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 4aa84f2ffa81f71e15e5cffc2cc6090dbee78f8e Mon Sep 17 00:00:00 2001
From: Pavel Begunkov <asml.silence(a)gmail.com>
Date: Thu, 7 Jan 2021 03:15:42 +0000
Subject: [PATCH] io_uring: dont kill fasync under completion_lock
CPU0 CPU1
---- ----
lock(&new->fa_lock);
local_irq_disable();
lock(&ctx->completion_lock);
lock(&new->fa_lock);
<Interrupt>
lock(&ctx->completion_lock);
*** DEADLOCK ***
Move kill_fasync() out of io_commit_cqring() to io_cqring_ev_posted(),
so it doesn't hold completion_lock while doing it. That saves from the
reported deadlock, and it's just nice to shorten the locking time and
untangle nested locks (compl_lock -> wq_head::lock).
Reported-by: syzbot+91ca3f25bd7f795f019c(a)syzkaller.appspotmail.com
Signed-off-by: Pavel Begunkov <asml.silence(a)gmail.com>
Signed-off-by: Jens Axboe <axboe(a)kernel.dk>
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 91e517ad1421..401316fe2ae2 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -1345,11 +1345,6 @@ static void __io_commit_cqring(struct io_ring_ctx *ctx)
/* order cqe stores with ring update */
smp_store_release(&rings->cq.tail, ctx->cached_cq_tail);
-
- if (wq_has_sleeper(&ctx->cq_wait)) {
- wake_up_interruptible(&ctx->cq_wait);
- kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
- }
}
static void io_put_identity(struct io_uring_task *tctx, struct io_kiocb *req)
@@ -1711,6 +1706,10 @@ static void io_cqring_ev_posted(struct io_ring_ctx *ctx)
wake_up(&ctx->sq_data->wait);
if (io_should_trigger_evfd(ctx))
eventfd_signal(ctx->cq_ev_fd, 1);
+ if (wq_has_sleeper(&ctx->cq_wait)) {
+ wake_up_interruptible(&ctx->cq_wait);
+ kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
+ }
}
static void io_cqring_ev_posted_iopoll(struct io_ring_ctx *ctx)
@@ -1721,6 +1720,10 @@ static void io_cqring_ev_posted_iopoll(struct io_ring_ctx *ctx)
}
if (io_should_trigger_evfd(ctx))
eventfd_signal(ctx->cq_ev_fd, 1);
+ if (wq_has_sleeper(&ctx->cq_wait)) {
+ wake_up_interruptible(&ctx->cq_wait);
+ kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
+ }
}
/* Returns true if there are no backlogged entries after the flush */
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 81b6d05ccad4f3d8a9dfb091fb46ad6978ee40e4 Mon Sep 17 00:00:00 2001
From: Pavel Begunkov <asml.silence(a)gmail.com>
Date: Mon, 4 Jan 2021 20:36:35 +0000
Subject: [PATCH] io_uring: synchronise IOPOLL on task_submit fail
io_req_task_submit() might be called for IOPOLL, do the fail path under
uring_lock to comply with IOPOLL synchronisation based solely on it.
Cc: stable(a)vger.kernel.org # 5.5+
Signed-off-by: Pavel Begunkov <asml.silence(a)gmail.com>
Signed-off-by: Jens Axboe <axboe(a)kernel.dk>
diff --git a/fs/io_uring.c b/fs/io_uring.c
index ca46f314640b..5be33fd8b6bc 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -2126,15 +2126,16 @@ static void io_req_task_cancel(struct callback_head *cb)
static void __io_req_task_submit(struct io_kiocb *req)
{
struct io_ring_ctx *ctx = req->ctx;
+ bool fail;
- if (!__io_sq_thread_acquire_mm(ctx) &&
- !__io_sq_thread_acquire_files(ctx)) {
- mutex_lock(&ctx->uring_lock);
+ fail = __io_sq_thread_acquire_mm(ctx) ||
+ __io_sq_thread_acquire_files(ctx);
+ mutex_lock(&ctx->uring_lock);
+ if (!fail)
__io_queue_sqe(req, NULL);
- mutex_unlock(&ctx->uring_lock);
- } else {
+ else
__io_req_task_cancel(req, -EFAULT);
- }
+ mutex_unlock(&ctx->uring_lock);
}
static void io_req_task_submit(struct callback_head *cb)
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 aebf5db917055b38f4945ed6d621d9f07a44ff30 Mon Sep 17 00:00:00 2001
From: Ming Lei <ming.lei(a)redhat.com>
Date: Mon, 21 Dec 2020 12:33:35 +0800
Subject: [PATCH] block: fix use-after-free in disk_part_iter_next
Make sure that bdgrab() is done on the 'block_device' instance before
referring to it for avoiding use-after-free.
Cc: <stable(a)vger.kernel.org>
Reported-by: syzbot+825f0f9657d4e528046e(a)syzkaller.appspotmail.com
Signed-off-by: Ming Lei <ming.lei(a)redhat.com>
Reviewed-by: Christoph Hellwig <hch(a)lst.de>
Signed-off-by: Jens Axboe <axboe(a)kernel.dk>
diff --git a/block/genhd.c b/block/genhd.c
index 73faec438e49..419548e92d82 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -246,15 +246,18 @@ struct block_device *disk_part_iter_next(struct disk_part_iter *piter)
part = rcu_dereference(ptbl->part[piter->idx]);
if (!part)
continue;
+ piter->part = bdgrab(part);
+ if (!piter->part)
+ continue;
if (!bdev_nr_sectors(part) &&
!(piter->flags & DISK_PITER_INCL_EMPTY) &&
!(piter->flags & DISK_PITER_INCL_EMPTY_PART0 &&
- piter->idx == 0))
+ piter->idx == 0)) {
+ bdput(piter->part);
+ piter->part = NULL;
continue;
+ }
- piter->part = bdgrab(part);
- if (!piter->part)
- continue;
piter->idx += inc;
break;
}
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 aebf5db917055b38f4945ed6d621d9f07a44ff30 Mon Sep 17 00:00:00 2001
From: Ming Lei <ming.lei(a)redhat.com>
Date: Mon, 21 Dec 2020 12:33:35 +0800
Subject: [PATCH] block: fix use-after-free in disk_part_iter_next
Make sure that bdgrab() is done on the 'block_device' instance before
referring to it for avoiding use-after-free.
Cc: <stable(a)vger.kernel.org>
Reported-by: syzbot+825f0f9657d4e528046e(a)syzkaller.appspotmail.com
Signed-off-by: Ming Lei <ming.lei(a)redhat.com>
Reviewed-by: Christoph Hellwig <hch(a)lst.de>
Signed-off-by: Jens Axboe <axboe(a)kernel.dk>
diff --git a/block/genhd.c b/block/genhd.c
index 73faec438e49..419548e92d82 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -246,15 +246,18 @@ struct block_device *disk_part_iter_next(struct disk_part_iter *piter)
part = rcu_dereference(ptbl->part[piter->idx]);
if (!part)
continue;
+ piter->part = bdgrab(part);
+ if (!piter->part)
+ continue;
if (!bdev_nr_sectors(part) &&
!(piter->flags & DISK_PITER_INCL_EMPTY) &&
!(piter->flags & DISK_PITER_INCL_EMPTY_PART0 &&
- piter->idx == 0))
+ piter->idx == 0)) {
+ bdput(piter->part);
+ piter->part = NULL;
continue;
+ }
- piter->part = bdgrab(part);
- if (!piter->part)
- continue;
piter->idx += inc;
break;
}
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 75353bcd2184010f08a3ed2f0da019bd9d604e1e Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld(a)intel.com>
Date: Thu, 24 Dec 2020 15:13:57 +0000
Subject: [PATCH] drm/i915: clear the shadow batch
The shadow batch is an internal object, which doesn't have any page
clearing, and since the batch_len can be smaller than the object, we
should take care to clear it.
Testcase: igt/gen9_exec_parse/shadow-peek
Fixes: 4f7af1948abc ("drm/i915: Support ro ppgtt mapped cmdparser shadow buffers")
Signed-off-by: Matthew Auld <matthew.auld(a)intel.com>
Reviewed-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224151358.401345-1-matth…
Cc: stable(a)vger.kernel.org
(cherry picked from commit eeb52ee6c4a429ec301faf1dc48988744960786e)
Signed-off-by: Jani Nikula <jani.nikula(a)intel.com>
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 93265951fdbb..b0899b665e85 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -1166,7 +1166,7 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
}
}
if (IS_ERR(src)) {
- unsigned long x, n;
+ unsigned long x, n, remain;
void *ptr;
/*
@@ -1177,14 +1177,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
* We don't care about copying too much here as we only
* validate up to the end of the batch.
*/
+ remain = length;
if (!(dst_obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ))
- length = round_up(length,
+ remain = round_up(remain,
boot_cpu_data.x86_clflush_size);
ptr = dst;
x = offset_in_page(offset);
- for (n = offset >> PAGE_SHIFT; length; n++) {
- int len = min(length, PAGE_SIZE - x);
+ for (n = offset >> PAGE_SHIFT; remain; n++) {
+ int len = min(remain, PAGE_SIZE - x);
src = kmap_atomic(i915_gem_object_get_page(src_obj, n));
if (needs_clflush)
@@ -1193,13 +1194,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
kunmap_atomic(src);
ptr += len;
- length -= len;
+ remain -= len;
x = 0;
}
}
i915_gem_object_unpin_pages(src_obj);
+ memset32(dst + length, 0, (dst_obj->base.size - length) / sizeof(u32));
+
/* dst_obj is returned with vmap pinned */
return dst;
}
@@ -1392,11 +1395,6 @@ static unsigned long *alloc_whitelist(u32 batch_length)
#define LENGTH_BIAS 2
-static bool shadow_needs_clflush(struct drm_i915_gem_object *obj)
-{
- return !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE);
-}
-
/**
* intel_engine_cmd_parser() - parse a batch buffer for privilege violations
* @engine: the engine on which the batch is to execute
@@ -1538,16 +1536,9 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine,
ret = 0; /* allow execution */
}
}
-
- if (shadow_needs_clflush(shadow->obj))
- drm_clflush_virt_range(batch_end, 8);
}
- if (shadow_needs_clflush(shadow->obj)) {
- void *ptr = page_mask_bits(shadow->obj->mm.mapping);
-
- drm_clflush_virt_range(ptr, (void *)(cmd + 1) - ptr);
- }
+ i915_gem_object_flush_map(shadow->obj);
if (!IS_ERR_OR_NULL(jump_whitelist))
kfree(jump_whitelist);
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 75353bcd2184010f08a3ed2f0da019bd9d604e1e Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld(a)intel.com>
Date: Thu, 24 Dec 2020 15:13:57 +0000
Subject: [PATCH] drm/i915: clear the shadow batch
The shadow batch is an internal object, which doesn't have any page
clearing, and since the batch_len can be smaller than the object, we
should take care to clear it.
Testcase: igt/gen9_exec_parse/shadow-peek
Fixes: 4f7af1948abc ("drm/i915: Support ro ppgtt mapped cmdparser shadow buffers")
Signed-off-by: Matthew Auld <matthew.auld(a)intel.com>
Reviewed-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224151358.401345-1-matth…
Cc: stable(a)vger.kernel.org
(cherry picked from commit eeb52ee6c4a429ec301faf1dc48988744960786e)
Signed-off-by: Jani Nikula <jani.nikula(a)intel.com>
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 93265951fdbb..b0899b665e85 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -1166,7 +1166,7 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
}
}
if (IS_ERR(src)) {
- unsigned long x, n;
+ unsigned long x, n, remain;
void *ptr;
/*
@@ -1177,14 +1177,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
* We don't care about copying too much here as we only
* validate up to the end of the batch.
*/
+ remain = length;
if (!(dst_obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ))
- length = round_up(length,
+ remain = round_up(remain,
boot_cpu_data.x86_clflush_size);
ptr = dst;
x = offset_in_page(offset);
- for (n = offset >> PAGE_SHIFT; length; n++) {
- int len = min(length, PAGE_SIZE - x);
+ for (n = offset >> PAGE_SHIFT; remain; n++) {
+ int len = min(remain, PAGE_SIZE - x);
src = kmap_atomic(i915_gem_object_get_page(src_obj, n));
if (needs_clflush)
@@ -1193,13 +1194,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
kunmap_atomic(src);
ptr += len;
- length -= len;
+ remain -= len;
x = 0;
}
}
i915_gem_object_unpin_pages(src_obj);
+ memset32(dst + length, 0, (dst_obj->base.size - length) / sizeof(u32));
+
/* dst_obj is returned with vmap pinned */
return dst;
}
@@ -1392,11 +1395,6 @@ static unsigned long *alloc_whitelist(u32 batch_length)
#define LENGTH_BIAS 2
-static bool shadow_needs_clflush(struct drm_i915_gem_object *obj)
-{
- return !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE);
-}
-
/**
* intel_engine_cmd_parser() - parse a batch buffer for privilege violations
* @engine: the engine on which the batch is to execute
@@ -1538,16 +1536,9 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine,
ret = 0; /* allow execution */
}
}
-
- if (shadow_needs_clflush(shadow->obj))
- drm_clflush_virt_range(batch_end, 8);
}
- if (shadow_needs_clflush(shadow->obj)) {
- void *ptr = page_mask_bits(shadow->obj->mm.mapping);
-
- drm_clflush_virt_range(ptr, (void *)(cmd + 1) - ptr);
- }
+ i915_gem_object_flush_map(shadow->obj);
if (!IS_ERR_OR_NULL(jump_whitelist))
kfree(jump_whitelist);
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 75353bcd2184010f08a3ed2f0da019bd9d604e1e Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld(a)intel.com>
Date: Thu, 24 Dec 2020 15:13:57 +0000
Subject: [PATCH] drm/i915: clear the shadow batch
The shadow batch is an internal object, which doesn't have any page
clearing, and since the batch_len can be smaller than the object, we
should take care to clear it.
Testcase: igt/gen9_exec_parse/shadow-peek
Fixes: 4f7af1948abc ("drm/i915: Support ro ppgtt mapped cmdparser shadow buffers")
Signed-off-by: Matthew Auld <matthew.auld(a)intel.com>
Reviewed-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224151358.401345-1-matth…
Cc: stable(a)vger.kernel.org
(cherry picked from commit eeb52ee6c4a429ec301faf1dc48988744960786e)
Signed-off-by: Jani Nikula <jani.nikula(a)intel.com>
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 93265951fdbb..b0899b665e85 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -1166,7 +1166,7 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
}
}
if (IS_ERR(src)) {
- unsigned long x, n;
+ unsigned long x, n, remain;
void *ptr;
/*
@@ -1177,14 +1177,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
* We don't care about copying too much here as we only
* validate up to the end of the batch.
*/
+ remain = length;
if (!(dst_obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ))
- length = round_up(length,
+ remain = round_up(remain,
boot_cpu_data.x86_clflush_size);
ptr = dst;
x = offset_in_page(offset);
- for (n = offset >> PAGE_SHIFT; length; n++) {
- int len = min(length, PAGE_SIZE - x);
+ for (n = offset >> PAGE_SHIFT; remain; n++) {
+ int len = min(remain, PAGE_SIZE - x);
src = kmap_atomic(i915_gem_object_get_page(src_obj, n));
if (needs_clflush)
@@ -1193,13 +1194,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
kunmap_atomic(src);
ptr += len;
- length -= len;
+ remain -= len;
x = 0;
}
}
i915_gem_object_unpin_pages(src_obj);
+ memset32(dst + length, 0, (dst_obj->base.size - length) / sizeof(u32));
+
/* dst_obj is returned with vmap pinned */
return dst;
}
@@ -1392,11 +1395,6 @@ static unsigned long *alloc_whitelist(u32 batch_length)
#define LENGTH_BIAS 2
-static bool shadow_needs_clflush(struct drm_i915_gem_object *obj)
-{
- return !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE);
-}
-
/**
* intel_engine_cmd_parser() - parse a batch buffer for privilege violations
* @engine: the engine on which the batch is to execute
@@ -1538,16 +1536,9 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine,
ret = 0; /* allow execution */
}
}
-
- if (shadow_needs_clflush(shadow->obj))
- drm_clflush_virt_range(batch_end, 8);
}
- if (shadow_needs_clflush(shadow->obj)) {
- void *ptr = page_mask_bits(shadow->obj->mm.mapping);
-
- drm_clflush_virt_range(ptr, (void *)(cmd + 1) - ptr);
- }
+ i915_gem_object_flush_map(shadow->obj);
if (!IS_ERR_OR_NULL(jump_whitelist))
kfree(jump_whitelist);
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 75353bcd2184010f08a3ed2f0da019bd9d604e1e Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld(a)intel.com>
Date: Thu, 24 Dec 2020 15:13:57 +0000
Subject: [PATCH] drm/i915: clear the shadow batch
The shadow batch is an internal object, which doesn't have any page
clearing, and since the batch_len can be smaller than the object, we
should take care to clear it.
Testcase: igt/gen9_exec_parse/shadow-peek
Fixes: 4f7af1948abc ("drm/i915: Support ro ppgtt mapped cmdparser shadow buffers")
Signed-off-by: Matthew Auld <matthew.auld(a)intel.com>
Reviewed-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224151358.401345-1-matth…
Cc: stable(a)vger.kernel.org
(cherry picked from commit eeb52ee6c4a429ec301faf1dc48988744960786e)
Signed-off-by: Jani Nikula <jani.nikula(a)intel.com>
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 93265951fdbb..b0899b665e85 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -1166,7 +1166,7 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
}
}
if (IS_ERR(src)) {
- unsigned long x, n;
+ unsigned long x, n, remain;
void *ptr;
/*
@@ -1177,14 +1177,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
* We don't care about copying too much here as we only
* validate up to the end of the batch.
*/
+ remain = length;
if (!(dst_obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ))
- length = round_up(length,
+ remain = round_up(remain,
boot_cpu_data.x86_clflush_size);
ptr = dst;
x = offset_in_page(offset);
- for (n = offset >> PAGE_SHIFT; length; n++) {
- int len = min(length, PAGE_SIZE - x);
+ for (n = offset >> PAGE_SHIFT; remain; n++) {
+ int len = min(remain, PAGE_SIZE - x);
src = kmap_atomic(i915_gem_object_get_page(src_obj, n));
if (needs_clflush)
@@ -1193,13 +1194,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
kunmap_atomic(src);
ptr += len;
- length -= len;
+ remain -= len;
x = 0;
}
}
i915_gem_object_unpin_pages(src_obj);
+ memset32(dst + length, 0, (dst_obj->base.size - length) / sizeof(u32));
+
/* dst_obj is returned with vmap pinned */
return dst;
}
@@ -1392,11 +1395,6 @@ static unsigned long *alloc_whitelist(u32 batch_length)
#define LENGTH_BIAS 2
-static bool shadow_needs_clflush(struct drm_i915_gem_object *obj)
-{
- return !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE);
-}
-
/**
* intel_engine_cmd_parser() - parse a batch buffer for privilege violations
* @engine: the engine on which the batch is to execute
@@ -1538,16 +1536,9 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine,
ret = 0; /* allow execution */
}
}
-
- if (shadow_needs_clflush(shadow->obj))
- drm_clflush_virt_range(batch_end, 8);
}
- if (shadow_needs_clflush(shadow->obj)) {
- void *ptr = page_mask_bits(shadow->obj->mm.mapping);
-
- drm_clflush_virt_range(ptr, (void *)(cmd + 1) - ptr);
- }
+ i915_gem_object_flush_map(shadow->obj);
if (!IS_ERR_OR_NULL(jump_whitelist))
kfree(jump_whitelist);
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 75353bcd2184010f08a3ed2f0da019bd9d604e1e Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld(a)intel.com>
Date: Thu, 24 Dec 2020 15:13:57 +0000
Subject: [PATCH] drm/i915: clear the shadow batch
The shadow batch is an internal object, which doesn't have any page
clearing, and since the batch_len can be smaller than the object, we
should take care to clear it.
Testcase: igt/gen9_exec_parse/shadow-peek
Fixes: 4f7af1948abc ("drm/i915: Support ro ppgtt mapped cmdparser shadow buffers")
Signed-off-by: Matthew Auld <matthew.auld(a)intel.com>
Reviewed-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224151358.401345-1-matth…
Cc: stable(a)vger.kernel.org
(cherry picked from commit eeb52ee6c4a429ec301faf1dc48988744960786e)
Signed-off-by: Jani Nikula <jani.nikula(a)intel.com>
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 93265951fdbb..b0899b665e85 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -1166,7 +1166,7 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
}
}
if (IS_ERR(src)) {
- unsigned long x, n;
+ unsigned long x, n, remain;
void *ptr;
/*
@@ -1177,14 +1177,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
* We don't care about copying too much here as we only
* validate up to the end of the batch.
*/
+ remain = length;
if (!(dst_obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ))
- length = round_up(length,
+ remain = round_up(remain,
boot_cpu_data.x86_clflush_size);
ptr = dst;
x = offset_in_page(offset);
- for (n = offset >> PAGE_SHIFT; length; n++) {
- int len = min(length, PAGE_SIZE - x);
+ for (n = offset >> PAGE_SHIFT; remain; n++) {
+ int len = min(remain, PAGE_SIZE - x);
src = kmap_atomic(i915_gem_object_get_page(src_obj, n));
if (needs_clflush)
@@ -1193,13 +1194,15 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj,
kunmap_atomic(src);
ptr += len;
- length -= len;
+ remain -= len;
x = 0;
}
}
i915_gem_object_unpin_pages(src_obj);
+ memset32(dst + length, 0, (dst_obj->base.size - length) / sizeof(u32));
+
/* dst_obj is returned with vmap pinned */
return dst;
}
@@ -1392,11 +1395,6 @@ static unsigned long *alloc_whitelist(u32 batch_length)
#define LENGTH_BIAS 2
-static bool shadow_needs_clflush(struct drm_i915_gem_object *obj)
-{
- return !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE);
-}
-
/**
* intel_engine_cmd_parser() - parse a batch buffer for privilege violations
* @engine: the engine on which the batch is to execute
@@ -1538,16 +1536,9 @@ int intel_engine_cmd_parser(struct intel_engine_cs *engine,
ret = 0; /* allow execution */
}
}
-
- if (shadow_needs_clflush(shadow->obj))
- drm_clflush_virt_range(batch_end, 8);
}
- if (shadow_needs_clflush(shadow->obj)) {
- void *ptr = page_mask_bits(shadow->obj->mm.mapping);
-
- drm_clflush_virt_range(ptr, (void *)(cmd + 1) - ptr);
- }
+ i915_gem_object_flush_map(shadow->obj);
if (!IS_ERR_OR_NULL(jump_whitelist))
kfree(jump_whitelist);
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 311bea3cb9ee20ef150ca76fc60a592bf6b159f5 Mon Sep 17 00:00:00 2001
From: Nick Desaulniers <ndesaulniers(a)google.com>
Date: Thu, 17 Dec 2020 16:24:32 -0800
Subject: [PATCH] arm64: link with -z norelro for LLD or aarch64-elf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
With GNU binutils 2.35+, linking with BFD produces warnings for vmlinux:
aarch64-linux-gnu-ld: warning: -z norelro ignored
BFD can produce this warning when the target emulation mode does not
support RELRO program headers, and -z relro or -z norelro is passed.
