diff options
Diffstat (limited to 'drivers/gpu')
34 files changed, 1007 insertions, 166 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 54ac8a845e9f..63208e5c1588 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1718,6 +1718,7 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, struct drm_file *file_priv); void amdgpu_driver_preclose_kms(struct drm_device *dev, struct drm_file *file_priv); +int amdgpu_suspend(struct amdgpu_device *adev); int amdgpu_device_suspend(struct drm_device *dev, bool suspend, bool fbcon); int amdgpu_device_resume(struct drm_device *dev, bool resume, bool fbcon); u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c index 4f973a9c7b87..8ec1967a850b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c @@ -305,8 +305,9 @@ static bool amdgpu_acpi_vfct_bios(struct amdgpu_device *adev) GOP_VBIOS_CONTENT *vbios; VFCT_IMAGE_HEADER *vhdr; - if (!ACPI_SUCCESS(acpi_get_table_with_size("VFCT", 1, &hdr, &tbl_size))) + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr))) return false; + tbl_size = hdr->length; if (tbl_size < sizeof(UEFI_ACPI_VFCT)) { DRM_ERROR("ACPI VFCT table present but broken (too short #1)\n"); goto out_unmap; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index f6394b5793c2..60bd4afe45c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1517,7 +1517,7 @@ static int amdgpu_fini(struct amdgpu_device *adev) return 0; } -static int amdgpu_suspend(struct amdgpu_device *adev) +int amdgpu_suspend(struct amdgpu_device *adev) { int i, r; @@ -2523,7 +2523,7 @@ static void amdgpu_debugfs_remove_files(struct amdgpu_device *adev) static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; bool pm_pg_lock, use_bank; @@ -2599,7 +2599,7 @@ end: static ssize_t amdgpu_debugfs_regs_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; bool pm_pg_lock, use_bank; @@ -2673,7 +2673,7 @@ static ssize_t amdgpu_debugfs_regs_write(struct file *f, const char __user *buf, static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2700,7 +2700,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2728,7 +2728,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2755,7 +2755,7 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2783,7 +2783,7 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2810,7 +2810,7 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2838,7 +2838,7 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user * static ssize_t amdgpu_debugfs_gca_config_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; uint32_t *config, no_regs = 0; @@ -2908,7 +2908,7 @@ static ssize_t amdgpu_debugfs_gca_config_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; int idx, r; int32_t value; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 7914f999b1bc..8cb937b2bfcc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -486,12 +486,15 @@ amdgpu_pci_remove(struct pci_dev *pdev) static void amdgpu_pci_shutdown(struct pci_dev *pdev) { + struct drm_device *dev = pci_get_drvdata(pdev); + struct amdgpu_device *adev = dev->dev_private; + /* if we are running in a VM, make sure the device * torn down properly on reboot/shutdown. * unfortunately we can't detect certain * hypervisors so just do this all the time. */ - amdgpu_pci_remove(pdev); + amdgpu_suspend(adev); } static int amdgpu_pmops_suspend(struct device *dev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 4c992826d2d6..a47628395914 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -280,7 +280,7 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring) static ssize_t amdgpu_debugfs_ring_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_ring *ring = (struct amdgpu_ring*)f->f_inode->i_private; + struct amdgpu_ring *ring = file_inode(f)->i_private; int r, i; uint32_t value, result, early[3]; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index bc70f80260d8..8e35c1ff59e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -1511,7 +1511,7 @@ static const struct drm_info_list amdgpu_ttm_debugfs_list[] = { static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -1555,7 +1555,7 @@ static const struct file_operations amdgpu_ttm_vram_fops = { static ssize_t amdgpu_ttm_gtt_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_d.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_d.h index 09a7df17570d..09a7df17570d 100755..100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_d.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_d.h diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_sh_mask.h index 1ddc4183a1c9..1ddc4183a1c9 100755..100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_11_2_sh_mask.h diff --git a/drivers/gpu/drm/amd/include/cgs_common.h b/drivers/gpu/drm/amd/include/cgs_common.h index e4a1697ec1d3..e4a1697ec1d3 100755..100644 --- a/drivers/gpu/drm/amd/include/cgs_common.h +++ b/drivers/gpu/drm/amd/include/cgs_common.h diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c index 26eff56b4a99..26eff56b4a99 100755..100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c index 768087ddb046..a293c8be232c 100644 --- a/drivers/gpu/drm/armada/armada_gem.c +++ b/drivers/gpu/drm/armada/armada_gem.c @@ -17,12 +17,11 @@ static int armada_gem_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { struct armada_gem_object *obj = drm_to_armada_gem(vma->vm_private_data); - unsigned long addr = (unsigned long)vmf->virtual_address; unsigned long pfn = obj->phys_addr >> PAGE_SHIFT; int ret; - pfn += (addr - vma->vm_start) >> PAGE_SHIFT; - ret = vm_insert_pfn(vma, addr, pfn); + pfn += (vmf->address - vma->vm_start) >> PAGE_SHIFT; + ret = vm_insert_pfn(vma, vmf->address, pfn); switch (ret) { case 0: diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c index caa4e4ca616d..bd311c77c254 100644 --- a/drivers/gpu/drm/drm_vm.c +++ b/drivers/gpu/drm/drm_vm.c @@ -124,8 +124,7 @@ static int drm_do_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) * Using vm_pgoff as a selector forces us to use this unusual * addressing scheme. */ - resource_size_t offset = (unsigned long)vmf->virtual_address - - vma->vm_start; + resource_size_t offset = vmf->address - vma->vm_start; resource_size_t baddr = map->offset + offset; struct drm_agp_mem *agpmem; struct page *page; @@ -195,7 +194,7 @@ static int drm_do_vm_shm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) if (!map) return VM_FAULT_SIGBUS; /* Nothing allocated */ - offset = (unsigned long)vmf->virtual_address - vma->vm_start; + offset = vmf->address - vma->vm_start; i = (unsigned long)map->handle + offset; page = vmalloc_to_page((void *)i); if (!page) @@ -301,7 +300,8 @@ static int drm_do_vm_dma_fault(struct vm_area_struct *vma, struct vm_fault *vmf) if (!dma->pagelist) return VM_FAULT_SIGBUS; /* Nothing allocated */ - offset = (unsigned long)vmf->virtual_address - vma->vm_start; /* vm_[pg]off[set] should be 0 */ + offset = vmf->address - vma->vm_start; + /* vm_[pg]off[set] should be 0 */ page_nr = offset >> PAGE_SHIFT; /* page_nr could just be vmf->pgoff */ page = virt_to_page((void *)dma->pagelist[page_nr]); @@ -337,7 +337,7 @@ static int drm_do_vm_sg_fault(struct vm_area_struct *vma, struct vm_fault *vmf) if (!entry->pagelist) return VM_FAULT_SIGBUS; /* Nothing allocated */ - offset = (unsigned long)vmf->virtual_address - vma->vm_start; + offset = vmf->address - vma->vm_start; map_offset = map->offset - (unsigned long)dev->sg->virtual; page_offset = (offset >> PAGE_SHIFT) + (map_offset >> PAGE_SHIFT); page = entry->pagelist[page_offset]; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c index 7d066a91d778..114dddbd297b 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c @@ -202,15 +202,14 @@ int etnaviv_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) } /* We don't use vmf->pgoff since that has the fake offset: */ - pgoff = ((unsigned long)vmf->virtual_address - - vma->vm_start) >> PAGE_SHIFT; + pgoff = (vmf->address - vma->vm_start) >> PAGE_SHIFT; page = pages[pgoff]; - VERB("Inserting %p pfn %lx, pa %lx", vmf->virtual_address, + VERB("Inserting %p pfn %lx, pa %lx", (void *)vmf->address, page_to_pfn(page), page_to_pfn(page) << PAGE_SHIFT); - ret = vm_insert_page(vma, (unsigned long)vmf->virtual_address, page); + ret = vm_insert_page(vma, vmf->address, page); out: switch (ret) { @@ -759,7 +758,7 @@ static struct page **etnaviv_gem_userptr_do_get_pages( down_read(&mm->mmap_sem); while (pinned < npages) { ret = get_user_pages_remote(task, mm, ptr, npages - pinned, - flags, pvec + pinned, NULL); + flags, pvec + pinned, NULL, NULL); if (ret < 0) break; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index ea7a18230888..57b81460fec8 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -455,8 +455,7 @@ int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) pgoff_t page_offset; int ret; - page_offset = ((unsigned long)vmf->virtual_address - - vma->vm_start) >> PAGE_SHIFT; + page_offset = (vmf->address - vma->vm_start) >> PAGE_SHIFT; if (page_offset >= (exynos_gem->size >> PAGE_SHIFT)) { DRM_ERROR("invalid page offset\n"); @@ -465,8 +464,7 @@ int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) } pfn = page_to_pfn(exynos_gem->pages[page_offset]); - ret = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, - __pfn_to_pfn_t(pfn, PFN_DEV)); + ret = vm_insert_mixed(vma, vmf->address, __pfn_to_pfn_t(pfn, PFN_DEV)); out: switch (ret) { diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c index 4071b2d1e8cf..8b44fa542562 100644 --- a/drivers/gpu/drm/gma500/framebuffer.c +++ b/drivers/gpu/drm/gma500/framebuffer.c @@ -125,7 +125,7 @@ static int psbfb_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) psbfb->gtt->offset; page_num = vma_pages(vma); - address = (unsigned long)vmf->virtual_address - (vmf->pgoff << PAGE_SHIFT); + address = vmf->address - (vmf->pgoff << PAGE_SHIFT); vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); diff --git a/drivers/gpu/drm/gma500/gem.c b/drivers/gpu/drm/gma500/gem.c index 6d1cb6b370b1..527c62917660 100644 --- a/drivers/gpu/drm/gma500/gem.c +++ b/drivers/gpu/drm/gma500/gem.c @@ -197,15 +197,14 @@ int psb_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) /* Page relative to the VMA start - we must calculate this ourselves because vmf->pgoff is the fake GEM offset */ - page_offset = ((unsigned long) vmf->virtual_address - vma->vm_start) - >> PAGE_SHIFT; + page_offset = (vmf->address - vma->vm_start) >> PAGE_SHIFT; /* CPU view of the page, don't go via the GART for CPU writes */ if (r->stolen) pfn = (dev_priv->stolen_base + r->offset) >> PAGE_SHIFT; else pfn = page_to_pfn(r->pages[page_offset]); - ret = vm_insert_pfn(vma, (unsigned long)vmf->virtual_address, pfn); + ret = vm_insert_pfn(vma, vmf->address, pfn); fail: mutex_unlock(&dev_priv->mmap_mutex); diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig index 5ddde7349fbd..183f5dc1c3f2 100644 --- a/drivers/gpu/drm/i915/Kconfig +++ b/drivers/gpu/drm/i915/Kconfig @@ -116,6 +116,7 @@ config DRM_I915_GVT_KVMGT tristate "Enable KVM/VFIO support for Intel GVT-g" depends on DRM_I915_GVT depends on KVM + depends on VFIO_MDEV && VFIO_MDEV_DEVICE default n help Choose this option if you want to enable KVMGT support for diff --git a/drivers/gpu/drm/i915/gvt/Makefile b/drivers/gpu/drm/i915/gvt/Makefile index 8a46a7f31d53..b123c20e2097 100644 --- a/drivers/gpu/drm/i915/gvt/Makefile +++ b/drivers/gpu/drm/i915/gvt/Makefile @@ -5,6 +5,4 @@ GVT_SOURCE := gvt.o aperture_gm.o handlers.o vgpu.o trace_points.o firmware.o \ ccflags-y += -I$(src) -I$(src)/$(GVT_DIR) -Wall i915-y += $(addprefix $(GVT_DIR)/, $(GVT_SOURCE)) - -CFLAGS_kvmgt.o := -Wno-unused-function obj-$(CONFIG_DRM_I915_GVT_KVMGT) += $(GVT_DIR)/kvmgt.o diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h index b1a7c8dd4b5f..ad0e9364ee70 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.h +++ b/drivers/gpu/drm/i915/gvt/gvt.h @@ -164,15 +164,17 @@ struct intel_vgpu { #if IS_ENABLED(CONFIG_DRM_I915_GVT_KVMGT) struct { - struct device *mdev; + struct mdev_device *mdev; struct vfio_region *region; int num_regions; struct eventfd_ctx *intx_trigger; struct eventfd_ctx *msi_trigger; struct rb_root cache; struct mutex cache_lock; - void *vfio_group; struct notifier_block iommu_notifier; + struct notifier_block group_notifier; + struct kvm *kvm; + struct work_struct release_work; } vdev; #endif }; diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index dc0365033157..4dd6722a7339 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -31,6 +31,7 @@ #include <linux/init.h> #include <linux/device.h> #include <linux/mm.h> +#include <linux/mmu_context.h> #include <linux/types.h> #include <linux/list.h> #include <linux/rbtree.h> @@ -39,24 +40,13 @@ #include <linux/uuid.h> #include <linux/kvm_host.h> #include <linux/vfio.h> +#include <linux/mdev.h> #include "i915_drv.h" #include "gvt.h" -static inline long kvmgt_pin_pages(struct device *dev, unsigned long *user_pfn, - long npage, int prot, unsigned long *phys_pfn) -{ - return 0; -} -static inline long kvmgt_unpin_pages(struct device *dev, unsigned long *pfn, - long npage) -{ - return 0; -} - static const struct intel_gvt_ops *intel_gvt_ops; - /* helper macros copied from vfio-pci */ #define VFIO_PCI_OFFSET_SHIFT 40 #define VFIO_PCI_OFFSET_TO_INDEX(off) (off >> VFIO_PCI_OFFSET_SHIFT) @@ -90,6 +80,15 @@ struct gvt_dma { kvm_pfn_t pfn; }; +static inline bool handle_valid(unsigned long handle) +{ + return !!