PCI/P2PDMA: Match interface changes to devm_memremap_pages()
Subsystems (3)
Logan Gunthorpe
maintainerPATCH 1/1 PCI/P2PDMA: Match interface changes to devm_memremap_pages()
"mm-hmm-mark-hmm_devmem_add-add_resource-export_symbol_gpl.patch" in the mm tree breaks p2pdma. The patch was written and reviewed before p2pdma was merged so the necessary changes were not done to the call site in that code.
Without this patch, all drivers will fail to register P2P resources because devm_memremap_pages() will return -EINVAL due to the 'kill' member of the pagemap structure not yet being set.
---
Ideally this patch should be squashed with the one mentioned above to avoid a bisect regression point.
drivers/pci/p2pdma.c | 10 ++--------
1 file changed, 2 insertions(+), 8 deletions(-)
drivers/pci/p2pdma.c +2 -8
@@ -82,10 +82,8 @@ static void pci_p2pdma_percpu_release(struct percpu_ref *ref)complete_all (&p2p->devmap_ref_done); } -static void pci_p2pdma_percpu_kill (void *data) +static void pci_p2pdma_percpu_kill (struct percpu_ref *ref) { -struct percpu_ref *ref = data; -/* * pci_p2pdma_add_resource() may be called multiple times * by a driver and may register the percpu_kill devm action multiple
@@ -198,6 +196,7 @@ int pci_p2pdma_add_resource(struct pci_dev *pdev, int bar, size_t size, pgmap->type =MEMORY_DEVICE_PCI_P2PDMA ; pgmap->pci_p2pdma_bus_offset =pci_bus_address (pdev, bar) -pci_resource_start (pdev, bar); + pgmap->kill = pci_p2pdma_percpu_kill; addr =devm_memremap_pages (&pdev->dev, pgmap);if (IS_ERR (addr)) {
@@ -211,11 +210,6 @@ int pci_p2pdma_add_resource(struct pci_dev *pdev, int bar, size_t size,if (error) goto pgmap_free; - error =devm_add_action_or_reset (&pdev->dev, pci_p2pdma_percpu_kill, - &pdev->p2pdma->devmap_ref); -if (error) - goto pgmap_free; -pci_info (pdev,"added peer-to-peer DMA memory %pR\n" , &pgmap->res);
--
2.19.0
Quoted 6 lines · Logan Gunthorpe
"mm-hmm-mark-hmm_devmem_add-add_resource-export_symbol_gpl.patch" in the mm tree breaks p2pdma. The patch was written and reviewed before p2pdma was merged so the necessary changes were not done to the call site in that code.
Quoted 7 lines
Without this patch, all drivers will fail to register P2P resources because
devm_memremap_pages()will return-EINVALdue to the 'kill' member of the pagemap structure not yet being set.Signed-off-by Logan Gunthorpe <[email protected]>Cc Andrew Morton <[email protected]>Cc Dan Williams <[email protected]>
Quoted 5 lines · Logan Gunthorpe
"mm-hmm-mark-hmm_devmem_add-add_resource-export_symbol_gpl.patch" in the mm tree breaks p2pdma. The patch was written and reviewed before p2pdma was merged so the necessary changes were not done to the call site in that code.
Quoted 8 lines
Without this patch, all drivers will fail to register P2P resources because
devm_memremap_pages()will return-EINVALdue to the 'kill' member of the pagemap structure not yet being set.Signed-off-by Logan Gunthorpe <[email protected]>Cc Andrew Morton <[email protected]>Cc Dan Williams <[email protected]>Cc Bjorn Helgaas <[email protected]>
Applied with Dan's reviewed-by to pci/peer-to-peer for v4.21, thanks!
If the mm patch you mention gets merged for v4.20, let me know and I can promote this to for-linus so v4.20 doesn't end up broken.