Alan Modra clarifies:
The default linker emulation for an aarch64-linux ld.bfd is
-maarch64linux, the default for an aarch64-elf linker is
-maarch64elf. They are not equivalent. If you choose -maarch64elf
you get an emulation that doesn't support -z relro.
The ARCH=arm64 kernel prefers -maarch64elf, but may fall back to
-maarch64linux based on the toolchain configuration.
LLD will always create RELRO program header regardless of target
emulation.
To avoid the above warning when linking with BFD, pass -z norelro only
when linking with LLD or with -maarch64linux.
Fixes: 3b92fa7485eb ("arm64: link with -z norelro regardless of CONFIG_RELOCATABLE")
Fixes: 3bbd3db86470 ("arm64: relocatable: fix inconsistencies in linker script and options")
Cc: <stable(a)vger.kernel.org> # 5.0.x-
Reported-by: kernelci.org bot <bot(a)kernelci.org>
Reported-by: Quentin Perret <qperret(a)google.com>
Signed-off-by: Nick Desaulniers <ndesaulniers(a)google.com>
Reviewed-by: Nathan Chancellor <natechancellor(a)gmail.com>
Acked-by: Ard Biesheuvel <ardb(a)kernel.org>
Cc: Alan Modra <amodra(a)gmail.com>
Cc: Fāng-ruì Sòng <maskray(a)google.com>
Link: https://lore.kernel.org/r/20201218002432.788499-1-ndesaulniers@google.com
Signed-off-by: Catalin Marinas <catalin.marinas(a)arm.com>
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index 6be9b3750250..90309208bb28 100644
--- a/arch/arm64/Makefile
+++ b/arch/arm64/Makefile
@@ -10,7 +10,7 @@
#
# Copyright (C) 1995-2001 by Russell King
-LDFLAGS_vmlinux :=--no-undefined -X -z norelro
+LDFLAGS_vmlinux :=--no-undefined -X
ifeq ($(CONFIG_RELOCATABLE), y)
# Pass --no-apply-dynamic-relocs to restore pre-binutils-2.27 behaviour
@@ -115,16 +115,20 @@ KBUILD_CPPFLAGS += -mbig-endian
CHECKFLAGS += -D__AARCH64EB__
# Prefer the baremetal ELF build target, but not all toolchains include
# it so fall back to the standard linux version if needed.
-KBUILD_LDFLAGS += -EB $(call ld-option, -maarch64elfb, -maarch64linuxb)
+KBUILD_LDFLAGS += -EB $(call ld-option, -maarch64elfb, -maarch64linuxb -z norelro)
UTS_MACHINE := aarch64_be
else
KBUILD_CPPFLAGS += -mlittle-endian
CHECKFLAGS += -D__AARCH64EL__
# Same as above, prefer ELF but fall back to linux target if needed.
-KBUILD_LDFLAGS += -EL $(call ld-option, -maarch64elf, -maarch64linux)
+KBUILD_LDFLAGS += -EL $(call ld-option, -maarch64elf, -maarch64linux -z norelro)
UTS_MACHINE := aarch64
endif
+ifeq ($(CONFIG_LD_IS_LLD), y)
+KBUILD_LDFLAGS += -z norelro
+endif
+
CHECKFLAGS += -D__aarch64__
ifeq ($(CONFIG_DYNAMIC_FTRACE_WITH_REGS),y)
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 311bea3cb9ee20ef150ca76fc60a592bf6b159f5 Mon Sep 17 00:00:00 2001
From: Nick Desaulniers <ndesaulniers(a)google.com>
Date: Thu, 17 Dec 2020 16:24:32 -0800
Subject: [PATCH] arm64: link with -z norelro for LLD or aarch64-elf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
With GNU binutils 2.35+, linking with BFD produces warnings for vmlinux:
aarch64-linux-gnu-ld: warning: -z norelro ignored
BFD can produce this warning when the target emulation mode does not
support RELRO program headers, and -z relro or -z norelro is passed.
Alan Modra clarifies:
The default linker emulation for an aarch64-linux ld.bfd is
-maarch64linux, the default for an aarch64-elf linker is
-maarch64elf. They are not equivalent. If you choose -maarch64elf
you get an emulation that doesn't support -z relro.
The ARCH=arm64 kernel prefers -maarch64elf, but may fall back to
-maarch64linux based on the toolchain configuration.
LLD will always create RELRO program header regardless of target
emulation.
To avoid the above warning when linking with BFD, pass -z norelro only
when linking with LLD or with -maarch64linux.
Fixes: 3b92fa7485eb ("arm64: link with -z norelro regardless of CONFIG_RELOCATABLE")
Fixes: 3bbd3db86470 ("arm64: relocatable: fix inconsistencies in linker script and options")
Cc: <stable(a)vger.kernel.org> # 5.0.x-
Reported-by: kernelci.org bot <bot(a)kernelci.org>
Reported-by: Quentin Perret <qperret(a)google.com>
Signed-off-by: Nick Desaulniers <ndesaulniers(a)google.com>
Reviewed-by: Nathan Chancellor <natechancellor(a)gmail.com>
Acked-by: Ard Biesheuvel <ardb(a)kernel.org>
Cc: Alan Modra <amodra(a)gmail.com>
Cc: Fāng-ruì Sòng <maskray(a)google.com>
Link: https://lore.kernel.org/r/20201218002432.788499-1-ndesaulniers@google.com
Signed-off-by: Catalin Marinas <catalin.marinas(a)arm.com>
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index 6be9b3750250..90309208bb28 100644
--- a/arch/arm64/Makefile
+++ b/arch/arm64/Makefile
@@ -10,7 +10,7 @@
#
# Copyright (C) 1995-2001 by Russell King
-LDFLAGS_vmlinux :=--no-undefined -X -z norelro
+LDFLAGS_vmlinux :=--no-undefined -X
ifeq ($(CONFIG_RELOCATABLE), y)
# Pass --no-apply-dynamic-relocs to restore pre-binutils-2.27 behaviour
@@ -115,16 +115,20 @@ KBUILD_CPPFLAGS += -mbig-endian
CHECKFLAGS += -D__AARCH64EB__
# Prefer the baremetal ELF build target, but not all toolchains include
# it so fall back to the standard linux version if needed.
-KBUILD_LDFLAGS += -EB $(call ld-option, -maarch64elfb, -maarch64linuxb)
+KBUILD_LDFLAGS += -EB $(call ld-option, -maarch64elfb, -maarch64linuxb -z norelro)
UTS_MACHINE := aarch64_be
else
KBUILD_CPPFLAGS += -mlittle-endian
CHECKFLAGS += -D__AARCH64EL__
# Same as above, prefer ELF but fall back to linux target if needed.
-KBUILD_LDFLAGS += -EL $(call ld-option, -maarch64elf, -maarch64linux)
+KBUILD_LDFLAGS += -EL $(call ld-option, -maarch64elf, -maarch64linux -z norelro)
UTS_MACHINE := aarch64
endif
+ifeq ($(CONFIG_LD_IS_LLD), y)
+KBUILD_LDFLAGS += -z norelro
+endif
+
CHECKFLAGS += -D__aarch64__
ifeq ($(CONFIG_DYNAMIC_FTRACE_WITH_REGS),y)
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 311bea3cb9ee20ef150ca76fc60a592bf6b159f5 Mon Sep 17 00:00:00 2001
From: Nick Desaulniers <ndesaulniers(a)google.com>
Date: Thu, 17 Dec 2020 16:24:32 -0800
Subject: [PATCH] arm64: link with -z norelro for LLD or aarch64-elf
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
With GNU binutils 2.35+, linking with BFD produces warnings for vmlinux:
aarch64-linux-gnu-ld: warning: -z norelro ignored
BFD can produce this warning when the target emulation mode does not
support RELRO program headers, and -z relro or -z norelro is passed.
Alan Modra clarifies:
The default linker emulation for an aarch64-linux ld.bfd is
-maarch64linux, the default for an aarch64-elf linker is
-maarch64elf. They are not equivalent. If you choose -maarch64elf
you get an emulation that doesn't support -z relro.
The ARCH=arm64 kernel prefers -maarch64elf, but may fall back to
-maarch64linux based on the toolchain configuration.
LLD will always create RELRO program header regardless of target
emulation.
To avoid the above warning when linking with BFD, pass -z norelro only
when linking with LLD or with -maarch64linux.
Fixes: 3b92fa7485eb ("arm64: link with -z norelro regardless of CONFIG_RELOCATABLE")
Fixes: 3bbd3db86470 ("arm64: relocatable: fix inconsistencies in linker script and options")
Cc: <stable(a)vger.kernel.org> # 5.0.x-
Reported-by: kernelci.org bot <bot(a)kernelci.org>
Reported-by: Quentin Perret <qperret(a)google.com>
Signed-off-by: Nick Desaulniers <ndesaulniers(a)google.com>
Reviewed-by: Nathan Chancellor <natechancellor(a)gmail.com>
Acked-by: Ard Biesheuvel <ardb(a)kernel.org>
Cc: Alan Modra <amodra(a)gmail.com>
Cc: Fāng-ruì Sòng <maskray(a)google.com>
Link: https://lore.kernel.org/r/20201218002432.788499-1-ndesaulniers@google.com
Signed-off-by: Catalin Marinas <catalin.marinas(a)arm.com>
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index 6be9b3750250..90309208bb28 100644
--- a/arch/arm64/Makefile
+++ b/arch/arm64/Makefile
@@ -10,7 +10,7 @@
#
# Copyright (C) 1995-2001 by Russell King
-LDFLAGS_vmlinux :=--no-undefined -X -z norelro
+LDFLAGS_vmlinux :=--no-undefined -X
ifeq ($(CONFIG_RELOCATABLE), y)
# Pass --no-apply-dynamic-relocs to restore pre-binutils-2.27 behaviour
@@ -115,16 +115,20 @@ KBUILD_CPPFLAGS += -mbig-endian
CHECKFLAGS += -D__AARCH64EB__
# Prefer the baremetal ELF build target, but not all toolchains include
# it so fall back to the standard linux version if needed.
-KBUILD_LDFLAGS += -EB $(call ld-option, -maarch64elfb, -maarch64linuxb)
+KBUILD_LDFLAGS += -EB $(call ld-option, -maarch64elfb, -maarch64linuxb -z norelro)
UTS_MACHINE := aarch64_be
else
KBUILD_CPPFLAGS += -mlittle-endian
CHECKFLAGS += -D__AARCH64EL__
# Same as above, prefer ELF but fall back to linux target if needed.
-KBUILD_LDFLAGS += -EL $(call ld-option, -maarch64elf, -maarch64linux)
+KBUILD_LDFLAGS += -EL $(call ld-option, -maarch64elf, -maarch64linux -z norelro)
UTS_MACHINE := aarch64
endif
+ifeq ($(CONFIG_LD_IS_LLD), y)
+KBUILD_LDFLAGS += -z norelro
+endif
+
CHECKFLAGS += -D__aarch64__
ifeq ($(CONFIG_DYNAMIC_FTRACE_WITH_REGS),y)
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 3d45f221ce627d13e2e6ef3274f06750c84a6542 Mon Sep 17 00:00:00 2001
From: Filipe Manana <fdmanana(a)suse.com>
Date: Wed, 2 Dec 2020 11:55:58 +0000
Subject: [PATCH] btrfs: fix deadlock when cloning inline extent and low on
free metadata space
When cloning an inline extent there are cases where we can not just copy
the inline extent from the source range to the target range (e.g. when the
target range starts at an offset greater than zero). In such cases we copy
the inline extent's data into a page of the destination inode and then
dirty that page. However, after that we will need to start a transaction
for each processed extent and, if we are ever low on available metadata
space, we may need to flush existing delalloc for all dirty inodes in an
attempt to release metadata space - if that happens we may deadlock:
* the async reclaim task queued a delalloc work to flush delalloc for
the destination inode of the clone operation;
* the task executing that delalloc work gets blocked waiting for the
range with the dirty page to be unlocked, which is currently locked
by the task doing the clone operation;
* the async reclaim task blocks waiting for the delalloc work to complete;
* the cloning task is waiting on the waitqueue of its reservation ticket
while holding the range with the dirty page locked in the inode's
io_tree;
* if metadata space is not released by some other task (like delalloc for
some other inode completing for example), the clone task waits forever
and as a consequence the delalloc work and async reclaim tasks will hang
forever as well. Releasing more space on the other hand may require
starting a transaction, which will hang as well when trying to reserve
metadata space, resulting in a deadlock between all these tasks.
When this happens, traces like the following show up in dmesg/syslog:
[87452.323003] INFO: task kworker/u16:11:1810830 blocked for more than 120 seconds.
[87452.323644] Tainted: G B W 5.10.0-rc4-btrfs-next-73 #1
[87452.324248] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[87452.324852] task:kworker/u16:11 state:D stack: 0 pid:1810830 ppid: 2 flags:0x00004000
[87452.325520] Workqueue: btrfs-flush_delalloc btrfs_work_helper [btrfs]
[87452.326136] Call Trace:
[87452.326737] __schedule+0x5d1/0xcf0
[87452.327390] schedule+0x45/0xe0
[87452.328174] lock_extent_bits+0x1e6/0x2d0 [btrfs]
[87452.328894] ? finish_wait+0x90/0x90
[87452.329474] btrfs_invalidatepage+0x32c/0x390 [btrfs]
[87452.330133] ? __mod_memcg_state+0x8e/0x160
[87452.330738] __extent_writepage+0x2d4/0x400 [btrfs]
[87452.331405] extent_write_cache_pages+0x2b2/0x500 [btrfs]
[87452.332007] ? lock_release+0x20e/0x4c0
[87452.332557] ? trace_hardirqs_on+0x1b/0xf0
[87452.333127] extent_writepages+0x43/0x90 [btrfs]
[87452.333653] ? lock_acquire+0x1a3/0x490
[87452.334177] do_writepages+0x43/0xe0
[87452.334699] ? __filemap_fdatawrite_range+0xa4/0x100
[87452.335720] __filemap_fdatawrite_range+0xc5/0x100
[87452.336500] btrfs_run_delalloc_work+0x17/0x40 [btrfs]
[87452.337216] btrfs_work_helper+0xf1/0x600 [btrfs]
[87452.337838] process_one_work+0x24e/0x5e0
[87452.338437] worker_thread+0x50/0x3b0
[87452.339137] ? process_one_work+0x5e0/0x5e0
[87452.339884] kthread+0x153/0x170
[87452.340507] ? kthread_mod_delayed_work+0xc0/0xc0
[87452.341153] ret_from_fork+0x22/0x30
[87452.341806] INFO: task kworker/u16:1:2426217 blocked for more than 120 seconds.
[87452.342487] Tainted: G B W 5.10.0-rc4-btrfs-next-73 #1
[87452.343274] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[87452.344049] task:kworker/u16:1 state:D stack: 0 pid:2426217 ppid: 2 flags:0x00004000
[87452.344974] Workqueue: events_unbound btrfs_async_reclaim_metadata_space [btrfs]
[87452.345655] Call Trace:
[87452.346305] __schedule+0x5d1/0xcf0
[87452.346947] ? kvm_clock_read+0x14/0x30
[87452.347676] ? wait_for_completion+0x81/0x110
[87452.348389] schedule+0x45/0xe0
[87452.349077] schedule_timeout+0x30c/0x580
[87452.349718] ? _raw_spin_unlock_irqrestore+0x3c/0x60
[87452.350340] ? lock_acquire+0x1a3/0x490
[87452.351006] ? try_to_wake_up+0x7a/0xa20
[87452.351541] ? lock_release+0x20e/0x4c0
[87452.352040] ? lock_acquired+0x199/0x490
[87452.352517] ? wait_for_completion+0x81/0x110
[87452.353000] wait_for_completion+0xab/0x110
[87452.353490] start_delalloc_inodes+0x2af/0x390 [btrfs]
[87452.353973] btrfs_start_delalloc_roots+0x12d/0x250 [btrfs]
[87452.354455] flush_space+0x24f/0x660 [btrfs]
[87452.355063] btrfs_async_reclaim_metadata_space+0x1bb/0x480 [btrfs]
[87452.355565] process_one_work+0x24e/0x5e0
[87452.356024] worker_thread+0x20f/0x3b0
[87452.356487] ? process_one_work+0x5e0/0x5e0
[87452.356973] kthread+0x153/0x170
[87452.357434] ? kthread_mod_delayed_work+0xc0/0xc0
[87452.357880] ret_from_fork+0x22/0x30
(...)
< stack traces of several tasks waiting for the locks of the inodes of the
clone operation >
(...)
[92867.444138] RSP: 002b:00007ffc3371bbe8 EFLAGS: 00000246 ORIG_RAX: 0000000000000052
[92867.444624] RAX: ffffffffffffffda RBX: 00007ffc3371bea0 RCX: 00007f61efe73f97
[92867.445116] RDX: 0000000000000000 RSI: 0000560fbd5d7a40 RDI: 0000560fbd5d8960
[92867.445595] RBP: 00007ffc3371beb0 R08: 0000000000000001 R09: 0000000000000003
[92867.446070] R10: 00007ffc3371b996 R11: 0000000000000246 R12: 0000000000000000
[92867.446820] R13: 000000000000001f R14: 00007ffc3371bea0 R15: 00007ffc3371beb0
[92867.447361] task:fsstress state:D stack: 0 pid:2508238 ppid:2508153 flags:0x00004000
[92867.447920] Call Trace:
[92867.448435] __schedule+0x5d1/0xcf0
[92867.448934] ? _raw_spin_unlock_irqrestore+0x3c/0x60
[92867.449423] schedule+0x45/0xe0
[92867.449916] __reserve_bytes+0x4a4/0xb10 [btrfs]
[92867.450576] ? finish_wait+0x90/0x90
[92867.451202] btrfs_reserve_metadata_bytes+0x29/0x190 [btrfs]
[92867.451815] btrfs_block_rsv_add+0x1f/0x50 [btrfs]
[92867.452412] start_transaction+0x2d1/0x760 [btrfs]
[92867.453216] clone_copy_inline_extent+0x333/0x490 [btrfs]
[92867.453848] ? lock_release+0x20e/0x4c0
[92867.454539] ? btrfs_search_slot+0x9a7/0xc30 [btrfs]
[92867.455218] btrfs_clone+0x569/0x7e0 [btrfs]
[92867.455952] btrfs_clone_files+0xf6/0x150 [btrfs]
[92867.456588] btrfs_remap_file_range+0x324/0x3d0 [btrfs]
[92867.457213] do_clone_file_range+0xd4/0x1f0
[92867.457828] vfs_clone_file_range+0x4d/0x230
[92867.458355] ? lock_release+0x20e/0x4c0
[92867.458890] ioctl_file_clone+0x8f/0xc0
[92867.459377] do_vfs_ioctl+0x342/0x750
[92867.459913] __x64_sys_ioctl+0x62/0xb0
[92867.460377] do_syscall_64+0x33/0x80
[92867.460842] entry_SYSCALL_64_after_hwframe+0x44/0xa9
(...)
< stack traces of more tasks blocked on metadata reservation like the clone
task above, because the async reclaim task has deadlocked >
(...)
Another thing to notice is that the worker task that is deadlocked when
trying to flush the destination inode of the clone operation is at
btrfs_invalidatepage(). This is simply because the clone operation has a
destination offset greater than the i_size and we only update the i_size
of the destination file after cloning an extent (just like we do in the
buffered write path).
Since the async reclaim path uses btrfs_start_delalloc_roots() to trigger
the flushing of delalloc for all inodes that have delalloc, add a runtime
flag to an inode to signal it should not be flushed, and for inodes with
that flag set, start_delalloc_inodes() will simply skip them. When the
cloning code needs to dirty a page to copy an inline extent, set that flag
on the inode and then clear it when the clone operation finishes.
This could be sporadically triggered with test case generic/269 from
fstests, which exercises many fsstress processes running in parallel with
several dd processes filling up the entire filesystem.
CC: stable(a)vger.kernel.org # 5.9+
Fixes: 05a5a7621ce6 ("Btrfs: implement full reflink support for inline extents")
Reviewed-by: Josef Bacik <josef(a)toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana(a)suse.com>
Reviewed-by: David Sterba <dsterba(a)suse.com>
Signed-off-by: David Sterba <dsterba(a)suse.com>
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 555cbcef6585..d9bf53d9ff90 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -42,6 +42,15 @@ enum {
* to an inode.
*/
BTRFS_INODE_NO_XATTRS,
+ /*
+ * Set when we are in a context where we need to start a transaction and
+ * have dirty pages with the respective file range locked. This is to
+ * ensure that when reserving space for the transaction, if we are low
+ * on available space and need to flush delalloc, we will not flush
+ * delalloc for this inode, because that could result in a deadlock (on
+ * the file range, inode's io_tree).
+ */
+ BTRFS_INODE_NO_DELALLOC_FLUSH,
};
/* in memory btrfs inode */
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 9dde7707873a..2674f24cf2e0 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -3074,7 +3074,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
u32 min_type);
int btrfs_start_delalloc_snapshot(struct btrfs_root *root);
-int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr);
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr,
+ bool in_reclaim_context);
int btrfs_set_extent_delalloc(struct btrfs_inode *inode, u64 start, u64 end,
unsigned int extra_bits,
struct extent_state **cached_state);
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index a98e33f232d5..324f646d6e5e 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -715,7 +715,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
* flush all outstanding I/O and inode extent mappings before the
* copy operation is declared as being finished
*/
- ret = btrfs_start_delalloc_roots(fs_info, U64_MAX);
+ ret = btrfs_start_delalloc_roots(fs_info, U64_MAX, false);
if (ret) {
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return ret;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 8e23780acfae..070716650df8 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -9390,7 +9390,8 @@ static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode
* some fairly slow code that needs optimization. This walks the list
* of all the inodes with pending delalloc and forces them to disk.
*/
-static int start_delalloc_inodes(struct btrfs_root *root, u64 *nr, bool snapshot)
+static int start_delalloc_inodes(struct btrfs_root *root, u64 *nr, bool snapshot,
+ bool in_reclaim_context)
{
struct btrfs_inode *binode;
struct inode *inode;
@@ -9411,6 +9412,11 @@ static int start_delalloc_inodes(struct btrfs_root *root, u64 *nr, bool snapshot
list_move_tail(&binode->delalloc_inodes,
&root->delalloc_inodes);
+
+ if (in_reclaim_context &&
+ test_bit(BTRFS_INODE_NO_DELALLOC_FLUSH, &binode->runtime_flags))
+ continue;
+
inode = igrab(&binode->vfs_inode);
if (!inode) {
cond_resched_lock(&root->delalloc_lock);
@@ -9464,10 +9470,11 @@ int btrfs_start_delalloc_snapshot(struct btrfs_root *root)
if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
return -EROFS;
- return start_delalloc_inodes(root, &nr, true);
+ return start_delalloc_inodes(root, &nr, true, false);
}
-int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr)
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr,
+ bool in_reclaim_context)
{
struct btrfs_root *root;
struct list_head splice;
@@ -9490,7 +9497,7 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr)
&fs_info->delalloc_roots);
spin_unlock(&fs_info->delalloc_root_lock);
- ret = start_delalloc_inodes(root, &nr, false);
+ ret = start_delalloc_inodes(root, &nr, false, in_reclaim_context);
btrfs_put_root(root);
if (ret < 0)
goto out;
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 703212ff50a5..dde49a791f3e 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -4951,7 +4951,7 @@ long btrfs_ioctl(struct file *file, unsigned int
case BTRFS_IOC_SYNC: {
int ret;
- ret = btrfs_start_delalloc_roots(fs_info, U64_MAX);
+ ret = btrfs_start_delalloc_roots(fs_info, U64_MAX, false);
if (ret)
return ret;
ret = btrfs_sync_fs(inode->i_sb, 1);
diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
index ab80896315be..b03e7891394e 100644
--- a/fs/btrfs/reflink.c
+++ b/fs/btrfs/reflink.c
@@ -89,6 +89,19 @@ static int copy_inline_to_page(struct btrfs_inode *inode,
if (ret)
goto out_unlock;
+ /*
+ * After dirtying the page our caller will need to start a transaction,
+ * and if we are low on metadata free space, that can cause flushing of
+ * delalloc for all inodes in order to get metadata space released.
+ * However we are holding the range locked for the whole duration of
+ * the clone/dedupe operation, so we may deadlock if that happens and no
+ * other task releases enough space. So mark this inode as not being
+ * possible to flush to avoid such deadlock. We will clear that flag
+ * when we finish cloning all extents, since a transaction is started
+ * after finding each extent to clone.
+ */
+ set_bit(BTRFS_INODE_NO_DELALLOC_FLUSH, &inode->runtime_flags);
+
if (comp_type == BTRFS_COMPRESS_NONE) {
char *map;
@@ -549,6 +562,8 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
out:
btrfs_free_path(path);
kvfree(buf);
+ clear_bit(BTRFS_INODE_NO_DELALLOC_FLUSH, &BTRFS_I(inode)->runtime_flags);
+
return ret;
}
diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index 64099565ab8f..67e55c5479b8 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -532,7 +532,7 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
loops = 0;
while ((delalloc_bytes || dio_bytes) && loops < 3) {
- btrfs_start_delalloc_roots(fs_info, items);
+ btrfs_start_delalloc_roots(fs_info, items, true);
loops++;
if (wait_ordered && !trans) {
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 9a664971569daf68254928149f580b4f5856d274 Mon Sep 17 00:00:00 2001
From: ethanwu <ethanwu(a)synology.com>
Date: Tue, 1 Dec 2020 17:25:12 +0800
Subject: [PATCH] btrfs: correctly calculate item size used when item key
collision happens
Item key collision is allowed for some item types, like dir item and
inode refs, but the overall item size is limited by the nodesize.
item size(ins_len) passed from btrfs_insert_empty_items to
btrfs_search_slot already contains size of btrfs_item.
When btrfs_search_slot reaches leaf, we'll see if we need to split leaf.
The check incorrectly reports that split leaf is required, because
it treats the space required by the newly inserted item as
btrfs_item + item data. But in item key collision case, only item data
is actually needed, the newly inserted item could merge into the existing
one. No new btrfs_item will be inserted.
And split_leaf return EOVERFLOW from following code:
if (extend && data_size + btrfs_item_size_nr(l, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(fs_info))
return -EOVERFLOW;
In most cases, when callers receive EOVERFLOW, they either return
this error or handle in different ways. For example, in normal dir item
creation the userspace will get errno EOVERFLOW; in inode ref case
INODE_EXTREF is used instead.
However, this is not the case for rename. To avoid the unrecoverable
situation in rename, btrfs_check_dir_item_collision is called in
early phase of rename. In this function, when item key collision is
detected leaf space is checked:
data_size = sizeof(*di) + name_len;
if (data_size + btrfs_item_size_nr(leaf, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root->fs_info))
the sizeof(struct btrfs_item) + btrfs_item_size_nr(leaf, slot) here
refers to existing item size, the condition here correctly calculates
the needed size for collision case rather than the wrong case above.
The consequence of inconsistent condition check between
btrfs_check_dir_item_collision and btrfs_search_slot when item key
collision happens is that we might pass check here but fail
later at btrfs_search_slot. Rename fails and volume is forced readonly
[436149.586170] ------------[ cut here ]------------
[436149.586173] BTRFS: Transaction aborted (error -75)
[436149.586196] WARNING: CPU: 0 PID: 16733 at fs/btrfs/inode.c:9870 btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586227] CPU: 0 PID: 16733 Comm: python Tainted: G D 4.18.0-rc5+ #1
[436149.586228] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/05/2016
[436149.586238] RIP: 0010:btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586254] RSP: 0018:ffffa327043a7ce0 EFLAGS: 00010286
[436149.586255] RAX: 0000000000000000 RBX: ffff8d8a17d13340 RCX: 0000000000000006
[436149.586256] RDX: 0000000000000007 RSI: 0000000000000096 RDI: ffff8d8a7fc164b0
[436149.586257] RBP: ffffa327043a7da0 R08: 0000000000000560 R09: 7265282064657472
[436149.586258] R10: 0000000000000000 R11: 6361736e61725420 R12: ffff8d8a0d4c8b08
[436149.586258] R13: ffff8d8a17d13340 R14: ffff8d8a33e0a540 R15: 00000000000001fe
[436149.586260] FS: 00007fa313933740(0000) GS:ffff8d8a7fc00000(0000) knlGS:0000000000000000
[436149.586261] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[436149.586262] CR2: 000055d8d9c9a720 CR3: 000000007aae0003 CR4: 00000000003606f0
[436149.586295] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[436149.586296] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[436149.586296] Call Trace:
[436149.586311] vfs_rename+0x383/0x920
[436149.586313] ? vfs_rename+0x383/0x920
[436149.586315] do_renameat2+0x4ca/0x590
[436149.586317] __x64_sys_rename+0x20/0x30
[436149.586324] do_syscall_64+0x5a/0x120
[436149.586330] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[436149.586332] RIP: 0033:0x7fa3133b1d37
[436149.586348] RSP: 002b:00007fffd3e43908 EFLAGS: 00000246 ORIG_RAX: 0000000000000052
[436149.586349] RAX: ffffffffffffffda RBX: 00007fa3133b1d30 RCX: 00007fa3133b1d37
[436149.586350] RDX: 000055d8da06b5e0 RSI: 000055d8da225d60 RDI: 000055d8da2c4da0
[436149.586351] RBP: 000055d8da2252f0 R08: 00007fa313782000 R09: 00000000000177e0
[436149.586351] R10: 000055d8da010680 R11: 0000000000000246 R12: 00007fa313840b00
Thanks to Hans van Kranenburg for information about crc32 hash collision
tools, I was able to reproduce the dir item collision with following
python script.
https://github.com/wutzuchieh/misc_tools/blob/master/crc32_forge.py Run
it under a btrfs volume will trigger the abort transaction. It simply
creates files and rename them to forged names that leads to
hash collision.
There are two ways to fix this. One is to simply revert the patch
878f2d2cb355 ("Btrfs: fix max dir item size calculation") to make the
condition consistent although that patch is correct about the size.
The other way is to handle the leaf space check correctly when
collision happens. I prefer the second one since it correct leaf
space check in collision case. This fix will not account
sizeof(struct btrfs_item) when the item already exists.
There are two places where ins_len doesn't contain
sizeof(struct btrfs_item), however.
1. extent-tree.c: lookup_inline_extent_backref
2. file-item.c: btrfs_csum_file_blocks
to make the logic of btrfs_search_slot more clear, we add a flag
search_for_extension in btrfs_path.
This flag indicates that ins_len passed to btrfs_search_slot doesn't
contain sizeof(struct btrfs_item). When key exists, btrfs_search_slot
will use the actual size needed to calculate the required leaf space.
CC: stable(a)vger.kernel.org # 4.4+
Reviewed-by: Filipe Manana <fdmanana(a)suse.com>
Signed-off-by: ethanwu <ethanwu(a)synology.com>
Signed-off-by: David Sterba <dsterba(a)suse.com>
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 07810891e204..cc89b63d65a4 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2555,8 +2555,14 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
* @p: Holds all btree nodes along the search path
* @root: The root node of the tree
* @key: The key we are looking for
- * @ins_len: Indicates purpose of search, for inserts it is 1, for
- * deletions it's -1. 0 for plain searches
+ * @ins_len: Indicates purpose of search:
+ * >0 for inserts it's size of item inserted (*)
+ * <0 for deletions
+ * 0 for plain searches, not modifying the tree
+ *
+ * (*) If size of item inserted doesn't include
+ * sizeof(struct btrfs_item), then p->search_for_extension must
+ * be set.
* @cow: boolean should CoW operations be performed. Must always be 1
* when modifying the tree.
*
@@ -2717,6 +2723,20 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
if (level == 0) {
p->slots[level] = slot;
+ /*
+ * Item key already exists. In this case, if we are
+ * allowed to insert the item (for example, in dir_item
+ * case, item key collision is allowed), it will be
+ * merged with the original item. Only the item size
+ * grows, no new btrfs item will be added. If
+ * search_for_extension is not set, ins_len already
+ * accounts the size btrfs_item, deduct it here so leaf
+ * space check will be correct.
+ */
+ if (ret == 0 && ins_len > 0 && !p->search_for_extension) {
+ ASSERT(ins_len >= sizeof(struct btrfs_item));
+ ins_len -= sizeof(struct btrfs_item);
+ }
if (ins_len > 0 &&
btrfs_leaf_free_space(b) < ins_len) {
if (write_lock_level < 1) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2674f24cf2e0..3935d297d198 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -368,6 +368,12 @@ struct btrfs_path {
unsigned int search_commit_root:1;
unsigned int need_commit_sem:1;
unsigned int skip_release_on_error:1;
+ /*
+ * Indicate that new item (btrfs_search_slot) is extending already
+ * existing item and ins_len contains only the data size and not item
+ * header (ie. sizeof(struct btrfs_item) is not included).
+ */
+ unsigned int search_for_extension:1;
};
#define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \
sizeof(struct btrfs_item))
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 56ea380f5a17..d79b8369e6aa 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -844,6 +844,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
want = extent_ref_type(parent, owner);
if (insert) {
extra_size = btrfs_extent_inline_ref_size(want);
+ path->search_for_extension = 1;
path->keep_locks = 1;
} else
extra_size = -1;
@@ -996,6 +997,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
out:
if (insert) {
path->keep_locks = 0;
+ path->search_for_extension = 0;
btrfs_unlock_up_safe(path, 1);
}
return err;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 1545c22ef280..6ccfc019ad90 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -1016,8 +1016,10 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
+ path->search_for_extension = 1;
ret = btrfs_search_slot(trans, root, &file_key, path,
csum_size, 1);
+ path->search_for_extension = 0;
if (ret < 0)
goto out;
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 9a664971569daf68254928149f580b4f5856d274 Mon Sep 17 00:00:00 2001
From: ethanwu <ethanwu(a)synology.com>
Date: Tue, 1 Dec 2020 17:25:12 +0800
Subject: [PATCH] btrfs: correctly calculate item size used when item key
collision happens
Item key collision is allowed for some item types, like dir item and
inode refs, but the overall item size is limited by the nodesize.
item size(ins_len) passed from btrfs_insert_empty_items to
btrfs_search_slot already contains size of btrfs_item.
When btrfs_search_slot reaches leaf, we'll see if we need to split leaf.
The check incorrectly reports that split leaf is required, because
it treats the space required by the newly inserted item as
btrfs_item + item data. But in item key collision case, only item data
is actually needed, the newly inserted item could merge into the existing
one. No new btrfs_item will be inserted.
And split_leaf return EOVERFLOW from following code:
if (extend && data_size + btrfs_item_size_nr(l, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(fs_info))
return -EOVERFLOW;
In most cases, when callers receive EOVERFLOW, they either return
this error or handle in different ways. For example, in normal dir item
creation the userspace will get errno EOVERFLOW; in inode ref case
INODE_EXTREF is used instead.
However, this is not the case for rename. To avoid the unrecoverable
situation in rename, btrfs_check_dir_item_collision is called in
early phase of rename. In this function, when item key collision is
detected leaf space is checked:
data_size = sizeof(*di) + name_len;
if (data_size + btrfs_item_size_nr(leaf, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root->fs_info))
the sizeof(struct btrfs_item) + btrfs_item_size_nr(leaf, slot) here
refers to existing item size, the condition here correctly calculates
the needed size for collision case rather than the wrong case above.
The consequence of inconsistent condition check between
btrfs_check_dir_item_collision and btrfs_search_slot when item key
collision happens is that we might pass check here but fail
later at btrfs_search_slot. Rename fails and volume is forced readonly
[436149.586170] ------------[ cut here ]------------
[436149.586173] BTRFS: Transaction aborted (error -75)
[436149.586196] WARNING: CPU: 0 PID: 16733 at fs/btrfs/inode.c:9870 btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586227] CPU: 0 PID: 16733 Comm: python Tainted: G D 4.18.0-rc5+ #1
[436149.586228] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/05/2016
[436149.586238] RIP: 0010:btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586254] RSP: 0018:ffffa327043a7ce0 EFLAGS: 00010286
[436149.586255] RAX: 0000000000000000 RBX: ffff8d8a17d13340 RCX: 0000000000000006
[436149.586256] RDX: 0000000000000007 RSI: 0000000000000096 RDI: ffff8d8a7fc164b0
[436149.586257] RBP: ffffa327043a7da0 R08: 0000000000000560 R09: 7265282064657472
[436149.586258] R10: 0000000000000000 R11: 6361736e61725420 R12: ffff8d8a0d4c8b08
[436149.586258] R13: ffff8d8a17d13340 R14: ffff8d8a33e0a540 R15: 00000000000001fe
[436149.586260] FS: 00007fa313933740(0000) GS:ffff8d8a7fc00000(0000) knlGS:0000000000000000
[436149.586261] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[436149.586262] CR2: 000055d8d9c9a720 CR3: 000000007aae0003 CR4: 00000000003606f0
[436149.586295] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[436149.586296] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[436149.586296] Call Trace:
[436149.586311] vfs_rename+0x383/0x920
[436149.586313] ? vfs_rename+0x383/0x920
[436149.586315] do_renameat2+0x4ca/0x590
[436149.586317] __x64_sys_rename+0x20/0x30
[436149.586324] do_syscall_64+0x5a/0x120
[436149.586330] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[436149.586332] RIP: 0033:0x7fa3133b1d37
[436149.586348] RSP: 002b:00007fffd3e43908 EFLAGS: 00000246 ORIG_RAX: 0000000000000052
[436149.586349] RAX: ffffffffffffffda RBX: 00007fa3133b1d30 RCX: 00007fa3133b1d37
[436149.586350] RDX: 000055d8da06b5e0 RSI: 000055d8da225d60 RDI: 000055d8da2c4da0
[436149.586351] RBP: 000055d8da2252f0 R08: 00007fa313782000 R09: 00000000000177e0
[436149.586351] R10: 000055d8da010680 R11: 0000000000000246 R12: 00007fa313840b00
Thanks to Hans van Kranenburg for information about crc32 hash collision
tools, I was able to reproduce the dir item collision with following
python script.
https://github.com/wutzuchieh/misc_tools/blob/master/crc32_forge.py Run
it under a btrfs volume will trigger the abort transaction. It simply
creates files and rename them to forged names that leads to
hash collision.
There are two ways to fix this. One is to simply revert the patch
878f2d2cb355 ("Btrfs: fix max dir item size calculation") to make the
condition consistent although that patch is correct about the size.
The other way is to handle the leaf space check correctly when
collision happens. I prefer the second one since it correct leaf
space check in collision case. This fix will not account
sizeof(struct btrfs_item) when the item already exists.
There are two places where ins_len doesn't contain
sizeof(struct btrfs_item), however.
1. extent-tree.c: lookup_inline_extent_backref
2. file-item.c: btrfs_csum_file_blocks
to make the logic of btrfs_search_slot more clear, we add a flag
search_for_extension in btrfs_path.
This flag indicates that ins_len passed to btrfs_search_slot doesn't
contain sizeof(struct btrfs_item). When key exists, btrfs_search_slot
will use the actual size needed to calculate the required leaf space.
CC: stable(a)vger.kernel.org # 4.4+
Reviewed-by: Filipe Manana <fdmanana(a)suse.com>
Signed-off-by: ethanwu <ethanwu(a)synology.com>
Signed-off-by: David Sterba <dsterba(a)suse.com>
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 07810891e204..cc89b63d65a4 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2555,8 +2555,14 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
* @p: Holds all btree nodes along the search path
* @root: The root node of the tree
* @key: The key we are looking for
- * @ins_len: Indicates purpose of search, for inserts it is 1, for
- * deletions it's -1. 0 for plain searches
+ * @ins_len: Indicates purpose of search:
+ * >0 for inserts it's size of item inserted (*)
+ * <0 for deletions
+ * 0 for plain searches, not modifying the tree
+ *
+ * (*) If size of item inserted doesn't include
+ * sizeof(struct btrfs_item), then p->search_for_extension must
+ * be set.
* @cow: boolean should CoW operations be performed. Must always be 1
* when modifying the tree.
*
@@ -2717,6 +2723,20 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
if (level == 0) {
p->slots[level] = slot;
+ /*
+ * Item key already exists. In this case, if we are
+ * allowed to insert the item (for example, in dir_item
+ * case, item key collision is allowed), it will be
+ * merged with the original item. Only the item size
+ * grows, no new btrfs item will be added. If
+ * search_for_extension is not set, ins_len already
+ * accounts the size btrfs_item, deduct it here so leaf
+ * space check will be correct.
+ */
+ if (ret == 0 && ins_len > 0 && !p->search_for_extension) {
+ ASSERT(ins_len >= sizeof(struct btrfs_item));
+ ins_len -= sizeof(struct btrfs_item);
+ }
if (ins_len > 0 &&
btrfs_leaf_free_space(b) < ins_len) {
if (write_lock_level < 1) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2674f24cf2e0..3935d297d198 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -368,6 +368,12 @@ struct btrfs_path {
unsigned int search_commit_root:1;
unsigned int need_commit_sem:1;
unsigned int skip_release_on_error:1;
+ /*
+ * Indicate that new item (btrfs_search_slot) is extending already
+ * existing item and ins_len contains only the data size and not item
+ * header (ie. sizeof(struct btrfs_item) is not included).
+ */
+ unsigned int search_for_extension:1;
};
#define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \
sizeof(struct btrfs_item))
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 56ea380f5a17..d79b8369e6aa 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -844,6 +844,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
want = extent_ref_type(parent, owner);
if (insert) {
extra_size = btrfs_extent_inline_ref_size(want);
+ path->search_for_extension = 1;
path->keep_locks = 1;
} else
extra_size = -1;
@@ -996,6 +997,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
out:
if (insert) {
path->keep_locks = 0;
+ path->search_for_extension = 0;
btrfs_unlock_up_safe(path, 1);
}
return err;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 1545c22ef280..6ccfc019ad90 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -1016,8 +1016,10 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
+ path->search_for_extension = 1;
ret = btrfs_search_slot(trans, root, &file_key, path,
csum_size, 1);
+ path->search_for_extension = 0;
if (ret < 0)
goto out;
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 9a664971569daf68254928149f580b4f5856d274 Mon Sep 17 00:00:00 2001
From: ethanwu <ethanwu(a)synology.com>
Date: Tue, 1 Dec 2020 17:25:12 +0800
Subject: [PATCH] btrfs: correctly calculate item size used when item key
collision happens
Item key collision is allowed for some item types, like dir item and
inode refs, but the overall item size is limited by the nodesize.
item size(ins_len) passed from btrfs_insert_empty_items to
btrfs_search_slot already contains size of btrfs_item.
When btrfs_search_slot reaches leaf, we'll see if we need to split leaf.
The check incorrectly reports that split leaf is required, because
it treats the space required by the newly inserted item as
btrfs_item + item data. But in item key collision case, only item data
is actually needed, the newly inserted item could merge into the existing
one. No new btrfs_item will be inserted.
And split_leaf return EOVERFLOW from following code:
if (extend && data_size + btrfs_item_size_nr(l, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(fs_info))
return -EOVERFLOW;
In most cases, when callers receive EOVERFLOW, they either return
this error or handle in different ways. For example, in normal dir item
creation the userspace will get errno EOVERFLOW; in inode ref case
INODE_EXTREF is used instead.
However, this is not the case for rename. To avoid the unrecoverable
situation in rename, btrfs_check_dir_item_collision is called in
early phase of rename. In this function, when item key collision is
detected leaf space is checked:
data_size = sizeof(*di) + name_len;
if (data_size + btrfs_item_size_nr(leaf, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root->fs_info))
the sizeof(struct btrfs_item) + btrfs_item_size_nr(leaf, slot) here
refers to existing item size, the condition here correctly calculates
the needed size for collision case rather than the wrong case above.
The consequence of inconsistent condition check between
btrfs_check_dir_item_collision and btrfs_search_slot when item key
collision happens is that we might pass check here but fail
later at btrfs_search_slot. Rename fails and volume is forced readonly
[436149.586170] ------------[ cut here ]------------
[436149.586173] BTRFS: Transaction aborted (error -75)
[436149.586196] WARNING: CPU: 0 PID: 16733 at fs/btrfs/inode.c:9870 btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586227] CPU: 0 PID: 16733 Comm: python Tainted: G D 4.18.0-rc5+ #1
[436149.586228] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/05/2016
[436149.586238] RIP: 0010:btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586254] RSP: 0018:ffffa327043a7ce0 EFLAGS: 00010286
[436149.586255] RAX: 0000000000000000 RBX: ffff8d8a17d13340 RCX: 0000000000000006
[436149.586256] RDX: 0000000000000007 RSI: 0000000000000096 RDI: ffff8d8a7fc164b0
[436149.586257] RBP: ffffa327043a7da0 R08: 0000000000000560 R09: 7265282064657472
[436149.586258] R10: 0000000000000000 R11: 6361736e61725420 R12: ffff8d8a0d4c8b08
[436149.586258] R13: ffff8d8a17d13340 R14: ffff8d8a33e0a540 R15: 00000000000001fe
[436149.586260] FS: 00007fa313933740(0000) GS:ffff8d8a7fc00000(0000) knlGS:0000000000000000
[436149.586261] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[436149.586262] CR2: 000055d8d9c9a720 CR3: 000000007aae0003 CR4: 00000000003606f0
[436149.586295] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[436149.586296] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[436149.586296] Call Trace:
[436149.586311] vfs_rename+0x383/0x920
[436149.586313] ? vfs_rename+0x383/0x920
[436149.586315] do_renameat2+0x4ca/0x590
[436149.586317] __x64_sys_rename+0x20/0x30
[436149.586324] do_syscall_64+0x5a/0x120
[436149.586330] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[436149.586332] RIP: 0033:0x7fa3133b1d37
[436149.586348] RSP: 002b:00007fffd3e43908 EFLAGS: 00000246 ORIG_RAX: 0000000000000052
[436149.586349] RAX: ffffffffffffffda RBX: 00007fa3133b1d30 RCX: 00007fa3133b1d37
[436149.586350] RDX: 000055d8da06b5e0 RSI: 000055d8da225d60 RDI: 000055d8da2c4da0
[436149.586351] RBP: 000055d8da2252f0 R08: 00007fa313782000 R09: 00000000000177e0
[436149.586351] R10: 000055d8da010680 R11: 0000000000000246 R12: 00007fa313840b00
Thanks to Hans van Kranenburg for information about crc32 hash collision
tools, I was able to reproduce the dir item collision with following
python script.
https://github.com/wutzuchieh/misc_tools/blob/master/crc32_forge.py Run
it under a btrfs volume will trigger the abort transaction. It simply
creates files and rename them to forged names that leads to
hash collision.
