: Xiang Chen
When rmmod the driver of the last device in the group, cached iovas
are not
used, and it is better to free them to save memories. And also export
function free_rcache_cached_iovas
From: Xiang Chen
When rmmod the driver of the last device in the domain, cached iovas are
not used and it is better to free them to save memories.
Signed-off-by: Xiang Chen
---
drivers/iommu/iommu.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/iommu/iommu.c
From: Xiang Chen
Add a function to free all rcached iovas including cpu rcache iovas and
global rcache iovas, and export it.
Signed-off-by: Xiang Chen
---
drivers/iommu/iova.c | 11 +++
include/linux/iova.h | 5 +
2 files changed, 16 insertions(+)
diff --git
From: Xiang Chen
Use function free_rcached_iovas() to free all rcached iovas instead.
Signed-off-by: Xiang Chen
---
drivers/iommu/iova.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index f595867..59926d5 100644
---
From: Xiang Chen
Add a function iommu_domain_to_iova() to get iova_domain from iommu domain.
Signed-off-by: Xiang Chen
---
drivers/iommu/dma-iommu.c | 7 +++
include/linux/dma-iommu.h | 6 ++
2 files changed, 13 insertions(+)
diff --git a/drivers/iommu/dma-iommu.c
From: Xiang Chen
When rmmod the driver of the last device in the group, cached iovas are not
used, and it is better to free them to save memories. And also export
function free_rcache_cached_iovas() and iommu_domain_to_iova().
Xiang Chen (4):
iommu/iova: add a function to free all rcached
在 2021/5/21 18:36, Robin Murphy 写道:
On 2021-05-21 04:05, chenxiang wrote:
From: Xiang Chen
The issue is reported by tool TscanCode, and it is possible to deference
null pointer when prev is NULL which is the initial value.
No it isn't. This is literally explained in the comment visible
From: Xiang Chen
The issue is reported by tool TscanCode, and it is possible to deference
null pointer when prev is NULL which is the initial value.
Signed-off-by: Xiang Chen
---
drivers/iommu/dma-iommu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
Hi Joerg,
在 2021/5/18 17:08, Joerg Roedel 写道:
On Mon, Apr 19, 2021 at 03:13:35PM +0800, chenxiang wrote:
From: Xiang Chen
It is not necessary to put free_iova_mem() inside of spinlock/unlock
iova_rbtree_lock which only leads to more completion for the spinlock.
It has a small promote
From: Xiang Chen
It is not necessary to put free_iova_mem() inside of spinlock/unlock
iova_rbtree_lock which only leads to more completion for the spinlock.
It has a small promote on the performance after the change. And also
rename private_free_iova() as remove_iova() because the function will
Hi Isaac,
在 2021/4/9 1:13, Isaac J. Manjarres 写道:
Implement the map_pages() callback for the ARM LPAE io-pgtable
format.
Signed-off-by: Isaac J. Manjarres
---
drivers/iommu/io-pgtable-arm.c | 42 ++
1 file changed, 32 insertions(+), 10 deletions(-)
diff
From: Xiang Chen
It is not necessary to put free_iova_mem() inside of spinlock/unlock
iova_rbtree_lock which only leads to more completion for the spinlock.
It has a small promote on the performance after the change. And also
rename private_free_iova() as remove_iova() because the function will
在 2021/4/16 16:53, John Garry 写道:
On 16/04/2021 04:30, chenxiang (M) wrote:
you need to make this:
if (iova)
free_iova_mem(iova);
as free_iova_mem(iova) dereferences iova:
if (iova->pfn_lo != IOVA_ANCHOR)
kmem_cache_free(iova_cache, iova)
So if iova were NULL, we cr
在 2021/4/15 17:49, John Garry 写道:
On 15/04/2021 04:52, chenxiang wrote:
From: Xiang Chen
It is not necessary to put free_iova_mem() inside of spinlock/unlock
iova_rbtree_lock which only leads to more completion for the spinlock.
