* [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars
@ 2026-04-21 17:43 Matt Evans
2026-04-21 18:14 ` Logan Gunthorpe
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: Matt Evans @ 2026-04-21 17:43 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Ankit Agrawal, Leon Romanovsky,
Alex Williamson, Niklas Schnelle
Cc: linux-pci, linux-kernel
Extend pcim_p2pdma_provider()'s checks to exclude functions that have
pdev->non_mappable_bars set.
Consumers such as VFIO were previously able to map these for access by
the CPU or P2P. Update the comment on non_mappable_bars to show it
refers to any access, not just userspace CPU access.
Fixes: 372d6d1b8ae3c ("PCI/P2PDMA: Refactor to separate core P2P functionality from memory allocation")
Signed-off-by: Matt Evans <mattev@meta.com>
---
This arises from Alex Williamson's suggestion to test
non_mappable_bars when getting the provider, with discussion here:
https://lore.kernel.org/kvm/20260415181623.1021090-1-mattev@meta.com/
The goal was to prevent a hole where VFIO could export DMABUFs for
BARs marked non-mappable, and to fix for all users of the provider
rather than just VFIO. Alex observed that non_mappable_bars should be
taken to mean BARs weren't usable by the CPU _or_ peers and,
considering that, its comment about userspace access wasn't quite
right.
drivers/pci/p2pdma.c | 3 ++-
include/linux/pci.h | 2 +-
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 7c898542af8d..4a783413f466 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -318,7 +318,8 @@ struct p2pdma_provider *pcim_p2pdma_provider(struct pci_dev *pdev, int bar)
{
struct pci_p2pdma *p2p;
- if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM))
+ if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM) ||
+ pdev->non_mappable_bars)
return NULL;
p2p = rcu_dereference_protected(pdev->p2pdma, 1);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 2c4454583c11..1e6802017d6b 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -508,7 +508,7 @@ struct pci_dev {
unsigned int no_command_memory:1; /* No PCI_COMMAND_MEMORY */
unsigned int rom_bar_overlap:1; /* ROM BAR disable broken */
unsigned int rom_attr_enabled:1; /* Display of ROM attribute enabled? */
- unsigned int non_mappable_bars:1; /* BARs can't be mapped to user-space */
+ unsigned int non_mappable_bars:1; /* BARs can't be mapped by CPU or peers */
pci_dev_flags_t dev_flags;
atomic_t enable_cnt; /* pci_enable_device has been called */
--
2.47.3
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars
2026-04-21 17:43 [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars Matt Evans
@ 2026-04-21 18:14 ` Logan Gunthorpe
2026-04-21 19:49 ` Leon Romanovsky
2026-04-22 9:19 ` Niklas Schnelle
2 siblings, 0 replies; 4+ messages in thread
From: Logan Gunthorpe @ 2026-04-21 18:14 UTC (permalink / raw)
To: Matt Evans, Bjorn Helgaas, Ankit Agrawal, Leon Romanovsky,
Alex Williamson, Niklas Schnelle
Cc: linux-pci, linux-kernel
On 2026-04-21 11:43, Matt Evans wrote:
> Extend pcim_p2pdma_provider()'s checks to exclude functions that have
> pdev->non_mappable_bars set.
>
> Consumers such as VFIO were previously able to map these for access by
> the CPU or P2P. Update the comment on non_mappable_bars to show it
> refers to any access, not just userspace CPU access.
>
> Fixes: 372d6d1b8ae3c ("PCI/P2PDMA: Refactor to separate core P2P functionality from memory allocation")
> Signed-off-by: Matt Evans <mattev@meta.com>
This makes sense to me, thanks.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars
2026-04-21 17:43 [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars Matt Evans
2026-04-21 18:14 ` Logan Gunthorpe
@ 2026-04-21 19:49 ` Leon Romanovsky
2026-04-22 9:19 ` Niklas Schnelle
2 siblings, 0 replies; 4+ messages in thread
From: Leon Romanovsky @ 2026-04-21 19:49 UTC (permalink / raw)
To: Matt Evans
Cc: Bjorn Helgaas, Logan Gunthorpe, Ankit Agrawal, Alex Williamson,
Niklas Schnelle, linux-pci, linux-kernel
On Tue, Apr 21, 2026 at 10:43:51AM -0700, Matt Evans wrote:
> Extend pcim_p2pdma_provider()'s checks to exclude functions that have
> pdev->non_mappable_bars set.