There are two ways to fix this. One is to simply revert the patch
878f2d2cb355 ("Btrfs: fix max dir item size calculation") to make the
condition consistent although that patch is correct about the size.
The other way is to handle the leaf space check correctly when
collision happens. I prefer the second one since it correct leaf
space check in collision case. This fix will not account
sizeof(struct btrfs_item) when the item already exists.
There are two places where ins_len doesn't contain
sizeof(struct btrfs_item), however.
1. extent-tree.c: lookup_inline_extent_backref
2. file-item.c: btrfs_csum_file_blocks
to make the logic of btrfs_search_slot more clear, we add a flag
search_for_extension in btrfs_path.
This flag indicates that ins_len passed to btrfs_search_slot doesn't
contain sizeof(struct btrfs_item). When key exists, btrfs_search_slot
will use the actual size needed to calculate the required leaf space.
CC: stable(a)vger.kernel.org # 4.4+
Reviewed-by: Filipe Manana <fdmanana(a)suse.com>
Signed-off-by: ethanwu <ethanwu(a)synology.com>
Signed-off-by: David Sterba <dsterba(a)suse.com>
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 07810891e204..cc89b63d65a4 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2555,8 +2555,14 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
* @p: Holds all btree nodes along the search path
* @root: The root node of the tree
* @key: The key we are looking for
- * @ins_len: Indicates purpose of search, for inserts it is 1, for
- * deletions it's -1. 0 for plain searches
+ * @ins_len: Indicates purpose of search:
+ * >0 for inserts it's size of item inserted (*)
+ * <0 for deletions
+ * 0 for plain searches, not modifying the tree
+ *
+ * (*) If size of item inserted doesn't include
+ * sizeof(struct btrfs_item), then p->search_for_extension must
+ * be set.
* @cow: boolean should CoW operations be performed. Must always be 1
* when modifying the tree.
*
@@ -2717,6 +2723,20 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
if (level == 0) {
p->slots[level] = slot;
+ /*
+ * Item key already exists. In this case, if we are
+ * allowed to insert the item (for example, in dir_item
+ * case, item key collision is allowed), it will be
+ * merged with the original item. Only the item size
+ * grows, no new btrfs item will be added. If
+ * search_for_extension is not set, ins_len already
+ * accounts the size btrfs_item, deduct it here so leaf
+ * space check will be correct.
+ */
+ if (ret == 0 && ins_len > 0 && !p->search_for_extension) {
+ ASSERT(ins_len >= sizeof(struct btrfs_item));
+ ins_len -= sizeof(struct btrfs_item);
+ }
if (ins_len > 0 &&
btrfs_leaf_free_space(b) < ins_len) {
if (write_lock_level < 1) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2674f24cf2e0..3935d297d198 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -368,6 +368,12 @@ struct btrfs_path {
unsigned int search_commit_root:1;
unsigned int need_commit_sem:1;
unsigned int skip_release_on_error:1;
+ /*
+ * Indicate that new item (btrfs_search_slot) is extending already
+ * existing item and ins_len contains only the data size and not item
+ * header (ie. sizeof(struct btrfs_item) is not included).
+ */
+ unsigned int search_for_extension:1;
};
#define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \
sizeof(struct btrfs_item))
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 56ea380f5a17..d79b8369e6aa 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -844,6 +844,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
want = extent_ref_type(parent, owner);
if (insert) {
extra_size = btrfs_extent_inline_ref_size(want);
+ path->search_for_extension = 1;
path->keep_locks = 1;
} else
extra_size = -1;
@@ -996,6 +997,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
out:
if (insert) {
path->keep_locks = 0;
+ path->search_for_extension = 0;
btrfs_unlock_up_safe(path, 1);
}
return err;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 1545c22ef280..6ccfc019ad90 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -1016,8 +1016,10 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
+ path->search_for_extension = 1;
ret = btrfs_search_slot(trans, root, &file_key, path,
csum_size, 1);
+ path->search_for_extension = 0;
if (ret < 0)
goto out;
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 9a664971569daf68254928149f580b4f5856d274 Mon Sep 17 00:00:00 2001
From: ethanwu <ethanwu(a)synology.com>
Date: Tue, 1 Dec 2020 17:25:12 +0800
Subject: [PATCH] btrfs: correctly calculate item size used when item key
collision happens
Item key collision is allowed for some item types, like dir item and
inode refs, but the overall item size is limited by the nodesize.
item size(ins_len) passed from btrfs_insert_empty_items to
btrfs_search_slot already contains size of btrfs_item.
When btrfs_search_slot reaches leaf, we'll see if we need to split leaf.
The check incorrectly reports that split leaf is required, because
it treats the space required by the newly inserted item as
btrfs_item + item data. But in item key collision case, only item data
is actually needed, the newly inserted item could merge into the existing
one. No new btrfs_item will be inserted.
And split_leaf return EOVERFLOW from following code:
if (extend && data_size + btrfs_item_size_nr(l, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(fs_info))
return -EOVERFLOW;
In most cases, when callers receive EOVERFLOW, they either return
this error or handle in different ways. For example, in normal dir item
creation the userspace will get errno EOVERFLOW; in inode ref case
INODE_EXTREF is used instead.
However, this is not the case for rename. To avoid the unrecoverable
situation in rename, btrfs_check_dir_item_collision is called in
early phase of rename. In this function, when item key collision is
detected leaf space is checked:
data_size = sizeof(*di) + name_len;
if (data_size + btrfs_item_size_nr(leaf, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root->fs_info))
the sizeof(struct btrfs_item) + btrfs_item_size_nr(leaf, slot) here
refers to existing item size, the condition here correctly calculates
the needed size for collision case rather than the wrong case above.
The consequence of inconsistent condition check between
btrfs_check_dir_item_collision and btrfs_search_slot when item key
collision happens is that we might pass check here but fail
later at btrfs_search_slot. Rename fails and volume is forced readonly
[436149.586170] ------------[ cut here ]------------
[436149.586173] BTRFS: Transaction aborted (error -75)
[436149.586196] WARNING: CPU: 0 PID: 16733 at fs/btrfs/inode.c:9870 btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586227] CPU: 0 PID: 16733 Comm: python Tainted: G D 4.18.0-rc5+ #1
[436149.586228] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/05/2016
[436149.586238] RIP: 0010:btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586254] RSP: 0018:ffffa327043a7ce0 EFLAGS: 00010286
[436149.586255] RAX: 0000000000000000 RBX: ffff8d8a17d13340 RCX: 0000000000000006
[436149.586256] RDX: 0000000000000007 RSI: 0000000000000096 RDI: ffff8d8a7fc164b0
[436149.586257] RBP: ffffa327043a7da0 R08: 0000000000000560 R09: 7265282064657472
[436149.586258] R10: 0000000000000000 R11: 6361736e61725420 R12: ffff8d8a0d4c8b08
[436149.586258] R13: ffff8d8a17d13340 R14: ffff8d8a33e0a540 R15: 00000000000001fe
[436149.586260] FS: 00007fa313933740(0000) GS:ffff8d8a7fc00000(0000) knlGS:0000000000000000
[436149.586261] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[436149.586262] CR2: 000055d8d9c9a720 CR3: 000000007aae0003 CR4: 00000000003606f0
[436149.586295] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[436149.586296] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[436149.586296] Call Trace:
[436149.586311] vfs_rename+0x383/0x920
[436149.586313] ? vfs_rename+0x383/0x920
[436149.586315] do_renameat2+0x4ca/0x590
[436149.586317] __x64_sys_rename+0x20/0x30
[436149.586324] do_syscall_64+0x5a/0x120
[436149.586330] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[436149.586332] RIP: 0033:0x7fa3133b1d37
[436149.586348] RSP: 002b:00007fffd3e43908 EFLAGS: 00000246 ORIG_RAX: 0000000000000052
[436149.586349] RAX: ffffffffffffffda RBX: 00007fa3133b1d30 RCX: 00007fa3133b1d37
[436149.586350] RDX: 000055d8da06b5e0 RSI: 000055d8da225d60 RDI: 000055d8da2c4da0
[436149.586351] RBP: 000055d8da2252f0 R08: 00007fa313782000 R09: 00000000000177e0
[436149.586351] R10: 000055d8da010680 R11: 0000000000000246 R12: 00007fa313840b00
Thanks to Hans van Kranenburg for information about crc32 hash collision
tools, I was able to reproduce the dir item collision with following
python script.
https://github.com/wutzuchieh/misc_tools/blob/master/crc32_forge.py Run
it under a btrfs volume will trigger the abort transaction. It simply
creates files and rename them to forged names that leads to
hash collision.
There are two ways to fix this. One is to simply revert the patch
878f2d2cb355 ("Btrfs: fix max dir item size calculation") to make the
condition consistent although that patch is correct about the size.
The other way is to handle the leaf space check correctly when
collision happens. I prefer the second one since it correct leaf
space check in collision case. This fix will not account
sizeof(struct btrfs_item) when the item already exists.
There are two places where ins_len doesn't contain
sizeof(struct btrfs_item), however.
1. extent-tree.c: lookup_inline_extent_backref
2. file-item.c: btrfs_csum_file_blocks
to make the logic of btrfs_search_slot more clear, we add a flag
search_for_extension in btrfs_path.
This flag indicates that ins_len passed to btrfs_search_slot doesn't
contain sizeof(struct btrfs_item). When key exists, btrfs_search_slot
will use the actual size needed to calculate the required leaf space.
CC: stable(a)vger.kernel.org # 4.4+
Reviewed-by: Filipe Manana <fdmanana(a)suse.com>
Signed-off-by: ethanwu <ethanwu(a)synology.com>
Signed-off-by: David Sterba <dsterba(a)suse.com>
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 07810891e204..cc89b63d65a4 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2555,8 +2555,14 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
* @p: Holds all btree nodes along the search path
* @root: The root node of the tree
* @key: The key we are looking for
- * @ins_len: Indicates purpose of search, for inserts it is 1, for
- * deletions it's -1. 0 for plain searches
+ * @ins_len: Indicates purpose of search:
+ * >0 for inserts it's size of item inserted (*)
+ * <0 for deletions
+ * 0 for plain searches, not modifying the tree
+ *
+ * (*) If size of item inserted doesn't include
+ * sizeof(struct btrfs_item), then p->search_for_extension must
+ * be set.
* @cow: boolean should CoW operations be performed. Must always be 1
* when modifying the tree.
*
@@ -2717,6 +2723,20 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
if (level == 0) {
p->slots[level] = slot;
+ /*
+ * Item key already exists. In this case, if we are
+ * allowed to insert the item (for example, in dir_item
+ * case, item key collision is allowed), it will be
+ * merged with the original item. Only the item size
+ * grows, no new btrfs item will be added. If
+ * search_for_extension is not set, ins_len already
+ * accounts the size btrfs_item, deduct it here so leaf
+ * space check will be correct.
+ */
+ if (ret == 0 && ins_len > 0 && !p->search_for_extension) {
+ ASSERT(ins_len >= sizeof(struct btrfs_item));
+ ins_len -= sizeof(struct btrfs_item);
+ }
if (ins_len > 0 &&
btrfs_leaf_free_space(b) < ins_len) {
if (write_lock_level < 1) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2674f24cf2e0..3935d297d198 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -368,6 +368,12 @@ struct btrfs_path {
unsigned int search_commit_root:1;
unsigned int need_commit_sem:1;
unsigned int skip_release_on_error:1;
+ /*
+ * Indicate that new item (btrfs_search_slot) is extending already
+ * existing item and ins_len contains only the data size and not item
+ * header (ie. sizeof(struct btrfs_item) is not included).
+ */
+ unsigned int search_for_extension:1;
};
#define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \
sizeof(struct btrfs_item))
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 56ea380f5a17..d79b8369e6aa 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -844,6 +844,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
want = extent_ref_type(parent, owner);
if (insert) {
extra_size = btrfs_extent_inline_ref_size(want);
+ path->search_for_extension = 1;
path->keep_locks = 1;
} else
extra_size = -1;
@@ -996,6 +997,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
out:
if (insert) {
path->keep_locks = 0;
+ path->search_for_extension = 0;
btrfs_unlock_up_safe(path, 1);
}
return err;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 1545c22ef280..6ccfc019ad90 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -1016,8 +1016,10 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
+ path->search_for_extension = 1;
ret = btrfs_search_slot(trans, root, &file_key, path,
csum_size, 1);
+ path->search_for_extension = 0;
if (ret < 0)
goto out;
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 9a664971569daf68254928149f580b4f5856d274 Mon Sep 17 00:00:00 2001
From: ethanwu <ethanwu(a)synology.com>
Date: Tue, 1 Dec 2020 17:25:12 +0800
Subject: [PATCH] btrfs: correctly calculate item size used when item key
collision happens
Item key collision is allowed for some item types, like dir item and
inode refs, but the overall item size is limited by the nodesize.
item size(ins_len) passed from btrfs_insert_empty_items to
btrfs_search_slot already contains size of btrfs_item.
When btrfs_search_slot reaches leaf, we'll see if we need to split leaf.
The check incorrectly reports that split leaf is required, because
it treats the space required by the newly inserted item as
btrfs_item + item data. But in item key collision case, only item data
is actually needed, the newly inserted item could merge into the existing
one. No new btrfs_item will be inserted.
And split_leaf return EOVERFLOW from following code:
if (extend && data_size + btrfs_item_size_nr(l, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(fs_info))
return -EOVERFLOW;
In most cases, when callers receive EOVERFLOW, they either return
this error or handle in different ways. For example, in normal dir item
creation the userspace will get errno EOVERFLOW; in inode ref case
INODE_EXTREF is used instead.
However, this is not the case for rename. To avoid the unrecoverable
situation in rename, btrfs_check_dir_item_collision is called in
early phase of rename. In this function, when item key collision is
detected leaf space is checked:
data_size = sizeof(*di) + name_len;
if (data_size + btrfs_item_size_nr(leaf, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root->fs_info))
the sizeof(struct btrfs_item) + btrfs_item_size_nr(leaf, slot) here
refers to existing item size, the condition here correctly calculates
the needed size for collision case rather than the wrong case above.
The consequence of inconsistent condition check between
btrfs_check_dir_item_collision and btrfs_search_slot when item key
collision happens is that we might pass check here but fail
later at btrfs_search_slot. Rename fails and volume is forced readonly
[436149.586170] ------------[ cut here ]------------
[436149.586173] BTRFS: Transaction aborted (error -75)
[436149.586196] WARNING: CPU: 0 PID: 16733 at fs/btrfs/inode.c:9870 btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586227] CPU: 0 PID: 16733 Comm: python Tainted: G D 4.18.0-rc5+ #1
[436149.586228] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/05/2016
[436149.586238] RIP: 0010:btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586254] RSP: 0018:ffffa327043a7ce0 EFLAGS: 00010286
[436149.586255] RAX: 0000000000000000 RBX: ffff8d8a17d13340 RCX: 0000000000000006
[436149.586256] RDX: 0000000000000007 RSI: 0000000000000096 RDI: ffff8d8a7fc164b0
[436149.586257] RBP: ffffa327043a7da0 R08: 0000000000000560 R09: 7265282064657472
[436149.586258] R10: 0000000000000000 R11: 6361736e61725420 R12: ffff8d8a0d4c8b08
[436149.586258] R13: ffff8d8a17d13340 R14: ffff8d8a33e0a540 R15: 00000000000001fe
[436149.586260] FS: 00007fa313933740(0000) GS:ffff8d8a7fc00000(0000) knlGS:0000000000000000
[436149.586261] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[436149.586262] CR2: 000055d8d9c9a720 CR3: 000000007aae0003 CR4: 00000000003606f0
[436149.586295] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[436149.586296] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[436149.586296] Call Trace:
[436149.586311] vfs_rename+0x383/0x920
[436149.586313] ? vfs_rename+0x383/0x920
[436149.586315] do_renameat2+0x4ca/0x590
[436149.586317] __x64_sys_rename+0x20/0x30
[436149.586324] do_syscall_64+0x5a/0x120
[436149.586330] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[436149.586332] RIP: 0033:0x7fa3133b1d37
[436149.586348] RSP: 002b:00007fffd3e43908 EFLAGS: 00000246 ORIG_RAX: 0000000000000052
[436149.586349] RAX: ffffffffffffffda RBX: 00007fa3133b1d30 RCX: 00007fa3133b1d37
[436149.586350] RDX: 000055d8da06b5e0 RSI: 000055d8da225d60 RDI: 000055d8da2c4da0
[436149.586351] RBP: 000055d8da2252f0 R08: 00007fa313782000 R09: 00000000000177e0
[436149.586351] R10: 000055d8da010680 R11: 0000000000000246 R12: 00007fa313840b00
Thanks to Hans van Kranenburg for information about crc32 hash collision
tools, I was able to reproduce the dir item collision with following
python script.
https://github.com/wutzuchieh/misc_tools/blob/master/crc32_forge.py Run
it under a btrfs volume will trigger the abort transaction. It simply
creates files and rename them to forged names that leads to
hash collision.
There are two ways to fix this. One is to simply revert the patch
878f2d2cb355 ("Btrfs: fix max dir item size calculation") to make the
condition consistent although that patch is correct about the size.
The other way is to handle the leaf space check correctly when
collision happens. I prefer the second one since it correct leaf
space check in collision case. This fix will not account
sizeof(struct btrfs_item) when the item already exists.
There are two places where ins_len doesn't contain
sizeof(struct btrfs_item), however.
1. extent-tree.c: lookup_inline_extent_backref
2. file-item.c: btrfs_csum_file_blocks
to make the logic of btrfs_search_slot more clear, we add a flag
search_for_extension in btrfs_path.
This flag indicates that ins_len passed to btrfs_search_slot doesn't
contain sizeof(struct btrfs_item). When key exists, btrfs_search_slot
will use the actual size needed to calculate the required leaf space.
CC: stable(a)vger.kernel.org # 4.4+
Reviewed-by: Filipe Manana <fdmanana(a)suse.com>
Signed-off-by: ethanwu <ethanwu(a)synology.com>
Signed-off-by: David Sterba <dsterba(a)suse.com>
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 07810891e204..cc89b63d65a4 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2555,8 +2555,14 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
* @p: Holds all btree nodes along the search path
* @root: The root node of the tree
* @key: The key we are looking for
- * @ins_len: Indicates purpose of search, for inserts it is 1, for
- * deletions it's -1. 0 for plain searches
+ * @ins_len: Indicates purpose of search:
+ * >0 for inserts it's size of item inserted (*)
+ * <0 for deletions
+ * 0 for plain searches, not modifying the tree
+ *
+ * (*) If size of item inserted doesn't include
+ * sizeof(struct btrfs_item), then p->search_for_extension must
+ * be set.
* @cow: boolean should CoW operations be performed. Must always be 1
* when modifying the tree.
*
@@ -2717,6 +2723,20 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
if (level == 0) {
p->slots[level] = slot;
+ /*
+ * Item key already exists. In this case, if we are
+ * allowed to insert the item (for example, in dir_item
+ * case, item key collision is allowed), it will be
+ * merged with the original item. Only the item size
+ * grows, no new btrfs item will be added. If
+ * search_for_extension is not set, ins_len already
+ * accounts the size btrfs_item, deduct it here so leaf
+ * space check will be correct.
+ */
+ if (ret == 0 && ins_len > 0 && !p->search_for_extension) {
+ ASSERT(ins_len >= sizeof(struct btrfs_item));
+ ins_len -= sizeof(struct btrfs_item);
+ }
if (ins_len > 0 &&
btrfs_leaf_free_space(b) < ins_len) {
if (write_lock_level < 1) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2674f24cf2e0..3935d297d198 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -368,6 +368,12 @@ struct btrfs_path {
unsigned int search_commit_root:1;
unsigned int need_commit_sem:1;
unsigned int skip_release_on_error:1;
+ /*
+ * Indicate that new item (btrfs_search_slot) is extending already
+ * existing item and ins_len contains only the data size and not item
+ * header (ie. sizeof(struct btrfs_item) is not included).
+ */
+ unsigned int search_for_extension:1;
};
#define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \
sizeof(struct btrfs_item))
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 56ea380f5a17..d79b8369e6aa 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -844,6 +844,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
want = extent_ref_type(parent, owner);
if (insert) {
extra_size = btrfs_extent_inline_ref_size(want);
+ path->search_for_extension = 1;
path->keep_locks = 1;
} else
extra_size = -1;
@@ -996,6 +997,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
out:
if (insert) {
path->keep_locks = 0;
+ path->search_for_extension = 0;
btrfs_unlock_up_safe(path, 1);
}
return err;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 1545c22ef280..6ccfc019ad90 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -1016,8 +1016,10 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
+ path->search_for_extension = 1;
ret = btrfs_search_slot(trans, root, &file_key, path,
csum_size, 1);
+ path->search_for_extension = 0;
if (ret < 0)
goto out;
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 9a664971569daf68254928149f580b4f5856d274 Mon Sep 17 00:00:00 2001
From: ethanwu <ethanwu(a)synology.com>
Date: Tue, 1 Dec 2020 17:25:12 +0800
Subject: [PATCH] btrfs: correctly calculate item size used when item key
collision happens
Item key collision is allowed for some item types, like dir item and
inode refs, but the overall item size is limited by the nodesize.
item size(ins_len) passed from btrfs_insert_empty_items to
btrfs_search_slot already contains size of btrfs_item.
When btrfs_search_slot reaches leaf, we'll see if we need to split leaf.
The check incorrectly reports that split leaf is required, because
it treats the space required by the newly inserted item as
btrfs_item + item data. But in item key collision case, only item data
is actually needed, the newly inserted item could merge into the existing
one. No new btrfs_item will be inserted.
And split_leaf return EOVERFLOW from following code:
if (extend && data_size + btrfs_item_size_nr(l, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(fs_info))
return -EOVERFLOW;
In most cases, when callers receive EOVERFLOW, they either return
this error or handle in different ways. For example, in normal dir item
creation the userspace will get errno EOVERFLOW; in inode ref case
INODE_EXTREF is used instead.
However, this is not the case for rename. To avoid the unrecoverable
situation in rename, btrfs_check_dir_item_collision is called in
early phase of rename. In this function, when item key collision is
detected leaf space is checked:
data_size = sizeof(*di) + name_len;
if (data_size + btrfs_item_size_nr(leaf, slot) +
sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root->fs_info))
the sizeof(struct btrfs_item) + btrfs_item_size_nr(leaf, slot) here
refers to existing item size, the condition here correctly calculates
the needed size for collision case rather than the wrong case above.