(handle & ~0xff); +} + +static int kvmgt_guest_init(struct mdev_device *mdev); +static void intel_vgpu_release_work(struct work_struct *work); +static bool kvmgt_guest_exit(struct kvmgt_guest_info *info); + static struct gvt_dma *__gvt_cache_find(struct intel_vgpu *vgpu, gfn_t gfn) { struct rb_node *node = vgpu->vdev.cache.rb_node; @@ -167,9 +166,10 @@ static void __gvt_cache_remove_entry(struct intel_vgpu *vgpu, static void gvt_cache_remove(struct intel_vgpu *vgpu, gfn_t gfn) { - struct device *dev = vgpu->vdev.mdev; + struct device *dev = &vgpu->vdev.mdev->dev; struct gvt_dma *this; - unsigned long pfn; + unsigned long g1; + int rc; mutex_lock(&vgpu->vdev.cache_lock); this = __gvt_cache_find(vgpu, gfn); @@ -178,8 +178,9 @@ static void gvt_cache_remove(struct intel_vgpu *vgpu, gfn_t gfn) return; } - pfn = this->pfn; - WARN_ON((kvmgt_unpin_pages(dev, &pfn, 1) != 1)); + g1 = gfn; + rc = vfio_unpin_pages(dev, &g1, 1); + WARN_ON(rc != 1); __gvt_cache_remove_entry(vgpu, this); mutex_unlock(&vgpu->vdev.cache_lock); } @@ -194,15 +195,15 @@ static void gvt_cache_destroy(struct intel_vgpu *vgpu) { struct gvt_dma *dma; struct rb_node *node = NULL; - struct device *dev = vgpu->vdev.mdev; - unsigned long pfn; + struct device *dev = &vgpu->vdev.mdev->dev; + unsigned long gfn; mutex_lock(&vgpu->vdev.cache_lock); while ((node = rb_first(&vgpu->vdev.cache))) { dma = rb_entry(node, struct gvt_dma, node); - pfn = dma->pfn; + gfn = dma->gfn; - kvmgt_unpin_pages(dev, &pfn, 1); + vfio_unpin_pages(dev, &gfn, 1); __gvt_cache_remove_entry(vgpu, dma); } mutex_unlock(&vgpu->vdev.cache_lock); @@ -226,7 +227,53 @@ static struct intel_vgpu_type *intel_gvt_find_vgpu_type(struct intel_gvt *gvt, return NULL; } +static ssize_t available_instance_show(struct kobject *kobj, struct device *dev, + char *buf) +{ + struct intel_vgpu_type *type; + unsigned int num = 0; + void *gvt = kdev_to_i915(dev)->gvt; + + type = intel_gvt_find_vgpu_type(gvt, kobject_name(kobj)); + if (!type) + num = 0; + else + num = type->avail_instance; + + return sprintf(buf, "%u\n", num); +} + +static ssize_t device_api_show(struct kobject *kobj, struct device *dev, + char *buf) +{ + return sprintf(buf, "%s\n", VFIO_DEVICE_API_PCI_STRING); +} + +static ssize_t description_show(struct kobject *kobj, struct device *dev, + char *buf) +{ + struct intel_vgpu_type *type; + void *gvt = kdev_to_i915(dev)->gvt; + + type = intel_gvt_find_vgpu_type(gvt, kobject_name(kobj)); + if (!type) + return 0; + + return sprintf(buf, "low_gm_size: %dMB\nhigh_gm_size: %dMB\n" + "fence: %d\n", + BYTES_TO_MB(type->low_gm_size), + BYTES_TO_MB(type->high_gm_size), + type->fence); +} + +static MDEV_TYPE_ATTR_RO(available_instance); +static MDEV_TYPE_ATTR_RO(device_api); +static MDEV_TYPE_ATTR_RO(description); + static struct attribute *type_attrs[] = { + &mdev_type_attr_available_instance.attr, + &mdev_type_attr_device_api.attr, + &mdev_type_attr_description.attr, NULL, }; @@ -322,7 +369,7 @@ static void kvmgt_protect_table_add(struct kvmgt_guest_info *info, gfn_t gfn) if (kvmgt_gfn_is_write_protected(info, gfn)) return; - p = kmalloc(sizeof(struct kvmgt_pgfn), GFP_ATOMIC); + p = kzalloc(sizeof(struct kvmgt_pgfn), GFP_ATOMIC); if (WARN(!p, "gfn: 0x%llx\n", gfn)) return; @@ -342,6 +389,720 @@ static void kvmgt_protect_table_del(struct kvmgt_guest_info *info, } } +static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) +{ + struct intel_vgpu *vgpu; + struct intel_vgpu_type *type; + struct device *pdev; + void *gvt; + + pdev = mdev->parent->dev; + gvt = kdev_to_i915(pdev)->gvt; + + type = intel_gvt_find_vgpu_type(gvt, kobject_name(kobj)); + if (!type) { + gvt_err("failed to find type %s to create\n", + kobject_name(kobj)); + return -EINVAL; + } + + vgpu = intel_gvt_ops->vgpu_create(gvt, type); + if (IS_ERR_OR_NULL(vgpu)) { + gvt_err("create intel vgpu failed\n"); + return -EINVAL; + } + + INIT_WORK(&vgpu->vdev.release_work, intel_vgpu_release_work); + + vgpu->vdev.mdev = mdev; + mdev_set_drvdata(mdev, vgpu); + + gvt_dbg_core("intel_vgpu_create succeeded for mdev: %s\n", + dev_name(&mdev->dev)); + return 0; +} + +static int intel_vgpu_remove(struct mdev_device *mdev) +{ + struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + + if (handle_valid(vgpu->handle)) + return -EBUSY; + + intel_gvt_ops->vgpu_destroy(vgpu); + return 0; +} + +static int intel_vgpu_iommu_notifier(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct intel_vgpu *vgpu = container_of(nb, + struct intel_vgpu, + vdev.iommu_notifier); + + if (action == VFIO_IOMMU_NOTIFY_DMA_UNMAP) { + struct vfio_iommu_type1_dma_unmap *unmap = data; + unsigned long gfn, end_gfn; + + gfn = unmap->iova >> PAGE_SHIFT; + end_gfn = gfn + unmap->size / PAGE_SIZE; + + while (gfn < end_gfn) + gvt_cache_remove(vgpu, gfn++); + } + + return NOTIFY_OK; +} + +static int intel_vgpu_group_notifier(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct intel_vgpu *vgpu = container_of(nb, + struct intel_vgpu, + vdev.group_notifier); + + /* the only action we care about */ + if (action == VFIO_GROUP_NOTIFY_SET_KVM) { + vgpu->vdev.kvm = data; + + if (!data) + schedule_work(&vgpu->vdev.release_work); + } + + return NOTIFY_OK; +} + +static int intel_vgpu_open(struct mdev_device *mdev) +{ + struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + unsigned long events; + int ret; + + vgpu->vdev.iommu_notifier.notifier_call = intel_vgpu_iommu_notifier; + vgpu->vdev.group_notifier.notifier_call = intel_vgpu_group_notifier; + + events = VFIO_IOMMU_NOTIFY_DMA_UNMAP; + ret = vfio_register_notifier(&mdev->dev, VFIO_IOMMU_NOTIFY, &events, + &vgpu->vdev.iommu_notifier); + if (ret != 0) { + gvt_err("vfio_register_notifier for iommu failed: %d\n", ret); + goto out; + } + + events = VFIO_GROUP_NOTIFY_SET_KVM; + ret = vfio_register_notifier(&mdev->dev, VFIO_GROUP_NOTIFY, &events, + &vgpu->vdev.group_notifier); + if (ret != 0) { + gvt_err("vfio_register_notifier for group failed: %d\n", ret); + goto undo_iommu; + } + + return kvmgt_guest_init(mdev); + +undo_iommu: + vfio_unregister_notifier(&mdev->dev, VFIO_IOMMU_NOTIFY, + &vgpu->vdev.iommu_notifier); +out: + return ret; +} + +static void __intel_vgpu_release(struct intel_vgpu *vgpu) +{ + struct kvmgt_guest_info *info; + + if (!handle_valid(vgpu->handle)) + return; + + vfio_unregister_notifier(&vgpu->vdev.mdev->dev, VFIO_IOMMU_NOTIFY, + &vgpu->vdev.iommu_notifier); + vfio_unregister_notifier(&vgpu->vdev.mdev->dev, VFIO_GROUP_NOTIFY, + &vgpu->vdev.group_notifier); + + info = (struct kvmgt_guest_info *)vgpu->handle; + kvmgt_guest_exit(info); + vgpu->handle = 0; +} + +static void intel_vgpu_release(struct mdev_device *mdev) +{ + struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + + __intel_vgpu_release(vgpu); +} + +static void intel_vgpu_release_work(struct work_struct *work) +{ + struct intel_vgpu *vgpu = container_of(work, struct intel_vgpu, + vdev.release_work); + __intel_vgpu_release(vgpu); +} + +static uint64_t intel_vgpu_get_bar0_addr(struct intel_vgpu *vgpu) +{ + u32 start_lo, start_hi; + u32 mem_type; + int pos = PCI_BASE_ADDRESS_0; + + start_lo = (*(u32 *)(vgpu->cfg_space.virtual_cfg_space + pos)) & + PCI_BASE_ADDRESS_MEM_MASK; + mem_type = (*(u32 *)(vgpu->cfg_space.virtual_cfg_space + pos)) & + PCI_BASE_ADDRESS_MEM_TYPE_MASK; + + switch (mem_type) { + case PCI_BASE_ADDRESS_MEM_TYPE_64: + start_hi = (*(u32 *)(vgpu->cfg_space.virtual_cfg_space + + pos + 4)); + break; + case PCI_BASE_ADDRESS_MEM_TYPE_32: + case PCI_BASE_ADDRESS_MEM_TYPE_1M: + /* 1M mem BAR treated as 32-bit BAR */ + default: + /* mem unknown type treated as 32-bit BAR */ + start_hi = 0; + break; + } + + return ((u64)start_hi << 32) | start_lo; +} + +static ssize_t intel_vgpu_rw(struct mdev_device *mdev, char *buf, + size_t count, loff_t *ppos, bool is_write) +{ + struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + unsigned int index = VFIO_PCI_OFFSET_TO_INDEX(*ppos); + uint64_t pos = *ppos & VFIO_PCI_OFFSET_MASK; + int ret = -EINVAL; + + + if (index >= VFIO_PCI_NUM_REGIONS) { + gvt_err("invalid index: %u\n", index); + return -EINVAL; + } + + switch (index) { + case VFIO_PCI_CONFIG_REGION_INDEX: + if (is_write) + ret = intel_gvt_ops->emulate_cfg_write(vgpu, pos, + buf, count); + else + ret = intel_gvt_ops->emulate_cfg_read(vgpu, pos, + buf, count); + break; + case VFIO_PCI_BAR0_REGION_INDEX: + case VFIO_PCI_BAR1_REGION_INDEX: + if (is_write) { + uint64_t bar0_start = intel_vgpu_get_bar0_addr(vgpu); + + ret = intel_gvt_ops->emulate_mmio_write(vgpu, + bar0_start + pos, buf, count); + } else { + uint64_t bar0_start = intel_vgpu_get_bar0_addr(vgpu); + + ret = intel_gvt_ops->emulate_mmio_read(vgpu, + bar0_start + pos, buf, count); + } + break; + case VFIO_PCI_BAR2_REGION_INDEX: + case VFIO_PCI_BAR3_REGION_INDEX: + case VFIO_PCI_BAR4_REGION_INDEX: + case VFIO_PCI_BAR5_REGION_INDEX: + case VFIO_PCI_VGA_REGION_INDEX: + case VFIO_PCI_ROM_REGION_INDEX: + default: + gvt_err("unsupported region: %u\n", index); + } + + return ret == 0 ? count : ret; +} + +static ssize_t intel_vgpu_read(struct mdev_device *mdev, char __user *buf, + size_t count, loff_t *ppos) +{ + unsigned int done = 0; + int ret; + + while (count) { + size_t filled; + + if (count >= 4 && !(*ppos % 4)) { + u32 val; + + ret = intel_vgpu_rw(mdev, (char *)&val, sizeof(val), + ppos, false); + if (ret <= 0) + goto read_err; + + if (copy_to_user(buf, &val, sizeof(val))) + goto read_err; + + filled = 4; + } else if (count >= 2 && !(*ppos % 2)) { + u16 val; + + ret = intel_vgpu_rw(mdev, (char *)&val, sizeof(val), + ppos, false); + if (ret <= 0) + goto read_err; + + if (copy_to_user(buf, &val, sizeof(val))) + goto read_err; + + filled = 2; + } else { + u8 val; + + ret = intel_vgpu_rw(mdev, &val, sizeof(val), ppos, + false); + if (ret <= 0) + goto read_err; + + if (copy_to_user(buf, &val, sizeof(val))) + goto read_err; + + filled = 1; + } + + count -= filled; + done += filled; + *ppos += filled; + buf += filled; + } + + return done; + +read_err: + return -EFAULT; +} + +static ssize_t intel_vgpu_write(struct mdev_device *mdev, + const char __user *buf, + size_t count, loff_t *ppos) +{ + unsigned int done = 0; + int ret; + + while (count) { + size_t filled; + + if (count >= 4 && !(*ppos % 4)) { + u32 val; + + if (copy_from_user(&val, buf, sizeof(val))) + goto write_err; + + ret = intel_vgpu_rw(mdev, (char *)&val, sizeof(val), + ppos, true); + if (ret <= 0) + goto write_err; + + filled = 4; + } else if (count >= 2 && !(*ppos % 2)) { + u16 val; + + if (copy_from_user(&val, buf, sizeof(val))) + goto write_err; + + ret = intel_vgpu_rw(mdev, (char *)&val, + sizeof(val), ppos, true); + if (ret <= 0) + goto write_err; + + filled = 2; + } else { + u8 val; + + if (copy_from_user(&val, buf, sizeof(val))) + goto write_err; + + ret = intel_vgpu_rw(mdev, &val, sizeof(val), + ppos, true); + if (ret <= 0) + goto write_err; + + filled = 1; + } + + count -= filled; + done += filled; + *ppos += filled; + buf += filled; + } + + return done; +write_err: + return -EFAULT; +} + +static int intel_vgpu_mmap(struct mdev_device *mdev, struct vm_area_struct *vma) +{ + unsigned int index; + u64 virtaddr; + unsigned long req_size, pgoff = 0; + pgprot_t pg_prot; + struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + + index = vma->vm_pgoff >> (VFIO_PCI_OFFSET_SHIFT - PAGE_SHIFT); + if (index >= VFIO_PCI_ROM_REGION_INDEX) + return -EINVAL; + + if (vma->vm_end < vma->vm_start) + return -EINVAL; + if ((vma->vm_flags & VM_SHARED) == 0) + return -EINVAL; + if (index != VFIO_PCI_BAR2_REGION_INDEX) + return -EINVAL; + + pg_prot = vma->vm_page_prot; + virtaddr = vma->vm_start; + req_size = vma->vm_end - vma->vm_start; + pgoff = vgpu_aperture_pa_base(vgpu) >> PAGE_SHIFT; + + return remap_pfn_range(vma, virtaddr, pgoff, req_size, pg_prot); +} + +static int intel_vgpu_get_irq_count(struct intel_vgpu *vgpu, int type) +{ + if (type == VFIO_PCI_INTX_IRQ_INDEX || type == VFIO_PCI_MSI_IRQ_INDEX) + return 1; + + return 0; +} + +static int intel_vgpu_set_intx_mask(struct intel_vgpu *vgpu, + unsigned int index, unsigned int start, + unsigned int count, uint32_t flags, + void *data) +{ + return 0; +} + +static