| Message ID | 20220517151915.2212838-1-Frank.Li@nxp.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Enable designware PCI EP EDMA locally | expand |
Hello Vinod, On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > Default Designware EDMA just probe remotely at host side. > This patch allow EDMA driver can probe at EP side. > > 1. Clean up patch > dmaengine: dw-edma: Detach the private data and chip info structures > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > dmaengine: dw-edma: Change rg_region to reg_base in struct > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > 2. Enhance EDMA driver to allow prode eDMA at EP side > dmaengine: dw-edma: Add support for chip specific flags > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > flags (this patch removed at v11 because dma tree already have fixed > patch) > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > dmaengine: dw-edma: Fix programming the source & dest addresses for > ep > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > 4. change pci-epf-test to use EDMA driver to transfer data. > PCI: endpoint: Add embedded DMA controller test > > 5. Using imx8dxl to do test, but some EP functions still have not > upstream yet. So below patch show how probe eDMA driver at EP > controller driver. > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 The series has been hanging out on review for over three months now. It has got to v11 and has been tested on at least two platforms. The original driver maintainer has been silent for all that time (most likely Gustavo dropped the driver maintaining role). Could you please merge it in seeing no comments have been posted for the last several weeks? The PCI Host/EP controller drivers maintainer suggested to get this series via the DMA-engine tree: https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ which is obviously right seeing it mainly concerns the DW eDMA driver. Though after that Lorenzo disappeared as quickly as popped up.) There is one more series depending on the changes in this patchset: https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ Me and Frank already settled all the conflicts and inter-dependencies, so at least his series is more than ready to be merged in into the kernel repo. It would be very good to get it accepted on this merge window so to have the kernel v5.19 with all this changes available. -Sergey > > -- > 2.35.1 >
On Mon, May 23, 2022 at 02:06:47PM +0300, Serge Semin wrote: > Hello Vinod, > > On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > > Default Designware EDMA just probe remotely at host side. > > This patch allow EDMA driver can probe at EP side. > > > > 1. Clean up patch > > dmaengine: dw-edma: Detach the private data and chip info structures > > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > > dmaengine: dw-edma: Change rg_region to reg_base in struct > > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > > > 2. Enhance EDMA driver to allow prode eDMA at EP side > > dmaengine: dw-edma: Add support for chip specific flags > > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > > flags (this patch removed at v11 because dma tree already have fixed > > patch) > > > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > > dmaengine: dw-edma: Fix programming the source & dest addresses for > > ep > > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > > > 4. change pci-epf-test to use EDMA driver to transfer data. > > PCI: endpoint: Add embedded DMA controller test > > > > 5. Using imx8dxl to do test, but some EP functions still have not > > upstream yet. So below patch show how probe eDMA driver at EP > > controller driver. > > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 > > The series has been hanging out on review for over three months now. > It has got to v11 and has been tested on at least two platforms. The > original driver maintainer has been silent for all that time (most > likely Gustavo dropped the driver maintaining role). Could you please > merge it in seeing no comments have been posted for the last several > weeks? The PCI Host/EP controller drivers maintainer suggested to get > this series via the DMA-engine tree: > https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ > which is obviously right seeing it mainly concerns the DW eDMA driver. > Though after that Lorenzo disappeared as quickly as popped up.) > > There is one more series depending on the changes in this > patchset: > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > Me