January 2022 - Linux-stable-mirror

FAILED: patch "[PATCH] mm/hmm.c: allow VM_MIXEDMAP to work with hmm_range_fault" failed to apply to 4.19-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 87c01d57fa23de82fff593a7d070933d08755801 Mon Sep 17 00:00:00 2001 From: Alistair Popple <apopple(a)nvidia.com> Date: Fri, 14 Jan 2022 14:09:31 -0800 Subject: [PATCH] mm/hmm.c: allow VM_MIXEDMAP to work with hmm_range_fault hmm_range_fault() can be used instead of get_user_pages() for devices which allow faulting however unlike get_user_pages() it will return an error when used on a VM_MIXEDMAP range. To make hmm_range_fault() more closely match get_user_pages() remove this restriction. This requires dealing with the !ARCH_HAS_PTE_SPECIAL case in hmm_vma_handle_pte(). Rather than replicating the logic of vm_normal_page() call it directly and do a check for the zero pfn similar to what get_user_pages() currently does. Also add a test to hmm selftest to verify functionality. Link: https://lkml.kernel.org/r/20211104012001.2555676-1-apopple@nvidia.com Fixes: da4c3c735ea4 ("mm/hmm/mirror: helper to snapshot CPU page table") Signed-off-by: Alistair Popple <apopple(a)nvidia.com> Reviewed-by: Jason Gunthorpe <jgg(a)nvidia.com> Cc: Jerome Glisse <jglisse(a)redhat.com> Cc: John Hubbard <jhubbard(a)nvidia.com> Cc: Zi Yan <ziy(a)nvidia.com> Cc: Ralph Campbell <rcampbell(a)nvidia.com> Cc: Felix Kuehling <Felix.Kuehling(a)amd.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> diff --git a/lib/test_hmm.c b/lib/test_hmm.c index e2ce8f9b7605..767538089a62 100644 --- a/lib/test_hmm.c +++ b/lib/test_hmm.c @@ -1086,9 +1086,33 @@ static long dmirror_fops_unlocked_ioctl(struct file *filp, return 0; } +static int dmirror_fops_mmap(struct file *file, struct vm_area_struct *vma) +{ + unsigned long addr; + + for (addr = vma->vm_start; addr < vma->vm_end; addr += PAGE_SIZE) { + struct page *page; + int ret; + + page = alloc_page(GFP_KERNEL | __GFP_ZERO); + if (!page) + return -ENOMEM; + + ret = vm_insert_page(vma, addr, page); + if (ret) { + __free_page(page); + return ret; + } + put_page(page); + } + + return 0; +} + static const struct file_operations dmirror_fops = { .open = dmirror_fops_open, .release = dmirror_fops_release, + .mmap = dmirror_fops_mmap, .unlocked_ioctl = dmirror_fops_unlocked_ioctl, .llseek = default_llseek, .owner = THIS_MODULE, diff --git a/mm/hmm.c b/mm/hmm.c index 842e26599238..bd56641c79d4 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -300,7 +300,8 @@ static int hmm_vma_handle_pte(struct mm_walk *walk, unsigned long addr, * Since each architecture defines a struct page for the zero page, just * fall through and treat it like a normal page. */ - if (pte_special(pte) && !pte_devmap(pte) && + if (!vm_normal_page(walk->vma, addr, pte) && + !pte_devmap(pte) && !is_zero_pfn(pte_pfn(pte))) { if (hmm_pte_need_fault(hmm_vma_walk, pfn_req_flags, 0)) { pte_unmap(ptep); @@ -518,7 +519,7 @@ static int hmm_vma_walk_test(unsigned long start, unsigned long end, struct hmm_range *range = hmm_vma_walk->range; struct vm_area_struct *vma = walk->vma; - if (!(vma->vm_flags & (VM_IO | VM_PFNMAP | VM_MIXEDMAP)) && + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)) && vma->vm_flags & VM_READ) return 0; diff --git a/tools/testing/selftests/vm/hmm-tests.c b/tools/testing/selftests/vm/hmm-tests.c index 864f126ffd78..203323967b50 100644 --- a/tools/testing/selftests/vm/hmm-tests.c +++ b/tools/testing/selftests/vm/hmm-tests.c @@ -1248,6 +1248,48 @@ TEST_F(hmm, anon_teardown) } } +/* + * Test memory snapshot without faulting in pages accessed by the device. + */ +TEST_F(hmm, mixedmap) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned char *m; + int ret; + + npages = 1; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(npages); + ASSERT_NE(buffer->mirror, NULL); + + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE, + self->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Simulate a device snapshotting CPU pagetables. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device saw. */ + m = buffer->mirror; + ASSERT_EQ(m[0], HMM_DMIRROR_PROT_READ); + + hmm_buffer_free(buffer); +} + /* * Test memory snapshot without faulting in pages accessed by the device. */

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] mm/hmm.c: allow VM_MIXEDMAP to work with hmm_range_fault" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 87c01d57fa23de82fff593a7d070933d08755801 Mon Sep 17 00:00:00 2001 From: Alistair Popple <apopple(a)nvidia.com> Date: Fri, 14 Jan 2022 14:09:31 -0800 Subject: [PATCH] mm/hmm.c: allow VM_MIXEDMAP to work with hmm_range_fault hmm_range_fault() can be used instead of get_user_pages() for devices which allow faulting however unlike get_user_pages() it will return an error when used on a VM_MIXEDMAP range. To make hmm_range_fault() more closely match get_user_pages() remove this restriction. This requires dealing with the !ARCH_HAS_PTE_SPECIAL case in hmm_vma_handle_pte(). Rather than replicating the logic of vm_normal_page() call it directly and do a check for the zero pfn similar to what get_user_pages() currently does. Also add a test to hmm selftest to verify functionality. Link: https://lkml.kernel.org/r/20211104012001.2555676-1-apopple@nvidia.com Fixes: da4c3c735ea4 ("mm/hmm/mirror: helper to snapshot CPU page table") Signed-off-by: Alistair Popple <apopple(a)nvidia.com> Reviewed-by: Jason Gunthorpe <jgg(a)nvidia.com> Cc: Jerome Glisse <jglisse(a)redhat.com> Cc: John Hubbard <jhubbard(a)nvidia.com> Cc: Zi Yan <ziy(a)nvidia.com> Cc: Ralph Campbell <rcampbell(a)nvidia.com> Cc: Felix Kuehling <Felix.Kuehling(a)amd.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> diff --git a/lib/test_hmm.c b/lib/test_hmm.c index e2ce8f9b7605..767538089a62 100644 --- a/lib/test_hmm.c +++ b/lib/test_hmm.c @@ -1086,9 +1086,33 @@ static long dmirror_fops_unlocked_ioctl(struct file *filp, return 0; } +static int dmirror_fops_mmap(struct file *file, struct vm_area_struct *vma) +{ + unsigned long addr; + + for (addr = vma->vm_start; addr < vma->vm_end; addr += PAGE_SIZE) { + struct page *page; + int ret; + + page = alloc_page(GFP_KERNEL | __GFP_ZERO); + if (!page) + return -ENOMEM; + + ret = vm_insert_page(vma, addr, page); + if (ret) { + __free_page(page); + return ret; + } + put_page(page); + } + + return 0; +} + static const struct file_operations dmirror_fops = { .open = dmirror_fops_open, .release = dmirror_fops_release, + .mmap = dmirror_fops_mmap, .unlocked_ioctl = dmirror_fops_unlocked_ioctl, .llseek = default_llseek, .owner = THIS_MODULE, diff --git a/mm/hmm.c b/mm/hmm.c index 842e26599238..bd56641c79d4 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -300,7 +300,8 @@ static int hmm_vma_handle_pte(struct mm_walk *walk, unsigned long addr, * Since each architecture defines a struct page for the zero page, just * fall through and treat it like a normal page. */ - if (pte_special(pte) && !pte_devmap(pte) && + if (!vm_normal_page(walk->vma, addr, pte) && + !pte_devmap(pte) && !is_zero_pfn(pte_pfn(pte))) { if (hmm_pte_need_fault(hmm_vma_walk, pfn_req_flags, 0)) { pte_unmap(ptep); @@ -518,7 +519,7 @@ static int hmm_vma_walk_test(unsigned long start, unsigned long end, struct hmm_range *range = hmm_vma_walk->range; struct vm_area_struct *vma = walk->vma; - if (!(vma->vm_flags & (VM_IO | VM_PFNMAP | VM_MIXEDMAP)) && + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)) && vma->vm_flags & VM_READ) return 0; diff --git a/tools/testing/selftests/vm/hmm-tests.c b/tools/testing/selftests/vm/hmm-tests.c index 864f126ffd78..203323967b50 100644 --- a/tools/testing/selftests/vm/hmm-tests.c +++ b/tools/testing/selftests/vm/hmm-tests.c @@ -1248,6 +1248,48 @@ TEST_F(hmm, anon_teardown) } } +/* + * Test memory snapshot without faulting in pages accessed by the device. + */ +TEST_F(hmm, mixedmap) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned char *m; + int ret; + + npages = 1; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(npages); + ASSERT_NE(buffer->mirror, NULL); + + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE, + self->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Simulate a device snapshotting CPU pagetables. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device saw. */ + m = buffer->mirror; + ASSERT_EQ(m[0], HMM_DMIRROR_PROT_READ); + + hmm_buffer_free(buffer); +} + /* * Test memory snapshot without faulting in pages accessed by the device. */