The consequence of inconsistent condition check between
btrfs_check_dir_item_collision and btrfs_search_slot when item key
collision happens is that we might pass check here but fail
later at btrfs_search_slot. Rename fails and volume is forced readonly
[436149.586170] ------------[ cut here ]------------
[436149.586173] BTRFS: Transaction aborted (error -75)
[436149.586196] WARNING: CPU: 0 PID: 16733 at fs/btrfs/inode.c:9870 btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586227] CPU: 0 PID: 16733 Comm: python Tainted: G D 4.18.0-rc5+ #1
[436149.586228] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/05/2016
[436149.586238] RIP: 0010:btrfs_rename2+0x1938/0x1b70 [btrfs]
[436149.586254] RSP: 0018:ffffa327043a7ce0 EFLAGS: 00010286
[436149.586255] RAX: 0000000000000000 RBX: ffff8d8a17d13340 RCX: 0000000000000006
[436149.586256] RDX: 0000000000000007 RSI: 0000000000000096 RDI: ffff8d8a7fc164b0
[436149.586257] RBP: ffffa327043a7da0 R08: 0000000000000560 R09: 7265282064657472
[436149.586258] R10: 0000000000000000 R11: 6361736e61725420 R12: ffff8d8a0d4c8b08
[436149.586258] R13: ffff8d8a17d13340 R14: ffff8d8a33e0a540 R15: 00000000000001fe
[436149.586260] FS: 00007fa313933740(0000) GS:ffff8d8a7fc00000(0000) knlGS:0000000000000000
[436149.586261] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[436149.586262] CR2: 000055d8d9c9a720 CR3: 000000007aae0003 CR4: 00000000003606f0
[436149.586295] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[436149.586296] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[436149.586296] Call Trace:
[436149.586311] vfs_rename+0x383/0x920
[436149.586313] ? vfs_rename+0x383/0x920
[436149.586315] do_renameat2+0x4ca/0x590
[436149.586317] __x64_sys_rename+0x20/0x30
[436149.586324] do_syscall_64+0x5a/0x120
[436149.586330] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[436149.586332] RIP: 0033:0x7fa3133b1d37
[436149.586348] RSP: 002b:00007fffd3e43908 EFLAGS: 00000246 ORIG_RAX: 0000000000000052
[436149.586349] RAX: ffffffffffffffda RBX: 00007fa3133b1d30 RCX: 00007fa3133b1d37
[436149.586350] RDX: 000055d8da06b5e0 RSI: 000055d8da225d60 RDI: 000055d8da2c4da0
[436149.586351] RBP: 000055d8da2252f0 R08: 00007fa313782000 R09: 00000000000177e0
[436149.586351] R10: 000055d8da010680 R11: 0000000000000246 R12: 00007fa313840b00
Thanks to Hans van Kranenburg for information about crc32 hash collision
tools, I was able to reproduce the dir item collision with following
python script.
https://github.com/wutzuchieh/misc_tools/blob/master/crc32_forge.py Run
it under a btrfs volume will trigger the abort transaction. It simply
creates files and rename them to forged names that leads to
hash collision.
There are two ways to fix this. One is to simply revert the patch
878f2d2cb355 ("Btrfs: fix max dir item size calculation") to make the
condition consistent although that patch is correct about the size.
The other way is to handle the leaf space check correctly when
collision happens. I prefer the second one since it correct leaf
space check in collision case. This fix will not account
sizeof(struct btrfs_item) when the item already exists.
There are two places where ins_len doesn't contain
sizeof(struct btrfs_item), however.
1. extent-tree.c: lookup_inline_extent_backref
2. file-item.c: btrfs_csum_file_blocks
to make the logic of btrfs_search_slot more clear, we add a flag
search_for_extension in btrfs_path.
This flag indicates that ins_len passed to btrfs_search_slot doesn't
contain sizeof(struct btrfs_item). When key exists, btrfs_search_slot
will use the actual size needed to calculate the required leaf space.
CC: stable(a)vger.kernel.org # 4.4+
Reviewed-by: Filipe Manana <fdmanana(a)suse.com>
Signed-off-by: ethanwu <ethanwu(a)synology.com>
Signed-off-by: David Sterba <dsterba(a)suse.com>
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 07810891e204..cc89b63d65a4 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2555,8 +2555,14 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
* @p: Holds all btree nodes along the search path
* @root: The root node of the tree
* @key: The key we are looking for
- * @ins_len: Indicates purpose of search, for inserts it is 1, for
- * deletions it's -1. 0 for plain searches
+ * @ins_len: Indicates purpose of search:
+ * >0 for inserts it's size of item inserted (*)
+ * <0 for deletions
+ * 0 for plain searches, not modifying the tree
+ *
+ * (*) If size of item inserted doesn't include
+ * sizeof(struct btrfs_item), then p->search_for_extension must
+ * be set.
* @cow: boolean should CoW operations be performed. Must always be 1
* when modifying the tree.
*
@@ -2717,6 +2723,20 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
if (level == 0) {
p->slots[level] = slot;
+ /*
+ * Item key already exists. In this case, if we are
+ * allowed to insert the item (for example, in dir_item
+ * case, item key collision is allowed), it will be
+ * merged with the original item. Only the item size
+ * grows, no new btrfs item will be added. If
+ * search_for_extension is not set, ins_len already
+ * accounts the size btrfs_item, deduct it here so leaf
+ * space check will be correct.
+ */
+ if (ret == 0 && ins_len > 0 && !p->search_for_extension) {
+ ASSERT(ins_len >= sizeof(struct btrfs_item));
+ ins_len -= sizeof(struct btrfs_item);
+ }
if (ins_len > 0 &&
btrfs_leaf_free_space(b) < ins_len) {
if (write_lock_level < 1) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2674f24cf2e0..3935d297d198 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -368,6 +368,12 @@ struct btrfs_path {
unsigned int search_commit_root:1;
unsigned int need_commit_sem:1;
unsigned int skip_release_on_error:1;
+ /*
+ * Indicate that new item (btrfs_search_slot) is extending already
+ * existing item and ins_len contains only the data size and not item
+ * header (ie. sizeof(struct btrfs_item) is not included).
+ */
+ unsigned int search_for_extension:1;
};
#define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \
sizeof(struct btrfs_item))
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 56ea380f5a17..d79b8369e6aa 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -844,6 +844,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
want = extent_ref_type(parent, owner);
if (insert) {
extra_size = btrfs_extent_inline_ref_size(want);
+ path->search_for_extension = 1;
path->keep_locks = 1;
} else
extra_size = -1;
@@ -996,6 +997,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
out:
if (insert) {
path->keep_locks = 0;
+ path->search_for_extension = 0;
btrfs_unlock_up_safe(path, 1);
}
return err;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 1545c22ef280..6ccfc019ad90 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -1016,8 +1016,10 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
}
btrfs_release_path(path);
+ path->search_for_extension = 1;
ret = btrfs_search_slot(trans, root, &file_key, path,
csum_size, 1);
+ path->search_for_extension = 0;
if (ret < 0)
goto out;
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 9ad9f45b3b91162b33abfe175ae75ab65718dbf5 Mon Sep 17 00:00:00 2001
From: Liu Yi L <yi.l.liu(a)intel.com>
Date: Thu, 7 Jan 2021 00:03:55 +0800
Subject: [PATCH] iommu/vt-d: Move intel_iommu info from struct intel_svm to
struct intel_svm_dev
'struct intel_svm' is shared by all devices bound to a give process,
but records only a single pointer to a 'struct intel_iommu'. Consequently,
cache invalidations may only be applied to a single DMAR unit, and are
erroneously skipped for the other devices.
In preparation for fixing this, rework the structures so that the iommu
pointer resides in 'struct intel_svm_dev', allowing 'struct intel_svm'
to track them in its device list.
Fixes: 1c4f88b7f1f9 ("iommu/vt-d: Shared virtual address in scalable mode")
Cc: Lu Baolu <baolu.lu(a)linux.intel.com>
Cc: Jacob Pan <jacob.jun.pan(a)linux.intel.com>
Cc: Raj Ashok <ashok.raj(a)intel.com>
Cc: David Woodhouse <dwmw2(a)infradead.org>
Reported-by: Guo Kaijie <Kaijie.Guo(a)intel.com>
Reported-by: Xin Zeng <xin.zeng(a)intel.com>
Signed-off-by: Guo Kaijie <Kaijie.Guo(a)intel.com>
Signed-off-by: Xin Zeng <xin.zeng(a)intel.com>
Signed-off-by: Liu Yi L <yi.l.liu(a)intel.com>
Tested-by: Guo Kaijie <Kaijie.Guo(a)intel.com>
Cc: stable(a)vger.kernel.org # v5.0+
Acked-by: Lu Baolu <baolu.lu(a)linux.intel.com>
Link: https://lore.kernel.org/r/1609949037-25291-2-git-send-email-yi.l.liu@intel.…
Signed-off-by: Will Deacon <will(a)kernel.org>
diff --git a/drivers/iommu/intel/svm.c b/drivers/iommu/intel/svm.c
index 9bcedd360235..790ef3497e7e 100644
--- a/drivers/iommu/intel/svm.c
+++ b/drivers/iommu/intel/svm.c
@@ -142,7 +142,7 @@ static void intel_flush_svm_range_dev (struct intel_svm *svm, struct intel_svm_d
}
desc.qw2 = 0;
desc.qw3 = 0;
- qi_submit_sync(svm->iommu, &desc, 1, 0);
+ qi_submit_sync(sdev->iommu, &desc, 1, 0);
if (sdev->dev_iotlb) {
desc.qw0 = QI_DEV_EIOTLB_PASID(svm->pasid) |
@@ -166,7 +166,7 @@ static void intel_flush_svm_range_dev (struct intel_svm *svm, struct intel_svm_d
}
desc.qw2 = 0;
desc.qw3 = 0;
- qi_submit_sync(svm->iommu, &desc, 1, 0);
+ qi_submit_sync(sdev->iommu, &desc, 1, 0);
}
}
@@ -211,7 +211,7 @@ static void intel_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
*/
rcu_read_lock();
list_for_each_entry_rcu(sdev, &svm->devs, list)
- intel_pasid_tear_down_entry(svm->iommu, sdev->dev,
+ intel_pasid_tear_down_entry(sdev->iommu, sdev->dev,
svm->pasid, true);
rcu_read_unlock();
@@ -364,6 +364,7 @@ int intel_svm_bind_gpasid(struct iommu_domain *domain, struct device *dev,
}
sdev->dev = dev;
sdev->sid = PCI_DEVID(info->bus, info->devfn);
+ sdev->iommu = iommu;
/* Only count users if device has aux domains */
if (iommu_dev_feature_enabled(dev, IOMMU_DEV_FEAT_AUX))
@@ -548,6 +549,7 @@ intel_svm_bind_mm(struct device *dev, unsigned int flags,
goto out;
}
sdev->dev = dev;
+ sdev->iommu = iommu;
ret = intel_iommu_enable_pasid(iommu, dev);
if (ret) {
@@ -577,7 +579,6 @@ intel_svm_bind_mm(struct device *dev, unsigned int flags,
kfree(sdev);
goto out;
}
- svm->iommu = iommu;
if (pasid_max > intel_pasid_max_id)
pasid_max = intel_pasid_max_id;
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index d956987ed032..94522685a0d9 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -758,6 +758,7 @@ struct intel_svm_dev {
struct list_head list;
struct rcu_head rcu;
struct device *dev;
+ struct intel_iommu *iommu;
struct svm_dev_ops *ops;
struct iommu_sva sva;
u32 pasid;
@@ -771,7 +772,6 @@ struct intel_svm {
struct mmu_notifier notifier;
struct mm_struct *mm;
- struct intel_iommu *iommu;
unsigned int flags;
u32 pasid;
int gpasid; /* In case that guest PASID is different from host PASID */
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 d16baa3f1453c14d680c5fee01cd122a22d0e0ce Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj(a)kernel.org>
Date: Tue, 5 Jan 2021 12:37:23 -0500
Subject: [PATCH] blk-iocost: fix NULL iocg deref from racing against
initialization
When initializing iocost for a queue, its rqos should be registered before
the blkcg policy is activated to allow policy data initiailization to lookup
the associated ioc. This unfortunately means that the rqos methods can be
called on bios before iocgs are attached to all existing blkgs.
While the race is theoretically possible on ioc_rqos_throttle(), it mostly
happened in ioc_rqos_merge() due to the difference in how they lookup ioc.
The former determines it from the passed in @rqos and then bails before
dereferencing iocg if the looked up ioc is disabled, which most likely is
the case if initialization is still in progress. The latter looked up ioc by
dereferencing the possibly NULL iocg making it a lot more prone to actually
triggering the bug.
* Make ioc_rqos_merge() use the same method as ioc_rqos_throttle() to look
up ioc for consistency.
* Make ioc_rqos_throttle() and ioc_rqos_merge() test for NULL iocg before
dereferencing it.
* Explain the danger of NULL iocgs in blk_iocost_init().
Signed-off-by: Tejun Heo <tj(a)kernel.org>
Reported-by: Jonathan Lemon <bsd(a)fb.com>
Cc: stable(a)vger.kernel.org # v5.4+
Signed-off-by: Jens Axboe <axboe(a)kernel.dk>
diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index ac6078a34939..98d656bdb42b 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -2551,8 +2551,8 @@ static void ioc_rqos_throttle(struct rq_qos *rqos, struct bio *bio)
bool use_debt, ioc_locked;
unsigned long flags;
- /* bypass IOs if disabled or for root cgroup */
- if (!ioc->enabled || !iocg->level)
+ /* bypass IOs if disabled, still initializing, or for root cgroup */
+ if (!ioc->enabled || !iocg || !iocg->level)
return;
/* calculate the absolute vtime cost */
@@ -2679,14 +2679,14 @@ static void ioc_rqos_merge(struct rq_qos *rqos, struct request *rq,
struct bio *bio)
{
struct ioc_gq *iocg = blkg_to_iocg(bio->bi_blkg);
- struct ioc *ioc = iocg->ioc;
+ struct ioc *ioc = rqos_to_ioc(rqos);
sector_t bio_end = bio_end_sector(bio);
struct ioc_now now;
u64 vtime, abs_cost, cost;
unsigned long flags;
- /* bypass if disabled or for root cgroup */
- if (!ioc->enabled || !iocg->level)
+ /* bypass if disabled, still initializing, or for root cgroup */
+ if (!ioc->enabled || !iocg || !iocg->level)
return;
abs_cost = calc_vtime_cost(bio, iocg, true);
@@ -2863,6 +2863,12 @@ static int blk_iocost_init(struct request_queue *q)
ioc_refresh_params(ioc, true);
spin_unlock_irq(&ioc->lock);
+ /*
+ * rqos must be added before activation to allow iocg_pd_init() to
+ * lookup the ioc from q. This means that the rqos methods may get
+ * called before policy activation completion, can't assume that the
+ * target bio has an iocg associated and need to test for NULL iocg.
+ */
rq_qos_add(q, rqos);
ret = blkcg_activate_policy(q, &blkcg_policy_iocost);
if (ret) {
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 a0195f314a25582b38993bf30db11c300f4f4611 Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu(a)intel.com>
Date: Thu, 17 Dec 2020 14:31:19 -0800
Subject: [PATCH] x86/resctrl: Don't move a task to the same resource group
Shakeel Butt reported in [1] that a user can request a task to be moved
to a resource group even if the task is already in the group. It just
wastes time to do the move operation which could be costly to send IPI
to a different CPU.
Add a sanity check to ensure that the move operation only happens when
the task is not already in the resource group.
[1] https://lore.kernel.org/lkml/CALvZod7E9zzHwenzf7objzGKsdBmVwTgEJ0nPgs0LUFU3…
Fixes: e02737d5b826 ("x86/intel_rdt: Add tasks files")
Reported-by: Shakeel Butt <shakeelb(a)google.com>
Signed-off-by: Fenghua Yu <fenghua.yu(a)intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre(a)intel.com>
Signed-off-by: Borislav Petkov <bp(a)suse.de>
Reviewed-by: Tony Luck <tony.luck(a)intel.com>
Cc: stable(a)vger.kernel.org
Link: https://lkml.kernel.org/r/962ede65d8e95be793cb61102cca37f7bb018e66.16082431…
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 1c6f8a60ac52..460f3e0df106 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -546,6 +546,13 @@ static void update_task_closid_rmid(struct task_struct *t)
static int __rdtgroup_move_task(struct task_struct *tsk,
struct rdtgroup *rdtgrp)
{
+ /* If the task is already in rdtgrp, no need to move the task. */
+ if ((rdtgrp->type == RDTCTRL_GROUP && tsk->closid == rdtgrp->closid &&
+ tsk->rmid == rdtgrp->mon.rmid) ||
+ (rdtgrp->type == RDTMON_GROUP && tsk->rmid == rdtgrp->mon.rmid &&
+ tsk->closid == rdtgrp->mon.parent->closid))
+ return 0;
+
/*
* Set the task's closid/rmid before the PQR_ASSOC MSR can be
* updated by them.
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 a0195f314a25582b38993bf30db11c300f4f4611 Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu(a)intel.com>
Date: Thu, 17 Dec 2020 14:31:19 -0800
Subject: [PATCH] x86/resctrl: Don't move a task to the same resource group
Shakeel Butt reported in [1] that a user can request a task to be moved
to a resource group even if the task is already in the group. It just
wastes time to do the move operation which could be costly to send IPI
to a different CPU.
Add a sanity check to ensure that the move operation only happens when
the task is not already in the resource group.
[1] https://lore.kernel.org/lkml/CALvZod7E9zzHwenzf7objzGKsdBmVwTgEJ0nPgs0LUFU3…
Fixes: e02737d5b826 ("x86/intel_rdt: Add tasks files")
Reported-by: Shakeel Butt <shakeelb(a)google.com>
Signed-off-by: Fenghua Yu <fenghua.yu(a)intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre(a)intel.com>
Signed-off-by: Borislav Petkov <bp(a)suse.de>
Reviewed-by: Tony Luck <tony.luck(a)intel.com>
Cc: stable(a)vger.kernel.org
Link: https://lkml.kernel.org/r/962ede65d8e95be793cb61102cca37f7bb018e66.16082431…
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 1c6f8a60ac52..460f3e0df106 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -546,6 +546,13 @@ static void update_task_closid_rmid(struct task_struct *t)
static int __rdtgroup_move_task(struct task_struct *tsk,
struct rdtgroup *rdtgrp)
{
+ /* If the task is already in rdtgrp, no need to move the task. */
+ if ((rdtgrp->type == RDTCTRL_GROUP && tsk->closid == rdtgrp->closid &&
+ tsk->rmid == rdtgrp->mon.rmid) ||
+ (rdtgrp->type == RDTMON_GROUP && tsk->rmid == rdtgrp->mon.rmid &&
+ tsk->closid == rdtgrp->mon.parent->closid))
+ return 0;
+
/*
* Set the task's closid/rmid before the PQR_ASSOC MSR can be
* updated by them.
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 a0195f314a25582b38993bf30db11c300f4f4611 Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu(a)intel.com>
Date: Thu, 17 Dec 2020 14:31:19 -0800
Subject: [PATCH] x86/resctrl: Don't move a task to the same resource group
Shakeel Butt reported in [1] that a user can request a task to be moved
to a resource group even if the task is already in the group. It just
wastes time to do the move operation which could be costly to send IPI
to a different CPU.
Add a sanity check to ensure that the move operation only happens when
the task is not already in the resource group.
[1] https://lore.kernel.org/lkml/CALvZod7E9zzHwenzf7objzGKsdBmVwTgEJ0nPgs0LUFU3…
Fixes: e02737d5b826 ("x86/intel_rdt: Add tasks files")
Reported-by: Shakeel Butt <shakeelb(a)google.com>
Signed-off-by: Fenghua Yu <fenghua.yu(a)intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre(a)intel.com>
Signed-off-by: Borislav Petkov <bp(a)suse.de>
Reviewed-by: Tony Luck <tony.luck(a)intel.com>
Cc: stable(a)vger.kernel.org
Link: https://lkml.kernel.org/r/962ede65d8e95be793cb61102cca37f7bb018e66.16082431…
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 1c6f8a60ac52..460f3e0df106 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -546,6 +546,13 @@ static void update_task_closid_rmid(struct task_struct *t)
static int __rdtgroup_move_task(struct task_struct *tsk,
struct rdtgroup *rdtgrp)
{
+ /* If the task is already in rdtgrp, no need to move the task. */
+ if ((rdtgrp->type == RDTCTRL_GROUP && tsk->closid == rdtgrp->closid &&
+ tsk->rmid == rdtgrp->mon.rmid) ||
+ (rdtgrp->type == RDTMON_GROUP && tsk->rmid == rdtgrp->mon.rmid &&
+ tsk->closid == rdtgrp->mon.parent->closid))
+ return 0;
+
/*
* Set the task's closid/rmid before the PQR_ASSOC MSR can be
* updated by them.
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 88bf56d04bc3564542049ec4ec168a8b60d0b48c Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs(a)linux.alibaba.com>
Date: Thu, 17 Dec 2020 23:41:18 +0800
Subject: [PATCH] kvm: check tlbs_dirty directly
In kvm_mmu_notifier_invalidate_range_start(), tlbs_dirty is used as:
need_tlb_flush |= kvm->tlbs_dirty;
with need_tlb_flush's type being int and tlbs_dirty's type being long.
It means that tlbs_dirty is always used as int and the higher 32 bits
is useless. We need to check tlbs_dirty in a correct way and this
change checks it directly without propagating it to need_tlb_flush.
Note: it's _extremely_ unlikely this neglecting of higher 32 bits can
cause problems in practice. It would require encountering tlbs_dirty
on a 4 billion count boundary, and KVM would need to be using shadow
paging or be running a nested guest.
Cc: stable(a)vger.kernel.org
Fixes: a4ee1ca4a36e ("KVM: MMU: delay flush all tlbs on sync_page path")
Signed-off-by: Lai Jiangshan <laijs(a)linux.alibaba.com>
Message-Id: <20201217154118.16497-1-jiangshanlai(a)gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini(a)redhat.com>
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 3abcb2ce5b7d..19dae28904f7 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -485,9 +485,8 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
kvm->mmu_notifier_count++;
need_tlb_flush = kvm_unmap_hva_range(kvm, range->start, range->end,
range->flags);
- need_tlb_flush |= kvm->tlbs_dirty;
/* we've to flush the tlb before the pages can be freed */
- if (need_tlb_flush)
+ if (need_tlb_flush || kvm->tlbs_dirty)
kvm_flush_remote_tlbs(kvm);
spin_unlock(&kvm->mmu_lock);
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 88bf56d04bc3564542049ec4ec168a8b60d0b48c Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs(a)linux.alibaba.com>
Date: Thu, 17 Dec 2020 23:41:18 +0800
Subject: [PATCH] kvm: check tlbs_dirty directly
In kvm_mmu_notifier_invalidate_range_start(), tlbs_dirty is used as:
need_tlb_flush |= kvm->tlbs_dirty;
with need_tlb_flush's type being int and tlbs_dirty's type being long.
