whiterose

linux unikernel
Log | Files | Refs | README | LICENSE | git clone https://git.ne02ptzero.me/git/whiterose

commit 38a2ca2cac28cd202a364e30f9fd252ac66619be
parent 34396bdfd2b7b41b9dcd2126eb1a57e43fe13368
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date:   Sat, 20 Apr 2019 12:20:58 -0700

Merge tag 'for-linus-20190420' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "A set of small fixes that should go into this series. This contains:

   - Removal of unused queue member (Hou)

   - Overflow bvec fix (Ming)

   - Various little io_uring tweaks (me)
       - kthread parking
       - Only call cpu_possible() for verified CPU
       - Drop unused 'file' argument to io_file_put()
       - io_uring_enter vs io_uring_register deadlock fix
       - CQ overflow fix

   - BFQ internal depth update fix (me)"

* tag 'for-linus-20190420' of git://git.kernel.dk/linux-block:
  block: make sure that bvec length can't be overflow
  block: kill all_q_node in request_queue
  io_uring: fix CQ overflow condition
  io_uring: fix possible deadlock between io_uring_{enter,register}
  io_uring: drop io_file_put() 'file' argument
  bfq: update internal depth state when queue depth changes
  io_uring: only test SQPOLL cpu after we've verified it
  io_uring: park SQPOLL thread if it's percpu

Diffstat:
Mblock/bfq-iosched.c | 8+++++++-
Mblock/blk-mq.c | 2++
Mfs/io_uring.c | 41++++++++++++++++++++++++++++++-----------
Minclude/linux/blkdev.h | 1-
Minclude/linux/bvec.h | 5+++--
Minclude/linux/elevator.h | 1+
6 files changed, 43 insertions(+), 15 deletions(-)