and Frank already settled all the conflicts and inter-dependencies, > so at least his series is more than ready to be merged in into the > kernel repo. It would be very good to get it accepted on this merge > window so to have the kernel v5.19 with all this changes available. Since the v5.19 merge window is already open, it seems doubtful that anybody would merge this so late in the cycle. If Gustavo isn't available or willing to merge it, it looks like Vinod (maintainer of drivers/dma) would be the next logical candidate. I suspect Vinod would appreciate an ack or reviewed-by from Kishon for the last patch because he maintains pci-epf-test.c. I have a couple trivial comments on the pci-epf-test.c (I'll respond there), but I'm not qualified to ack it. Bjorn
On Mon, May 23, 2022 at 1:02 PM Bjorn Helgaas <helgaas@kernel.org> wrote: > > On Mon, May 23, 2022 at 02:06:47PM +0300, Serge Semin wrote: > > Hello Vinod, > > > > On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > > > Default Designware EDMA just probe remotely at host side. > > > This patch allow EDMA driver can probe at EP side. > > > > > > 1. Clean up patch > > > dmaengine: dw-edma: Detach the private data and chip info structures > > > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > > > dmaengine: dw-edma: Change rg_region to reg_base in struct > > > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > > > > > 2. Enhance EDMA driver to allow prode eDMA at EP side > > > dmaengine: dw-edma: Add support for chip specific flags > > > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > > > flags (this patch removed at v11 because dma tree already have fixed > > > patch) > > > > > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > > > dmaengine: dw-edma: Fix programming the source & dest addresses for > > > ep > > > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > > > > > 4. change pci-epf-test to use EDMA driver to transfer data. > > > PCI: endpoint: Add embedded DMA controller test > > > > > > 5. Using imx8dxl to do test, but some EP functions still have not > > > upstream yet. So below patch show how probe eDMA driver at EP > > > controller driver. > > > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 > > > > The series has been hanging out on review for over three months now. > > It has got to v11 and has been tested on at least two platforms. The > > original driver maintainer has been silent for all that time (most > > likely Gustavo dropped the driver maintaining role). Could you please > > merge it in seeing no comments have been posted for the last several > > weeks? The PCI Host/EP controller drivers maintainer suggested to get > > this series via the DMA-engine tree: > > https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ > > which is obviously right seeing it mainly concerns the DW eDMA driver. > > Though after that Lorenzo disappeared as quickly as popped up.) > > > > There is one more series depending on the changes in this > > patchset: > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > Me and Frank already settled all the conflicts and inter-dependencies, > > so at least his series is more than ready to be merged in into the > > kernel repo. It would be very good to get it accepted on this merge > > window so to have the kernel v5.19 with all this changes available. > > Since the v5.19 merge window is already open, it seems doubtful that > anybody would merge this so late in the cycle. > > If Gustavo isn't available or willing to merge it, it looks like Vinod > (maintainer of drivers/dma) would be the next logical candidate. I think the last patch should not block other patches from merging. The last patch about pci-epf-test.c is totally independent from other patches. I prefer to merge all the dma patches first. best regards Frank Li > > I suspect Vinod would appreciate an ack or reviewed-by from Kishon for > the last patch because he maintains pci-epf-test.c. > > I have a couple trivial comments on the pci-epf-test.c (I'll respond > there), but I'm not qualified to ack it. > > Bjorn