It means that tlbs_dirty is always used as int and the higher 32 bits
is useless. We need to check tlbs_dirty in a correct way and this
change checks it directly without propagating it to need_tlb_flush.
Note: it's _extremely_ unlikely this neglecting of higher 32 bits can
cause problems in practice. It would require encountering tlbs_dirty
on a 4 billion count boundary, and KVM would need to be using shadow
paging or be running a nested guest.
Cc: stable(a)vger.kernel.org
Fixes: a4ee1ca4a36e ("KVM: MMU: delay flush all tlbs on sync_page path")
Signed-off-by: Lai Jiangshan <laijs(a)linux.alibaba.com>
Message-Id: <20201217154118.16497-1-jiangshanlai(a)gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini(a)redhat.com>
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 3abcb2ce5b7d..19dae28904f7 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -485,9 +485,8 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
kvm->mmu_notifier_count++;
need_tlb_flush = kvm_unmap_hva_range(kvm, range->start, range->end,
range->flags);
- need_tlb_flush |= kvm->tlbs_dirty;
/* we've to flush the tlb before the pages can be freed */
- if (need_tlb_flush)
+ if (need_tlb_flush || kvm->tlbs_dirty)
kvm_flush_remote_tlbs(kvm);
spin_unlock(&kvm->mmu_lock);
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 0a88fa221ce911c331bf700d2214c5b2f77414d3 Mon Sep 17 00:00:00 2001
From: Manish Narani <manish.narani(a)xilinx.com>
Date: Tue, 17 Nov 2020 12:43:35 +0530
Subject: [PATCH] usb: gadget: u_ether: Fix MTU size mismatch with RX packet
size
Fix the MTU size issue with RX packet size as the host sends the packet
with extra bytes containing ethernet header. This causes failure when
user sets the MTU size to the maximum i.e. 15412. In this case the
ethernet packet received will be of length 15412 plus the ethernet header
length. This patch fixes the issue where there is a check that RX packet
length must not be more than max packet length.
Fixes: bba787a860fa ("usb: gadget: ether: Allow jumbo frames")
Signed-off-by: Manish Narani <manish.narani(a)xilinx.com>
Cc: stable <stable(a)vger.kernel.org>
Link: https://lore.kernel.org/r/1605597215-122027-1-git-send-email-manish.narani@…
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/gadget/function/u_ether.c b/drivers/usb/gadget/function/u_ether.c
index 31ea76adcc0d..c019f2b0c0af 100644
--- a/drivers/usb/gadget/function/u_ether.c
+++ b/drivers/usb/gadget/function/u_ether.c
@@ -45,9 +45,10 @@
#define UETH__VERSION "29-May-2008"
/* Experiments show that both Linux and Windows hosts allow up to 16k
- * frame sizes. Set the max size to 15k+52 to prevent allocating 32k
+ * frame sizes. Set the max MTU size to 15k+52 to prevent allocating 32k
* blocks and still have efficient handling. */
-#define GETHER_MAX_ETH_FRAME_LEN 15412
+#define GETHER_MAX_MTU_SIZE 15412
+#define GETHER_MAX_ETH_FRAME_LEN (GETHER_MAX_MTU_SIZE + ETH_HLEN)
struct eth_dev {
/* lock is held while accessing port_usb
@@ -786,7 +787,7 @@ struct eth_dev *gether_setup_name(struct usb_gadget *g,
/* MTU range: 14 - 15412 */
net->min_mtu = ETH_HLEN;
- net->max_mtu = GETHER_MAX_ETH_FRAME_LEN;
+ net->max_mtu = GETHER_MAX_MTU_SIZE;
dev->gadget = g;
SET_NETDEV_DEV(net, &g->dev);
@@ -848,7 +849,7 @@ struct net_device *gether_setup_name_default(const char *netname)
/* MTU range: 14 - 15412 */
net->min_mtu = ETH_HLEN;
- net->max_mtu = GETHER_MAX_ETH_FRAME_LEN;
+ net->max_mtu = GETHER_MAX_MTU_SIZE;
return net;
}
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 0a88fa221ce911c331bf700d2214c5b2f77414d3 Mon Sep 17 00:00:00 2001
From: Manish Narani <manish.narani(a)xilinx.com>
Date: Tue, 17 Nov 2020 12:43:35 +0530
Subject: [PATCH] usb: gadget: u_ether: Fix MTU size mismatch with RX packet
size
Fix the MTU size issue with RX packet size as the host sends the packet
with extra bytes containing ethernet header. This causes failure when
user sets the MTU size to the maximum i.e. 15412. In this case the
ethernet packet received will be of length 15412 plus the ethernet header
length. This patch fixes the issue where there is a check that RX packet
length must not be more than max packet length.
Fixes: bba787a860fa ("usb: gadget: ether: Allow jumbo frames")
Signed-off-by: Manish Narani <manish.narani(a)xilinx.com>
Cc: stable <stable(a)vger.kernel.org>
Link: https://lore.kernel.org/r/1605597215-122027-1-git-send-email-manish.narani@…
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/gadget/function/u_ether.c b/drivers/usb/gadget/function/u_ether.c
index 31ea76adcc0d..c019f2b0c0af 100644
--- a/drivers/usb/gadget/function/u_ether.c
+++ b/drivers/usb/gadget/function/u_ether.c
@@ -45,9 +45,10 @@
#define UETH__VERSION "29-May-2008"
/* Experiments show that both Linux and Windows hosts allow up to 16k
- * frame sizes. Set the max size to 15k+52 to prevent allocating 32k
+ * frame sizes. Set the max MTU size to 15k+52 to prevent allocating 32k
* blocks and still have efficient handling. */
-#define GETHER_MAX_ETH_FRAME_LEN 15412
+#define GETHER_MAX_MTU_SIZE 15412
+#define GETHER_MAX_ETH_FRAME_LEN (GETHER_MAX_MTU_SIZE + ETH_HLEN)
struct eth_dev {
/* lock is held while accessing port_usb
@@ -786,7 +787,7 @@ struct eth_dev *gether_setup_name(struct usb_gadget *g,
/* MTU range: 14 - 15412 */
net->min_mtu = ETH_HLEN;
- net->max_mtu = GETHER_MAX_ETH_FRAME_LEN;
+ net->max_mtu = GETHER_MAX_MTU_SIZE;
dev->gadget = g;
SET_NETDEV_DEV(net, &g->dev);
@@ -848,7 +849,7 @@ struct net_device *gether_setup_name_default(const char *netname)
/* MTU range: 14 - 15412 */
net->min_mtu = ETH_HLEN;
- net->max_mtu = GETHER_MAX_ETH_FRAME_LEN;
+ net->max_mtu = GETHER_MAX_MTU_SIZE;
return net;
}
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 c318840fb2a42ce25febc95c4c19357acf1ae5ca Mon Sep 17 00:00:00 2001
From: Alan Stern <stern(a)rowland.harvard.edu>
Date: Wed, 30 Dec 2020 11:20:44 -0500
Subject: [PATCH] USB: Gadget: dummy-hcd: Fix shift-out-of-bounds bug
The dummy-hcd driver was written under the assumption that all the
parameters in URBs sent to its root hub would be valid. With URBs
sent from userspace via usbfs, that assumption can be violated.
In particular, the driver doesn't fully check the port-feature values
stored in the wValue entry of Clear-Port-Feature and Set-Port-Feature
requests. Values that are too large can cause the driver to perform
an invalid left shift of more than 32 bits. Ironically, two of those
left shifts are unnecessary, because they implement Set-Port-Feature
requests that hubs are not required to support, according to section
11.24.2.13 of the USB-2.0 spec.
This patch adds the appropriate checks for the port feature selector
values and removes the unnecessary feature settings. It also rejects
requests to set the TEST feature or to set or clear the INDICATOR and
C_OVERCURRENT features, as none of these are relevant to dummy-hcd's
root-hub emulation.
CC: <stable(a)vger.kernel.org>
Reported-and-tested-by: syzbot+5925509f78293baa7331(a)syzkaller.appspotmail.com
Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu>
Link: https://lore.kernel.org/r/20201230162044.GA727759@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index ab5e978b5052..1a953f44183a 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -2118,9 +2118,21 @@ static int dummy_hub_control(
dum_hcd->port_status &= ~USB_PORT_STAT_POWER;
set_link_state(dum_hcd);
break;
- default:
+ case USB_PORT_FEAT_ENABLE:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
dum_hcd->port_status &= ~(1 << wValue);
set_link_state(dum_hcd);
+ break;
+ default:
+ /* Disallow INDICATOR and C_OVER_CURRENT */
+ goto error;
}
break;
case GetHubDescriptor:
@@ -2281,18 +2293,17 @@ static int dummy_hub_control(
*/
dum_hcd->re_timeout = jiffies + msecs_to_jiffies(50);
fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3, and ignored for USB-2 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ break;
default:
- if (hcd->speed == HCD_USB3) {
- if ((dum_hcd->port_status &
- USB_SS_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- } else
- if ((dum_hcd->port_status &
- USB_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- set_link_state(dum_hcd);
+ /* Disallow TEST, INDICATOR, and C_OVER_CURRENT */
+ goto error;
}
break;
case GetPortErrorCount:
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 c318840fb2a42ce25febc95c4c19357acf1ae5ca Mon Sep 17 00:00:00 2001
From: Alan Stern <stern(a)rowland.harvard.edu>
Date: Wed, 30 Dec 2020 11:20:44 -0500
Subject: [PATCH] USB: Gadget: dummy-hcd: Fix shift-out-of-bounds bug
The dummy-hcd driver was written under the assumption that all the
parameters in URBs sent to its root hub would be valid. With URBs
sent from userspace via usbfs, that assumption can be violated.
In particular, the driver doesn't fully check the port-feature values
stored in the wValue entry of Clear-Port-Feature and Set-Port-Feature
requests. Values that are too large can cause the driver to perform
an invalid left shift of more than 32 bits. Ironically, two of those
left shifts are unnecessary, because they implement Set-Port-Feature
requests that hubs are not required to support, according to section
11.24.2.13 of the USB-2.0 spec.
This patch adds the appropriate checks for the port feature selector
values and removes the unnecessary feature settings. It also rejects
requests to set the TEST feature or to set or clear the INDICATOR and
C_OVERCURRENT features, as none of these are relevant to dummy-hcd's
root-hub emulation.
CC: <stable(a)vger.kernel.org>
Reported-and-tested-by: syzbot+5925509f78293baa7331(a)syzkaller.appspotmail.com
Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu>
Link: https://lore.kernel.org/r/20201230162044.GA727759@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index ab5e978b5052..1a953f44183a 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -2118,9 +2118,21 @@ static int dummy_hub_control(
dum_hcd->port_status &= ~USB_PORT_STAT_POWER;
set_link_state(dum_hcd);
break;
- default:
+ case USB_PORT_FEAT_ENABLE:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
dum_hcd->port_status &= ~(1 << wValue);
set_link_state(dum_hcd);
+ break;
+ default:
+ /* Disallow INDICATOR and C_OVER_CURRENT */
+ goto error;
}
break;
case GetHubDescriptor:
@@ -2281,18 +2293,17 @@ static int dummy_hub_control(
*/
dum_hcd->re_timeout = jiffies + msecs_to_jiffies(50);
fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3, and ignored for USB-2 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ break;
default:
- if (hcd->speed == HCD_USB3) {
- if ((dum_hcd->port_status &
- USB_SS_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- } else
- if ((dum_hcd->port_status &
- USB_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- set_link_state(dum_hcd);
+ /* Disallow TEST, INDICATOR, and C_OVER_CURRENT */
+ goto error;
}
break;
case GetPortErrorCount:
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 c318840fb2a42ce25febc95c4c19357acf1ae5ca Mon Sep 17 00:00:00 2001
From: Alan Stern <stern(a)rowland.harvard.edu>
Date: Wed, 30 Dec 2020 11:20:44 -0500
Subject: [PATCH] USB: Gadget: dummy-hcd: Fix shift-out-of-bounds bug
The dummy-hcd driver was written under the assumption that all the
parameters in URBs sent to its root hub would be valid. With URBs
sent from userspace via usbfs, that assumption can be violated.
In particular, the driver doesn't fully check the port-feature values
stored in the wValue entry of Clear-Port-Feature and Set-Port-Feature
requests. Values that are too large can cause the driver to perform
an invalid left shift of more than 32 bits. Ironically, two of those
left shifts are unnecessary, because they implement Set-Port-Feature
requests that hubs are not required to support, according to section
11.24.2.13 of the USB-2.0 spec.
This patch adds the appropriate checks for the port feature selector
values and removes the unnecessary feature settings. It also rejects
requests to set the TEST feature or to set or clear the INDICATOR and
C_OVERCURRENT features, as none of these are relevant to dummy-hcd's
root-hub emulation.
CC: <stable(a)vger.kernel.org>
Reported-and-tested-by: syzbot+5925509f78293baa7331(a)syzkaller.appspotmail.com
Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu>
Link: https://lore.kernel.org/r/20201230162044.GA727759@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index ab5e978b5052..1a953f44183a 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -2118,9 +2118,21 @@ static int dummy_hub_control(
dum_hcd->port_status &= ~USB_PORT_STAT_POWER;
set_link_state(dum_hcd);
break;
- default:
+ case USB_PORT_FEAT_ENABLE:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
dum_hcd->port_status &= ~(1 << wValue);
set_link_state(dum_hcd);
+ break;
+ default:
+ /* Disallow INDICATOR and C_OVER_CURRENT */
+ goto error;
}
break;
case GetHubDescriptor:
@@ -2281,18 +2293,17 @@ static int dummy_hub_control(
*/
dum_hcd->re_timeout = jiffies + msecs_to_jiffies(50);
fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3, and ignored for USB-2 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ break;
default:
- if (hcd->speed == HCD_USB3) {
- if ((dum_hcd->port_status &
- USB_SS_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- } else
- if ((dum_hcd->port_status &
- USB_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- set_link_state(dum_hcd);
+ /* Disallow TEST, INDICATOR, and C_OVER_CURRENT */
+ goto error;
}
break;
case GetPortErrorCount:
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 c318840fb2a42ce25febc95c4c19357acf1ae5ca Mon Sep 17 00:00:00 2001
From: Alan Stern <stern(a)rowland.harvard.edu>
Date: Wed, 30 Dec 2020 11:20:44 -0500
Subject: [PATCH] USB: Gadget: dummy-hcd: Fix shift-out-of-bounds bug
The dummy-hcd driver was written under the assumption that all the
parameters in URBs sent to its root hub would be valid. With URBs
sent from userspace via usbfs, that assumption can be violated.
In particular, the driver doesn't fully check the port-feature values
stored in the wValue entry of Clear-Port-Feature and Set-Port-Feature
requests. Values that are too large can cause the driver to perform
an invalid left shift of more than 32 bits. Ironically, two of those
left shifts are unnecessary, because they implement Set-Port-Feature
requests that hubs are not required to support, according to section
11.24.2.13 of the USB-2.0 spec.
This patch adds the appropriate checks for the port feature selector
values and removes the unnecessary feature settings. It also rejects
requests to set the TEST feature or to set or clear the INDICATOR and
C_OVERCURRENT features, as none of these are relevant to dummy-hcd's
root-hub emulation.
CC: <stable(a)vger.kernel.org>
Reported-and-tested-by: syzbot+5925509f78293baa7331(a)syzkaller.appspotmail.com
Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu>
Link: https://lore.kernel.org/r/20201230162044.GA727759@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index ab5e978b5052..1a953f44183a 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -2118,9 +2118,21 @@ static int dummy_hub_control(
dum_hcd->port_status &= ~USB_PORT_STAT_POWER;
set_link_state(dum_hcd);
break;
- default:
+ case USB_PORT_FEAT_ENABLE:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
dum_hcd->port_status &= ~(1 << wValue);
set_link_state(dum_hcd);
+ break;
+ default:
+ /* Disallow INDICATOR and C_OVER_CURRENT */
+ goto error;
}
break;
case GetHubDescriptor:
@@ -2281,18 +2293,17 @@ static int dummy_hub_control(
*/
dum_hcd->re_timeout = jiffies + msecs_to_jiffies(50);
fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3, and ignored for USB-2 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ break;
default:
- if (hcd->speed == HCD_USB3) {
- if ((dum_hcd->port_status &
- USB_SS_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- } else
- if ((dum_hcd->port_status &
- USB_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- set_link_state(dum_hcd);
+ /* Disallow TEST, INDICATOR, and C_OVER_CURRENT */
+ goto error;
}
break;
case GetPortErrorCount:
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 c318840fb2a42ce25febc95c4c19357acf1ae5ca Mon Sep 17 00:00:00 2001
From: Alan Stern <stern(a)rowland.harvard.edu>
Date: Wed, 30 Dec 2020 11:20:44 -0500
Subject: [PATCH] USB: Gadget: dummy-hcd: Fix shift-out-of-bounds bug
The dummy-hcd driver was written under the assumption that all the
parameters in URBs sent to its root hub would be valid. With URBs
sent from userspace via usbfs, that assumption can be violated.
In particular, the driver doesn't fully check the port-feature values
stored in the wValue entry of Clear-Port-Feature and Set-Port-Feature
requests. Values that are too large can cause the driver to perform
an invalid left shift of more than 32 bits. Ironically, two of those
left shifts are unnecessary, because they implement Set-Port-Feature
requests that hubs are not required to support, according to section
11.24.2.13 of the USB-2.0 spec.
This patch adds the appropriate checks for the port feature selector
values and removes the unnecessary feature settings. It also rejects
requests to set the TEST feature or to set or clear the INDICATOR and
C_OVERCURRENT features, as none of these are relevant to dummy-hcd's
root-hub emulation.
CC: <stable(a)vger.kernel.org>
Reported-and-tested-by: syzbot+5925509f78293baa7331(a)syzkaller.appspotmail.com
Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu>
Link: https://lore.kernel.org/r/20201230162044.GA727759@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index ab5e978b5052..1a953f44183a 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -2118,9 +2118,21 @@ static int dummy_hub_control(
dum_hcd->port_status &= ~USB_PORT_STAT_POWER;
set_link_state(dum_hcd);
break;
- default:
+ case USB_PORT_FEAT_ENABLE:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
dum_hcd->port_status &= ~(1 << wValue);
set_link_state(dum_hcd);
+ break;
+ default:
+ /* Disallow INDICATOR and C_OVER_CURRENT */
+ goto error;
}
break;
case GetHubDescriptor:
@@ -2281,18 +2293,17 @@ static int dummy_hub_control(
*/
dum_hcd->re_timeout = jiffies + msecs_to_jiffies(50);
fallthrough;
+ case USB_PORT_FEAT_C_CONNECTION:
+ case USB_PORT_FEAT_C_RESET:
+ case USB_PORT_FEAT_C_ENABLE:
+ case USB_PORT_FEAT_C_SUSPEND:
+ /* Not allowed for USB-3, and ignored for USB-2 */
+ if (hcd->speed == HCD_USB3)
+ goto error;
+ break;
default:
- if (hcd->speed == HCD_USB3) {
- if ((dum_hcd->port_status &
- USB_SS_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- } else
- if ((dum_hcd->port_status &
- USB_PORT_STAT_POWER) != 0) {
- dum_hcd->port_status |= (1 << wValue);
- }
- set_link_state(dum_hcd);
+ /* Disallow TEST, INDICATOR, and C_OVER_CURRENT */
+ goto error;
}
break;
case GetPortErrorCount:
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 e5f4ca3fce90a37b23a77bfcc86800d484a80514 Mon Sep 17 00:00:00 2001
From: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Date: Thu, 10 Dec 2020 11:50:08 +0300
Subject: [PATCH] usb: dwc3: ulpi: Fix USB2.0 HS/FS/LS PHY suspend regression
First of all the commit e0082698b689 ("usb: dwc3: ulpi: conditionally
resume ULPI PHY") introduced the Suspend USB2.0 HS/FS/LS PHY regression,
as by design of the fix any attempt to read/write from/to the PHY control
registers will completely disable the PHY suspension, which consequently
will increase the USB bus power consumption. Secondly the fix won't work
well for the very first attempt of the ULPI PHY control registers IO,
because after disabling the USB2.0 PHY suspension functionality it will
still take some time for the bus to resume from the sleep state if one has
been reached before it. So the very first PHY register read/write
operation will take more time than the busy-loop provides and the IO
timeout error might be returned anyway.
Here we suggest to fix the denoted problems in the following way. First of
all let's not disable the Suspend USB2.0 HS/FS/LS PHY functionality so to
make the controller and the USB2.0 bus more power efficient. Secondly
instead of that we'll extend the PHY IO op wait procedure with 1 - 1.2 ms
sleep if the PHY suspension is enabled (1ms should be enough as by LPM
specification it is at most how long it takes for the USB2.0 bus to resume
from L1 (Sleep) state). Finally in case if the USB2.0 PHY suspension
functionality has been disabled on the DWC USB3 controller setup procedure
we'll compensate the USB bus resume process latency by extending the
busy-loop attempts counter.
Fixes: e0082698b689 ("usb: dwc3: ulpi: conditionally resume ULPI PHY")
Acked-by: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
Signed-off-by: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Link: https://lore.kernel.org/r/20201210085008.13264-4-Sergey.Semin@baikalelectro…
Cc: stable <stable(a)vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/dwc3/ulpi.c b/drivers/usb/dwc3/ulpi.c
index 54c877f7b51d..f23f4c9a557e 100644
--- a/drivers/usb/dwc3/ulpi.c
+++ b/drivers/usb/dwc3/ulpi.c
@@ -24,7 +24,7 @@
static int dwc3_ulpi_busyloop(struct dwc3 *dwc, u8 addr, bool read)
{
unsigned long ns = 5L * DWC3_ULPI_BASE_DELAY;
- unsigned int count = 1000;
+ unsigned int count = 10000;
u32 reg;
if (addr >= ULPI_EXT_VENDOR_SPECIFIC)
@@ -33,6 +33,10 @@ static int dwc3_ulpi_busyloop(struct dwc3 *dwc, u8 addr, bool read)
if (read)
ns += DWC3_ULPI_BASE_DELAY;
+ reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
+ if (reg & DWC3_GUSB2PHYCFG_SUSPHY)
+ usleep_range(1000, 1200);
+
while (count--) {
ndelay(ns);
reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYACC(0));
@@ -50,12 +54,6 @@ static int dwc3_ulpi_read(struct device *dev, u8 addr)
u32 reg;
int ret;
- reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
- if (reg & DWC3_GUSB2PHYCFG_SUSPHY) {
- reg &= ~DWC3_GUSB2PHYCFG_SUSPHY;
- dwc3_writel(dwc->regs, DWC3_GUSB2PHYCFG(0), reg);
- }
-
reg = DWC3_GUSB2PHYACC_NEWREGREQ | DWC3_ULPI_ADDR(addr);
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
@@ -73,12 +71,6 @@ static int dwc3_ulpi_write(struct device *dev, u8 addr, u8 val)
struct dwc3 *dwc = dev_get_drvdata(dev);
u32 reg;
- reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
- if (reg & DWC3_GUSB2PHYCFG_SUSPHY) {
- reg &= ~DWC3_GUSB2PHYCFG_SUSPHY;
- dwc3_writel(dwc->regs, DWC3_GUSB2PHYCFG(0), reg);
- }
-
reg = DWC3_GUSB2PHYACC_NEWREGREQ | DWC3_ULPI_ADDR(addr);
reg |= DWC3_GUSB2PHYACC_WRITE | val;
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
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 e5f4ca3fce90a37b23a77bfcc86800d484a80514 Mon Sep 17 00:00:00 2001
From: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Date: Thu, 10 Dec 2020 11:50:08 +0300
Subject: [PATCH] usb: dwc3: ulpi: Fix USB2.0 HS/FS/LS PHY suspend regression
First of all the commit e0082698b689 ("usb: dwc3: ulpi: conditionally
resume ULPI PHY") introduced the Suspend USB2.0 HS/FS/LS PHY regression,
as by design of the fix any attempt to read/write from/to the PHY control
registers will completely disable the PHY suspension, which consequently
will increase the USB bus power consumption. Secondly the fix won't work
well for the very first attempt of the ULPI PHY control registers IO,
because after disabling the USB2.0 PHY suspension functionality it will
still take some time for the bus to resume from the sleep state if one has
been reached before it. So the very first PHY register read/write
operation will take more time than the busy-loop provides and the IO
timeout error might be returned anyway.