>
> Consumers such as VFIO were previously able to map these for access by
> the CPU or P2P. Update the comment on non_mappable_bars to show it
> refers to any access, not just userspace CPU access.
>
> Fixes: 372d6d1b8ae3c ("PCI/P2PDMA: Refactor to separate core P2P functionality from memory allocation")
I don't object to the patch, but this Fixes line doesn't look correct.
non_mappable_bars applies only to s390, which doesn't support p2p. That
wasn't prevented before 372d6d1b8ae3c refactoring too.
Thanks
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars
2026-04-21 17:43 [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars Matt Evans
2026-04-21 18:14 ` Logan Gunthorpe
2026-04-21 19:49 ` Leon Romanovsky
@ 2026-04-22 9:19 ` Niklas Schnelle
2 siblings, 0 replies; 4+ messages in thread
From: Niklas Schnelle @ 2026-04-22 9:19 UTC (permalink / raw)
To: Matt Evans, Bjorn Helgaas, Logan Gunthorpe, Ankit Agrawal,
Leon Romanovsky, Alex Williamson
Cc: linux-pci, linux-kernel
On Tue, 2026-04-21 at 10:43 -0700, Matt Evans wrote:
> Extend pcim_p2pdma_provider()'s checks to exclude functions that have
> pdev->non_mappable_bars set.
>
> Consumers such as VFIO were previously able to map these for access by
> the CPU or P2P. Update the comment on non_mappable_bars to show it
> refers to any access, not just userspace CPU access.
>
> Fixes: 372d6d1b8ae3c ("PCI/P2PDMA: Refactor to separate core P2P functionality from memory allocation")
> Signed-off-by: Matt Evans <mattev@meta.com>
> ---
>
> This arises from Alex Williamson's suggestion to test
> non_mappable_bars when getting the provider, with discussion here:
>
> https://lore.kernel.org/kvm/20260415181623.1021090-1-mattev@meta.com/
>
> The goal was to prevent a hole where VFIO could export DMABUFs for
> BARs marked non-mappable, and to fix for all users of the provider
> rather than just VFIO. Alex observed that non_mappable_bars should be
> taken to mean BARs weren't usable by the CPU _or_ peers and,
> considering that, its comment about userspace access wasn't quite
> right.
>
>
> drivers/pci/p2pdma.c | 3 ++-
> include/linux/pci.h | 2 +-
> 2 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
> index 7c898542af8d..4a783413f466 100644
> --- a/drivers/pci/p2pdma.c
> +++ b/drivers/pci/p2pdma.c
> @@ -318,7 +318,8 @@ struct p2pdma_provider *pcim_p2pdma_provider(struct pci_dev *pdev, int bar)
> {
> struct pci_p2pdma *p2p;
>
> - if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM))
> + if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM) ||
> + pdev->non_mappable_bars)
> return NULL;
>
> p2p = rcu_dereference_protected(pdev->p2pdma, 1);
> diff --git a/include/linux/pci.h b/include/linux/pci.h
> index 2c4454583c11..1e6802017d6b 100644
> --- a/include/linux/pci.h
> +++ b/include/linux/pci.h
> @@ -508,7 +508,7 @@ struct pci_dev {
> unsigned int no_command_memory:1; /* No PCI_COMMAND_MEMORY */
> unsigned int rom_bar_overlap:1; /* ROM BAR disable broken */
> unsigned int rom_attr_enabled:1; /* Display of ROM attribute enabled? */
> - unsigned int non_mappable_bars:1; /* BARs can't be mapped to user-space */
> + unsigned int non_mappable_bars:1; /* BARs can't be mapped by CPU or peers */
> pci_dev_flags_t dev_flags;
> atomic_t enable_cnt; /* pci_enable_device has been called */
>
Sorry for the late reply. For s390's ISM device this definitely makes
sense. The BAR can indeed not be accessed by P2P and extending this
seems reasonable to me.
Feel free to add my:
Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com>
Thanks,
Niklas
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-04-22 9:21 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-21 17:43 [PATCH] PCI/P2PDMA: Avoid returning a provider for non_mappable_bars Matt Evans
2026-04-21 18:14 ` Logan Gunthorpe
2026-04-21 19:49 ` Leon Romanovsky
2026-04-22 9:19 ` Niklas Schnelle
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox