RE: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-09-04 Thread Sathya Perla
-Original Message- From: Yijing Wang [mailto:wangyij...@huawei.com] [Adding the Emulex driver developers to Cc for some input on the device, and why it might use wrong request ids] On Mon, Aug 25, 2014 at 02:44:59PM +0800, Yijing Wang wrote: We found some strange devices in

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-09-04 Thread Yijing Wang
Could you also please send the me the FW version (output of ethtool -i eth0) Hi Sathya, thanks for your help. FW version is 4.6.247.5 driver version is 4.4.161.0s Hi Wang, this issue (a FW bug) was fixed in the 4.9 FW series. The HP qualified FW (ver 4.9.416.0) is available at

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-28 Thread Linda Knippers
On 8/25/2014 9:51 PM, Yijing Wang wrote: On 2014/8/25 23:04, David Woodhouse wrote: On Mon, 2014-08-25 at 12:11 +, Sathya Perla wrote: Hi Wang, from the kernel log I can see that the faulting address 0xbdf7 falls in the RMRR range the BIOS requested: [0.111343] DMAR: RMRR base:

RE: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread bharat.bhus...@freescale.com
-Original Message- From: iommu-boun...@lists.linux-foundation.org [mailto:iommu- boun...@lists.linux-foundation.org] On Behalf Of Yijing Wang Sent: Monday, August 25, 2014 12:15 PM To: Joerg Roedel Cc: iommu@lists.linux-foundation.org; David Woodhouse; Jiang Liu Subject: [PATCH

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread Yijing Wang
+/* We found some strange devices in HP c7000 and other platforms, they + * can not be enumerated by OS, and they did DMA read/write without + * driver management. if we open iommu in these platforms, the DMA read/write + * will be blocked by IOMMU hardware. Currently, we only

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread Joerg Roedel
[Adding the Emulex driver developers to Cc for some input on the device, and why it might use wrong request ids] On Mon, Aug 25, 2014 at 02:44:59PM +0800, Yijing Wang wrote: We found some strange devices in HP C7000 and Huawei Storage Server. These devices can not be enumerated by OS, but they

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread Yijing Wang
On 2014/8/25 20:11, Sathya Perla wrote: -Original Message- From: Yijing Wang [mailto:wangyij...@huawei.com] On 2014/8/25 17:32, Sathya Perla wrote: -Original Message- From: Joerg Roedel [mailto:j...@8bytes.org] [Adding the Emulex driver developers to Cc for some input on

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread David Woodhouse
On Mon, 2014-08-25 at 12:11 +, Sathya Perla wrote: Hi Wang, from the kernel log I can see that the faulting address 0xbdf7 falls in the RMRR range the BIOS requested: [0.111343] DMAR: RMRR base: 0x00bdf6f000 end: 0x00bdf7efff We can't see which *devices* that RMRR was

RE: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread Sathya Perla
-Original Message- From: Yijing Wang [mailto:wangyij...@huawei.com] On 2014/8/25 17:32, Sathya Perla wrote: -Original Message- From: Joerg Roedel [mailto:j...@8bytes.org] [Adding the Emulex driver developers to Cc for some input on the device, and why it might use

RE: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread Sathya Perla
-Original Message- From: Joerg Roedel [mailto:j...@8bytes.org] [Adding the Emulex driver developers to Cc for some input on the device, and why it might use wrong request ids] On Mon, Aug 25, 2014 at 02:44:59PM +0800, Yijing Wang wrote: We found some strange devices in HP C7000

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread Yijing Wang
On 2014/8/25 23:04, David Woodhouse wrote: On Mon, 2014-08-25 at 12:11 +, Sathya Perla wrote: Hi Wang, from the kernel log I can see that the faulting address 0xbdf7 falls in the RMRR range the BIOS requested: [0.111343] DMAR: RMRR base: 0x00bdf6f000 end: 0x00bdf7efff

Re: [PATCH v2] iommu/vt-d: Fix broken device issue when using iommu=pt

2014-08-25 Thread Yijing Wang
On 2014/8/25 17:15, Joerg Roedel wrote: [Adding the Emulex driver developers to Cc for some input on the device, and why it might use wrong request ids] Thanks! On Mon, Aug 25, 2014 at 02:44:59PM +0800, Yijing Wang wrote: We found some strange devices in HP C7000 and Huawei Storage