From: Jason Gunthorpe <jgg@ziepe.ca> To: linux-mm@kvack.org, Jerome Glisse <jglisse@redhat.com>, Ralph Campbell <rcampbell@nvidia.com>, John Hubbard <jhubbard@nvidia.com>, Felix.Kuehling@amd.com Cc: "Juergen Gross" <jgross@suse.com>, "David Zhou" <David1.Zhou@amd.com>, "Mike Marciniszyn" <mike.marciniszyn@intel.com>, "Stefano Stabellini" <sstabellini@kernel.org>, "Philip Yang" <Philip.Yang@amd.com>, "Oleksandr Andrushchenko" <oleksandr_andrushchenko@epam.com>, linux-rdma@vger.kernel.org, nouveau@lists.freedesktop.org, "Dennis Dalessandro" <dennis.dalessandro@intel.com>, amd-gfx@lists.freedesktop.org, "Christoph Hellwig" <hch@infradead.org>, "Jason Gunthorpe" <jgg@mellanox.com>, dri-devel@lists.freedesktop.org, "Alex Deucher" <alexander.deucher@amd.com>, xen-devel@lists.xenproject.org, "Boris Ostrovsky" <boris.ostrovsky@oracle.com>, "Petr Cvek" <petrcvekcz@gmail.com>, "Christian König" <christian.koenig@amd.com>, "Ben Skeggs" <bskeggs@redhat.com> Subject: [PATCH v3 03/14] mm/hmm: allow hmm_range to be used with a mmu_interval_notifier or hmm_mirror Date: Tue, 12 Nov 2019 16:22:20 -0400 [thread overview] Message-ID: <20191112202231.3856-4-jgg@ziepe.ca> (raw) In-Reply-To: <20191112202231.3856-1-jgg@ziepe.ca> From: Jason Gunthorpe <jgg@mellanox.com> hmm_mirror's handling of ranges does not use a sequence count which results in this bug: CPU0 CPU1 hmm_range_wait_until_valid(range) valid == true hmm_range_fault(range) hmm_invalidate_range_start() range->valid = false hmm_invalidate_range_end() range->valid = true hmm_range_valid(range) valid == true Where the hmm_range_valid() should not have succeeded. Adding the required sequence count would make it nearly identical to the new mmu_interval_notifier. Instead replace the hmm_mirror stuff with mmu_interval_notifier. Co-existence of the two APIs is the first step. Reviewed-by: Jérôme Glisse <jglisse@redhat.com> Tested-by: Philip Yang <Philip.Yang@amd.com> Tested-by: Ralph Campbell <rcampbell@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- include/linux/hmm.h | 5 +++++ mm/hmm.c | 25 +++++++++++++++++++------ 2 files changed, 24 insertions(+), 6 deletions(-) diff --git a/include/linux/hmm.h b/include/linux/hmm.h index 3fec513b9c00f1..fbb35c78637e57 100644 --- a/include/linux/hmm.h +++ b/include/linux/hmm.h @@ -145,6 +145,9 @@ enum hmm_pfn_value_e { /* * struct hmm_range - track invalidation lock on virtual address range * + * @notifier: an optional mmu_interval_notifier + * @notifier_seq: when notifier is used this is the result of + * mmu_interval_read_begin() * @hmm: the core HMM structure this range is active against * @vma: the vm area struct for the range * @list: all range lock are on a list @@ -159,6 +162,8 @@ enum hmm_pfn_value_e { * @valid: pfns array did not change since it has been fill by an HMM function */ struct hmm_range { + struct mmu_interval_notifier *notifier; + unsigned long notifier_seq; struct hmm *hmm; struct list_head list; unsigned long start; diff --git a/mm/hmm.c b/mm/hmm.c index 6b0136665407a3..8d060c5dabe37b 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -858,6 +858,14 @@ void hmm_range_unregister(struct hmm_range *range) } EXPORT_SYMBOL(hmm_range_unregister); +static bool needs_retry(struct hmm_range *range) +{ + if (range->notifier) + return mmu_interval_check_retry(range->notifier, + range->notifier_seq); + return !range->valid; +} + static const struct mm_walk_ops hmm_walk_ops = { .pud_entry = hmm_vma_walk_pud, .pmd_entry = hmm_vma_walk_pmd, @@ -898,18 +906,23 @@ long hmm_range_fault(struct hmm_range *range, unsigned int flags) const unsigned long device_vma = VM_IO | VM_PFNMAP | VM_MIXEDMAP; unsigned long start = range->start, end; struct hmm_vma_walk hmm_vma_walk; - struct hmm *hmm = range->hmm; + struct mm_struct *mm; struct vm_area_struct *vma; int ret; - lockdep_assert_held(&hmm->mmu_notifier.mm->mmap_sem); + if (range->notifier) + mm = range->notifier->mm; + else + mm = range->hmm->mmu_notifier.mm; + + lockdep_assert_held(&mm->mmap_sem); do { /* If range is no longer valid force retry. */ - if (!range->valid) + if (needs_retry(range)) return -EBUSY; - vma = find_vma(hmm->mmu_notifier.mm, start); + vma = find_vma(mm, start); if (vma == NULL || (vma->vm_flags & device_vma)) return -EFAULT; @@ -939,7 +952,7 @@ long hmm_range_fault(struct hmm_range *range, unsigned int flags) start = hmm_vma_walk.last; /* Keep trying while the range is valid. */ - } while (ret == -EBUSY && range->valid); + } while (ret == -EBUSY && !needs_retry(range)); if (ret) { unsigned long i; @@ -997,7 +1010,7 @@ long hmm_range_dma_map(struct hmm_range *range, struct device *device, continue; /* Check if range is being invalidated */ - if (!range->valid) { + if (needs_retry(range)) { ret = -EBUSY; goto unmap; } -- 2.24.0 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel
WARNING: multiple messages have this Message-ID (diff)
From: Jason Gunthorpe <jgg@ziepe.ca> To: linux-mm@kvack.org, Jerome Glisse <jglisse@redhat.com>, Ralph Campbell <rcampbell@nvidia.com>, John Hubbard <jhubbard@nvidia.com>, Felix.Kuehling@amd.com Cc: "Juergen Gross" <jgross@suse.com>, "David Zhou" <David1.Zhou@amd.com>, "Mike Marciniszyn" <mike.marciniszyn@intel.com>, "Stefano Stabellini" <sstabellini@kernel.org>, "Philip Yang" <Philip.Yang@amd.com>, "Oleksandr Andrushchenko" <oleksandr_andrushchenko@epam.com>, linux-rdma@vger.kernel.org, nouveau@lists.freedesktop.org, "Dennis Dalessandro" <dennis.dalessandro@intel.com>, amd-gfx@lists.freedesktop.org, "Christoph Hellwig" <hch@infradead.org>, "Jason Gunthorpe" <jgg@mellanox.com>, dri-devel@lists.freedesktop.org, "Alex Deucher" <alexander.deucher@amd.com>, xen-devel@lists.xenproject.org, "Boris Ostrovsky" <boris.ostrovsky@oracle.com>, "Petr Cvek" <petrcvekcz@gmail.com>, "Christian König" <christian.koenig@amd.com>, "Ben Skeggs" <bskeggs@redhat.com> Subject: [PATCH v3 03/14] mm/hmm: allow hmm_range to be used with a mmu_interval_notifier or hmm_mirror Date: Tue, 12 Nov 2019 16:22:20 -0400 [thread overview] Message-ID: <20191112202231.3856-4-jgg@ziepe.ca> (raw) Message-ID: <20191112202220.JhkSQHdAulome_lQRc8h5H8AklxtmneX8ZLJmPqlrXQ@z> (raw) In-Reply-To: <20191112202231.3856-1-jgg@ziepe.ca> From: Jason Gunthorpe <jgg@mellanox.com> hmm_mirror's handling of ranges does not use a sequence count which results in this bug: CPU0 CPU1 hmm_range_wait_until_valid(range) valid == true hmm_range_fault(range) hmm_invalidate_range_start() range->valid = false hmm_invalidate_range_end() range->valid = true hmm_range_valid(range) valid == true Where the hmm_range_valid() should not have succeeded. Adding the required sequence count would make it nearly identical to the new mmu_interval_notifier. Instead replace the hmm_mirror