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] blk-mq: fix tag_get wait task can't be awakened" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 Mon Sep 17 00:00:00 2001 From: Laibin Qiu <qiulaibin(a)huawei.com> Date: Thu, 13 Jan 2022 10:55:36 +0800 Subject: [PATCH] blk-mq: fix tag_get wait task can't be awakened In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U); tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx. Fix this by recalculating wake_batch when inc or dec active_queues. Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei <ming.lei(a)redhat.com> Suggested-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Laibin Qiu <qiulaibin(a)huawei.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index e55a6834c9a6..845f74e8dd7b 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -16,6 +16,21 @@ #include "blk-mq-sched.h" #include "blk-mq-tag.h" +/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -24,18 +39,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } } + users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; } @@ -56,6 +79,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users; if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -68,7 +92,9 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) return; } - atomic_dec(&tags->active_queues); + users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); blk_mq_tag_wakeup_all(tags, false); } diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index fc0357a6e19b..95df357ec009 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -415,6 +415,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); } +/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 2709ab825499..6220fa67fb7e 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -457,10 +457,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node); -static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i; if (sbq->wake_batch != wake_batch) { @@ -476,6 +475,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } } +static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] blk-mq: fix tag_get wait task can't be awakened" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 Mon Sep 17 00:00:00 2001 From: Laibin Qiu <qiulaibin(a)huawei.com> Date: Thu, 13 Jan 2022 10:55:36 +0800 Subject: [PATCH] blk-mq: fix tag_get wait task can't be awakened In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U); tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx. Fix this by recalculating wake_batch when inc or dec active_queues. Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei <ming.lei(a)redhat.com> Suggested-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Laibin Qiu <qiulaibin(a)huawei.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index e55a6834c9a6..845f74e8dd7b 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -16,6 +16,21 @@ #include "blk-mq-sched.h" #include "blk-mq-tag.h" +/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -24,18 +39,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } } + users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; } @@ -56,6 +79,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users; if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -68,7 +92,9 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) return; } - atomic_dec(&tags->active_queues); + users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); blk_mq_tag_wakeup_all(tags, false); } diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index fc0357a6e19b..95df357ec009 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -415,6 +415,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); } +/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 2709ab825499..6220fa67fb7e 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -457,10 +457,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node); -static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i; if (sbq->wake_batch != wake_batch) { @@ -476,6 +475,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } } +static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] blk-mq: fix tag_get wait task can't be awakened" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 Mon Sep 17 00:00:00 2001 From: Laibin Qiu <qiulaibin(a)huawei.com> Date: Thu, 13 Jan 2022 10:55:36 +0800 Subject: [PATCH] blk-mq: fix tag_get wait task can't be awakened In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U); tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx. Fix this by recalculating wake_batch when inc or dec active_queues. Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei <ming.lei(a)redhat.com> Suggested-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Laibin Qiu <qiulaibin(a)huawei.