On Mon, May 23, 2022 at 01:41:48PM -0500, Zhi Li wrote: > On Mon, May 23, 2022 at 1:02 PM Bjorn Helgaas <helgaas@kernel.org> wrote: > > On Mon, May 23, 2022 at 02:06:47PM +0300, Serge Semin wrote: > > > Hello Vinod, > > > > > > On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > > > > Default Designware EDMA just probe remotely at host side. > > > > This patch allow EDMA driver can probe at EP side. > > > > > > > > 1. Clean up patch > > > > dmaengine: dw-edma: Detach the private data and chip info structures > > > > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > > > > dmaengine: dw-edma: Change rg_region to reg_base in struct > > > > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > > > > > > > 2. Enhance EDMA driver to allow prode eDMA at EP side > > > > dmaengine: dw-edma: Add support for chip specific flags > > > > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > > > > flags (this patch removed at v11 because dma tree already have fixed > > > > patch) > > > > > > > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > > > > dmaengine: dw-edma: Fix programming the source & dest addresses for > > > > ep > > > > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > > > > > > > 4. change pci-epf-test to use EDMA driver to transfer data. > > > > PCI: endpoint: Add embedded DMA controller test > > > > > > > > 5. Using imx8dxl to do test, but some EP functions still have not > > > > upstream yet. So below patch show how probe eDMA driver at EP > > > > controller driver. > > > > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 > > > > > > The series has been hanging out on review for over three months now. > > > It has got to v11 and has been tested on at least two platforms. The > > > original driver maintainer has been silent for all that time (most > > > likely Gustavo dropped the driver maintaining role). Could you please > > > merge it in seeing no comments have been posted for the last several > > > weeks? The PCI Host/EP controller drivers maintainer suggested to get > > > this series via the DMA-engine tree: > > > https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ > > > which is obviously right seeing it mainly concerns the DW eDMA driver. > > > Though after that Lorenzo disappeared as quickly as popped up.) > > > > > > There is one more series depending on the changes in this > > > patchset: > > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > Me and Frank already settled all the conflicts and inter-dependencies, > > > so at least his series is more than ready to be merged in into the > > > kernel repo. It would be very good to get it accepted on this merge > > > window so to have the kernel v5.19 with all this changes available. > > > > Since the v5.19 merge window is already open, it seems doubtful that > > anybody would merge this so late in the cycle. > > > > If Gustavo isn't available or willing to merge it, it looks like Vinod > > (maintainer of drivers/dma) would be the next logical candidate. > > I think the last patch should not block other patches from merging. > The last patch about pci-epf-test.c is totally independent from other patches. > > I prefer to merge all the dma patches first. Absolutely. Given an ack from Kishon, it would make sense for Vinod to merge them all together since they're logically related, but I have no objection to merging any of the drivers/dma patches separately. Bjorn
Hi Bjorn On Mon, May 23, 2022 at 05:12:56PM -0500, Bjorn Helgaas wrote: > On Mon, May 23, 2022 at 01:41:48PM -0500, Zhi Li wrote: > > On Mon, May 23, 2022 at 1:02 PM Bjorn Helgaas <helgaas@kernel.org> wrote: > > > On Mon, May 23, 2022 at 02:06:47PM +0300, Serge Semin wrote: > > > > Hello Vinod, > > > > > > > > On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > > > > > Default Designware EDMA just probe remotely at host side. > > > > > This patch allow EDMA driver can probe at EP side. > > > > > > > > > > 1. Clean up patch > > > > > dmaengine: dw-edma: Detach the private data and chip info structures > > > > > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > > > > > dmaengine: dw-edma: Change rg_region to reg_base in struct > > > > > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > > > > > > > > > 2. Enhance EDMA driver to allow prode eDMA at EP side > > > > > dmaengine: dw-edma: Add support for chip specific flags > > > > > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > > > > > flags (this patch removed at v11 because dma tree already have fixed > > > > > patch) > > > > > > > > > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > > > > > dmaengine: dw-edma: Fix programming the source & dest addresses for > > > > > ep > > > > > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > > > > > > > > > 4. change pci-epf-test to use EDMA driver to transfer data. > > > > > PCI: endpoint: Add embedded DMA controller test > > > > > > > > > > 5. Using imx8dxl to do test, but some EP functions still have not > > > > > upstream yet. So below patch show how probe eDMA driver at EP > > > > > controller driver. > > > > > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 > > > > > > > > The series has been hanging out on review for over three months now. > > > > It has got to v11 and has been tested on at least two platforms. The > > > > original driver maintainer has been silent for all that time (most > > > > likely Gustavo dropped the driver maintaining role). Could you please > > > > merge it in seeing no comments have been posted for the last several > > > > weeks? The PCI Host/EP controller drivers maintainer suggested to get > > > > this series via the DMA-engine tree: > > > > https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ > > > > which is obviously right seeing it mainly concerns the DW eDMA driver. > > > > Though after that Lorenzo disappeared as quickly as popped up.) > > > > > > > > There is one more series depending on the changes in this > > > > patchset: > > > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > > Me and Frank already settled all the conflicts and inter-dependencies, > > > > so at least his series is more than ready to be merged in into the > > > > kernel repo. It would be very good to get it accepted on this merge > > > > window so to have the kernel v5.19 with all this changes available. > > > > > > Since the v5.19 merge window is already open, it seems doubtful that > > > anybody would merge this so late in the cycle. In this case it would be safer to merge this whole series through your repo. See my series: "[PATCH v2 00/26] dmaengine: dw-edma: Add RP/EP local DMA controllers support" https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ depends in the Frank' patchset. Meanwhile my patchset is also based on the DW PCIe modifications introduced in the set of the series: "[PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support" https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ and "[PATCH v3 00/13] PCI: dwc: Various fixes and cleanups" https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ So to speak in order to have more coherent repos with least merge, logical problems the next order of the merging would be preferable: 1) Frank's patchset (ready to be merged in): [PATCH v11 0/8] Enable designware PCI EP EDMA locally https://lore.kernel.org/linux-pci/20220517151915.2212838-1-Frank.Li@nxp.com 2) My series (ready to be merged in): [PATCH v3 00/13] PCI: dwc: Various fixes and cleanups https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ 3) My series (still in review, I need to fix some Rob' and Manivannan' notes) [PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ 4) Me series (ready to be merged in, but depends on the prev patchsets): https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ Seeing the Frank patches won't make it into the mainline repo on this merge window, it would be great to collect all the changes in a single repository. Seeing Lorenzo disappeared as fast as popped up your repo is the best candidate since the DW eDMA block is a part of the DW PCIe controller. > > > > > > If Gustavo isn't available or willing to merge it, it looks like Vinod > > > (maintainer of drivers/dma) would be the next logical candidate. > > > > I think the last patch should not block other patches from merging. > > The last patch about pci-epf-test.c is totally independent from other patches. > > > > I prefer to merge all the dma patches first. > > Absolutely. > > Given an ack from Kishon, it would make sense for Vinod to merge them > all together since they're logically related, but I have no objection > to merging any of the drivers/dma patches separately. Please see my comment above. Thanks -Sergey > > Bjorn
On Tue, May 24, 2022 at 08:48:50AM +0300, Serge Semin wrote: > On Mon, May 23, 2022 at 05:12:56PM -0500, Bjorn Helgaas wrote: > > On Mon, May 23, 2022 at 01:41:48PM -0500, Zhi Li wrote: > > > On Mon, May 23, 2022 at 1:02 PM Bjorn Helgaas <helgaas@kernel.org> wrote: > > > > On Mon, May 23, 2022 at 02:06:47PM +0300, Serge Semin wrote: > > > > > Hello Vinod, > > > > > > > > > > On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > > > > > > Default Designware EDMA just probe remotely at host side. > > > > > > This patch allow EDMA driver can probe at EP side. > > > > > > > > > > > > 1. Clean up patch > > > > > > dmaengine: dw-edma: Detach the private data and chip info structures > > > > > > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > > > > > > dmaengine: dw-edma: Change rg_region to reg_base in struct > > > > > > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > > > > > > > > > > > 2. Enhance EDMA driver to allow prode eDMA at EP side > > > > > > dmaengine: dw-edma: Add support for chip specific flags > > > > > > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > > > > > > flags (this patch removed at v11 because dma tree already have fixed > > > > > > patch) > > > > > > > > > > > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > > > > > > dmaengine: dw-edma: Fix programming the source & dest addresses for > > > > > > ep > > > > > > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > > > > > > > > > > > 4. change pci-epf-test to use EDMA driver to transfer data. > > > > > > PCI: endpoint: Add embedded DMA controller test > > > > > > > > > > > > 5. Using imx8dxl to do test, but some EP functions still have not > > > > > > upstream yet. So below patch show how probe eDMA driver at EP > > > > > > controller driver. > > > > > > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 > > > > > > > > > > The series has been hanging out on review for over three months now. > > > > > It has got to v11 and has been tested on at least two platforms. The > > > > > original driver maintainer has been silent for all that time (most > > > > > likely Gustavo dropped the driver maintaining role). Could you please > > > > > merge it in seeing no comments have been posted for the last several > > > > > weeks? The PCI Host/EP controller drivers maintainer suggested to get > > > > > this series via the DMA-engine tree: > > > > > https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ > > > > > which is obviously right seeing it mainly concerns the DW eDMA driver. > > > > > Though after that Lorenzo disappeared as quickly as popped up.) > > > > > > > > > > There is one more series depending on the changes in this > > > > > patchset: > > > > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > > > Me and Frank already settled all the conflicts and inter-dependencies, > > > > > so at least his series is more than ready to be merged in into the > > > > > kernel repo. It would be very good to get it accepted on this merge > > > > > window so to have the kernel v5.19 with all this changes available. > > > > > > > > > Since the v5.19 merge window is already open, it seems doubtful that > > > > anybody would merge this so late in the cycle. > > In this case it would be safer to merge this whole series through your > repo. See my series: > "[PATCH v2 00/26] dmaengine: dw-edma: Add RP/EP local DMA controllers support" > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > depends in the Frank' patchset. Meanwhile my patchset is also based on the > DW PCIe modifications introduced in the set of the series: > "[PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support" > https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ > and > "[PATCH v3 00/13] PCI: dwc: Various fixes and cleanups" > https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ > > So to speak in order to have more coherent repos with least merge, > logical problems the next order of the merging would be preferable: > 1) Frank's patchset (ready to be merged in): > [PATCH v11 0/8] Enable designware PCI EP EDMA locally > https://lore.kernel.org/linux-pci/20220517151915.2212838-1-Frank.Li@nxp.com > 2) My series (ready to be merged in): > [PATCH v3 00/13] PCI: dwc: Various fixes and cleanups > https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ > 3) My series (still in review, I need to fix some Rob' and Manivannan' notes) > [PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support > https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ > 4) Me series (ready to be merged in, but depends on the prev patchsets): > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > Seeing the Frank patches won't make it into the mainline repo on > this merge window, it would be great to collect all the changes in a > single repository. Seeing Lorenzo disappeared as fast as popped up > your repo is the best candidate since the DW eDMA block is a part of > the DW PCIe controller. I'll be happy to merge this via the PCI tree, given acks from Gustavo and/or Vinod. Bjorn