stuff with mmu_interval_notifier. Co-existence of the two APIs is the first step. Reviewed-by: Jérôme Glisse <jglisse@redhat.com> Tested-by: Philip Yang <Philip.Yang@amd.com> Tested-by: Ralph Campbell <rcampbell@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> --- include/linux/hmm.h | 5 +++++ mm/hmm.c | 25 +++++++++++++++++++------ 2 files changed, 24 insertions(+), 6 deletions(-) diff --git a/include/linux/hmm.h b/include/linux/hmm.h index 3fec513b9c00f1..fbb35c78637e57 100644 --- a/include/linux/hmm.h +++ b/include/linux/hmm.h @@ -145,6 +145,9 @@ enum hmm_pfn_value_e { /* * struct hmm_range - track invalidation lock on virtual address range * + * @notifier: an optional mmu_interval_notifier + * @notifier_seq: when notifier is used this is the result of + * mmu_interval_read_begin() * @hmm: the core HMM structure this range is active against * @vma: the vm area struct for the range * @list: all range lock are on a list @@ -159,6 +162,8 @@ enum hmm_pfn_value_e { * @valid: pfns array did not change since it has been fill by an HMM function */ struct hmm_range { + struct mmu_interval_notifier *notifier; + unsigned long notifier_seq; struct hmm *hmm; struct list_head list; unsigned long start; diff --git a/mm/hmm.c b/mm/hmm.c index 6b0136665407a3..8d060c5dabe37b 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -858,6 +858,14 @@ void hmm_range_unregister(struct hmm_range *range) } EXPORT_SYMBOL(hmm_range_unregister); +static bool needs_retry(struct hmm_range *range) +{ + if (range->notifier) + return mmu_interval_check_retry(range->notifier, + range->notifier_seq); + return !range->valid; +} + static const struct mm_walk_ops hmm_walk_ops = { .pud_entry = hmm_vma_walk_pud, .pmd_entry = hmm_vma_walk_pmd, @@ -898,18 +906,23 @@ long hmm_range_fault(struct hmm_range *range, unsigned int flags) const unsigned long device_vma = VM_IO | VM_PFNMAP | VM_MIXEDMAP; unsigned long start = range->start, end; struct hmm_vma_walk hmm_vma_walk; - struct hmm *hmm = range->hmm; + struct mm_struct *mm; struct vm_area_struct *vma; int ret; - lockdep_assert_held(&hmm->mmu_notifier.mm->mmap_sem); + if (range->notifier) + mm = range->notifier->mm; + else + mm = range->hmm->mmu_notifier.mm; + + lockdep_assert_held(&mm->mmap_sem); do { /* If range is no longer valid force retry. */ - if (!range->valid) + if (needs_retry(range)) return -EBUSY; - vma = find_vma(hmm->mmu_notifier.mm, start); + vma = find_vma(mm, start); if (vma == NULL || (vma->vm_flags & device_vma)) return -EFAULT; @@ -939,7 +952,7 @@ long hmm_range_fault(struct hmm_range *range, unsigned int flags) start = hmm_vma_walk.last; /* Keep trying while the range is valid. */ - } while (ret == -EBUSY && range->valid); + } while (ret == -EBUSY && !needs_retry(range)); if (ret) { unsigned long i; @@ -997,7 +1010,7 @@ long hmm_range_dma_map(struct hmm_range *range, struct device *device, continue; /* Check if range is being invalidated */ - if (!range->valid) { + if (needs_retry(range)) { ret = -EBUSY; goto unmap; } -- 2.24.0 _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx
next prev parent reply other threads:[~2019-11-12 20:22 UTC|newest] Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-11-12 20:22 [PATCH hmm v3 00/14] Consolidate the mmu notifier interval_tree and locking Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 01/14] mm/mmu_notifier: define the header pre-processor parts even if disabled Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-13 13:52 ` Christoph Hellwig 2019-11-13 13:52 ` Christoph Hellwig 2019-11-12 20:22 ` [PATCH v3 02/14] mm/mmu_notifier: add an interval tree notifier Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-13 13:59 ` Christoph Hellwig 2019-11-13 13:59 ` Christoph Hellwig 2019-11-13 16:46 ` Jason Gunthorpe 2019-11-13 16:46 ` Jason Gunthorpe 2019-11-23 0:54 ` Ralph Campbell 2019-11-23 0:54 ` Ralph Campbell 2019-11-23 23:59 ` Jason Gunthorpe 2019-11-23 23:59 ` Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe [this message] 2019-11-12 20:22 ` [PATCH v3 03/14] mm/hmm: allow hmm_range to be used with a mmu_interval_notifier or hmm_mirror Jason Gunthorpe 2019-11-13 14:00 ` Christoph Hellwig 2019-11-13 14:00 ` Christoph Hellwig 2019-11-12 20:22 ` [PATCH v3 04/14] mm/hmm: define the pre-processor related parts of hmm.h even if disabled Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-13 14:01 ` Christoph Hellwig 2019-11-13 14:01 ` Christoph Hellwig 2019-11-12 20:22 ` [PATCH v3 05/14] RDMA/odp: Use mmu_interval_notifier_insert() Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 06/14] RDMA/hfi1: Use mmu_interval_notifier_insert for user_exp_rcv Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 07/14] drm/radeon: use mmu_interval_notifier_insert Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 08/14] nouveau: use mmu_notifier directly for invalidate_range_start Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 09/14] nouveau: use mmu_interval_notifier instead of hmm_mirror Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 10/14] drm/amdgpu: Call find_vma under mmap_sem Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 11/14] drm/amdgpu: Use mmu_interval_insert instead of hmm_mirror Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-12 20:22 ` [PATCH v3 12/14] drm/amdgpu: Use mmu_interval_notifier " Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-19 19:59 ` Philip Yang 2019-11-19 19:59 ` Philip Yang 2019-11-12 20:22 ` [PATCH v3 13/14] mm/hmm: remove hmm_mirror and related Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe 2019-11-13 14:02 ` Christoph Hellwig 2019-11-13 14:02 ` Christoph Hellwig 2019-11-12 20:22 ` [PATCH v3 14/14] xen/gntdev: use mmu_interval_notifier_insert Jason Gunthorpe 2019-11-12 20:22 ` Jason Gunthorpe
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=20191112202231.3856-4-jgg@ziepe.ca \ --to=jgg@ziepe.ca \ --cc=David1.Zhou@amd.com \ --cc=Felix.Kuehling@amd.com \ --cc=Philip.Yang@amd.com \ --cc=alexander.deucher@amd.com \ --cc=amd-gfx@lists.freedesktop.org \ --cc=boris.ostrovsky@oracle.com \ --cc=bskeggs@redhat.com \ --cc=christian.koenig@amd.com \ --cc=dennis.dalessandro@intel.com \ --cc=dri-devel@lists.freedesktop.org \ --cc=hch@infradead.org \ --cc=jgg@mellanox.com \ --cc=jglisse@redhat.com \ --cc=jgross@suse.com \ --cc=jhubbard@nvidia.com \ --cc=linux-mm@kvack.org \ --cc=linux-rdma@vger.kernel.org \ --cc=mike.marciniszyn@intel.com \ --cc=nouveau@lists.freedesktop.org \ --cc=oleksandr_andrushchenko@epam.com \ --cc=petrcvekcz@gmail.com \ --cc=rcampbell@nvidia.com \ --cc=sstabellini@kernel.org \ --cc=xen-devel@lists.xenproject.org \ /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 a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).