Merge tag 'vfio-v7.2-rc1' of https://github.com/awilliam/linux-vfio

Pull VFIO updates from Alex Williamson:

 - Fix out-of-tree vfio selftest builds with make O= (Jason Gunthorpe)

 - Allow vfio selftests to build when ARCH=x86 is used for 64-bit x86
   builds (David Matlack)

 - Tighten vfio selftest infrastructure with stricter builds, safer path
   handling, sysfs helpers, and reusable device/VF-token setup. Build on
   that to add the SR-IOV UAPI selftest across supported IOMMU modes
   (Raghavendra Rao Ananta)

 - Conclude earlier vfio PCI BAR work already taken as v7.1 fixes by
   replacing vfio_pci_core_setup_barmap() and direct barmap[] access
   with vfio_pci_core_get_iomap(). Fix resulting sparse warnings (Matt
   Evans)

 - Simplify hisi_acc vfio-pci variant driver device-info reads by using
   the mailbox's new direct command-based read helper (Weili Qian)

 - Avoid duplicate reset handling in the Xe vfio-pci variant driver
   reset-done path (GuoHan Zhao)

 - Resolve a lockdep circular dependency splat by tracking active VFs
   with a private sriov_active flag rather than calling pci_num_vf()
   under memory_lock (Raghavendra Rao Ananta)

 - Add CXL DVSEC-based readiness polling for Blackwell-Next in the
   nvgrace-gpu vfio-pci variant driver, including interruptible,
   lockless waits to support worst case spec defined timeouts (Ankit
   Agrawal)

 - Prevent vfio_mig_get_next_state() from spinning forever on blocked
   migration state transition (Junrui Luo)

 - Fix a qat vfio variant driver migration resume race by taking the
   migration file lock before boundary checks (Giovanni Cabiddu)

 - Add explicit dependencies between vfio selftest output object files
   and output directories to ensure directories are always created
   (David Matlack)

* tag 'vfio-v7.2-rc1' of https://github.com/awilliam/linux-vfio:
  vfio: selftests: Ensure libvfio output dirs are always created
  vfio/qat: fix f_pos race in qat_vf_resume_write()
  vfio: prevent infinite loop in vfio_mig_get_next_state() on blocked arc
  vfio/nvgrace-gpu: Add Blackwell-Next GPU readiness check via CXL DVSEC
  vfio/pci: Use a private flag to prevent power state change with VFs
  vfio/pci: Fix sparse warning in vfio_pci_core_get_iomap()
  vfio/xe: avoid duplicate reset in xe_vfio_pci_reset_done
  hisi_acc_vfio_pci: simplify the command for reading device information
  vfio/pci: Replace vfio_pci_core_setup_barmap() with vfio_pci_core_get_iomap()
  vfio: selftests: Add tests to validate SR-IOV UAPI
  vfio: selftests: Add helpers to alloc/free vfio_pci_device
  vfio: selftests: Add helper to set/override a vf_token
  vfio: selftests: Expose more vfio_pci_device functions
  vfio: selftests: Extend container/iommufd setup for passing vf_token
  vfio: selftests: Introduce a sysfs lib
  vfio: selftests: Introduce snprintf_assert()
  vfio: selftests: Add -Wall and -Werror to the Makefile
  vfio: selftests: Allow builds when ARCH=x86
  vfio: selftests: Fix out-of-tree build with make O=

1 files changed, 20 insertions, 1 deletions

diff --git a/include/linux/vfio_pci_core.h b/include/linux/vfio_pci_core.h
index 89165b769e5c..5fc6ce4dd786 100644
--- a/include/linux/vfio_pci_core.h
+++ b/include/linux/vfio_pci_core.h
@@ -127,6 +127,7 @@ struct vfio_pci_core_device {
 	bool			needs_pm_restore:1;
 	bool			pm_intx_masked:1;
 	bool			pm_runtime_engaged:1;
+	bool			sriov_active;
 	struct pci_saved_state	*pci_saved_state;
 	struct pci_saved_state	*pm_save;
 	int			ioeventfds_nr;
@@ -188,7 +189,6 @@ int vfio_pci_core_match_token_uuid(struct vfio_device *core_vdev,
 int vfio_pci_core_enable(struct vfio_pci_core_device *vdev);
 void vfio_pci_core_disable(struct vfio_pci_core_device *vdev);
 void vfio_pci_core_finish_enable(struct vfio_pci_core_device *vdev);
-int vfio_pci_core_setup_barmap(struct vfio_pci_core_device *vdev, int bar);
 pci_ers_result_t vfio_pci_core_aer_err_detected(struct pci_dev *pdev,
 						pci_channel_state_t state);
 ssize_t vfio_pci_core_do_io_rw(struct vfio_pci_core_device *vdev, bool test_mem,
@@ -234,6 +234,25 @@ static inline bool is_aligned_for_order(struct vm_area_struct *vma,
 			   !IS_ALIGNED(pfn, 1 << order)));
 }
 
+/*
+ * Returns a BAR's iomap base or an ERR_PTR() if, for example, the
+ * BAR isn't valid, its resource wasn't acquired, or its iomap
+ * failed.  This shall only be used after vfio_pci_core_enable()
+ * has set up the BAR maps and before vfio_pci_core_disable()
+ * tears them down.
+ */
+static inline void __iomem __must_check *
+vfio_pci_core_get_iomap(struct vfio_pci_core_device *vdev, unsigned int bar)
+{
+	if (WARN_ON_ONCE(bar >= PCI_STD_NUM_BARS))
+		return IOMEM_ERR_PTR(-EINVAL);
+
+	if (WARN_ON_ONCE(!vdev->barmap[bar]))
+		return IOMEM_ERR_PTR(-ENODEV);
+
+	return vdev->barmap[bar];
+}
+
 int vfio_pci_dma_buf_iommufd_map(struct dma_buf_attachment *attachment,
 				 struct phys_vec *phys);