It has a small promote on the performance after the change
From: Xiang Chen
It is not necessary to put free_iova_mem() inside of spinlock/unlock
iova_rbtree_lock which only leads to more completion for the spinlock.
It has a small promote on the performance after the change. And also
rename private_free_iova() as remove_iova() because the function will
Hi Robin,
在 2021/4/14 21:17, Robin Murphy 写道:
On 2021-04-14 07:38, chenxiang wrote:
From: Xiang Chen
It is not necessary to put free_iova_mem() inside of spinlock/unlock
iova_rbtree_lock which only leads to more completion for the spinlock.
It has a small promote on the performance after
From: Xiang Chen
It is not necessary to put free_iova_mem() inside of spinlock/unlock
iova_rbtree_lock which only leads to more completion for the spinlock.
It has a small promote on the performance after the change.
Signed-off-by: Xiang Chen
---
drivers/iommu/iova.c | 5 +++--
1 file
Hi,
在 2021/2/12 18:50, Joerg Roedel 写道:
On Tue, Feb 09, 2021 at 06:06:20PM +0530, Sai Prakash Ranjan wrote:
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index 5e7fe519430a..6064187d9bb6 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -87,6 +87,7 @@ struct
Hi Isaac,
在 2021/3/31 11:00, Isaac J. Manjarres 写道:
When unmapping a buffer from an IOMMU domain, the IOMMU framework unmaps
the buffer at a granule of the largest page size that is supported by
the IOMMU hardware and fits within the buffer. For every block that
is unmapped, the IOMMU
From: Xiang Chen
It may send a invalid tlb sync for smmuv3 if iotlb_gather is not valid
(iotlb_gather->pgsize = 0). So add a check to avoid invalid iotlb sync
for it.
Signed-off-by: Xiang Chen
---
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 3 +++
1 file changed, 3 insertions(+)
diff --git
在 2021/3/30 17:25, Will Deacon 写道:
On Tue, Mar 30, 2021 at 10:04:53AM +0100, Robin Murphy wrote:
On 2021-03-30 02:22, chenxiang (M) wrote:
Hi Will,
在 2021/3/29 22:45, Will Deacon 写道:
On Sat, Mar 27, 2021 at 02:23:10PM +0800, chenxiang wrote:
From: Xiang Chen
Currently it will send
Hi Will,
在 2021/3/29 22:45, Will Deacon 写道:
On Sat, Mar 27, 2021 at 02:23:10PM +0800, chenxiang wrote:
From: Xiang Chen
Currently it will send a iotlb sync at end of iommu unmap even if
iotlb_gather is not valid (iotlb_gather->pgsize = 0). Actually it is not
necessary, so add a ch
From: Xiang Chen
Currently it will send a iotlb sync at end of iommu unmap even if
iotlb_gather is not valid (iotlb_gather->pgsize = 0). Actually it is not
necessary, so add a check to avoid invalid iotlb sync.
Signed-off-by: Xiang Chen
---
include/linux/iommu.h | 3 +++
1 file changed, 3
From: Xiang Chen
After the change of patch ("iommu: Switch gather->end to the
inclusive end"), the performace drops from 1600+K IOPS to 1200K in our
kunpeng ARM64 platform.
We find that the range [start1, end1) actually is joint from the range
[end1, end2), but it is considered as disjoint
Hi Eric,
在 2021/3/22 17:05, Auger Eric 写道:
Hi Chenxiang,
On 3/22/21 7:40 AM, chenxiang (M) wrote:
Hi Eric,
在 2021/3/20 1:36, Auger Eric 写道:
Hi Chenxiang,
On 3/4/21 8:55 AM, chenxiang (M) wrote:
Hi Eric,
在 2021/2/24 4:56, Eric Auger 写道:
Implement domain-selective, pasid selective
Hi Eric,
在 2021/3/20 1:36, Auger Eric 写道:
Hi Chenxiang,
On 3/4/21 8:55 AM, chenxiang (M) wrote:
Hi Eric,
在 2021/2/24 4:56, Eric Auger 写道:
Implement domain-selective, pasid selective and page-selective
IOTLB invalidations.