int intel_vgpu_set_intx_unmask(struct intel_vgpu *vgpu, + unsigned int index, unsigned int start, + unsigned int count, uint32_t flags, void *data) +{ + return 0; +} + +static int intel_vgpu_set_intx_trigger(struct intel_vgpu *vgpu, + unsigned int index, unsigned int start, unsigned int count, + uint32_t flags, void *data) +{ + return 0; +} + +static int intel_vgpu_set_msi_trigger(struct intel_vgpu *vgpu, + unsigned int index, unsigned int start, unsigned int count, + uint32_t flags, void *data) +{ + struct eventfd_ctx *trigger; + + if (flags & VFIO_IRQ_SET_DATA_EVENTFD) { + int fd = *(int *)data; + + trigger = eventfd_ctx_fdget(fd); + if (IS_ERR(trigger)) { + gvt_err("eventfd_ctx_fdget failed\n"); + return PTR_ERR(trigger); + } + vgpu->vdev.msi_trigger = trigger; + } + + return 0; +} + +static int intel_vgpu_set_irqs(struct intel_vgpu *vgpu, uint32_t flags, + unsigned int index, unsigned int start, unsigned int count, + void *data) +{ + int (*func)(struct intel_vgpu *vgpu, unsigned int index, + unsigned int start, unsigned int count, uint32_t flags, + void *data) = NULL; + + switch (index) { + case VFIO_PCI_INTX_IRQ_INDEX: + switch (flags & VFIO_IRQ_SET_ACTION_TYPE_MASK) { + case VFIO_IRQ_SET_ACTION_MASK: + func = intel_vgpu_set_intx_mask; + break; + case VFIO_IRQ_SET_ACTION_UNMASK: + func = intel_vgpu_set_intx_unmask; + break; + case VFIO_IRQ_SET_ACTION_TRIGGER: + func = intel_vgpu_set_intx_trigger; + break; + } + break; + case VFIO_PCI_MSI_IRQ_INDEX: + switch (flags & VFIO_IRQ_SET_ACTION_TYPE_MASK) { + case VFIO_IRQ_SET_ACTION_MASK: + case VFIO_IRQ_SET_ACTION_UNMASK: + /* XXX Need masking support exported */ + break; + case VFIO_IRQ_SET_ACTION_TRIGGER: + func = intel_vgpu_set_msi_trigger; + break; + } + break; + } + + if (!func) + return -ENOTTY; + + return func(vgpu, index, start, count, flags, data); +} + +static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd, + unsigned long arg) +{ + struct intel_vgpu *vgpu = mdev_get_drvdata(mdev); + unsigned long minsz; + + gvt_dbg_core("vgpu%d ioctl, cmd: %d\n", vgpu->id, cmd); + + if (cmd == VFIO_DEVICE_GET_INFO) { + struct vfio_device_info info; + + minsz = offsetofend(struct vfio_device_info, num_irqs); + + if (copy_from_user(&info, (void __user *)arg, minsz)) + return -EFAULT; + + if (info.argsz < minsz) + return -EINVAL; + + info.flags = VFIO_DEVICE_FLAGS_PCI; + info.flags |= VFIO_DEVICE_FLAGS_RESET; + info.num_regions = VFIO_PCI_NUM_REGIONS; + info.num_irqs = VFIO_PCI_NUM_IRQS; + + return copy_to_user((void __user *)arg, &info, minsz) ? + -EFAULT : 0; + + } else if (cmd == VFIO_DEVICE_GET_REGION_INFO) { + struct vfio_region_info info; + struct vfio_info_cap caps = { .buf = NULL, .size = 0 }; + int i, ret; + struct vfio_region_info_cap_sparse_mmap *sparse = NULL; + size_t size; + int nr_areas = 1; + int cap_type_id; + + minsz = offsetofend(struct vfio_region_info, offset); + + if (copy_from_user(&info, (void __user *)arg, minsz)) + return -EFAULT; + + if (info.argsz < minsz) + return -EINVAL; + + switch (info.index) { + case VFIO_PCI_CONFIG_REGION_INDEX: + info.offset = VFIO_PCI_INDEX_TO_OFFSET(info.index); + info.size = INTEL_GVT_MAX_CFG_SPACE_SZ; + info.flags = VFIO_REGION_INFO_FLAG_READ | + VFIO_REGION_INFO_FLAG_WRITE; + break; + case VFIO_PCI_BAR0_REGION_INDEX: + info.offset = VFIO_PCI_INDEX_TO_OFFSET(info.index); + info.size = vgpu->cfg_space.bar[info.index].size; + if (!info.size) { + info.flags = 0; + break; + } + + info.flags = VFIO_REGION_INFO_FLAG_READ | + VFIO_REGION_INFO_FLAG_WRITE; + break; + case VFIO_PCI_BAR1_REGION_INDEX: + info.offset = VFIO_PCI_INDEX_TO_OFFSET(info.index); + info.size = 0; + info.flags = 0; + break; + case VFIO_PCI_BAR2_REGION_INDEX: + info.offset = VFIO_PCI_INDEX_TO_OFFSET(info.index); + info.flags = VFIO_REGION_INFO_FLAG_CAPS | + VFIO_REGION_INFO_FLAG_MMAP | + VFIO_REGION_INFO_FLAG_READ | + VFIO_REGION_INFO_FLAG_WRITE; + info.size = gvt_aperture_sz(vgpu->gvt); + + size = sizeof(*sparse) + + (nr_areas * sizeof(*sparse->areas)); + sparse = kzalloc(size, GFP_KERNEL); + if (!sparse) + return -ENOMEM; + + sparse->nr_areas = nr_areas; + cap_type_id = VFIO_REGION_INFO_CAP_SPARSE_MMAP; + sparse->areas[0].offset = + PAGE_ALIGN(vgpu_aperture_offset(vgpu)); + sparse->areas[0].size = vgpu_aperture_sz(vgpu); + if (!caps.buf) { + kfree(caps.buf); + caps.buf = NULL; + caps.size = 0; + } + break; + + case VFIO_PCI_BAR3_REGION_INDEX ... VFIO_PCI_BAR5_REGION_INDEX: + info.offset = VFIO_PCI_INDEX_TO_OFFSET(info.index); + info.size = 0; + + info.flags = 0; + gvt_dbg_core("get region info bar:%d\n", info.index); + break; + + case VFIO_PCI_ROM_REGION_INDEX: + case VFIO_PCI_VGA_REGION_INDEX: + gvt_dbg_core("get region info index:%d\n", info.index); + break; + default: + { + struct vfio_region_info_cap_type cap_type; + + if (info.index >= VFIO_PCI_NUM_REGIONS + + vgpu->vdev.num_regions) + return -EINVAL; + + i = info.index - VFIO_PCI_NUM_REGIONS; + + info.offset = + VFIO_PCI_INDEX_TO_OFFSET(info.index); + info.size = vgpu->vdev.region[i].size; + info.flags = vgpu->vdev.region[i].flags; + + cap_type.type = vgpu->vdev.region[i].type; + cap_type.subtype = vgpu->vdev.region[i].subtype; + + ret = vfio_info_add_capability(&caps, + VFIO_REGION_INFO_CAP_TYPE, + &cap_type); + if (ret) + return ret; + } + } + + if ((info.flags & VFIO_REGION_INFO_FLAG_CAPS) && sparse) { + switch (cap_type_id) { + case VFIO_REGION_INFO_CAP_SPARSE_MMAP: + ret = vfio_info_add_capability(&caps, + VFIO_REGION_INFO_CAP_SPARSE_MMAP, + sparse); + kfree(sparse); + if (ret) + return ret; + break; + default: + return -EINVAL; + } + } + + if (caps.size) { + if (info.argsz < sizeof(info) + caps.size) { + info.argsz = sizeof(info) + caps.size; + info.cap_offset = 0; + } else { + vfio_info_cap_shift(&caps, sizeof(info)); + if (copy_to_user((void __user *)arg + + sizeof(info), caps.buf, + caps.size)) { + kfree(caps.buf); + return -EFAULT; + } + info.cap_offset = sizeof(info); + } + + kfree(caps.buf); + } + + return copy_to_user((void __user *)arg, &info, minsz) ? + -EFAULT : 0; + } else if (cmd == VFIO_DEVICE_GET_IRQ_INFO) { + struct vfio_irq_info info; + + minsz = offsetofend(struct vfio_irq_info, count); + + if (copy_from_user(&info, (void __user *)arg, minsz)) + return -EFAULT; + + if (info.argsz < minsz || info.index >= VFIO_PCI_NUM_IRQS) + return -EINVAL; + + switch (info.index) { + case VFIO_PCI_INTX_IRQ_INDEX: + case VFIO_PCI_MSI_IRQ_INDEX: + break; + default: + return -EINVAL; + } + + info.flags = VFIO_IRQ_INFO_EVENTFD; + + info.count = intel_vgpu_get_irq_count(vgpu, info.index); + + if (info.index == VFIO_PCI_INTX_IRQ_INDEX) + info.flags |= (VFIO_IRQ_INFO_MASKABLE | + VFIO_IRQ_INFO_AUTOMASKED); + else + info.flags |= VFIO_IRQ_INFO_NORESIZE; + + return copy_to_user((void __user *)arg, &info, minsz) ? + -EFAULT : 0; + } else if (cmd == VFIO_DEVICE_SET_IRQS) { + struct vfio_irq_set hdr; + u8 *data = NULL; + int ret = 0; + size_t data_size = 0; + + minsz = offsetofend(struct vfio_irq_set, count); + + if (copy_from_user(&hdr, (void __user *)arg, minsz)) + return -EFAULT; + + if (!