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c @@ -5396,7 +5396,7 @@ static unsigned int bfq_update_depths(struct bfq_data *bfqd, return min_shallow; } -static int bfq_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int index) +static void bfq_depth_updated(struct blk_mq_hw_ctx *hctx) { struct bfq_data *bfqd = hctx->queue->elevator->elevator_data; struct blk_mq_tags *tags = hctx->sched_tags; @@ -5404,6 +5404,11 @@ static int bfq_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int index) min_shallow = bfq_update_depths(bfqd, &tags->bitmap_tags); sbitmap_queue_min_shallow_depth(&tags->bitmap_tags, min_shallow); +} + +static int bfq_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int index) +{ + bfq_depth_updated(hctx); return 0; } @@ -5826,6 +5831,7 @@ static struct elevator_type iosched_bfq_mq = { .requests_merged = bfq_requests_merged, .request_merged = bfq_request_merged, .has_work = bfq_has_work, + .depth_updated = bfq_depth_updated, .init_hctx = bfq_init_hctx, .init_sched = bfq_init_queue, .exit_sched = bfq_exit_queue, diff --git a/block/blk-mq.c b/block/blk-mq.c @@ -3135,6 +3135,8 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr) } if (ret) break; + if (q->elevator && q->elevator->type->ops.depth_updated) + q->elevator->type->ops.depth_updated(hctx); } if (!ret) diff --git a/fs/io_uring.c b/fs/io_uring.c @@ -338,7 +338,7 @@ static struct io_uring_cqe *io_get_cqring(struct io_ring_ctx *ctx) tail = ctx->cached_cq_tail; /* See comment at the top of the file */ smp_rmb(); - if (tail + 1 == READ_ONCE(ring->r.head)) + if (tail - READ_ONCE(ring->r.head) == ring->ring_entries) return NULL; ctx->cached_cq_tail++; @@ -682,11 +682,9 @@ static void io_iopoll_req_issued(struct io_kiocb *req) list_add_tail(&req->list, &ctx->poll_list); } -static void io_file_put(struct io_submit_state *state, struct file *file) +static void io_file_put(struct io_submit_state *state) { - if (!state) { - fput(file); - } else if (state->file) { + if (state->file) { int diff = state->has_refs - state->used_refs; if (diff) @@ -711,7 +709,7 @@ static struct file *io_file_get(struct io_submit_state *state, int fd) state->ios_left--; return state->file; } - io_file_put(state, NULL); + io_file_put(state); } state->file = fget_many(fd, state->ios_left); if (!state->file) @@ -1671,7 +1669,7 @@ out: static void io_submit_state_end(struct io_submit_state *state) { blk_finish_plug(&state->plug); - io_file_put(state, NULL); + io_file_put(state); if (state->free_reqs) kmem_cache_free_bulk(req_cachep, state->free_reqs, &state->reqs[state->cur_req]); @@ -1920,6 +1918,10 @@ static int io_sq_thread(void *data) unuse_mm(cur_mm); mmput(cur_mm); } + + if (kthread_should_park()) + kthread_parkme(); + return 0; } @@ -2054,6 +2056,7 @@ static void io_sq_thread_stop(struct io_ring_ctx *ctx) if (ctx->sqo_thread) { ctx->sqo_stop = 1; mb(); + kthread_park(ctx->sqo_thread); kthread_stop(ctx->sqo_thread); ctx->sqo_thread = NULL; } @@ -2236,10 +2239,6 @@ static int io_sq_offload_start(struct io_ring_ctx *ctx, mmgrab(current->mm); ctx->sqo_mm = current->mm; - ctx->sq_thread_idle = msecs_to_jiffies(p->sq_thread_idle); - if (!ctx->sq_thread_idle) - ctx->sq_thread_idle = HZ; - ret = -EINVAL; if (!cpu_possible(p->sq_thread_cpu)) goto err; @@ -2249,10 +2248,18 @@ static int io_sq_offload_start(struct io_ring_ctx *ctx, if (!capable(CAP_SYS_ADMIN)) goto err; + ctx->sq_thread_idle = msecs_to_jiffies(p->sq_thread_idle); + if (!ctx->sq_thread_idle) + ctx->sq_thread_idle = HZ; + if (p->flags & IORING_SETUP_SQ_AFF) { int cpu; cpu = array_index_nospec(p->sq_thread_cpu, NR_CPUS); + ret = -EINVAL; + if (!cpu_possible(p->sq_thread_cpu)) + goto err; + ctx->sqo_thread = kthread_create_on_cpu(io_sq_thread, ctx, cpu, "io_uring-sq"); @@ -2922,11 +2929,23 @@ SYSCALL_DEFINE2(io_uring_setup, u32, entries, static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode, void __user *arg, unsigned nr_args) + __releases(ctx->uring_lock) + __acquires(ctx->uring_lock) { int ret; percpu_ref_kill(&ctx->refs); + + /* + * Drop uring mutex before waiting for references to exit. If another + * thread is currently inside io_uring_enter() it might need to grab + * the uring_lock to make progress. If we hold it here across the drain + * wait, then we can deadlock. It's safe to drop the mutex here, since + * no new references will come in after we've killed the percpu ref. + */ + mutex_unlock(&ctx->uring_lock); wait_for_completion(&ctx->ctx_done); + mutex_lock(&ctx->uring_lock); switch (opcode) { case IORING_REGISTER_BUFFERS: diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h @@ -548,7 +548,6 @@ struct request_queue { struct rcu_head rcu_head; wait_queue_head_t mq_freeze_wq; struct percpu_ref q_usage_counter; - struct list_head all_q_node; struct blk_mq_tag_set *tag_set; struct list_head tag_set_list; diff --git a/include/linux/bvec.h b/include/linux/bvec.h @@ -160,8 +160,9 @@ static inline void bvec_advance(const struct bio_vec *bvec, bv->bv_page = nth_page(bv->bv_page, 1); bv->bv_offset = 0; } else { - bv->bv_page = bvec->bv_page; - bv->bv_offset = bvec->bv_offset; + bv->bv_page = bvec_nth_page(bvec->bv_page, bvec->bv_offset / + PAGE_SIZE); + bv->bv_offset = bvec->bv_offset & ~PAGE_MASK; } bv->bv_len = min_t(unsigned int, PAGE_SIZE - bv->bv_offset, bvec->bv_len - iter_all->done); diff --git a/include/linux/elevator.h b/include/linux/elevator.h @@ -31,6 +31,7 @@ struct elevator_mq_ops { void (*exit_sched)(struct elevator_queue *); int (*init_hctx)(struct blk_mq_hw_ctx *, unsigned int); void (*exit_hctx)(struct blk_mq_hw_ctx *, unsigned int); + void (*depth_updated)(struct blk_mq_hw_ctx *); bool (*allow_merge)(struct request_queue *, struct request *, struct bio *); bool (*bio_merge)(struct blk_mq_hw_ctx *, struct bio *);