Quoted 46 lines
---
Ideally this patch should be squashed with the one mentioned above to avoid a bisect regression point.
drivers/pci/p2pdma.c| 10 ++--------1 file changed, 2 insertions(+), 8 deletions(-)drivers/pci/p2pdma.c +2
-9@@ -82,10 +82,8 @@ static void pci_p2pdma_percpu_release(struct percpu_ref *ref)complete_all (&p2p->devmap_ref_done); } -static void pci_p2pdma_percpu_kill (void *data) +static void pci_p2pdma_percpu_kill (struct percpu_ref *ref) { -struct percpu_ref *ref = data; -/* * pci_p2pdma_add_resource() may be called multiple times * by a driver and may register the percpu_kill devm action multiple @@ -198,6 +196,7 @@ int pci_p2pdma_add_resource(struct pci_dev *pdev, int bar, size_t size, pgmap->type =MEMORY_DEVICE_PCI_P2PDMA ; pgmap->pci_p2pdma_bus_offset =pci_bus_address (pdev, bar) -pci_resource_start (pdev, bar); + pgmap->kill = pci_p2pdma_percpu_kill; addr =devm_memremap_pages (&pdev->dev, pgmap);if (IS_ERR (addr)) {@@ -211,11 +210,6 @@ int pci_p2pdma_add_resource(struct pci_dev *pdev, int bar, size_t size,if (error) goto pgmap_free; - error =devm_add_action_or_reset (&pdev->dev, pci_p2pdma_percpu_kill, - &pdev->p2pdma->devmap_ref); -if (error) - goto pgmap_free; -pci_info (pdev,"added peer-to-peer DMA memory %pR\n" , &pgmap->res); --2.19.0
Logan Gunthorpe
maintainerQuoted 15 lines · Bjorn Helgaas
Quoted 5 lines · Logan Gunthorpe
"mm-hmm-mark-hmm_devmem_add-add_resource-export_symbol_gpl.patch" in the mm tree breaks p2pdma. The patch was written and reviewed before p2pdma was merged so the necessary changes were not done to the call site in that code.
Quoted 8 lines
Without this patch, all drivers will fail to register P2P resources because
devm_memremap_pages()will return-EINVALdue to the 'kill' member of the pagemap structure not yet being set.Signed-off-by Logan Gunthorpe <[email protected]>Cc Andrew Morton <[email protected]>Cc Dan Williams <[email protected]>Cc Bjorn Helgaas <[email protected]>
Quoted 4 lines
Applied with Dan's reviewed-by to pci/peer-to-peer for v4.21, thanks!
If the mm patch you mention gets merged for v4.20, let me know and I can promote this to for-linus so v4.20 doesn't end up broken.
Thanks Bjorn, but I think Andrew has also picked it up in the mm tree with the patch in question. My hope is that he squashes the two but I'm not sure what his intentions are. I mostly copied you for information purposes as this patch shouldn't even compile without Dan's patch.
Though, I guess we'll find out what goes in after the merge window -- you may not need to do anything.
Logan
Quoted 23 lines · Logan Gunthorpe
Quoted 15 lines · Bjorn Helgaas
Quoted 5 lines · Logan Gunthorpe
"mm-hmm-mark-hmm_devmem_add-add_resource-export_symbol_gpl.patch" in the mm tree breaks p2pdma. The patch was written and reviewed before p2pdma was merged so the necessary changes were not done to the call site in that code.
Quoted 8 lines
Without this patch, all drivers will fail to register P2P resources because
devm_memremap_pages()will return-EINVALdue to the 'kill' member of the pagemap structure not yet being set.Signed-off-by Logan Gunthorpe <[email protected]>Cc Andrew Morton <[email protected]>Cc Dan Williams <[email protected]>Cc Bjorn Helgaas <[email protected]>Quoted 4 lines
Applied with Dan's reviewed-by to pci/peer-to-peer for v4.21, thanks!
If the mm patch you mention gets merged for v4.20, let me know and I can promote this to for-linus so v4.20 doesn't end up broken.
Quoted 7 lines
Thanks Bjorn, but I think Andrew has also picked it up in the mm tree with the patch in question. My hope is that he squashes the two but I'm not sure what his intentions are. I mostly copied you for information purposes as this patch shouldn't even compile without Dan's patch.
Though, I guess we'll find out what goes in after the merge window -- you may not need to do anything.
OK, I dropped this because I don't have the matching mm patch. I agree, it should be squashed so we avoid a bisection hole. Let me know if I need to do anything else with this.
Bjorn