On Tue, May 24, 2022 at 10:52:01AM -0500, Bjorn Helgaas wrote: > On Tue, May 24, 2022 at 08:48:50AM +0300, Serge Semin wrote: > > On Mon, May 23, 2022 at 05:12:56PM -0500, Bjorn Helgaas wrote: > > > On Mon, May 23, 2022 at 01:41:48PM -0500, Zhi Li wrote: > > > > On Mon, May 23, 2022 at 1:02 PM Bjorn Helgaas <helgaas@kernel.org> wrote: > > > > > On Mon, May 23, 2022 at 02:06:47PM +0300, Serge Semin wrote: > > > > > > Hello Vinod, > > > > > > > > > > > > On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > > > > > > > Default Designware EDMA just probe remotely at host side. > > > > > > > This patch allow EDMA driver can probe at EP side. > > > > > > > > > > > > > > 1. Clean up patch > > > > > > > dmaengine: dw-edma: Detach the private data and chip info structures > > > > > > > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > > > > > > > dmaengine: dw-edma: Change rg_region to reg_base in struct > > > > > > > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > > > > > > > > > > > > > 2. Enhance EDMA driver to allow prode eDMA at EP side > > > > > > > dmaengine: dw-edma: Add support for chip specific flags > > > > > > > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > > > > > > > flags (this patch removed at v11 because dma tree already have fixed > > > > > > > patch) > > > > > > > > > > > > > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > > > > > > > dmaengine: dw-edma: Fix programming the source & dest addresses for > > > > > > > ep > > > > > > > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > > > > > > > > > > > > > 4. change pci-epf-test to use EDMA driver to transfer data. > > > > > > > PCI: endpoint: Add embedded DMA controller test > > > > > > > > > > > > > > 5. Using imx8dxl to do test, but some EP functions still have not > > > > > > > upstream yet. So below patch show how probe eDMA driver at EP > > > > > > > controller driver. > > > > > > > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 > > > > > > > > > > > > The series has been hanging out on review for over three months now. > > > > > > It has got to v11 and has been tested on at least two platforms. The > > > > > > original driver maintainer has been silent for all that time (most > > > > > > likely Gustavo dropped the driver maintaining role). Could you please > > > > > > merge it in seeing no comments have been posted for the last several > > > > > > weeks? The PCI Host/EP controller drivers maintainer suggested to get > > > > > > this series via the DMA-engine tree: > > > > > > https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ > > > > > > which is obviously right seeing it mainly concerns the DW eDMA driver. > > > > > > Though after that Lorenzo disappeared as quickly as popped up.) > > > > > > > > > > > > There is one more series depending on the changes in this > > > > > > patchset: > > > > > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > > > > Me and Frank already settled all the conflicts and inter-dependencies, > > > > > > so at least his series is more than ready to be merged in into the > > > > > > kernel repo. It would be very good to get it accepted on this merge > > > > > > window so to have the kernel v5.19 with all this changes available. > > > > > > > > > > > > Since the v5.19 merge window is already open, it seems doubtful that > > > > > anybody would merge this so late in the cycle. > > > > In this case it would be safer to merge this whole series through your > > repo. See my series: > > "[PATCH v2 00/26] dmaengine: dw-edma: Add RP/EP local DMA controllers support" > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > depends in the Frank' patchset. Meanwhile my patchset is also based on the > > DW PCIe modifications introduced in the set of the series: > > "[PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support" > > https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ > > and > > "[PATCH v3 00/13] PCI: dwc: Various fixes and cleanups" > > https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ > > > > So to speak in order to have more coherent repos with least merge, > > logical problems the next order of the merging would be preferable: > > 1) Frank's patchset (ready to be merged in): > > [PATCH v11 0/8] Enable designware PCI EP EDMA locally > > https://lore.kernel.org/linux-pci/20220517151915.2212838-1-Frank.Li@nxp.com > > 2) My series (ready to be merged in): > > [PATCH v3 00/13] PCI: dwc: Various fixes and cleanups > > https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ > > 3) My series (still in review, I need to fix some Rob' and Manivannan' notes) > > [PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support > > https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ > > 4) Me series (ready to be merged in, but depends on the prev