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index e55a6834c9a6..845f74e8dd7b 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -16,6 +16,21 @@ #include "blk-mq-sched.h" #include "blk-mq-tag.h" +/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -24,18 +39,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } } + users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; } @@ -56,6 +79,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users; if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -68,7 +92,9 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) return; } - atomic_dec(&tags->active_queues); + users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); blk_mq_tag_wakeup_all(tags, false); } diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index fc0357a6e19b..95df357ec009 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -415,6 +415,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); } +/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 2709ab825499..6220fa67fb7e 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -457,10 +457,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node); -static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i; if (sbq->wake_batch != wake_batch) { @@ -476,6 +475,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } } +static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] blk-mq: fix tag_get wait task can't be awakened" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 Mon Sep 17 00:00:00 2001 From: Laibin Qiu <qiulaibin(a)huawei.com> Date: Thu, 13 Jan 2022 10:55:36 +0800 Subject: [PATCH] blk-mq: fix tag_get wait task can't be awakened In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U); tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx. Fix this by recalculating wake_batch when inc or dec active_queues. Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei <ming.lei(a)redhat.com> Suggested-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Laibin Qiu <qiulaibin(a)huawei.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index e55a6834c9a6..845f74e8dd7b 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -16,6 +16,21 @@ #include "blk-mq-sched.h" #include "blk-mq-tag.h" +/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -24,18 +39,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } } + users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; } @@ -56,6 +79,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users; if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -68,7 +92,9 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) return; } - atomic_dec(&tags->active_queues); + users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); blk_mq_tag_wakeup_all(tags, false); } diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index fc0357a6e19b..95df357ec009 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -415,6 +415,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); } +/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 2709ab825499..6220fa67fb7e 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -457,10 +457,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node); -static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i; if (sbq->wake_batch != wake_batch) { @@ -476,6 +475,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } } +static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] blk-mq: fix tag_get wait task can't be awakened" failed to apply to 4.19-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 Mon Sep 17 00:00:00 2001 From: Laibin Qiu <qiulaibin(a)huawei.com> Date: Thu, 13 Jan 2022 10:55:36 +0800 Subject: [PATCH] blk-mq: fix tag_get wait task can't be awakened In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U); tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx. Fix this by recalculating wake_batch when inc or dec active_queues. Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei <ming.lei(a)redhat.com> Suggested-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Laibin Qiu <qiulaibin(a)huawei.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index e55a6834c9a6..845f74e8dd7b 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -16,6 +16,21 @@ #include "blk-mq-sched.h" #include "blk-mq-tag.h" +/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -24,18 +39,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } } + users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; } @@ -56,6 +79,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users; if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -68,7 +92,9 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) return; } - atomic_dec(&tags->active_queues); + users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); blk_mq_tag_wakeup_all(tags, false); } diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index fc0357a6e19b..95df357ec009 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -415,6 +415,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); } +/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 2709ab825499..6220fa67fb7e 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -457,10 +457,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node); -static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i; if (sbq->wake_batch != wake_batch) { @@ -476,6 +475,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } } +static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] blk-mq: fix tag_get wait task can't be awakened" failed to apply to 4.9-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.9-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 Mon Sep 17 00:00:00 2001 From: Laibin Qiu <qiulaibin(a)huawei.com> Date: Thu, 13 Jan 2022 10:55:36 +0800 Subject: [PATCH] blk-mq: fix tag_get wait task can't be awakened In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U); tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx. Fix this by recalculating wake_batch when inc or dec active_queues. Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei <ming.lei(a)redhat.com> Suggested-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Laibin Qiu <qiulaibin(a)huawei.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index e55a6834c9a6..845f74e8dd7b 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -16,6 +16,21 @@ #include "blk-mq-sched.h" #include "blk-mq-tag.h" +/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -24,18 +39,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } } + users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; } @@ -56,6 +79,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users; if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -68,7 +92,9 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) return; } - atomic_dec(&tags->active_queues); + users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); blk_mq_tag_wakeup_all(tags, false); } diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index fc0357a6e19b..95df357ec009 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -415,6 +415,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); } +/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 2709ab825499..6220fa67fb7e 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -457,10 +457,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node); -static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i; if (sbq->wake_batch != wake_batch) { @@ -476,6 +475,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } } +static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] blk-mq: fix tag_get wait task can't be awakened" failed to apply to 4.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 180dccb0dba4f5e84a4a70c1be1d34cbb6528b32 Mon Sep 17 00:00:00 2001 From: Laibin Qiu <qiulaibin(a)huawei.com> Date: Thu, 13 Jan 2022 10:55:36 +0800 Subject: [PATCH] blk-mq: fix tag_get wait task can't be awakened In case of shared tags, there might be more than one hctx which allocates from the same tags, and each hctx is limited to allocate at most: hctx_max_depth = max((bt->sb.depth + users - 1) / users, 4U); tag idle detection is lazy, and may be delayed for 30sec, so there could be just one real active hctx(queue) but all others are actually idle and still accounted as active because of the lazy idle detection. Then if wake_batch is > hctx_max_depth, driver tag allocation may wait forever on this real active hctx. Fix this by recalculating wake_batch when inc or dec active_queues. Fixes: 0d2602ca30e41 ("blk-mq: improve support for shared tags maps") Suggested-by: Ming Lei <ming.lei(a)redhat.com> Suggested-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Laibin Qiu <qiulaibin(a)huawei.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://lore.kernel.org/r/20220113025536.1479653-1-qiulaibin@huawei.