From: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> To: intel-xe@lists.freedesktop.org Cc: "Matthew Brost" <matthew.brost@intel.com>, "Thomas Hellström" <thomas.hellstrom@linux.intel.com>, dri-devel@lists.freedesktop.org Subject: [PATCH 3/4] drm/xe/vm: Perform accounting of userptr pinned pages Date: Fri, 18 Aug 2023 17:08:44 +0200 [thread overview] Message-ID: <20230818150845.96679-4-thomas.hellstrom@linux.intel.com> (raw) In-Reply-To: <20230818150845.96679-1-thomas.hellstrom@linux.intel.com> Account these pages against RLIMIT_MEMLOCK following how RDMA does this with CAP_IPC_LOCK bypassing the limit. Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> --- drivers/gpu/drm/xe/xe_vm.c | 43 ++++++++++++++++++++++++++++++++++++-- 1 file changed, 41 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index ecbcad696b60..d9c000689002 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -34,6 +34,33 @@ #define TEST_VM_ASYNC_OPS_ERROR +/* + * Perform userptr PIN accounting against RLIMIT_MEMLOCK for now, similarly + * to how RDMA does this. + */ +static int xe_vma_mlock_alloc(struct xe_vma *vma, unsigned long num_pages) +{ + unsigned long lock_limit, new_pinned; + struct mm_struct *mm = vma->userptr.notifier.mm; + + if (!can_do_mlock()) + return -EPERM; + + lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; + new_pinned = atomic64_add_return(num_pages, &mm->pinned_vm); + if (new_pinned > lock_limit && !capable(CAP_IPC_LOCK)) { + atomic64_sub(num_pages, &mm->pinned_vm); + return -ENOMEM; + } + + return 0; +} + +static void xe_vma_mlock_free(struct xe_vma *vma, unsigned long num_pages) +{ + atomic64_sub(num_pages, &vma->userptr.notifier.mm->pinned_vm); +} + /** * xe_vma_userptr_check_repin() - Advisory check for repin needed * @vma: The userptr vma @@ -89,9 +116,17 @@ int xe_vma_userptr_pin_pages(struct xe_vma *vma) !read_only); pages = vma->userptr.pinned_pages; } else { + if (xe_vma_is_pinned(vma)) { + ret = xe_vma_mlock_alloc(vma, num_pages); + if (ret) + return ret; + } + pages = kvmalloc_array(num_pages, sizeof(*pages), GFP_KERNEL); - if (!pages) - return -ENOMEM; + if (!pages) { + ret = -ENOMEM; + goto out_account; + } } pinned = ret = 0; @@ -187,6 +222,9 @@ int xe_vma_userptr_pin_pages(struct xe_vma *vma) mm_closed: kvfree(pages); vma->userptr.pinned_pages = NULL; +out_account: + if (xe_vma_is_pinned(vma)) + xe_vma_mlock_free(vma, num_pages); return ret; } @@ -1004,6 +1042,7 @@ static void xe_vma_destroy_late(struct xe_vma *vma) unpin_user_pages_dirty_lock(vma->userptr.pinned_pages, vma->userptr.num_pinned, !read_only); + xe_vma_mlock_free(vma, xe_vma_size(vma) >> PAGE_SHIFT); kvfree(vma->userptr.pinned_pages); } -- 2.41.0
WARNING: multiple messages have this Message-ID (diff)
From: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> To: intel-xe@lists.freedesktop.org Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>, dri-devel@lists.freedesktop.org, Daniel Vetter <daniel@ffwll.ch> Subject: [Intel-xe] [PATCH 3/4] drm/xe/vm: Perform accounting of userptr pinned pages Date: Fri, 18 Aug 2023 17:08:44 +0200 [thread overview] Message-ID: <20230818150845.96679-4-thomas.hellstrom@linux.intel.com> (raw) In-Reply-To: <20230818150845.96679-1-thomas.hellstrom@linux.intel.com> Account these pages against RLIMIT_MEMLOCK following how RDMA does this with CAP_IPC_LOCK bypassing the limit. Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> --- drivers/gpu/drm/xe/xe_vm.c | 43 ++++++++++++++++++++++++++++++++++++-- 1 file changed, 41 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index ecbcad696b60..d9c000689002 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -34,6 +34,33 @@ #define TEST_VM_ASYNC_OPS_ERROR +/* + * Perform userptr PIN accounting against RLIMIT_MEMLOCK for now, similarly + * to how RDMA does this. + */ +static int xe_vma_mlock_alloc(struct xe_vma *vma, unsigned long num_pages) +{ + unsigned long lock_limit, new_pinned; + struct mm_struct *mm = vma->userptr.notifier.mm; + + if (!can_do_mlock()) + return -EPERM; + + lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; + new_pinned = atomic64_add_return(num_pages, &mm->pinned_vm); + if (new_pinned > lock_limit && !capable(CAP_IPC_LOCK)) { + atomic64_sub(num_pages, &mm->pinned_vm); + return -ENOMEM; + } + + return 0; +} + +static void xe_vma_mlock_free(struct xe_vma *vma, unsigned long num_pages) +{ + atomic64_sub(num_pages, &vma->userptr.notifier.mm->pinned_vm); +} + /** * xe_vma_userptr_check_repin() - Advisory check for repin needed * @vma: The userptr vma @@ -89,9 +116,17 @@ int xe_vma_userptr_pin_pages(struct xe_vma *vma) !read_only); pages = vma->userptr.pinned_pages; } else { + if (xe_vma_is_pinned(vma)) { + ret = xe_vma_mlock_alloc(vma, num_pages); + if (ret) + return ret; + } + pages = kvmalloc_array(num_pages, sizeof(*pages), GFP_KERNEL); - if (!pages) - return -ENOMEM; + if (!pages) { + ret = -ENOMEM; + goto out_account; + } } pinned = ret = 0; @@ -187,6 +222,9 @@ int xe_vma_userptr_pin_pages(struct xe_vma *vma) mm_closed: kvfree(pages); vma->userptr.pinned_pages = NULL; +out_account: + if (xe_vma_is_pinned(vma)) + xe_vma_mlock_free(vma, num_pages); return ret; } @@ -1004,6 +1042,7 @@ static void xe_vma_destroy_late(struct xe_vma *vma) unpin_user_pages_dirty_lock(vma->userptr.pinned_pages, vma->userptr.num_pinned, !read_only); + xe_vma_mlock_free(vma, xe_vma_size(vma) >> PAGE_SHIFT); kvfree(vma->userptr.pinned_pages); } -- 2.41.0
next prev parent reply other threads:[~2023-08-18 15:10 UTC|newest] Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top 2023-08-18 15:08 [PATCH 0/4] drm/xe: Support optional pinning of userptr pages Thomas Hellström 2023-08-18 15:08 ` [Intel-xe] " Thomas Hellström 2023-08-18 15:08 ` [PATCH 1/4] drm/xe/vm: Use onion unwind for xe_vma_userptr_pin_pages() Thomas Hellström 2023-08-18 15:08 ` [Intel-xe] " Thomas Hellström 2023-08-18 18:15 ` Matthew Brost 2023-08-18 18:15 ` [Intel-xe] " Matthew Brost 2023-08-18 15:08 ` [PATCH 2/4] drm/xe/vm: Implement userptr page pinning Thomas Hellström 2023-08-18 15:08 ` [Intel-xe] " Thomas Hellström 2023-08-20 4:06 ` Matthew Brost 2023-08-20 4:06 ` [Intel-xe] " Matthew Brost 2023-08-22 8:23 ` Thomas Hellström 2023-08-22 8:23 ` [Intel-xe] " Thomas Hellström 2023-08-18 15:08 ` Thomas Hellström [this message] 2023-08-18 15:08 ` [Intel-xe] [PATCH 3/4] drm/xe/vm: Perform accounting of userptr pinned pages Thomas Hellström 2023-08-20 3:43 ` Matthew Brost 2023-08-20 3:43 ` [Intel-xe] " Matthew Brost 2023-08-22 8:10 ` Thomas Hellström 2023-08-22 8:10 ` [Intel-xe] " Thomas Hellström 2023-08-18 15:08 ` [PATCH 4/4] drm/xe/uapi: Support pinning of userptr vmas Thomas Hellström 2023-08-18 15:08 ` [Intel-xe] " Thomas Hellström 2023-08-20 3:54 ` Matthew Brost 2023-08-20 3:54 ` [Intel-xe] " Matthew Brost 2023-08-22 8:25 ` Thomas Hellström 2023-08-22 8:25 ` [Intel-xe] " Thomas Hellström 2023-08-18 15:12 ` [Intel-xe] ✓ CI.Patch_applied: success for drm/xe: Support optional pinning of userptr pages Patchwork 2023-08-18 15:12 ` [Intel-xe] ✗ CI.checkpatch: warning " Patchwork 2023-08-18 15:14 ` [Intel-xe] ✓ CI.KUnit: success " Patchwork 2023-08-18 15:17 ` [Intel-xe] ✓ CI.Build: " Patchwork 2023-08-18 15:18 ` [Intel-xe] ✗ CI.Hooks: failure " Patchwork
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20230818150845.96679-4-thomas.hellstrom@linux.intel.com \ --to=thomas.hellstrom@linux.intel.com \ --cc=dri-devel@lists.freedesktop.org \ --cc=intel-xe@lists.freedesktop.org \ --cc=matthew.brost@intel.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.