patchsets): > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > > Seeing the Frank patches won't make it into the mainline repo on > > this merge window, it would be great to collect all the changes in a > > single repository. Seeing Lorenzo disappeared as fast as popped up > > your repo is the best candidate since the DW eDMA block is a part of > > the DW PCIe controller. > > I'll be happy to merge this via the PCI tree, given acks from Gustavo > and/or Vinod. Great! Thanks. I'll add my note regarding this to the v12 patchset thread. Alas Gustavo hasn't shown any sign of activity for more than a year (last commit on Feb 2021 and no more acks and rb tags afterwards). So we need to wait for the Vinod' ack. -Sergey > > Bjorn
On Wed, May 25, 2022 at 4:07 AM Serge Semin <fancer.lancer@gmail.com> wrote: > > On Tue, May 24, 2022 at 10:52:01AM -0500, Bjorn Helgaas wrote: > > On Tue, May 24, 2022 at 08:48:50AM +0300, Serge Semin wrote: > > > On Mon, May 23, 2022 at 05:12:56PM -0500, Bjorn Helgaas wrote: > > > > On Mon, May 23, 2022 at 01:41:48PM -0500, Zhi Li wrote: > > > > > On Mon, May 23, 2022 at 1:02 PM Bjorn Helgaas <helgaas@kernel.org> wrote: > > > > > > On Mon, May 23, 2022 at 02:06:47PM +0300, Serge Semin wrote: > > > > > > > Hello Vinod, > > > > > > > > > > > > > > On Tue, May 17, 2022 at 10:19:07AM -0500, Frank Li wrote: > > > > > > > > Default Designware EDMA just probe remotely at host side. > > > > > > > > This patch allow EDMA driver can probe at EP side. > > > > > > > > > > > > > > > > 1. Clean up patch > > > > > > > > dmaengine: dw-edma: Detach the private data and chip info structures > > > > > > > > dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip > > > > > > > > dmaengine: dw-edma: Change rg_region to reg_base in struct > > > > > > > > dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct > > > > > > > > > > > > > > > > 2. Enhance EDMA driver to allow prode eDMA at EP side > > > > > > > > dmaengine: dw-edma: Add support for chip specific flags > > > > > > > > dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific > > > > > > > > flags (this patch removed at v11 because dma tree already have fixed > > > > > > > > patch) > > > > > > > > > > > > > > > > 3. Bugs fix at EDMA driver when probe eDMA at EP side > > > > > > > > dmaengine: dw-edma: Fix programming the source & dest addresses for > > > > > > > > ep > > > > > > > > dmaengine: dw-edma: Don't rely on the deprecated "direction" member > > > > > > > > > > > > > > > > 4. change pci-epf-test to use EDMA driver to transfer data. > > > > > > > > PCI: endpoint: Add embedded DMA controller test > > > > > > > > > > > > > > > > 5. Using imx8dxl to do test, but some EP functions still have not > > > > > > > > upstream yet. So below patch show how probe eDMA driver at EP > > > > > > > > controller driver. > > > > > > > > https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 > > > > > > > > > > > > > > The series has been hanging out on review for over three months now. > > > > > > > It has got to v11 and has been tested on at least two platforms. The > > > > > > > original driver maintainer has been silent for all that time (most > > > > > > > likely Gustavo dropped the driver maintaining role). Could you please > > > > > > > merge it in seeing no comments have been posted for the last several > > > > > > > weeks? The PCI Host/EP controller drivers maintainer suggested to get > > > > > > > this series via the DMA-engine tree: > > > > > > > https://lore.kernel.org/linux-pci/YnqlRShJzvma2SKM@lpieralisi/ > > > > > > > which is obviously right seeing it mainly concerns the DW eDMA driver. > > > > > > > Though after that Lorenzo disappeared as quickly as popped up.) > > > > > > > > > > > > > > There is one more series depending on the changes in this > > > > > > > patchset: > > > > > > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > > > > > Me and Frank already settled all the conflicts and inter-dependencies, > > > > > > > so at least his series is more than ready to be merged in into the > > > > > > > kernel repo. It would be very good to get it accepted on this merge > > > > > > > window so to have the kernel v5.19 with all this changes available. > > > > > > > > > > > > > > > Since the v5.19 merge window is already open, it seems doubtful that > > > > > > anybody would merge this so late in the cycle. > > > > > > In this case it would be safer to merge this whole series through your > > > repo. See my series: > > > "[PATCH v2 00/26] dmaengine: dw-edma: Add RP/EP