| Message ID | 20220525223316.388490-1-willmcvicker@google.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Lorenzo Pieralisi |
| Headers | show |
| Series | [v1,1/1] PCI: dwc: Fix MSI msi_msg dma mapping | expand |
On Wed, May 25, 2022 at 3:33 PM Will McVicker <willmcvicker@google.com> wrote: > > As of commit 07940c369a6b ("PCI: dwc: Fix MSI page leakage in > suspend/resume"), the PCIe designware host driver has been using the > driver data allocation for the msi_msg dma mapping which can result in > a DMA_MAPPING_ERROR due to the DMA overflow check in > dma_direct_map_page() when the address is greater than 32-bits (reported > in [1]). The commit was trying to address a memory leak on > suspend/resume by moving the MSI mapping to dw_pcie_host_init(), but > subsequently dropped the page allocation thinking it wasn't needed. > > To fix the DMA mapping issue as well as make msi_msg DMA'able, let's > switch back to allocating a 32-bit page for the msi_msg. To avoid the > suspend/resume leak, we can allocate the page in dw_pcie_host_init() > since that function shouldn't be called during suspend/resume. > > [1] https://lore.kernel.org/all/Yo0soniFborDl7+C@google.com/ > > Signed-off-by: Will McVicker <willmcvicker@google.com> > --- > drivers/pci/controller/dwc/pcie-designware-host.c | 14 ++++++++------ > drivers/pci/controller/dwc/pcie-designware.h | 2 +- > 2 files changed, 9 insertions(+), 7 deletions(-) > > diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c > index 2fa86f32d964..3655c6f88bf1 100644 > --- a/drivers/pci/controller/dwc/pcie-designware-host.c > +++ b/drivers/pci/controller/dwc/pcie-designware-host.c > @@ -267,8 +267,9 @@ static void dw_pcie_free_msi(struct pcie_port *pp) > struct dw_pcie *pci = to_dw_pcie_from_pp(pp); > struct device *dev = pci->dev; > > - dma_unmap_single_attrs(dev, pp->msi_data, sizeof(pp->msi_msg), > - DMA_FROM_DEVICE, DMA_ATTR_SKIP_CPU_SYNC); > + dma_unmap_page(dev, pp->msi_data, PAGE_SIZE, DMA_FROM_DEVICE); > + if (pp->msi_page) > + __free_page(pp->msi_page); > } > } > > @@ -392,12 +393,13 @@ int dw_pcie_host_init(struct pcie_port *pp) > if (ret) > dev_warn(pci->dev, "Failed to set DMA mask to 32-bit. Devices with only 32-bit MSI support may not work properly\n"); > > - pp->msi_data = dma_map_single_attrs(pci->dev, &pp->msi_msg, > - sizeof(pp->msi_msg), > - DMA_FROM_DEVICE, > - DMA_ATTR_SKIP_CPU_SYNC); > + pp->msi_page = alloc_page(GFP_DMA32); > + pp->msi_data = dma_map_page(pci->dev, pp->msi_page, 0, PAGE_SIZE, > + DMA_FROM_DEVICE); > if (dma_mapping_error(pci->dev, pp->msi_data)) { > dev_err(pci->dev, "Failed to map MSI data\n"); > + __free_page(pp->msi_page); > + pp->msi_page = NULL; > pp->msi_data = 0; > goto err_free_msi; > } > diff --git a/drivers/pci/controller/dwc/pcie-designware.h b/drivers/pci/controller/dwc/pcie-designware.h > index 7d6e9b7576be..b5f528536358 100644 > --- a/drivers/pci/controller/dwc/pcie-designware.h > +++ b/drivers/pci/controller/dwc/pcie-designware.h > @@ -190,8 +190,8 @@ struct pcie_port { > int msi_irq; > struct irq_domain *irq_domain; > struct irq_domain *msi_domain; > - u16 msi_msg; > dma_addr_t msi_data; > + struct page *msi_page; > struct irq_chip *msi_irq_chip; > u32 num_vectors; > u32 irq_mask[MAX_MSI_CTRLS]; > -- > 2.36.1.124.g0e6072fb45-goog > Friendly ping for review. Thanks, Will