(hdr.flags & VFIO_IRQ_SET_DATA_NONE)) { + int max = intel_vgpu_get_irq_count(vgpu, hdr.index); + + ret = vfio_set_irqs_validate_and_prepare(&hdr, max, + VFIO_PCI_NUM_IRQS, &data_size); + if (ret) { + gvt_err("intel:vfio_set_irqs_validate_and_prepare failed\n"); + return -EINVAL; + } + if (data_size) { + data = memdup_user((void __user *)(arg + minsz), + data_size); + if (IS_ERR(data)) + return PTR_ERR(data); + } + } + + ret = intel_vgpu_set_irqs(vgpu, hdr.flags, hdr.index, + hdr.start, hdr.count, data); + kfree(data); + + return ret; + } else if (cmd == VFIO_DEVICE_RESET) { + intel_gvt_ops->vgpu_reset(vgpu); + return 0; + } + + return 0; +} + +static const struct parent_ops intel_vgpu_ops = { + .supported_type_groups = intel_vgpu_type_groups, + .create = intel_vgpu_create, + .remove = intel_vgpu_remove, + + .open = intel_vgpu_open, + .release = intel_vgpu_release, + + .read = intel_vgpu_read, + .write = intel_vgpu_write, + .mmap = intel_vgpu_mmap, + .ioctl = intel_vgpu_ioctl, +}; + static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops) { if (!intel_gvt_init_vgpu_type_groups(gvt)) @@ -349,22 +1110,28 @@ static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops) intel_gvt_ops = ops; - /* MDEV is not yet available */ - return -ENODEV; + return mdev_register_device(dev, &intel_vgpu_ops); } static void kvmgt_host_exit(struct device *dev, void *gvt) { intel_gvt_cleanup_vgpu_type_groups(gvt); + mdev_unregister_device(dev); } static int kvmgt_write_protect_add(unsigned long handle, u64 gfn) { - struct kvmgt_guest_info *info = (struct kvmgt_guest_info *)handle; - struct kvm *kvm = info->kvm; + struct kvmgt_guest_info *info; + struct kvm *kvm; struct kvm_memory_slot *slot; int idx; + if (!handle_valid(handle)) + return -ESRCH; + + info = (struct kvmgt_guest_info *)handle; + kvm = info->kvm; + idx = srcu_read_lock(&kvm->srcu); slot = gfn_to_memslot(kvm, gfn); @@ -384,11 +1151,17 @@ out: static int kvmgt_write_protect_remove(unsigned long handle, u64 gfn) { - struct kvmgt_guest_info *info = (struct kvmgt_guest_info *)handle; - struct kvm *kvm = info->kvm; + struct kvmgt_guest_info *info; + struct kvm *kvm; struct kvm_memory_slot *slot; int idx; + if (!handle_valid(handle)) + return 0; + + info = (struct kvmgt_guest_info *)handle; + kvm = info->kvm; + idx = srcu_read_lock(&kvm->srcu); slot = gfn_to_memslot(kvm, gfn); @@ -476,6 +1249,85 @@ static int kvmgt_detect_host(void) return kvmgt_check_guest() ? -ENODEV : 0; } +static bool __kvmgt_vgpu_exist(struct intel_vgpu *vgpu, struct kvm *kvm) +{ + struct intel_vgpu *itr; + struct kvmgt_guest_info *info; + int id; + bool ret = false; + + mutex_lock(&vgpu->gvt->lock); + for_each_active_vgpu(vgpu->gvt, itr, id) { + if (!handle_valid(itr->handle)) + continue; + + info = (struct kvmgt_guest_info *)itr->handle; + if (kvm && kvm == info->kvm) { + ret = true; + goto out; + } + } +out: + mutex_unlock(&vgpu->gvt->lock); + return ret; +} + +static int kvmgt_guest_init(struct mdev_device *mdev) +{ + struct kvmgt_guest_info *info; + struct intel_vgpu *vgpu; + struct kvm *kvm; + + vgpu = mdev_get_drvdata(mdev); + if (handle_valid(vgpu->handle)) + return -EEXIST; + + kvm = vgpu->vdev.kvm; + if (!kvm || kvm->mm != current->mm) { + gvt_err("KVM is required to use Intel vGPU\n"); + return -ESRCH; + } + + if (__kvmgt_vgpu_exist(vgpu, kvm)) + return -EEXIST; + + info = vzalloc(sizeof(struct kvmgt_guest_info)); + if (!info) + return -ENOMEM; + + vgpu->handle = (unsigned long)info; + info->vgpu = vgpu; + info->kvm = kvm; + + kvmgt_protect_table_init(info); + gvt_cache_init(vgpu); + + info->track_node.track_write = kvmgt_page_track_write; + info->track_node.track_flush_slot = kvmgt_page_track_flush_slot; + kvm_page_track_register_notifier(kvm, &info->track_node); + + return 0; +} + +static bool kvmgt_guest_exit(struct kvmgt_guest_info *info) +{ + struct intel_vgpu *vgpu; + + if (!info) { + gvt_err("kvmgt_guest_info invalid\n"); + return false; + } + + vgpu = info->vgpu; + + kvm_page_track_unregister_notifier(info->kvm, &info->track_node); + kvmgt_protect_table_destroy(info); + gvt_cache_destroy(vgpu); + vfree(info); + + return true; +} + static int kvmgt_attach_vgpu(void *vgpu, unsigned long *handle) { /* nothing to do here */ @@ -489,63 +1341,72 @@ static void kvmgt_detach_vgpu(unsigned long handle) static int kvmgt_inject_msi(unsigned long handle, u32 addr, u16 data) { - struct kvmgt_guest_info *info = (struct kvmgt_guest_info *)handle; - struct intel_vgpu *vgpu = info->vgpu; + struct kvmgt_guest_info *info; + struct intel_vgpu *vgpu; - if (vgpu->vdev.msi_trigger) - return eventfd_signal(vgpu->vdev.msi_trigger, 1) == 1; + if (!handle_valid(handle)) + return -ESRCH; - return false; + info = (struct kvmgt_guest_info *)handle; + vgpu = info->vgpu; + + if (eventfd_signal(vgpu->vdev.msi_trigger, 1) == 1) + return 0; + + return -EFAULT; } static unsigned long kvmgt_gfn_to_pfn(unsigned long handle, unsigned long gfn) { unsigned long pfn; - struct kvmgt_guest_info *info = (struct kvmgt_guest_info *)handle; + struct kvmgt_guest_info *info; + struct device *dev; int rc; + if (!handle_valid(handle)) + return INTEL_GVT_INVALID_ADDR; + + info = (struct kvmgt_guest_info *)handle; pfn = gvt_cache_find(info->vgpu, gfn); if (pfn != 0) return pfn; - rc = kvmgt_pin_pages(info->vgpu->vdev.mdev, &gfn, 1, - IOMMU_READ | IOMMU_WRITE, &pfn); + pfn = INTEL_GVT_INVALID_ADDR; + dev = &info->vgpu->vdev.mdev->dev; + rc = vfio_pin_pages(dev, &gfn, 1, IOMMU_READ | IOMMU_WRITE, &pfn); if (rc != 1) { - gvt_err("vfio_pin_pages failed for gfn: 0x%lx\n", gfn); - return 0; + gvt_err("vfio_pin_pages failed for gfn 0x%lx: %d\n", gfn, rc); + return