diff options
author | Dave Airlie <airlied@redhat.com> | 2020-04-03 06:09:51 +1000 |
---|---|---|
committer | Dave Airlie <airlied@redhat.com> | 2020-04-03 09:07:49 +1000 |
commit | 0e7e6198af28c1573267aba1be33dd0b7fb35691 (patch) | |
tree | ec51021f5684005c7d6358b81ab7de690a3834e0 /drivers/gpu/drm/vmwgfx | |
parent | 59e7a8cc2dcf335116d500d684bfb34d1d97a6fe (diff) | |
parent | 9431042dbc8ce490d49c7f9d5142805b6249208b (diff) |
Merge branch 'ttm-transhuge' of git://people.freedesktop.org/~thomash/linux into drm-next
Huge page-table entries for TTM
In order to reduce CPU usage [1] and in theory TLB misses this patchset enables
huge- and giant page-table entries for TTM and TTM-enabled graphics drivers.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Hellstrom (VMware) <thomas_os@shipmail.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20200325073102.6129-1-thomas_os@shipmail.org
Diffstat (limited to 'drivers/gpu/drm/vmwgfx')
-rw-r--r-- | drivers/gpu/drm/vmwgfx/Makefile | 1 | ||||
-rw-r--r-- | drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 13 | ||||
-rw-r--r-- | drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 12 | ||||
-rw-r--r-- | drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c | 76 | ||||
-rw-r--r-- | drivers/gpu/drm/vmwgfx/vmwgfx_thp.c | 166 | ||||
-rw-r--r-- | drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c | 2 | ||||
-rw-r--r-- | drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c | 5 |
7 files changed, 271 insertions, 4 deletions
diff --git a/drivers/gpu/drm/vmwgfx/Makefile b/drivers/gpu/drm/vmwgfx/Makefile index 5c3515e8cce1..31f85f09f1fc 100644 --- a/drivers/gpu/drm/vmwgfx/Makefile +++ b/drivers/gpu/drm/vmwgfx/Makefile @@ -11,4 +11,5 @@ vmwgfx-y := vmwgfx_execbuf.o vmwgfx_gmr.o vmwgfx_kms.o vmwgfx_drv.o \ vmwgfx_validation.o vmwgfx_page_dirty.o vmwgfx_streamoutput.o \ ttm_object.o ttm_lock.o +vmwgfx-$(CONFIG_TRANSPARENT_HUGEPAGE) += vmwgfx_thp.o obj-$(CONFIG_DRM_VMWGFX) := vmwgfx.o diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 71e45b568511..c2247a893ed4 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -1247,6 +1247,18 @@ static void vmw_remove(struct pci_dev *pdev) pci_disable_device(pdev); } +static unsigned long +vmw_get_unmapped_area(struct file *file, unsigned long uaddr, + unsigned long len, unsigned long pgoff, + unsigned long flags) +{ + struct drm_file *file_priv = file->private_data; + struct vmw_private *dev_priv = vmw_priv(file_priv->minor->dev); + + return drm_get_unmapped_area(file, uaddr, len, pgoff, flags, + &dev_priv->vma_manager); +} + static int vmwgfx_pm_notifier(struct notifier_block *nb, unsigned long val, void *ptr) { @@ -1418,6 +1430,7 @@ static const struct file_operations vmwgfx_driver_fops = { .compat_ioctl = vmw_compat_ioctl, #endif .llseek = noop_llseek, + .get_unmapped_area = vmw_get_unmapped_area, }; static struct drm_driver driver = { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 5ddbcb9f6df4..8cdcd6e5f9e1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -1000,6 +1000,7 @@ extern int vmw_mmap(struct file *filp, struct vm_area_struct *vma); extern void vmw_validation_mem_init_ttm(struct vmw_private *dev_priv, size_t gran); + /** * TTM buffer object driver - vmwgfx_ttm_buffer.c */ @@ -1510,6 +1511,17 @@ void vmw_bo_dirty_unmap(struct vmw_buffer_object *vbo, pgoff_t start, pgoff_t end); vm_fault_t vmw_bo_vm_fault(struct vm_fault *vmf); vm_fault_t vmw_bo_vm_mkwrite(struct vm_fault *vmf); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +vm_fault_t vmw_bo_vm_huge_fault(struct vm_fault *vmf, + enum page_entry_size pe_size); +#endif + +/* Transparent hugepage support - vmwgfx_thp.c */ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +extern const struct ttm_mem_type_manager_func vmw_thp_func; +#else +#define vmw_thp_func ttm_bo_manager_func +#endif /** * VMW_DEBUG_KMS - Debug output for kernel mode-setting diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c index 60cfbfadd3f2..d4d66532f9c9 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c @@ -473,11 +473,11 @@ vm_fault_t vmw_bo_vm_fault(struct vm_fault *vmf) * a lot of unnecessary write faults. */ if (vbo->dirty && vbo->dirty->method == VMW_BO_DIRTY_MKWRITE) - prot = vma->vm_page_prot; + prot = vm_get_page_prot(vma->vm_flags & ~VM_SHARED); else prot = vm_get_page_prot(vma->vm_flags); - ret = ttm_bo_vm_fault_reserved(vmf, prot, num_prefault); + ret = ttm_bo_vm_fault_reserved(vmf, prot, num_prefault, 1); if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) return ret; @@ -486,3 +486,75 @@ out_unlock: return ret; } + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +vm_fault_t vmw_bo_vm_huge_fault(struct vm_fault *vmf, + enum page_entry_size pe_size) +{ + struct vm_area_struct *vma = vmf->vma; + struct ttm_buffer_object *bo = (struct ttm_buffer_object *) + vma->vm_private_data; + struct vmw_buffer_object *vbo = + container_of(bo, struct vmw_buffer_object, base); + pgprot_t prot; + vm_fault_t ret; + pgoff_t fault_page_size; + bool write = vmf->flags & FAULT_FLAG_WRITE; + bool is_cow_mapping = + (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; + + switch (pe_size) { + case PE_SIZE_PMD: + fault_page_size = HPAGE_PMD_SIZE >> PAGE_SHIFT; + break; +#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD + case PE_SIZE_PUD: + fault_page_size = HPAGE_PUD_SIZE >> PAGE_SHIFT; + break; +#endif + default: + WARN_ON_ONCE(1); + return VM_FAULT_FALLBACK; + } + + /* Always do write dirty-tracking and COW on PTE level. */ + if (write && (READ_ONCE(vbo->dirty) || is_cow_mapping)) + return VM_FAULT_FALLBACK; + + ret = ttm_bo_vm_reserve(bo, vmf); + if (ret) + return ret; + + if (vbo->dirty) { + pgoff_t allowed_prefault; + unsigned long page_offset; + + page_offset = vmf->pgoff - + drm_vma_node_start(&bo->base.vma_node); + if (page_offset >= bo->num_pages || + vmw_resources_clean(vbo, page_offset, + page_offset + PAGE_SIZE, + &allowed_prefault)) { + ret = VM_FAULT_SIGBUS; + goto out_unlock; + } + + /* + * Write protect, so we get a new fault on write, and can + * split. + */ + prot = vm_get_page_prot(vma->vm_flags & ~VM_SHARED); + } else { + prot = vm_get_page_prot(vma->vm_flags); + } + + ret = ttm_bo_vm_fault_reserved(vmf, prot, 1, fault_page_size); + if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) + return ret; + +out_unlock: + dma_resv_unlock(bo->base.resv); + + return ret; +} +#endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c new file mode 100644 index 000000000000..b7c816ba7166 --- /dev/null +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_thp.c @@ -0,0 +1,166 @@ +// SPDX-License-Identifier: GPL-2.0 OR MIT +/* + * Huge page-table-entry support for IO memory. + * + * Copyright (C) 2007-2019 Vmware, Inc. All rights reservedd. + */ +#include "vmwgfx_drv.h" +#include <drm/ttm/ttm_module.h> +#include <drm/ttm/ttm_bo_driver.h> +#include <drm/ttm/ttm_placement.h> + +/** + * struct vmw_thp_manager - Range manager implementing huge page alignment + * + * @mm: The underlying range manager. Protected by @lock. + * @lock: Manager lock. + */ +struct vmw_thp_manager { + struct drm_mm mm; + spinlock_t lock; +}; + +static int vmw_thp_insert_aligned(struct drm_mm *mm, struct drm_mm_node *node, + unsigned long align_pages, + const struct ttm_place *place, + struct ttm_mem_reg *mem, + unsigned long lpfn, + enum drm_mm_insert_mode mode) +{ + if (align_pages >= mem->page_alignment && + (!mem->page_alignment || align_pages % mem->page_alignment == 0)) { + return drm_mm_insert_node_in_range(mm, node, + mem->num_pages, + align_pages, 0, + place->fpfn, lpfn, mode); + } + + return -ENOSPC; +} + +static int vmw_thp_get_node(struct ttm_mem_type_manager *man, + struct ttm_buffer_object *bo, + const struct ttm_place *place, + struct ttm_mem_reg *mem) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + struct drm_mm *mm = &rman->mm; + struct drm_mm_node *node; + unsigned long align_pages; + unsigned long lpfn; + enum drm_mm_insert_mode mode = DRM_MM_INSERT_BEST; + int ret; + + node = kzalloc(sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + + lpfn = place->lpfn; + if (!lpfn) + lpfn = man->size; + + mode = DRM_MM_INSERT_BEST; + if (place->flags & TTM_PL_FLAG_TOPDOWN) + mode = DRM_MM_INSERT_HIGH; + + spin_lock(&rman->lock); + if (IS_ENABLED(CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD)) { + align_pages = (HPAGE_PUD_SIZE >> PAGE_SHIFT); + if (mem->num_pages >= align_pages) { + ret = vmw_thp_insert_aligned(mm, node, align_pages, + place, mem, lpfn, mode); + if (!ret) + goto found_unlock; + } + } + + align_pages = (HPAGE_PMD_SIZE >> PAGE_SHIFT); + if (mem->num_pages >= align_pages) { + ret = vmw_thp_insert_aligned(mm, node, align_pages, place, mem, + lpfn, mode); + if (!ret) + goto found_unlock; + } + + ret = drm_mm_insert_node_in_range(mm, node, mem->num_pages, + mem->page_alignment, 0, + place->fpfn, lpfn, mode); +found_unlock: + spin_unlock(&rman->lock); + + if (unlikely(ret)) { + kfree(node); + } else { + mem->mm_node = node; + mem->start = node->start; + } + + return 0; +} + + + +static void vmw_thp_put_node(struct ttm_mem_type_manager *man, + struct ttm_mem_reg *mem) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + + if (mem->mm_node) { + spin_lock(&rman->lock); + drm_mm_remove_node(mem->mm_node); + spin_unlock(&rman->lock); + + kfree(mem->mm_node); + mem->mm_node = NULL; + } +} + +static int vmw_thp_init(struct ttm_mem_type_manager *man, + unsigned long p_size) +{ + struct vmw_thp_manager *rman; + + rman = kzalloc(sizeof(*rman), GFP_KERNEL); + if (!rman) + return -ENOMEM; + + drm_mm_init(&rman->mm, 0, p_size); + spin_lock_init(&rman->lock); + man->priv = rman; + return 0; +} + +static int vmw_thp_takedown(struct ttm_mem_type_manager *man) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + struct drm_mm *mm = &rman->mm; + + spin_lock(&rman->lock); + if (drm_mm_clean(mm)) { + drm_mm_takedown(mm); + spin_unlock(&rman->lock); + kfree(rman); + man->priv = NULL; + return 0; + } + spin_unlock(&rman->lock); + return -EBUSY; +} + +static void vmw_thp_debug(struct ttm_mem_type_manager *man, + struct drm_printer *printer) +{ + struct vmw_thp_manager *rman = (struct vmw_thp_manager *) man->priv; + + spin_lock(&rman->lock); + drm_mm_print(&rman->mm, printer); + spin_unlock(&rman->lock); +} + +const struct ttm_mem_type_manager_func vmw_thp_func = { + .init = vmw_thp_init, + .takedown = vmw_thp_takedown, + .get_node = vmw_thp_get_node, + .put_node = vmw_thp_put_node, + .debug = vmw_thp_debug +}; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c index 3f3b2c7a208a..bf0bc4697959 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c @@ -749,7 +749,7 @@ static int vmw_init_mem_type(struct ttm_bo_device *bdev, uint32_t type, break; case TTM_PL_VRAM: /* "On-card" video ram */ - man->func = &ttm_bo_manager_func; + man->func = &vmw_thp_func; man->gpu_offset = 0; man->flags = TTM_MEMTYPE_FLAG_FIXED | TTM_MEMTYPE_FLAG_MAPPABLE; man->available_caching = TTM_PL_FLAG_CACHED; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c index aa7e50f63b94..3c03b1746661 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c @@ -34,7 +34,10 @@ int vmw_mmap(struct file *filp, struct vm_area_struct *vma) .page_mkwrite = vmw_bo_vm_mkwrite, .fault = vmw_bo_vm_fault, .open = ttm_bo_vm_open, - .close = ttm_bo_vm_close + .close = ttm_bo_vm_close, +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + .huge_fault = vmw_bo_vm_huge_fault, +#endif }; struct drm_file *file_priv = filp->private_data; struct vmw_private *dev_priv = vmw_priv(file_priv->minor->dev); |