On Wed, May 25, 2022 at 10:33:16PM +0000, Will McVicker wrote: > As of commit 07940c369a6b ("PCI: dwc: Fix MSI page leakage in > suspend/resume"), the PCIe designware host driver has been using the > driver data allocation for the msi_msg dma mapping which can result in > a DMA_MAPPING_ERROR due to the DMA overflow check in > dma_direct_map_page() when the address is greater than 32-bits (reported > in [1]). The commit was trying to address a memory leak on > suspend/resume by moving the MSI mapping to dw_pcie_host_init(), but > subsequently dropped the page allocation thinking it wasn't needed. > > To fix the DMA mapping issue as well as make msi_msg DMA'able, let's > switch back to allocating a 32-bit page for the msi_msg. To avoid the > suspend/resume leak, we can allocate the page in dw_pcie_host_init() > since that function shouldn't be called during suspend/resume. > > [1] https://lore.kernel.org/all/Yo0soniFborDl7+C@google.com/ > > Signed-off-by: Will McVicker <willmcvicker@google.com> > --- > drivers/pci/controller/dwc/pcie-designware-host.c | 14 ++++++++------ > drivers/pci/controller/dwc/pcie-designware.h | 2 +- > 2 files changed, 9 insertions(+), 7 deletions(-) Reviewed-by: Rob Herring <robh@kernel.org>
On Wed, May 25, 2022 at 10:33:16PM +0000, Will McVicker wrote: > As of commit 07940c369a6b ("PCI: dwc: Fix MSI page leakage in > suspend/resume"), the PCIe designware host driver has been using the > driver data allocation for the msi_msg dma mapping which can result in > a DMA_MAPPING_ERROR due to the DMA overflow check in > dma_direct_map_page() when the address is greater than 32-bits (reported > in [1]). The commit was trying to address a memory leak on > suspend/resume by moving the MSI mapping to dw_pcie_host_init(), but > subsequently dropped the page allocation thinking it wasn't needed. > > To fix the DMA mapping issue as well as make msi_msg DMA'able, let's > switch back to allocating a 32-bit page for the msi_msg. To avoid the > suspend/resume leak, we can allocate the page in dw_pcie_host_init() > since that function shouldn't be called during suspend/resume. > > [1] https://lore.kernel.org/all/Yo0soniFborDl7+C@google.com/ > > Signed-off-by: Will McVicker <willmcvicker@google.com> Applied to pci/ctrl/dwc for v5.20, thanks! > --- > drivers/pci/controller/dwc/pcie-designware-host.c | 14 ++++++++------ > drivers/pci/controller/dwc/pcie-designware.h | 2 +- > 2 files changed, 9 insertions(+), 7 deletions(-) > > diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c > index 2fa86f32d964..3655c6f88bf1 100644 > --- a/drivers/pci/controller/dwc/pcie-designware-host.c > +++ b/drivers/pci/controller/dwc/pcie-designware-host.c > @@ -267,8 +267,9 @@ static void dw_pcie_free_msi(struct pcie_port *pp) > struct dw_pcie *pci = to_dw_pcie_from_pp(pp); > struct device *dev = pci->dev; > > - dma_unmap_single_attrs(dev, pp->msi_data, sizeof(pp->msi_msg), > - DMA_FROM_DEVICE, DMA_ATTR_SKIP_CPU_SYNC); > + dma_unmap_page(dev, pp->msi_data, PAGE_SIZE, DMA_FROM_DEVICE); > + if (pp->msi_page) > + __free_page(pp->msi_page); > } > } > > @@ -392,12 +393,13 @@ int dw_pcie_host_init(struct pcie_port *pp) > if (ret) > dev_warn(pci->dev, "Failed to set DMA mask to 32-bit. Devices with only 32-bit MSI support may not work properly\n"); > > - pp->msi_data = dma_map_single_attrs(pci->dev, &pp->msi_msg, > - sizeof(pp->msi_msg), > - DMA_FROM_DEVICE, > - DMA_ATTR_SKIP_CPU_SYNC); > + pp->msi_page = alloc_page(GFP_DMA32); > + pp->msi_data = dma_map_page(pci->dev, pp->msi_page, 0, PAGE_SIZE, > + DMA_FROM_DEVICE); > if (dma_mapping_error(pci->dev, pp->msi_data)) { > dev_err(pci->dev, "Failed to map