INTEL_GVT_INVALID_ADDR; } gvt_cache_add(info->vgpu, gfn, pfn); return pfn; } -static void *kvmgt_gpa_to_hva(unsigned long handle, unsigned long gpa) +static int kvmgt_rw_gpa(unsigned long handle, unsigned long gpa, + void *buf, unsigned long len, bool write) { - unsigned long pfn; - gfn_t gfn = gpa_to_gfn(gpa); + struct kvmgt_guest_info *info; + struct kvm *kvm; + int ret; + bool kthread = current->mm == NULL; - pfn = kvmgt_gfn_to_pfn(handle, gfn); - if (!pfn) - return NULL; + if (!handle_valid(handle)) + return -ESRCH; - return (char *)pfn_to_kaddr(pfn) + offset_in_page(gpa); -} + info = (struct kvmgt_guest_info *)handle; + kvm = info->kvm; -static int kvmgt_rw_gpa(unsigned long handle, unsigned long gpa, - void *buf, unsigned long len, bool write) -{ - void *hva = NULL; + if (kthread) + use_mm(kvm->mm); - hva = kvmgt_gpa_to_hva(handle, gpa); - if (!hva) - return -EFAULT; + ret = write ? kvm_write_guest(kvm, gpa, buf, len) : + kvm_read_guest(kvm, gpa, buf, len); - if (write) - memcpy(hva, buf, len); - else - memcpy(buf, hva, len); + if (kthread) + unuse_mm(kvm->mm); - return 0; + return ret; } static int kvmgt_read_gpa(unsigned long handle, unsigned long gpa, diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index d0dcaf35b429..412f3513f269 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1796,8 +1796,7 @@ int i915_gem_fault(struct vm_area_struct *area, struct vm_fault *vmf) int ret; /* We don't use vmf->pgoff since that has the fake offset */ - page_offset = ((unsigned long)vmf->virtual_address - area->vm_start) >> - PAGE_SHIFT; + page_offset = (vmf->address - area->vm_start) >> PAGE_SHIFT; trace_i915_gem_object_fault(obj, page_offset, true, write); diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c index 57194471f8cd..b8f403faadbb 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.c +++ b/drivers/gpu/drm/i915/i915_gem_request.c @@ -972,7 +972,7 @@ bool __i915_spin_request(const struct drm_i915_gem_request *req, if (busywait_stop(timeout_us, cpu)) break; - cpu_relax_lowlatency(); + cpu_relax(); } while (!need_resched()); return false; diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c index a6fc1bdc48af..401006b4c6a3 100644 --- a/drivers/gpu/drm/i915/i915_gem_shrinker.c +++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c @@ -35,31 +35,22 @@ #include "i915_drv.h" #include "i915_trace.h" -static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task) +static bool i915_gem_shrinker_lock(struct drm_device *dev, bool *unlock) { - if (!mutex_is_locked(mutex)) + switch (mutex_trylock_recursive(&dev->struct_mutex)) { + case MUTEX_TRYLOCK_FAILED: return false; -#if defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_MUTEX_SPIN_ON_OWNER) - return mutex->owner == task; -#else - /* Since UP may be pre-empted, we cannot assume that we own the lock */ - return false; -#endif -} - -static bool i915_gem_shrinker_lock(struct drm_device *dev, bool *unlock) -{ - if (!mutex_trylock(&dev->struct_mutex)) { - if (!mutex_is_locked_by(&dev->struct_mutex, current)) - return false; + case MUTEX_TRYLOCK_SUCCESS: + *unlock = true; + return true; + case MUTEX_TRYLOCK_RECURSIVE: *unlock = false; - } else { - *unlock = true; + return true; } - return true; + BUG(); } static bool any_vma_pinned(struct drm_i915_gem_object *obj) diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c index 107ddf51065e..d068af2ec3a3 100644 --- a/drivers/gpu/drm/i915/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -515,7 +515,7 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work) obj->userptr.ptr + pinned * PAGE_SIZE, npages - pinned, flags, - pvec + pinned, NULL); + pvec + pinned, NULL, NULL); if (ret < 0) break; diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index cd06cfd94687..d8bc59c7e261 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -225,16 +225,14 @@ int msm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) } /* We don't use vmf->pgoff since that has the fake offset: */ - pgoff = ((unsigned long)vmf->virtual_address - - vma->vm_start) >> PAGE_SHIFT; + pgoff = (vmf->address - vma->vm_start) >> PAGE_SHIFT; pfn = page_to_pfn(pages[pgoff]); - VERB("Inserting %p pfn %lx, pa %lx", vmf->virtual_address, + VERB("Inserting %p pfn %lx, pa %lx", (void *)vmf->address, pfn, pfn << PAGE_SHIFT); - ret = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, - __pfn_to_pfn_t(pfn, PFN_DEV)); + ret = vm_insert_mixed(vma, vmf->address, __pfn_to_pfn_t(pfn, PFN_DEV)); out_unlock: mutex_unlock(&dev->struct_mutex); diff --git a/drivers/gpu/drm/msm/msm_gem_shrinker.c b/drivers/gpu/drm/msm/msm_gem_shrinker.c index 192b2d3a79cb..ab1dd020eb04 100644 --- a/drivers/gpu/drm/msm/msm_gem_shrinker.c +++ b/drivers/gpu/drm/msm/msm_gem_shrinker.c @@ -18,33 +18,24 @@ #include "msm_drv.h" #include "msm_gem.h" -static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task) +static bool msm_gem_shrinker_lock(struct drm_device *dev, bool *unlock) { - if (!mutex_is_locked(mutex)) + switch (mutex_trylock_recursive(&dev->struct_mutex)) { + case MUTEX_TRYLOCK_FAILED: return false; -#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_MUTEXES) - return mutex->owner == task; -#else - /* Since UP may be pre-empted, we cannot assume that we own the lock */ - return false; -#endif -} + case MUTEX_TRYLOCK_SUCCESS: + *unlock = true; + return true; -static bool msm_gem_shrinker_lock(struct drm_device *dev, bool *unlock) -{ - if (!mutex_trylock(&dev->struct_mutex)) { - if (!mutex_is_locked_by(&dev->struct_mutex, current)) - return false; + case MUTEX_TRYLOCK_RECURSIVE: *unlock = false; - } else { - *unlock = true; + return true; } - return true; + BUG(); } - static unsigned long msm_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc) { diff --git a/drivers/gpu/drm/omapdrm/omap_gem.c b/drivers/gpu/drm/omapdrm/omap_gem.c index d4e1e11466f8..4a90c690f09e 100644 --- a/drivers/gpu/drm/omapdrm/omap_gem.c +++ b/drivers/gpu/drm/omapdrm/omap_gem.c @@ -398,8 +398,7 @@ static int fault_1d(struct drm_gem_object *obj, pgoff_t pgoff; /* We don't use vmf->pgoff since that has the fake offset: */ - pgoff = ((unsigned long)vmf->virtual_address - - vma->vm_start) >> PAGE_SHIFT; + pgoff = (vmf->address - vma->vm_start) >> PAGE_SHIFT; if (omap_obj->pages) { omap_gem_cpu_sync(obj, pgoff); @@ -409,11 +408,10 @@ static int fault_1d(struct drm_gem_object *obj, pfn = (omap_obj->paddr >> PAGE_SHIFT) + pgoff; } - VERB("Inserting %p pfn %lx, pa %lx", vmf->virtual_address, + VERB("Inserting %p pfn %lx, pa %lx", (void *)vmf->address, pfn, pfn << PAGE_SHIFT); - return vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, - __pfn_to_pfn_t(pfn, PFN_DEV)); + return vm_insert_mixed(vma, vmf->address, __pfn_to_pfn_t(pfn, PFN_DEV)); } /* Special handling for the case of faulting in 2d tiled buffers */ @@ -427,7 +425,7 @@ static int fault_2d(struct drm_gem_object *obj, struct page *pages[64]; /* XXX is this too much to have on stack? */ unsigned long pfn; pgoff_t pgoff, base_pgoff; - void __user *vaddr; + unsigned long vaddr; int i, ret, slots; /* @@ -447,8 +445,7 @@ static int fault_2d(struct drm_gem_object *obj, const int m = 1 + ((omap_obj->width << fmt) / PAGE_SIZE); /* We don't use vmf->pgoff since that has the fake offset: */ - pgoff = ((unsigned long)vmf->virtual_address - - vma->vm_start) >> PAGE_SHIFT; + pgoff = (vmf->address - vma->vm_start) >> PAGE_SHIFT; /* * Actual address we start mapping at is rounded down to previous slot @@ -459,7 +456,7 @@ static int fault_2d(struct drm_gem_object *obj, /* figure out buffer width in slots */ slots = omap_obj->width >> priv->usergart[fmt].slot_shift; - vaddr = vmf->virtual_address - ((pgoff - base_pgoff) << PAGE_SHIFT); + vaddr = vmf->address - ((pgoff - base_pgoff) << PAGE_SHIFT); entry = &priv->usergart[fmt].entry[priv->usergart[fmt].last]; @@ -503,12 +500,11 @@ static int fault_2d(struct drm_gem_object *obj, pfn = entry->paddr >> PAGE_SHIFT; - VERB("Inserting %p pfn %lx, pa %lx", vmf->virtual_address, + VERB("Inserting %p pfn %lx, pa %lx", (void *)vmf->address, pfn, pfn << PAGE_SHIFT); for (i = n; i > 0; i--) { - vm_insert_mixed(vma, (unsigned long)vaddr, - __pfn_to_pfn_t(pfn, PFN_DEV)); + vm_insert_mixed(vma, vaddr, __pfn_to_pfn_t(pfn, PFN_DEV)); pfn += priv->usergart[fmt].stride_pfn; vaddr += PAGE_SIZE * m; } diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c index 21b6732425c5..c829cfb02fc4 100644 --- a/drivers/gpu/drm/radeon/radeon_bios.c +++ b/drivers/gpu/drm/radeon/radeon_bios.c @@ -603,8 +603,9 @@ static bool radeon_acpi_vfct_bios(struct radeon_device *rdev) GOP_VBIOS_CONTENT *vbios; VFCT_IMAGE_HEADER *vhdr; - if (!ACPI_SUCCESS(acpi_get_table_with_size("VFCT", 1, &hdr, &tbl_size))) + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr))) return false; + tbl_size = hdr->length; if (tbl_size < sizeof(UEFI_ACPI_VFCT)) { DRM_ERROR("ACPI VFCT table present but broken (too short #1)\n"); goto out_unmap; diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c index c08e5279eeac..7d853e6b5ff0 100644 --- a/drivers/gpu/drm/tegra/gem.c +++ b/drivers/gpu/drm/tegra/gem.c @@ -452,10 +452,10 @@ static int tegra_bo_fault(struct vm_area_struct *vma, struct vm_fault *vmf) if (!bo->pages) return VM_FAULT_SIGBUS; - offset = ((unsigned long)vmf->virtual_address - vma->vm_start) >> PAGE_SHIFT; + offset = (vmf->address - vma->vm_start) >> PAGE_SHIFT; page = bo->pages[offset]; - err = vm_insert_page(vma, (unsigned long)vmf->virtual_address, page); + err = vm_insert_page(vma, vmf->address, page); switch (err) { case -EAGAIN: case 0: diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c index 4748aedc933a..68ef993ab431 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_vm.c +++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c @@ -101,7 +101,7 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) struct page *page; int ret; int i; - unsigned long address = (unsigned long)vmf->virtual_address; + unsigned long address = vmf->address; int retval = VM_FAULT_NOPAGE; struct ttm_mem_type_manager *man = &bdev->man[bo->mem.mem_type]; diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c index 818e70712b18..3c0c4bd3f750 100644 --- a/drivers/gpu/drm/udl/udl_gem.c +++ b/drivers/gpu/drm/udl/udl_gem.c @@ -107,14 +107,13 @@ int udl_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) unsigned int page_offset; int ret = 0; - page_offset = ((unsigned long)vmf->virtual_address - vma->vm_start) >> - PAGE_SHIFT; + page_offset = (vmf->address - vma->vm_start) >> PAGE_SHIFT; if (!obj->pages) return VM_FAULT_SIGBUS; page = obj->pages[page_offset]; - ret = vm_insert_page(vma, (unsigned long)vmf->virtual_address, page); + ret = vm_insert_page(vma, vmf->address, page); switch (ret) { case -EAGAIN: case 0: diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c index f36c14729b55..477e07f0ecb6 100644 --- a/drivers/gpu/drm/vgem/vgem_drv.c +++ b/drivers/gpu/drm/vgem/vgem_drv.c @@ -54,7 +54,7 @@ static int vgem_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { struct drm_vgem_gem_object *obj = vma->vm_private_data; /* We don't use vmf->pgoff since that has the fake offset */ - unsigned long vaddr = (unsigned long)vmf->virtual_address; + unsigned long vaddr = vmf->address; struct page *page; page = shmem_read_mapping_page(file_inode(obj->base.filp)->i_mapping, diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c index cb75f0663ba0..11288ffa4af6 100644 --- a/drivers/gpu/drm/virtio/virtgpu_plane.c +++ b/drivers/gpu/drm/virtio/virtgpu_plane.c @@ -88,8 +88,8 @@ static void virtio_gpu_primary_plane_update(struct drm_plane *plane, (vgdev, handle, 0, cpu_to_le32(plane->state->src_w >> 16), cpu_to_le32(plane->state->src_h >> 16), - plane->state->src_x >> 16, - plane->state->src_y >> 16, NULL); + cpu_to_le32(plane->state->src_x >> 16), + cpu_to_le32(plane->state->src_y >> 16), NULL); } } else { handle = 0; diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c index 974f9410474b..43ea0dc957d2 100644 --- a/drivers/gpu/drm/virtio/virtgpu_vq.c +++ b/drivers/gpu/drm/virtio/virtgpu_vq.c @@ -109,8 +109,10 @@ void virtio_gpu_free_vbufs(struct virtio_gpu_device *vgdev) spin_lock(&vgdev->free_vbufs_lock); for (i = 0; i < count; i++) { - if (WARN_ON(list_empty(&vgdev->free_vbufs))) + if (WARN_ON(list_empty(&vgdev->free_vbufs))) { + spin_unlock(&vgdev->free_vbufs_lock); return; + } vbuf = list_first_entry(&vgdev->free_vbufs, struct virtio_gpu_vbuffer, list); list_del(&vbuf->list); @@ -295,6 +297,8 @@ void virtio_gpu_dequeue_cursor_func(struct work_struct *work) static int virtio_gpu_queue_ctrl_buffer_locked(struct virtio_gpu_device *vgdev, struct virtio_gpu_vbuffer *vbuf) + __releases(&vgdev->ctrlq.qlock) + __acquires(&vgdev->ctrlq.qlock) { struct virtqueue *vq = vgdev->ctrlq.vq; struct scatterlist *sgs[3], vcmd, vout, vresp; |