local DMA controllers support" > > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > depends in the Frank' patchset. Meanwhile my patchset is also based on the > > > DW PCIe modifications introduced in the set of the series: > > > "[PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support" > > > https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ > > > and > > > "[PATCH v3 00/13] PCI: dwc: Various fixes and cleanups" > > > https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ > > > > > > So to speak in order to have more coherent repos with least merge, > > > logical problems the next order of the merging would be preferable: > > > 1) Frank's patchset (ready to be merged in): > > > [PATCH v11 0/8] Enable designware PCI EP EDMA locally > > > https://lore.kernel.org/linux-pci/20220517151915.2212838-1-Frank.Li@nxp.com > > > 2) My series (ready to be merged in): > > > [PATCH v3 00/13] PCI: dwc: Various fixes and cleanups > > > https://lore.kernel.org/linux-pci/20220517125058.18488-1-Sergey.Semin@baikalelectronics.ru/ > > > 3) My series (still in review, I need to fix some Rob' and Manivannan' notes) > > > [PATCH v2 00/17] PCI: dwc: Add dma-ranges/YAML-schema/Baikal-T1 support > > > https://lore.kernel.org/linux-pci/20220503214638.1895-1-Sergey.Semin@baikalelectronics.ru/ > > > 4) Me series (ready to be merged in, but depends on the prev patchsets): > > > https://lore.kernel.org/linux-pci/20220503225104.12108-1-Sergey.Semin@baikalelectronics.ru/ > > > > > > Seeing the Frank patches won't make it into the mainline repo on > > > this merge window, it would be great to collect all the changes in a > > > single repository. Seeing Lorenzo disappeared as fast as popped up > > > your repo is the best candidate since the DW eDMA block is a part of > > > the DW PCIe controller. > > > > > I'll be happy to merge this via the PCI tree, given acks from Gustavo > > and/or Vinod. @Vinod Koul Friendly ping Vinod? > > Great! Thanks. I'll add my note regarding this to the v12 patchset > thread. Alas Gustavo hasn't shown any sign of activity for more than a > year (last commit on Feb 2021 and no more acks and rb tags afterwards). > So we need to wait for the Vinod' ack. > > -Sergey > > > > > Bjorn
Default Designware EDMA just probe remotely at host side. This patch allow EDMA driver can probe at EP side. 1. Clean up patch dmaengine: dw-edma: Detach the private data and chip info structures dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip dmaengine: dw-edma: Change rg_region to reg_base in struct dmaengine: dw-edma: rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct 2. Enhance EDMA driver to allow prode eDMA at EP side dmaengine: dw-edma: Add support for chip specific flags dmaengine: dw-edma: Add DW_EDMA_CHIP_32BIT_DBI for chip specific flags (this patch removed at v11 because dma tree already have fixed patch) 3. Bugs fix at EDMA driver when probe eDMA at EP side dmaengine: dw-edma: Fix programming the source & dest addresses for ep dmaengine: dw-edma: Don't rely on the deprecated "direction" member 4. change pci-epf-test to use EDMA driver to transfer data. PCI: endpoint: Add embedded DMA controller test 5. Using imx8dxl to do test, but some EP functions still have not upstream yet. So below patch show how probe eDMA driver at EP controller driver. https://lore.kernel.org/linux-pci/20220309120149.GB134091@thinkpad/T/#m979eb506c73ab3cfca2e7a43635ecdaec18d8097 Frank Li (6): dmaengine: dw-edma: Remove unused field irq in struct dw_edma_chip dmaengine: dw-edma: Detach the private data and chip info structures dmaengine: dw-edma: Change rg_region to reg_base in struct dw_edma_chip dmaengine: dw-edma: Rename wr(rd)_ch_cnt to ll_wr(rd)_cnt in struct dw_edma_chip dmaengine: dw-edma: Add support for chip specific flags PCI: endpoint: Enable DMA tests for endpoints with DMA capabilities Serge Semin (2): dmaengine: dw-edma: Drop dma_slave_config.direction field usage dmaengine: dw-edma: Fix eDMA Rd/Wr-channels and DMA-direction semantics drivers/dma/dw-edma/dw-edma-core.c | 141 +++++++++++------- drivers/dma/dw-edma/dw-edma-core.h | 31 +--- drivers/dma/dw-edma/dw-edma-pcie.c | 83 +++++------ drivers/dma/dw-edma/dw-edma-v0-core.c | 41 ++--- drivers/dma/dw-edma/dw-edma-v0-core.h | 4 +- drivers/dma/dw-edma/dw-edma-v0-debugfs.c | 18 +-- drivers/dma/dw-edma/dw-edma-v0-debugfs.h | 8 +- drivers/pci/endpoint/functions/pci-epf-test.c | 108 ++++++++++++-- include/linux/dma/edma.h | 59 +++++++- 9 files changed, 313 insertions(+), 180 deletions(-)