Here we suggest to fix the denoted problems in the following way. First of
all let's not disable the Suspend USB2.0 HS/FS/LS PHY functionality so to
make the controller and the USB2.0 bus more power efficient. Secondly
instead of that we'll extend the PHY IO op wait procedure with 1 - 1.2 ms
sleep if the PHY suspension is enabled (1ms should be enough as by LPM
specification it is at most how long it takes for the USB2.0 bus to resume
from L1 (Sleep) state). Finally in case if the USB2.0 PHY suspension
functionality has been disabled on the DWC USB3 controller setup procedure
we'll compensate the USB bus resume process latency by extending the
busy-loop attempts counter.
Fixes: e0082698b689 ("usb: dwc3: ulpi: conditionally resume ULPI PHY")
Acked-by: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
Signed-off-by: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Link: https://lore.kernel.org/r/20201210085008.13264-4-Sergey.Semin@baikalelectro…
Cc: stable <stable(a)vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/dwc3/ulpi.c b/drivers/usb/dwc3/ulpi.c
index 54c877f7b51d..f23f4c9a557e 100644
--- a/drivers/usb/dwc3/ulpi.c
+++ b/drivers/usb/dwc3/ulpi.c
@@ -24,7 +24,7 @@
static int dwc3_ulpi_busyloop(struct dwc3 *dwc, u8 addr, bool read)
{
unsigned long ns = 5L * DWC3_ULPI_BASE_DELAY;
- unsigned int count = 1000;
+ unsigned int count = 10000;
u32 reg;
if (addr >= ULPI_EXT_VENDOR_SPECIFIC)
@@ -33,6 +33,10 @@ static int dwc3_ulpi_busyloop(struct dwc3 *dwc, u8 addr, bool read)
if (read)
ns += DWC3_ULPI_BASE_DELAY;
+ reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
+ if (reg & DWC3_GUSB2PHYCFG_SUSPHY)
+ usleep_range(1000, 1200);
+
while (count--) {
ndelay(ns);
reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYACC(0));
@@ -50,12 +54,6 @@ static int dwc3_ulpi_read(struct device *dev, u8 addr)
u32 reg;
int ret;
- reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
- if (reg & DWC3_GUSB2PHYCFG_SUSPHY) {
- reg &= ~DWC3_GUSB2PHYCFG_SUSPHY;
- dwc3_writel(dwc->regs, DWC3_GUSB2PHYCFG(0), reg);
- }
-
reg = DWC3_GUSB2PHYACC_NEWREGREQ | DWC3_ULPI_ADDR(addr);
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
@@ -73,12 +71,6 @@ static int dwc3_ulpi_write(struct device *dev, u8 addr, u8 val)
struct dwc3 *dwc = dev_get_drvdata(dev);
u32 reg;
- reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
- if (reg & DWC3_GUSB2PHYCFG_SUSPHY) {
- reg &= ~DWC3_GUSB2PHYCFG_SUSPHY;
- dwc3_writel(dwc->regs, DWC3_GUSB2PHYCFG(0), reg);
- }
-
reg = DWC3_GUSB2PHYACC_NEWREGREQ | DWC3_ULPI_ADDR(addr);
reg |= DWC3_GUSB2PHYACC_WRITE | val;
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
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 e5f4ca3fce90a37b23a77bfcc86800d484a80514 Mon Sep 17 00:00:00 2001
From: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Date: Thu, 10 Dec 2020 11:50:08 +0300
Subject: [PATCH] usb: dwc3: ulpi: Fix USB2.0 HS/FS/LS PHY suspend regression
First of all the commit e0082698b689 ("usb: dwc3: ulpi: conditionally
resume ULPI PHY") introduced the Suspend USB2.0 HS/FS/LS PHY regression,
as by design of the fix any attempt to read/write from/to the PHY control
registers will completely disable the PHY suspension, which consequently
will increase the USB bus power consumption. Secondly the fix won't work
well for the very first attempt of the ULPI PHY control registers IO,
because after disabling the USB2.0 PHY suspension functionality it will
still take some time for the bus to resume from the sleep state if one has
been reached before it. So the very first PHY register read/write
operation will take more time than the busy-loop provides and the IO
timeout error might be returned anyway.
Here we suggest to fix the denoted problems in the following way. First of
all let's not disable the Suspend USB2.0 HS/FS/LS PHY functionality so to
make the controller and the USB2.0 bus more power efficient. Secondly
instead of that we'll extend the PHY IO op wait procedure with 1 - 1.2 ms
sleep if the PHY suspension is enabled (1ms should be enough as by LPM
specification it is at most how long it takes for the USB2.0 bus to resume
from L1 (Sleep) state). Finally in case if the USB2.0 PHY suspension
functionality has been disabled on the DWC USB3 controller setup procedure
we'll compensate the USB bus resume process latency by extending the
busy-loop attempts counter.
Fixes: e0082698b689 ("usb: dwc3: ulpi: conditionally resume ULPI PHY")
Acked-by: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
Signed-off-by: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Link: https://lore.kernel.org/r/20201210085008.13264-4-Sergey.Semin@baikalelectro…
Cc: stable <stable(a)vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/dwc3/ulpi.c b/drivers/usb/dwc3/ulpi.c
index 54c877f7b51d..f23f4c9a557e 100644
--- a/drivers/usb/dwc3/ulpi.c
+++ b/drivers/usb/dwc3/ulpi.c
@@ -24,7 +24,7 @@
static int dwc3_ulpi_busyloop(struct dwc3 *dwc, u8 addr, bool read)
{
unsigned long ns = 5L * DWC3_ULPI_BASE_DELAY;
- unsigned int count = 1000;
+ unsigned int count = 10000;
u32 reg;
if (addr >= ULPI_EXT_VENDOR_SPECIFIC)
@@ -33,6 +33,10 @@ static int dwc3_ulpi_busyloop(struct dwc3 *dwc, u8 addr, bool read)
if (read)
ns += DWC3_ULPI_BASE_DELAY;
+ reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
+ if (reg & DWC3_GUSB2PHYCFG_SUSPHY)
+ usleep_range(1000, 1200);
+
while (count--) {
ndelay(ns);
reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYACC(0));
@@ -50,12 +54,6 @@ static int dwc3_ulpi_read(struct device *dev, u8 addr)
u32 reg;
int ret;
- reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
- if (reg & DWC3_GUSB2PHYCFG_SUSPHY) {
- reg &= ~DWC3_GUSB2PHYCFG_SUSPHY;
- dwc3_writel(dwc->regs, DWC3_GUSB2PHYCFG(0), reg);
- }
-
reg = DWC3_GUSB2PHYACC_NEWREGREQ | DWC3_ULPI_ADDR(addr);
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
@@ -73,12 +71,6 @@ static int dwc3_ulpi_write(struct device *dev, u8 addr, u8 val)
struct dwc3 *dwc = dev_get_drvdata(dev);
u32 reg;
- reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
- if (reg & DWC3_GUSB2PHYCFG_SUSPHY) {
- reg &= ~DWC3_GUSB2PHYCFG_SUSPHY;
- dwc3_writel(dwc->regs, DWC3_GUSB2PHYCFG(0), reg);
- }
-
reg = DWC3_GUSB2PHYACC_NEWREGREQ | DWC3_ULPI_ADDR(addr);
reg |= DWC3_GUSB2PHYACC_WRITE | val;
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
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 fca3f138105727c3a22edda32d02f91ce1bf11c9 Mon Sep 17 00:00:00 2001
From: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Date: Thu, 10 Dec 2020 11:50:07 +0300
Subject: [PATCH] usb: dwc3: ulpi: Replace CPU-based busyloop with
Protocol-based one
Originally the procedure of the ULPI transaction finish detection has been
developed as a simple busy-loop with just decrementing counter and no
delays. It's wrong since on different systems the loop will take a
different time to complete. So if the system bus and CPU are fast enough
to overtake the ULPI bus and the companion PHY reaction, then we'll get to
take a false timeout error. Fix this by converting the busy-loop procedure
to take the standard bus speed, address value and the registers access
mode into account for the busy-loop delay calculation.
Here is the way the fix works. It's known that the ULPI bus is clocked
with 60MHz signal. In accordance with [1] the ULPI bus protocol is created
so to spend 5 and 6 clock periods for immediate register write and read
operations respectively, and 6 and 7 clock periods - for the extended
register writes and reads. Based on that we can easily pre-calculate the
time which will be needed for the controller to perform a requested IO
operation. Note we'll still preserve the attempts counter in case if the
DWC USB3 controller has got some internals delays.
[1] UTMI+ Low Pin Interface (ULPI) Specification, Revision 1.1,
October 20, 2004, pp. 30 - 36.
Fixes: 88bc9d194ff6 ("usb: dwc3: add ULPI interface support")
Acked-by: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
Signed-off-by: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Link: https://lore.kernel.org/r/20201210085008.13264-3-Sergey.Semin@baikalelectro…
Cc: stable <stable(a)vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/dwc3/ulpi.c b/drivers/usb/dwc3/ulpi.c
index 3cc4f4970c05..54c877f7b51d 100644
--- a/drivers/usb/dwc3/ulpi.c
+++ b/drivers/usb/dwc3/ulpi.c
@@ -7,6 +7,8 @@
* Author: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
*/
+#include <linux/delay.h>
+#include <linux/time64.h>
#include <linux/ulpi/regs.h>
#include "core.h"
@@ -17,12 +19,22 @@
DWC3_GUSB2PHYACC_ADDR(ULPI_ACCESS_EXTENDED) | \
DWC3_GUSB2PHYACC_EXTEND_ADDR(a) : DWC3_GUSB2PHYACC_ADDR(a))
-static int dwc3_ulpi_busyloop(struct dwc3 *dwc)
+#define DWC3_ULPI_BASE_DELAY DIV_ROUND_UP(NSEC_PER_SEC, 60000000L)
+
+static int dwc3_ulpi_busyloop(struct dwc3 *dwc, u8 addr, bool read)
{
+ unsigned long ns = 5L * DWC3_ULPI_BASE_DELAY;
unsigned int count = 1000;
u32 reg;
+ if (addr >= ULPI_EXT_VENDOR_SPECIFIC)
+ ns += DWC3_ULPI_BASE_DELAY;
+
+ if (read)
+ ns += DWC3_ULPI_BASE_DELAY;
+
while (count--) {
+ ndelay(ns);
reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYACC(0));
if (reg & DWC3_GUSB2PHYACC_DONE)
return 0;
@@ -47,7 +59,7 @@ static int dwc3_ulpi_read(struct device *dev, u8 addr)
reg = DWC3_GUSB2PHYACC_NEWREGREQ | DWC3_ULPI_ADDR(addr);
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
- ret = dwc3_ulpi_busyloop(dwc);
+ ret = dwc3_ulpi_busyloop(dwc, addr, true);
if (ret)
return ret;
@@ -71,7 +83,7 @@ static int dwc3_ulpi_write(struct device *dev, u8 addr, u8 val)
reg |= DWC3_GUSB2PHYACC_WRITE | val;
dwc3_writel(dwc->regs, DWC3_GUSB2PHYACC(0), reg);
- return dwc3_ulpi_busyloop(dwc);
+ return dwc3_ulpi_busyloop(dwc, addr, false);
}
static const struct ulpi_ops dwc3_ulpi_ops = {
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 ce722da66d3e9384aa2de9d33d584ee154e5e157 Mon Sep 17 00:00:00 2001
From: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Date: Thu, 10 Dec 2020 11:50:06 +0300
Subject: [PATCH] usb: dwc3: ulpi: Use VStsDone to detect PHY regs access
completion
In accordance with [1] the DWC_usb3 core sets the GUSB2PHYACCn.VStsDone
bit when the PHY vendor control access is done and clears it when the
application initiates a new transaction. The doc doesn't say anything
about the GUSB2PHYACCn.VStsBsy flag serving for the same purpose. Moreover
we've discovered that the VStsBsy flag can be cleared before the VStsDone
bit. So using the former as a signal of the PHY control registers
completion might be dangerous. Let's have the VStsDone flag utilized
instead then.
[1] Synopsys DesignWare Cores SuperSpeed USB 3.0 xHCI Host Controller
Databook, 2.70a, December 2013, p.388
Fixes: 88bc9d194ff6 ("usb: dwc3: add ULPI interface support")
Acked-by: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
Signed-off-by: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Link: https://lore.kernel.org/r/20201210085008.13264-2-Sergey.Semin@baikalelectro…
Cc: stable <stable(a)vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
index 2f95f08ca511..1b241f937d8f 100644
--- a/drivers/usb/dwc3/core.h
+++ b/drivers/usb/dwc3/core.h
@@ -285,6 +285,7 @@
/* Global USB2 PHY Vendor Control Register */
#define DWC3_GUSB2PHYACC_NEWREGREQ BIT(25)
+#define DWC3_GUSB2PHYACC_DONE BIT(24)
#define DWC3_GUSB2PHYACC_BUSY BIT(23)
#define DWC3_GUSB2PHYACC_WRITE BIT(22)
#define DWC3_GUSB2PHYACC_ADDR(n) (n << 16)
diff --git a/drivers/usb/dwc3/ulpi.c b/drivers/usb/dwc3/ulpi.c
index aa213c9815f6..3cc4f4970c05 100644
--- a/drivers/usb/dwc3/ulpi.c
+++ b/drivers/usb/dwc3/ulpi.c
@@ -24,7 +24,7 @@ static int dwc3_ulpi_busyloop(struct dwc3 *dwc)
while (count--) {
reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYACC(0));
- if (!(reg & DWC3_GUSB2PHYACC_BUSY))
+ if (reg & DWC3_GUSB2PHYACC_DONE)
return 0;
cpu_relax();
}
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 ce722da66d3e9384aa2de9d33d584ee154e5e157 Mon Sep 17 00:00:00 2001
From: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Date: Thu, 10 Dec 2020 11:50:06 +0300
Subject: [PATCH] usb: dwc3: ulpi: Use VStsDone to detect PHY regs access
completion
In accordance with [1] the DWC_usb3 core sets the GUSB2PHYACCn.VStsDone
bit when the PHY vendor control access is done and clears it when the
application initiates a new transaction. The doc doesn't say anything
about the GUSB2PHYACCn.VStsBsy flag serving for the same purpose. Moreover
we've discovered that the VStsBsy flag can be cleared before the VStsDone
bit. So using the former as a signal of the PHY control registers
completion might be dangerous. Let's have the VStsDone flag utilized
instead then.
[1] Synopsys DesignWare Cores SuperSpeed USB 3.0 xHCI Host Controller
Databook, 2.70a, December 2013, p.388
Fixes: 88bc9d194ff6 ("usb: dwc3: add ULPI interface support")
Acked-by: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
Signed-off-by: Serge Semin <Sergey.Semin(a)baikalelectronics.ru>
Link: https://lore.kernel.org/r/20201210085008.13264-2-Sergey.Semin@baikalelectro…
Cc: stable <stable(a)vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
index 2f95f08ca511..1b241f937d8f 100644
--- a/drivers/usb/dwc3/core.h
+++ b/drivers/usb/dwc3/core.h
@@ -285,6 +285,7 @@
/* Global USB2 PHY Vendor Control Register */
#define DWC3_GUSB2PHYACC_NEWREGREQ BIT(25)
+#define DWC3_GUSB2PHYACC_DONE BIT(24)
#define DWC3_GUSB2PHYACC_BUSY BIT(23)
#define DWC3_GUSB2PHYACC_WRITE BIT(22)
#define DWC3_GUSB2PHYACC_ADDR(n) (n << 16)
diff --git a/drivers/usb/dwc3/ulpi.c b/drivers/usb/dwc3/ulpi.c
index aa213c9815f6..3cc4f4970c05 100644
--- a/drivers/usb/dwc3/ulpi.c
+++ b/drivers/usb/dwc3/ulpi.c
@@ -24,7 +24,7 @@ static int dwc3_ulpi_busyloop(struct dwc3 *dwc)
while (count--) {
reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYACC(0));
- if (!(reg & DWC3_GUSB2PHYACC_BUSY))
+ if (reg & DWC3_GUSB2PHYACC_DONE)
return 0;
cpu_relax();
}
In rare cases a task may be exiting while io_ring_exit_work() trying to
cancel/wait its requests. It's ok for __io_sq_thread_acquire_mm()
because of SQPOLL check, but is not for __io_sq_thread_acquire_files().
Play safe and fail for both of them.
Cc: stable(a)vger.kernel.org # 5.5+
Signed-off-by: Pavel Begunkov <asml.silence(a)gmail.com>
---
fs/io_uring.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 7af74c1ec909..b0e6d8e607a3 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -1106,6 +1106,9 @@ static void io_sq_thread_drop_mm_files(void)
static int __io_sq_thread_acquire_files(struct io_ring_ctx *ctx)
{
+ if (current->flags & PF_EXITING)
+ return -EFAULT;
+
if (!current->files) {
struct files_struct *files;
struct nsproxy *nsproxy;
@@ -1133,6 +1136,8 @@ static int __io_sq_thread_acquire_mm(struct io_ring_ctx *ctx)
{
struct mm_struct *mm;
+ if (current->flags & PF_EXITING)
+ return -EFAULT;
if (current->mm)
return 0;
--
2.24.0
__io_req_task_submit() run by task_work can set mm and files, but
io_sq_thread() in some cases, and because __io_sq_thread_acquire_mm()
and __io_sq_thread_acquire_files() do a simple current->mm/files check
it may end up submitting IO with mm/files of another task.
We also need to drop it after in the end to drop potentially grabbed
references to them.
Cc: stable(a)vger.kernel.org # 5.9+
Signed-off-by: Pavel Begunkov <asml.silence(a)gmail.com>
---
fs/io_uring.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 2f305c097bd5..7af74c1ec909 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -7056,6 +7056,7 @@ static int io_sq_thread(void *data)
if (sqt_spin || !time_after(jiffies, timeout)) {
io_run_task_work();
+ io_sq_thread_drop_mm_files();
cond_resched();
if (sqt_spin)
timeout = jiffies + sqd->sq_thread_idle;
@@ -7093,6 +7094,7 @@ static int io_sq_thread(void *data)
}
io_run_task_work();
+ io_sq_thread_drop_mm_files();
if (cur_css)
io_sq_thread_unassociate_blkcg();
--
2.24.0
This is the start of the stable review cycle for the 4.19.165 release.
There are 35 patches in this series, all will be posted as a response
to this one. If anyone has any issues with these being applied, please
let me know.
Responses should be made by Wed, 06 Jan 2021 15:56:52 +0000.
Anything received after that time might be too late.
The whole patch series can be found in one patch at:
https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.19.165-r…
or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.19.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
Linux 4.19.165-rc1
Hyeongseok Kim <hyeongseok(a)gmail.com>
dm verity: skip verity work if I/O error when system is shutting down
Takashi Iwai <tiwai(a)suse.de>
ALSA: pcm: Clear the full allocated memory at hw_params
Jessica Yu <jeyu(a)kernel.org>
module: delay kobject uevent until after module init call
Trond Myklebust <trond.myklebust(a)hammerspace.com>
NFSv4: Fix a pNFS layout related use-after-free race when freeing the inode
Qinglang Miao <miaoqinglang(a)huawei.com>
powerpc: sysdev: add missing iounmap() on error in mpic_msgr_probe()
Jan Kara <jack(a)suse.cz>
quota: Don't overflow quota file offsets
Miroslav Benes <mbenes(a)suse.cz>
module: set MODULE_STATE_GOING state when a module fails to load
Dinghao Liu <dinghao.liu(a)zju.edu.cn>
rtc: sun6i: Fix memleak in sun6i_rtc_clk_init
Boqun Feng <boqun.feng(a)gmail.com>
fcntl: Fix potential deadlock in send_sig{io, urg}()
Takashi Iwai <tiwai(a)suse.de>
ALSA: rawmidi: Access runtime->avail always in spinlock
Takashi Iwai <tiwai(a)suse.de>
ALSA: seq: Use bool for snd_seq_queue internal flags
Mauro Carvalho Chehab <mchehab+huawei(a)kernel.org>
media: gp8psk: initialize stats at power control logic
Anant Thazhemadam <anant.thazhemadam(a)gmail.com>
misc: vmw_vmci: fix kernel info-leak by initializing dbells in vmci_ctx_get_chkpt_doorbells()
Rustam Kovhaev <rkovhaev(a)gmail.com>
reiserfs: add check for an invalid ih_entry_count
Anant Thazhemadam <anant.thazhemadam(a)gmail.com>
Bluetooth: hci_h5: close serdev device and free hu in h5_close
Peter Zijlstra <peterz(a)infradead.org>
asm-generic/tlb: avoid potential double flush
Peter Zijlstra <peterz(a)infradead.org>
mm/mmu_gather: invalidate TLB correctly on batch allocation failure and flush
Aneesh Kumar K.V <aneesh.kumar(a)linux.ibm.com>
powerpc/mmu_gather: enable RCU_TABLE_FREE even for !SMP case
Peter Zijlstra <peterz(a)infradead.org>
asm-generic/tlb, arch: Invert CONFIG_HAVE_RCU_TABLE_INVALIDATE
Will Deacon <will.deacon(a)arm.com>
asm-generic/tlb: Track which levels of the page tables have been cleared
Peter Zijlstra <peterz(a)infradead.org>
asm-generic/tlb: Track freeing of page-table directories in struct mmu_gather
Johan Hovold <johan(a)kernel.org>
of: fix linker-section match-table corruption
Damien Le Moal <damien.lemoal(a)wdc.com>
null_blk: Fix zone size initialization
Souptick Joarder <jrdr.linux(a)gmail.com>
xen/gntdev.c: Mark pages as dirty
Christophe Leroy <christophe.leroy(a)csgroup.eu>
powerpc/bitops: Fix possible undefined behaviour with fls() and fls64()
Paolo Bonzini <pbonzini(a)redhat.com>
KVM: x86: reinstate vendor-agnostic check on SPEC_CTRL cpuid bits
Paolo Bonzini <pbonzini(a)redhat.com>
KVM: SVM: relax conditions for allowing MSR_IA32_SPEC_CTRL accesses
Petr Vorel <petr.vorel(a)gmail.com>
uapi: move constants from <linux/kernel.h> to <linux/const.h>
Jan Kara <jack(a)suse.cz>
ext4: don't remount read-only with errors=continue on reboot
Eric Auger <eric.auger(a)redhat.com>
vfio/pci: Move dummy_resources_list init in vfio_pci_probe()
Eric Biggers <ebiggers(a)google.com>
ubifs: prevent creating duplicate encrypted filenames
Eric Biggers <ebiggers(a)google.com>
f2fs: prevent creating duplicate encrypted filenames
Eric Biggers <ebiggers(a)google.com>
ext4: prevent creating duplicate encrypted filenames
Eric Biggers <ebiggers(a)google.com>
fscrypt: add fscrypt_is_nokey_name()
Kevin Vigor <kvigor(a)gmail.com>
md/raid10: initialize r10_bio->read_slot before use.