MSI data\n"); > + __free_page(pp->msi_page); > + pp->msi_page = NULL; > pp->msi_data = 0; > goto err_free_msi; > } > diff --git a/drivers/pci/controller/dwc/pcie-designware.h b/drivers/pci/controller/dwc/pcie-designware.h > index 7d6e9b7576be..b5f528536358 100644 > --- a/drivers/pci/controller/dwc/pcie-designware.h > +++ b/drivers/pci/controller/dwc/pcie-designware.h > @@ -190,8 +190,8 @@ struct pcie_port { > int msi_irq; > struct irq_domain *irq_domain; > struct irq_domain *msi_domain; > - u16 msi_msg; > dma_addr_t msi_data; > + struct page *msi_page; > struct irq_chip *msi_irq_chip; > u32 num_vectors; > u32 irq_mask[MAX_MSI_CTRLS]; > -- > 2.36.1.124.g0e6072fb45-goog >
diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c index 2fa86f32d964..3655c6f88bf1 100644 --- a/drivers/pci/controller/dwc/pcie-designware-host.c +++ b/drivers/pci/controller/dwc/pcie-designware-host.c @@ -267,8 +267,9 @@ static void dw_pcie_free_msi(struct pcie_port *pp) struct dw_pcie *pci = to_dw_pcie_from_pp(pp); struct device *dev = pci->dev; - dma_unmap_single_attrs(dev, pp->msi_data, sizeof(pp->msi_msg), - DMA_FROM_DEVICE, DMA_ATTR_SKIP_CPU_SYNC); + dma_unmap_page(dev, pp->msi_data, PAGE_SIZE, DMA_FROM_DEVICE); + if (pp->msi_page) + __free_page(pp->msi_page); } } @@ -392,12 +393,13 @@ int dw_pcie_host_init(struct pcie_port *pp) if (ret) dev_warn(pci->dev, "Failed to set DMA mask to 32-bit. Devices with only 32-bit MSI support may not work properly\n"); - pp->msi_data = dma_map_single_attrs(pci->dev, &pp->msi_msg, - sizeof(pp->msi_msg), - DMA_FROM_DEVICE, - DMA_ATTR_SKIP_CPU_SYNC); + pp->msi_page = alloc_page(GFP_DMA32); + pp->msi_data = dma_map_page(pci->dev, pp->msi_page, 0, PAGE_SIZE, + DMA_FROM_DEVICE); if (dma_mapping_error(pci->dev, pp->msi_data)) { dev_err(pci->dev, "Failed to map MSI data\n"); + __free_page(pp->msi_page); + pp->msi_page = NULL; pp->msi_data = 0; goto err_free_msi; } diff --git a/drivers/pci/controller/dwc/pcie-designware.h b/drivers/pci/controller/dwc/pcie-designware.h index 7d6e9b7576be..b5f528536358 100644 --- a/drivers/pci/controller/dwc/pcie-designware.h +++ b/drivers/pci/controller/dwc/pcie-designware.h @@ -190,8 +190,8 @@ struct pcie_port { int msi_irq; struct irq_domain *irq_domain; struct irq_domain *msi_domain; - u16 msi_msg; dma_addr_t msi_data; + struct page *msi_page; struct irq_chip *msi_irq_chip; u32 num_vectors; u32 irq_mask[MAX_MSI_CTRLS];
As of commit 07940c369a6b ("PCI: dwc: Fix MSI page leakage in suspend/resume"), the PCIe designware host driver has been using the driver data allocation for the msi_msg dma mapping which can result in a DMA_MAPPING_ERROR due to the DMA overflow check in dma_direct_map_page() when the address is greater than 32-bits (reported in [1]). The commit was trying to address a memory leak on suspend/resume by moving the MSI mapping to dw_pcie_host_init(), but subsequently dropped the page allocation thinking it wasn't needed. To fix the DMA mapping issue as well as make msi_msg DMA'able, let's switch back to allocating a 32-bit page for the msi_msg. To avoid the suspend/resume leak, we can allocate the page in dw_pcie_host_init() since that function shouldn't be called during suspend/resume. [1] https://lore.kernel.org/all/Yo0soniFborDl7+C@google.com/ Signed-off-by: Will McVicker <willmcvicker@google.com> --- drivers/pci/controller/dwc/pcie-designware-host.c | 14 ++++++++------ drivers/pci/controller/dwc/pcie-designware.h | 2 +- 2 files changed, 9 insertions(+), 7 deletions(-)