Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-28 Thread Burakov, Anatoly
On 28-Jun-19 2:47 PM, Burakov, Anatoly wrote: On 28-Jun-19 12:38 PM, Takeshi T Yoshimura wrote: To: "Mo, YufengX" , dev@dpdk.org From: "Burakov, Anatoly" Date: 06/26/2019 06:43PM Cc: d...@ibm.com, prad...@us.ibm.com, Takeshi Yoshimura Subject: [EXTERNAL] Re: [dpdk-de

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-28 Thread Burakov, Anatoly
On 28-Jun-19 12:38 PM, Takeshi T Yoshimura wrote: To: "Mo, YufengX" , dev@dpdk.org From: "Burakov, Anatoly" Date: 06/26/2019 06:43PM Cc: d...@ibm.com, prad...@us.ibm.com, Takeshi Yoshimura Subject: [EXTERNAL] Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-28 Thread Takeshi T Yoshimura
>To: "Mo, YufengX" , dev@dpdk.org >From: "Burakov, Anatoly" >Date: 06/26/2019 06:43PM >Cc: d...@ibm.com, prad...@us.ibm.com, Takeshi Yoshimura > >Subject: [EXTERNAL] Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA >area in ppc64le > >On 18-Jun-

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-26 Thread Burakov, Anatoly
On 18-Jun-19 3:37 AM, Mo, YufengX wrote: From: Takeshi Yoshimura In ppc64le, expanding DMA areas always fail because we cannot remove a DMA window. As a result, we cannot allocate more than one memseg in ppc64le. This is because vfio_spapr_dma_mem_map() doesn't unmap all the mapped DMA before r

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-17 Thread Mo, YufengX
Sorry, missed send this mail. Ignore it pls. > -Original Message- > From: Mo, YufengX > Sent: Tuesday, June 18, 2019 10:38 AM > To: Mo, YufengX ; dev@dpdk.org > Cc: d...@ibm.com; prad...@us.ibm.com; Takeshi Yoshimura > Subject: [dpdk-dev] [PATCH] vfio: fix expanding D

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-14 Thread David Marchand
Before submitting further revisions, please check the documentation at http://doc.dpdk.org/guides/contributing/patches.html You are supposed to version your patches and prune old superseded patches in patchwork. Thanks. -- David Marchand On Thu, Jun 13, 2019 at 4:23 AM Takeshi Yoshimura wrote

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-13 Thread David Christensen
Adding the vfio maintainer on the To: line. On 6/12/19 7:22 PM, Takeshi Yoshimura wrote: In ppc64le, expanding DMA areas always fail because we cannot remove a DMA window. As a result, we cannot allocate more than one memseg in ppc64le. This is because vfio_spapr_dma_mem_map() doesn't unmap al

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-12 Thread Takeshi T Yoshimura
>宛先: dev@dpdk.org >送信元: Takeshi Yoshimura >日付: 2019/06/13 11:22AM >Cc: d...@ibm.com, prad...@us.ibm.com, Takeshi Yoshimura > >件名: [EXTERNAL] [PATCH] vfio: fix expanding DMA area in ppc64le > >In ppc64le, expanding DMA areas always fail because we cannot remove >a DMA window. As a result, we cannot

[dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-12 Thread Takeshi Yoshimura
In ppc64le, expanding DMA areas always fail because we cannot remove a DMA window. As a result, we cannot allocate more than one memseg in ppc64le. This is because vfio_spapr_dma_mem_map() doesn't unmap all the mapped DMA before removing the window. This patch fixes this incorrect behavior. I adde

Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-12 Thread Aaron Conole
Takeshi Yoshimura writes: > In ppc64le, expanding DMA areas always fail because we cannot remove > a DMA window. As a result, we cannot allocate more than one memseg in > ppc64le. This is because vfio_spapr_dma_mem_map() doesn't unmap all > the mapped DMA before removing the window. This patch fi

[dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le

2019-06-11 Thread Takeshi Yoshimura
In ppc64le, expanding DMA areas always fail because we cannot remove a DMA window. As a result, we cannot allocate more than one memseg in ppc64le. This is because vfio_spapr_dma_mem_map() doesn't unmap all the mapped DMA before removing the window. This patch fixes this incorrect behavior. I adde