Signed-off-by: Eric Auger
---
v13 -> v14:
- Add dom
在 2021/3/19 19:02, Robin Murphy 写道:
On 2021-03-19 01:02, chenxiang (M) wrote:
Hi,
在 2021/3/18 19:01, Robin Murphy 写道:
On 2021-03-18 09:55, chenxiang (M) wrote:
Hi will,
在 2021/3/18 17:34, Will Deacon 写道:
On Thu, Mar 18, 2021 at 11:21:24AM +0800, chenxiang wrote:
From: Xiang Chen
Fix
Hi,
在 2021/3/18 19:01, Robin Murphy 写道:
On 2021-03-18 09:55, chenxiang (M) wrote:
Hi will,
在 2021/3/18 17:34, Will Deacon 写道:
On Thu, Mar 18, 2021 at 11:21:24AM +0800, chenxiang wrote:
From: Xiang Chen
Fix a type "SAC" to "DAC" in the comment of function
iommu_dma_a
Hi will,
在 2021/3/18 17:34, Will Deacon 写道:
On Thu, Mar 18, 2021 at 11:21:24AM +0800, chenxiang wrote:
From: Xiang Chen
Fix a type "SAC" to "DAC" in the comment of function
iommu_dma_alloc_iova().
Signed-off-by: Xiang Chen
---
drivers/iommu/dma-iommu.c | 2 +-
From: Xiang Chen
Currently it only support one page map/unmap once a time for dma-map
benchmark, but there are some other scenaries which need to support for
multi-page map/unmap: for those multi-pages interfaces such as
dma_alloc_coherent() and dma_map_sg(), the time spent on multi-pages
From: Xiang Chen
Fix a type "SAC" to "DAC" in the comment of function
iommu_dma_alloc_iova().
Signed-off-by: Xiang Chen
---
drivers/iommu/dma-iommu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
index
Hi Eric,
在 2021/2/24 4:56, Eric Auger 写道:
Implement domain-selective, pasid selective and page-selective
IOTLB invalidations.
Signed-off-by: Eric Auger
---
v13 -> v14:
- Add domain invalidation
- do global inval when asid is not provided with addr
granularity
v7 -> v8:
- ASID based
Hi Robin,
在 2021/1/29 20:03, Robin Murphy 写道:
On 2021-01-29 09:48, Leizhen (ThunderTown) wrote:
Currently, we are thinking about the solution to the problem.
However, because the end time of v5.11 is approaching, this patch is
sent first.
However, that commit was made for a reason - how
Hi Eric,
在 2020/11/18 19:21, Eric Auger 写道:
Implement domain-selective and page-selective IOTLB invalidations.
Signed-off-by: Eric Auger
---
v7 -> v8:
- ASID based invalidation using iommu_inv_pasid_info
- check ARCHID/PASID flags in addr based invalidation
- use __arm_smmu_tlb_inv_context
在 2021/1/12 4:01, Robin Murphy 写道:
On 2021-01-07 02:45, chenxiang (M) wrote:
Hi Will, Robin or other guys,
When debugging SMMU/SVA issue on huawei ARM64 board, we find that it
lacks of enough debugfs for ARM SMMU driver (such as
the value of STE/CD which we need to check sometimes
Hi Will, Robin or other guys,
When debugging SMMU/SVA issue on huawei ARM64 board, we find that it
lacks of enough debugfs for ARM SMMU driver (such as
the value of STE/CD which we need to check sometimes). Currently it
creates top-level iommu directory in debugfs, but there is no debugfs
44 matches
Mail list logo