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index e55a6834c9a6..845f74e8dd7b 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -16,6 +16,21 @@ #include "blk-mq-sched.h" #include "blk-mq-tag.h" +/* + * Recalculate wakeup batch when tag is shared by hctx. + */ +static void blk_mq_update_wake_batch(struct blk_mq_tags *tags, + unsigned int users) +{ + if (!users) + return; + + sbitmap_queue_recalculate_wake_batch(&tags->bitmap_tags, + users); + sbitmap_queue_recalculate_wake_batch(&tags->breserved_tags, + users); +} + /* * If a previously inactive queue goes active, bump the active user count. * We need to do this before try to allocate driver tag, then even if fail @@ -24,18 +39,26 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { + unsigned int users; + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && - !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) { + return true; + } } else { - if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && - !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) - atomic_inc(&hctx->tags->active_queues); + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) { + return true; + } } + users = atomic_inc_return(&hctx->tags->active_queues); + + blk_mq_update_wake_batch(hctx->tags, users); + return true; } @@ -56,6 +79,7 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; + unsigned int users; if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; @@ -68,7 +92,9 @@ void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) return; } - atomic_dec(&tags->active_queues); + users = atomic_dec_return(&tags->active_queues); + + blk_mq_update_wake_batch(tags, users); blk_mq_tag_wakeup_all(tags, false); } diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h index fc0357a6e19b..95df357ec009 100644 --- a/include/linux/sbitmap.h +++ b/include/linux/sbitmap.h @@ -415,6 +415,17 @@ static inline void sbitmap_queue_free(struct sbitmap_queue *sbq) sbitmap_free(&sbq->sb); } +/** + * sbitmap_queue_recalculate_wake_batch() - Recalculate wake batch + * @sbq: Bitmap queue to recalculate wake batch. + * @users: Number of shares. + * + * Like sbitmap_queue_update_wake_batch(), this will calculate wake batch + * by depth. This interface is for HCTX shared tags or queue shared tags. + */ +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users); + /** * sbitmap_queue_resize() - Resize a &struct sbitmap_queue. * @sbq: Bitmap queue to resize. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index 2709ab825499..6220fa67fb7e 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -457,10 +457,9 @@ int sbitmap_queue_init_node(struct sbitmap_queue *sbq, unsigned int depth, } EXPORT_SYMBOL_GPL(sbitmap_queue_init_node); -static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, - unsigned int depth) +static inline void __sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int wake_batch) { - unsigned int wake_batch = sbq_calc_wake_batch(sbq, depth); int i; if (sbq->wake_batch != wake_batch) { @@ -476,6 +475,26 @@ static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, } } +static void sbitmap_queue_update_wake_batch(struct sbitmap_queue *sbq, + unsigned int depth) +{ + unsigned int wake_batch; + + wake_batch = sbq_calc_wake_batch(sbq, depth); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} + +void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq, + unsigned int users) +{ + unsigned int wake_batch; + + wake_batch = clamp_val((sbq->sb.depth + users - 1) / + users, 4, SBQ_WAKE_BATCH); + __sbitmap_queue_update_wake_batch(sbq, wake_batch); +} +EXPORT_SYMBOL_GPL(sbitmap_queue_recalculate_wake_batch); + void sbitmap_queue_resize(struct sbitmap_queue *sbq, unsigned int depth) { sbitmap_queue_update_wake_batch(sbq, depth);

2 years, 11 months

1
0
0 0

FAILED: patch "[PATCH] ath11k: add support for WCN6855 hw2.1" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From d1147a316b53df9cb0152e415ec41dcb6ea62c1c Mon Sep 17 00:00:00 2001 From: Baochen Qiang <quic_bqiang(a)quicinc.com> Date: Mon, 29 Nov 2021 10:56:12 +0800 Subject: [PATCH] ath11k: add support for WCN6855 hw2.1 Ath11k fails to probe WCN6855 hw2.1 chip: [ 6.983821] ath11k_pci 0000:06:00.0: enabling device (0000 -> 0002) [ 6.983841] ath11k_pci 0000:06:00.0: Unsupported WCN6855 SOC hardware version: 18 17 This is caused by the wrong bit mask setting of hardware major version: for QCA6390/QCN6855, it should be BIT8-11, not BIT8-16, so change the definition to GENMASK(11, 8). Also, add a separate entry for WCN6855 hw2.1 in ath11k_hw_params. Please note that currently WCN6855 hw2.1 shares the same firmwares as hw2.0, so users of this chip need to create a symlink as below: ln -s hw2.0 hw2.1 Tested-on: WCN6855 hw2.1 PCI WLAN.HSP.1.1-01720.1-QCAHSPSWPL_V1_V2_SILICONZ_LITE-1 Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-01720.1-QCAHSPSWPL_V1_V2_SILICONZ_LITE-1 Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1 Fixes: 18ac1665e785 ("ath11k: pci: check TCSR_SOC_HW_VERSION") Signed-off-by: Baochen Qiang <quic_bqiang(a)quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo(a)quicinc.com> Link: https://lore.kernel.org/r/20211129025613.21594-1-quic_bqiang@quicinc.com diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c index dd1a1bb078c3..5f6bdee26f31 100644 --- a/drivers/net/wireless/ath/ath11k/core.c +++ b/drivers/net/wireless/ath/ath11k/core.c @@ -284,6 +284,59 @@ static const struct ath11k_hw_params ath11k_hw_params[] = { .max_fft_bins = 0, }, + .interface_modes = BIT(NL80211_IFTYPE_STATION) | + BIT(NL80211_IFTYPE_AP), + .supports_monitor = false, + .supports_shadow_regs = true, + .idle_ps = true, + .supports_sta_ps = true, + .cold_boot_calib = false, + .supports_suspend = true, + .hal_desc_sz = sizeof(struct hal_rx_desc_wcn6855), + .fix_l1ss = false, + .credit_flow = true, + .max_tx_ring = DP_TCL_NUM_RING_MAX_QCA6390, + .hal_params = &ath11k_hw_hal_params_qca6390, + .supports_dynamic_smps_6ghz = false, + .alloc_cacheable_memory = false, + .wakeup_mhi = true, + }, + { + .name = "wcn6855 hw2.1", + .hw_rev = ATH11K_HW_WCN6855_HW21, + .fw = { + .dir = "WCN6855/hw2.1", + .board_size = 256 * 1024, + .cal_offset = 128 * 1024, + }, + .max_radios = 3, + .bdf_addr = 0x4B0C0000, + .hw_ops = &wcn6855_ops, + .ring_mask = &ath11k_hw_ring_mask_qca6390, + .internal_sleep_clock = true, + .regs = &wcn6855_regs, + .qmi_service_ins_id = ATH11K_QMI_WLFW_SERVICE_INS_ID_V01_QCA6390, + .host_ce_config = ath11k_host_ce_config_qca6390, + .ce_count = 9, + .target_ce_config = ath11k_target_ce_config_wlan_qca6390, + .target_ce_count = 9, + .svc_to_ce_map = ath11k_target_service_to_ce_map_wlan_qca6390, + .svc_to_ce_map_len = 14, + .single_pdev_only = true, + .rxdma1_enable = false, + .num_rxmda_per_pdev = 2, + .rx_mac_buf_ring = true, + .vdev_start_delay = true, + .htt_peer_map_v2 = false, + + .spectral = { + .fft_sz = 0, + .fft_pad_sz = 0, + .summary_pad_sz = 0, + .fft_hdr_len = 0, + .max_fft_bins = 0, + }, + .interface_modes = BIT(NL80211_IFTYPE_STATION) | BIT(NL80211_IFTYPE_AP), .supports_monitor = false, diff --git a/drivers/net/wireless/ath/ath11k/core.h b/drivers/net/wireless/ath/ath11k/core.h index 0103cfd0508d..1d13e6124693 100644 --- a/drivers/net/wireless/ath/ath11k/core.h +++ b/drivers/net/wireless/ath/ath11k/core.h @@ -117,6 +117,7 @@ enum ath11k_hw_rev { ATH11K_HW_IPQ6018_HW10, ATH11K_HW_QCN9074_HW10, ATH11K_HW_WCN6855_HW20, + ATH11K_HW_WCN6855_HW21, }; enum ath11k_firmware_mode { diff --git a/drivers/net/wireless/ath/ath11k/mhi.c b/drivers/net/wireless/ath/ath11k/mhi.c index d0f94a785a59..9a8e61ca9964 100644 --- a/drivers/net/wireless/ath/ath11k/mhi.c +++ b/drivers/net/wireless/ath/ath11k/mhi.c @@ -366,6 +366,7 @@ int ath11k_mhi_register(struct ath11k_pci *ab_pci) break; case ATH11K_HW_QCA6390_HW20: case ATH11K_HW_WCN6855_HW20: + case ATH11K_HW_WCN6855_HW21: ath11k_mhi_config = &ath11k_mhi_config_qca6390; break; default: diff --git a/drivers/net/wireless/ath/ath11k/pci.c b/drivers/net/wireless/ath/ath11k/pci.c index 1898b1a33960..0a2b2ba4ae07 100644 --- a/drivers/net/wireless/ath/ath11k/pci.c +++ b/drivers/net/wireless/ath/ath11k/pci.c @@ -26,7 +26,7 @@ #define WINDOW_RANGE_MASK GENMASK(18, 0) #define TCSR_SOC_HW_VERSION 0x0224 -#define TCSR_SOC_HW_VERSION_MAJOR_MASK GENMASK(16, 8) +#define TCSR_SOC_HW_VERSION_MAJOR_MASK GENMASK(11, 8) #define TCSR_SOC_HW_VERSION_MINOR_MASK GENMASK(7, 0) /* BAR0 + 4k is always accessible, and no @@ -1409,9 +1409,21 @@ static int ath11k_pci_probe(struct pci_dev *pdev, &soc_hw_version_minor); switch (soc_hw_version_major) { case 2: - ab->hw_rev = ATH11K_HW_WCN6855_HW20; + switch (soc_hw_version_minor) { + case 0x00: + case 0x01: + ab->hw_rev = ATH11K_HW_WCN6855_HW20; + break; + case 0x10: + case 0x11: + ab->hw_rev = ATH11K_HW_WCN6855_HW21; + break; + default: + goto unsupported_wcn6855_soc; + } break; default: +unsupported_wcn6855_soc: dev_err(&pdev->dev, "Unsupported WCN6855 SOC hardware version: %d %d\n", soc_hw_version_major, soc_hw_version_minor); ret = -EOPNOTSUPP;

2 years, 11 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror January 2022