-------------
Diffstat:
Makefile | 4 +-
arch/Kconfig | 3 -
arch/powerpc/Kconfig | 2 +-
arch/powerpc/include/asm/bitops.h | 23 +++++-
arch/powerpc/include/asm/book3s/32/pgalloc.h | 8 ---
arch/powerpc/include/asm/book3s/64/pgalloc.h | 2 -
arch/powerpc/include/asm/nohash/32/pgalloc.h | 8 ---
arch/powerpc/include/asm/nohash/64/pgalloc.h | 9 +--
arch/powerpc/include/asm/tlb.h | 11 +++
arch/powerpc/mm/pgtable-book3s64.c | 7 --
arch/powerpc/sysdev/mpic_msgr.c | 2 +-
arch/sparc/include/asm/tlb_64.h | 9 +++
arch/x86/Kconfig | 1 -
arch/x86/kvm/cpuid.h | 14 ++++
arch/x86/kvm/svm.c | 9 +--
arch/x86/kvm/vmx.c | 6 +-
drivers/block/null_blk_zoned.c | 20 ++++--
drivers/bluetooth/hci_h5.c | 8 ++-
drivers/md/dm-verity-target.c | 12 +++-
drivers/md/raid10.c | 3 +-
drivers/media/usb/dvb-usb/gp8psk.c | 2 +-
drivers/misc/vmw_vmci/vmci_context.c | 2 +-
drivers/rtc/rtc-sun6i.c | 8 ++-
drivers/vfio/pci/vfio_pci.c | 3 +-
drivers/xen/gntdev.c | 17 +++--
fs/crypto/hooks.c | 10 +--
fs/ext4/namei.c | 3 +
fs/ext4/super.c | 14 ++--
fs/f2fs/f2fs.h | 2 +
fs/fcntl.c | 10 +--
fs/nfs/nfs4super.c | 2 +-
fs/nfs/pnfs.c | 33 ++++++++-
fs/nfs/pnfs.h | 5 ++
fs/quota/quota_tree.c | 8 +--
fs/reiserfs/stree.c | 6 ++
fs/ubifs/dir.c | 17 +++--
include/asm-generic/tlb.h | 103 +++++++++++++++++++++++----
include/linux/fscrypt_notsupp.h | 5 ++
include/linux/fscrypt_supp.h | 29 ++++++++
include/linux/of.h | 1 +
include/uapi/linux/const.h | 5 ++
include/uapi/linux/ethtool.h | 2 +-
include/uapi/linux/kernel.h | 9 +--
include/uapi/linux/lightnvm.h | 2 +-
include/uapi/linux/mroute6.h | 2 +-
include/uapi/linux/netfilter/x_tables.h | 2 +-
include/uapi/linux/netlink.h | 2 +-
include/uapi/linux/sysctl.h | 2 +-
kernel/module.c | 6 +-
mm/memory.c | 20 +++---
sound/core/pcm_native.c | 9 ++-
sound/core/rawmidi.c | 49 +++++++++----
sound/core/seq/seq_queue.h | 8 +--
53 files changed, 398 insertions(+), 161 deletions(-)
On Sat, Jan 09, 2021 at 05:23:22PM -0800, Linus Torvalds wrote:
>Ack, I think 436e980e2ed5 ("kbuild: don't hardcode depmod path") is
>stable material even if it doesn't fix a bug.
>
>Not only does the fix for that commit not make sense without the
>commit in the first place, but any environment that sets depmod
>somewhere else might well be an environment that still wants stable
>kernels.
>
>It may not be the traditional case, but there's little reason for the
>kernel build to force that /sbin/depmod location.
I'll take it, thanks!
--
Thanks,
Sasha
The page_huge_active() can be called from scan_movable_pages() which
do not hold a reference count to the HugeTLB page. So when we call
page_huge_active() from scan_movable_pages(), the HugeTLB page can
be freed parallel. Then we will trigger a BUG_ON which is in the
page_huge_active() when CONFIG_DEBUG_VM is enabled. Just remove the
VM_BUG_ON_PAGE.
Fixes: 7e1f049efb86 ("mm: hugetlb: cleanup using paeg_huge_active()")
Signed-off-by: Muchun Song <songmuchun(a)bytedance.com>
Reviewed-by: Mike Kravetz <mike.kravetz(a)oracle.com>
Acked-by: Michal Hocko <mhocko(a)suse.com>
Cc: stable(a)vger.kernel.org
---
mm/hugetlb.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index e7ed30afbb8f..5940bf0c49b9 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1360,8 +1360,7 @@ struct hstate *size_to_hstate(unsigned long size)
*/
bool page_huge_active(struct page *page)
{
- VM_BUG_ON_PAGE(!PageHuge(page), page);
- return PageHead(page) && PagePrivate(&page[1]);
+ return PageHeadHuge(page) && PagePrivate(&page[1]);
}
/* never called for tail page */
--
2.11.0
There is a race between isolate_huge_page() and __free_huge_page().
CPU0: CPU1:
if (PageHuge(page))
put_page(page)
__free_huge_page(page)
spin_lock(&hugetlb_lock)
update_and_free_page(page)
set_compound_page_dtor(page,
NULL_COMPOUND_DTOR)
spin_unlock(&hugetlb_lock)
isolate_huge_page(page)
// trigger BUG_ON
VM_BUG_ON_PAGE(!PageHead(page), page)
spin_lock(&hugetlb_lock)
page_huge_active(page)
// trigger BUG_ON
VM_BUG_ON_PAGE(!PageHuge(page), page)
spin_unlock(&hugetlb_lock)
When we isolate a HugeTLB page on CPU0. Meanwhile, we free it to the
buddy allocator on CPU1. Then, we can trigger a BUG_ON on CPU0. Because
it is already freed to the buddy allocator.
Fixes: c8721bbbdd36 ("mm: memory-hotplug: enable memory hotplug to handle hugepage")
Signed-off-by: Muchun Song <songmuchun(a)bytedance.com>
Reviewed-by: Mike Kravetz <mike.kravetz(a)oracle.com>
Acked-by: Michal Hocko <mhocko(a)suse.com>
Cc: stable(a)vger.kernel.org
---
mm/hugetlb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index a176ceed55f1..e7ed30afbb8f 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -5575,9 +5575,9 @@ bool isolate_huge_page(struct page *page, struct list_head *list)
{
bool ret = true;
- VM_BUG_ON_PAGE(!PageHead(page), page);
spin_lock(&hugetlb_lock);
- if (!page_huge_active(page) || !get_page_unless_zero(page)) {
+ if (!PageHeadHuge(page) || !page_huge_active(page) ||
+ !get_page_unless_zero(page)) {
ret = false;
goto unlock;
}
--
2.11.0
Hi Greg, Sasha,
This series contains some scripts/gdb/ fixes that are already present in
newer stable kernels.
Thanks!
André Draszik (1):
scripts/gdb: make lx-dmesg command work (reliably)
Du Changbin (1):
scripts/gdb: fix lx-version string output
Leonard Crestez (2):
scripts/gdb: lx-dmesg: cast log_buf to void* for addr fetch
scripts/gdb: lx-dmesg: use explicit encoding=utf8 errors=replace
scripts/gdb/linux/dmesg.py | 22 +++++++++++++++-------
scripts/gdb/linux/proc.py | 2 +-
2 files changed, 16 insertions(+), 8 deletions(-)
--
2.25.1
This is the start of the stable review cycle for the 5.10.6 release.
There are 20 patches in this series, all will be posted as a response
to this one. If anyone has any issues with these being applied, please
let me know.
Responses should be made by Sat, 09 Jan 2021 14:30:35 +0000.
Anything received after that time might be too late.
The whole patch series can be found in one patch at:
https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.10.6-rc1…
or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <gregkh(a)linuxfoundation.org>
Linux 5.10.6-rc1
Zhang Xiaohui <ruc_zhangxiaohui(a)163.com>
mwifiex: Fix possible buffer overflows in mwifiex_cmd_802_11_ad_hoc_start
Eric W. Biederman <ebiederm(a)xmission.com>
exec: Transform exec_update_mutex into a rw_semaphore
Eric W. Biederman <ebiederm(a)xmission.com>
rwsem: Implement down_read_interruptible
Eric W. Biederman <ebiederm(a)xmission.com>
rwsem: Implement down_read_killable_nested
peterz(a)infradead.org <peterz(a)infradead.org>
perf: Break deadlock involving exec_update_mutex
Miklos Szeredi <mszeredi(a)redhat.com>
fuse: fix bad inode
Jason Gunthorpe <jgg(a)ziepe.ca>
RDMA/siw,rxe: Make emulated devices virtual in the device tree
Christoph Hellwig <hch(a)lst.de>
RDMA/core: remove use of dma_virt_ops
Stanley Chu <stanley.chu(a)mediatek.com>
scsi: ufs: Re-enable WriteBooster after device reset
Adrian Hunter <adrian.hunter(a)intel.com>
scsi: ufs: Allow an error return value from ->device_reset()
Imre Deak <imre.deak(a)intel.com>
drm/i915/tgl: Fix Combo PHY DPLL fractional divider for 38.4MHz ref clock
Takashi Iwai <tiwai(a)suse.de>
ALSA: hda/hdmi: Fix incorrect mutex unlock in silent_stream_disable()
Kailang Yang <kailang(a)realtek.com>
ALSA: hda/realtek - Modify Dell platform name
Edward Vear <edwardvear(a)gmail.com>
Bluetooth: Fix attempting to set RPA timeout when unsupported
Josh Poimboeuf <jpoimboe(a)redhat.com>
kdev_t: always inline major/minor helper functions
Rasmus Villemoes <rasmus.villemoes(a)prevas.dk>
dt-bindings: rtc: add reset-source property
Uwe Kleine-König <u.kleine-koenig(a)pengutronix.de>
rtc: pcf2127: only use watchdog when explicitly available
Uwe Kleine-König <u.kleine-koenig(a)pengutronix.de>
rtc: pcf2127: move watchdog initialisation to a separate function
Felix Fietkau <nbd(a)nbd.name>
Revert "mtd: spinand: Fix OOB read"
Alex Deucher <alexdeucher(a)gmail.com>
Revert "drm/amd/display: Fix memory leaks in S3 resume"
-------------
Diffstat:
Documentation/devicetree/bindings/rtc/rtc.yaml | 5 ++
Makefile | 4 +-
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 +-
drivers/gpu/drm/i915/display/intel_dpll_mgr.c | 41 ++++++++-----
drivers/infiniband/core/device.c | 43 ++++++-------
drivers/infiniband/core/rw.c | 5 +-
drivers/infiniband/sw/rdmavt/Kconfig | 1 -
drivers/infiniband/sw/rdmavt/mr.c | 6 +-
drivers/infiniband/sw/rdmavt/vt.c | 8 ---
drivers/infiniband/sw/rxe/Kconfig | 1 -
drivers/infiniband/sw/rxe/rxe_net.c | 12 ----
drivers/infiniband/sw/rxe/rxe_verbs.c | 8 ---
drivers/infiniband/sw/rxe/rxe_verbs.h | 1 -
drivers/infiniband/sw/siw/Kconfig | 1 -
drivers/infiniband/sw/siw/siw.h | 1 -
drivers/infiniband/sw/siw/siw_main.c | 26 +-------
drivers/mtd/nand/spi/core.c | 4 --
drivers/net/wireless/marvell/mwifiex/join.c | 2 +
drivers/nvme/target/rdma.c | 3 +-
drivers/rtc/rtc-pcf2127.c | 57 ++++++++++--------
drivers/scsi/ufs/ufs-mediatek.c | 4 +-
drivers/scsi/ufs/ufs-qcom.c | 6 +-
drivers/scsi/ufs/ufshcd.h | 16 +++--
fs/exec.c | 12 ++--
fs/fuse/acl.c | 6 ++
fs/fuse/dir.c | 37 ++++++++++--
fs/fuse/file.c | 19 +++---
fs/fuse/fuse_i.h | 12 ++++
fs/fuse/inode.c | 4 +-
fs/fuse/readdir.c | 4 +-
fs/fuse/xattr.c | 9 +++
fs/proc/base.c | 10 ++--
include/linux/kdev_t.h | 22 +++----
include/linux/rwsem.h | 3 +
include/linux/sched/signal.h | 11 ++--
include/rdma/ib_verbs.h | 73 ++++++++++++++++-------
init/init_task.c | 2 +-
kernel/events/core.c | 52 ++++++++--------
kernel/fork.c | 6 +-
kernel/kcmp.c | 30 +++++-----
kernel/locking/rwsem.c | 40 +++++++++++++
kernel/pid.c | 4 +-
net/bluetooth/hci_core.c | 2 +-
sound/pci/hda/patch_hdmi.c | 2 +-
sound/pci/hda/patch_realtek.c | 2 +-
45 files changed, 363 insertions(+), 257 deletions(-)
Since commit 55567976629e ("genirq/irqdomain: Allow partial trimming of
irq_data hierarchy") the irq_data chain is valided.
The irq_domain_trim_hierarchy() function doesn't consider the irq + ipi
domain hierarchy as valid, since the ipi domain has the irq domain set
as parent, but the parent domain has no chip set. Hence the boot ends in
a kernel panic.
Set the chip for the parent domain as it is done in the mips gic irq
driver, to have a valid irq_data chain.
Fixes: 3838a547fda2 ("irqchip: mips-cpu: Introduce IPI IRQ domain support")
Cc: <stable(a)vger.kernel.org> # v5.10+
Signed-off-by: Mathias Kresin <dev(a)kresin.me>
---
drivers/irqchip/irq-mips-cpu.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/drivers/irqchip/irq-mips-cpu.c b/drivers/irqchip/irq-mips-cpu.c
index 95d4fd8f7a96..0bbb0b2d0dd5 100644
--- a/drivers/irqchip/irq-mips-cpu.c
+++ b/drivers/irqchip/irq-mips-cpu.c
@@ -197,6 +197,13 @@ static int mips_cpu_ipi_alloc(struct irq_domain *domain, unsigned int virq,
if (ret)
return ret;
+ ret = irq_domain_set_hwirq_and_chip(domain->parent, virq + i, hwirq,
+ &mips_mt_cpu_irq_controller,
+ NULL);
+
+ if (ret)
+ return ret;
+
ret = irq_set_irq_type(virq + i, IRQ_TYPE_LEVEL_HIGH);
if (ret)
return ret;
--
2.25.1
On Fri, 8 Jan 2021 10:29:01 -0600, Suman Anna wrote:
> The TI PRUSS INTC irqchip driver handles the local interrupt controller
> which is a child device of it's parent PRUSS/ICSSG device. The driver
> was upstreamed in parallel with the PRUSS platform driver, and was
> configurable independently previously. The PRUSS interrupt controller
> is an integral part of the overall PRUSS software architecture, and is
> not useful at all by itself.
>
> [...]
Applied to irq/irqchip-next, thanks!
[1/1] irqchip: Simplify the TI_PRUSS_INTC Kconfig
commit: b8e594fa20d2e33d40c7a8c7c106549a35c38972
Cheers,
M.
--
Without deviation from the norm, progress is not possible.
bfq_setup_cooperator() uses bfqd->in_serv_last_pos so detect whether it
makes sense to merge current bfq queue with the in-service queue.
However if the in-service queue is freshly scheduled and didn't dispatch
any requests yet, bfqd->in_serv_last_pos is stale and contains value
from the previously scheduled bfq queue which can thus result in a bogus
decision that the two queues should be merged. This bug can be observed
for example with the following fio jobfile:
[global]
direct=0
ioengine=sync
invalidate=1
size=1g
rw=read
[reader]
numjobs=4
directory=/mnt
where the 4 processes will end up in the one shared bfq queue although
they do IO to physically very distant files (for some reason I was able to
observe this only with slice_idle=1ms setting).
Fix the problem by invalidating bfqd->in_serv_last_pos when switching
in-service queue.
Fixes: 058fdecc6de7 ("block, bfq: fix in-service-queue check for queue merging")
CC: stable(a)vger.kernel.org
Signed-off-by: Jan Kara <jack(a)suse.cz>
---
block/bfq-iosched.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 3d411716d7ee..50017275915f 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -2937,6 +2937,7 @@ static void __bfq_set_in_service_queue(struct bfq_data *bfqd,
}
bfqd->in_service_queue = bfqq;
+ bfqd->in_serv_last_pos = 0;
}
/*
--
2.16.4
The device link device's name was of the form:
<supplier-dev-name>--<consumer-dev-name>
This can cause name collision as reported here [1] as device names are
not globally unique. Since device names have to be unique within the
bus/class, add the bus/class name as a prefix to the device names used to
construct the device link device name.
So the devuce link device's name will be of the form:
<supplier-bus-name>:<supplier-dev-name>--<consumer-bus-name>:<consumer-dev-name>
[1] - https://lore.kernel.org/lkml/20201229033440.32142-1-michael@walle.cc/
Cc: stable(a)vger.kernel.org
Fixes: 287905e68dd2 ("driver core: Expose device link details in sysfs")
Reported-by: Michael Walle <michael(a)walle.cc>
Signed-off-by: Saravana Kannan <saravanak(a)google.com>
---
Documentation/ABI/testing/sysfs-class-devlink | 4 +--
.../ABI/testing/sysfs-devices-consumer | 5 ++--
.../ABI/testing/sysfs-devices-supplier | 5 ++--
drivers/base/core.c | 27 ++++++++++---------
include/linux/device.h | 12 +++++++++
5 files changed, 35 insertions(+), 18 deletions(-)
diff --git a/Documentation/ABI/testing/sysfs-class-devlink b/Documentation/ABI/testing/sysfs-class-devlink
index b662f747c83e..8a21ce515f61 100644
--- a/Documentation/ABI/testing/sysfs-class-devlink
+++ b/Documentation/ABI/testing/sysfs-class-devlink
@@ -5,8 +5,8 @@ Description:
Provide a place in sysfs for the device link objects in the
kernel at any given time. The name of a device link directory,
denoted as ... above, is of the form <supplier>--<consumer>
- where <supplier> is the supplier device name and <consumer> is
- the consumer device name.
+ where <supplier> is the supplier bus:device name and <consumer>
+ is the consumer bus:device name.
What: /sys/class/devlink/.../auto_remove_on
Date: May 2020
diff --git a/Documentation/ABI/testing/sysfs-devices-consumer b/Documentation/ABI/testing/sysfs-devices-consumer
index 1f06d74d1c3c..0809fda092e6 100644
--- a/Documentation/ABI/testing/sysfs-devices-consumer
+++ b/Documentation/ABI/testing/sysfs-devices-consumer
@@ -4,5 +4,6 @@ Contact: Saravana Kannan <saravanak(a)google.com>
Description:
The /sys/devices/.../consumer:<consumer> are symlinks to device
links where this device is the supplier. <consumer> denotes the
- name of the consumer in that device link. There can be zero or
- more of these symlinks for a given device.
+ name of the consumer in that device link and is of the form
+ bus:device name. There can be zero or more of these symlinks
+ for a given device.
diff --git a/Documentation/ABI/testing/sysfs-devices-supplier b/Documentation/ABI/testing/sysfs-devices-supplier
index a919e0db5e90..207f5972e98d 100644
--- a/Documentation/ABI/testing/sysfs-devices-supplier
+++ b/Documentation/ABI/testing/sysfs-devices-supplier
@@ -4,5 +4,6 @@ Contact: Saravana Kannan <saravanak(a)google.com>
Description:
The /sys/devices/.../supplier:<supplier> are symlinks to device
links where this device is the consumer. <supplier> denotes the
- name of the supplier in that device link. There can be zero or
- more of these symlinks for a given device.
+ name of the supplier in that device link and is of the form
+ bus:device name. There can be zero or more of these symlinks
+ for a given device.
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 25e08e5f40bd..47a6faf1605a 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -456,7 +456,9 @@ static int devlink_add_symlinks(struct device *dev,
struct device *con = link->consumer;
char *buf;
- len = max(strlen(dev_name(sup)), strlen(dev_name(con)));
+ len = max(strlen(dev_bus_name(sup)) + strlen(dev_name(sup)),
+ strlen(dev_bus_name(con)) + strlen(dev_name(con)));
+ len += strlen(":");
len += strlen("supplier:") + 1;
buf = kzalloc(len, GFP_KERNEL);
if (!buf)
@@ -470,12 +472,12 @@ static int devlink_add_symlinks(struct device *dev,
if (ret)
goto err_con;
- snprintf(buf, len, "consumer:%s", dev_name(con));
+ snprintf(buf, len, "consumer:%s:%s", dev_bus_name(con), dev_name(con));
ret = sysfs_create_link(&sup->kobj, &link->link_dev.kobj, buf);
if (ret)
goto err_con_dev;
- snprintf(buf, len, "supplier:%s", dev_name(sup));
+ snprintf(buf, len, "supplier:%s:%s", dev_bus_name(sup), dev_name(sup));
ret = sysfs_create_link(&con->kobj, &link->link_dev.kobj, buf);
if (ret)
goto err_sup_dev;
@@ -483,7 +485,7 @@ static int devlink_add_symlinks(struct device *dev,
goto out;
err_sup_dev:
- snprintf(buf, len, "consumer:%s", dev_name(con));
+ snprintf(buf, len, "consumer:%s:%s", dev_bus_name(con), dev_name(con));
sysfs_remove_link(&sup->kobj, buf);
err_con_dev:
sysfs_remove_link(&link->link_dev.kobj, "consumer");
@@ -506,7 +508,9 @@ static void devlink_remove_symlinks(struct device *dev,
sysfs_remove_link(&link->link_dev.kobj, "consumer");
sysfs_remove_link(&link->link_dev.kobj, "supplier");
- len = max(strlen(dev_name(sup)), strlen(dev_name(con)));
+ len = max(strlen(dev_bus_name(sup)) + strlen(dev_name(sup)),
+ strlen(dev_bus_name(con)) + strlen(dev_name(con)));
+ len += strlen(":");
len += strlen("supplier:") + 1;
buf = kzalloc(len, GFP_KERNEL);
if (!buf) {
@@ -514,9 +518,9 @@ static void devlink_remove_symlinks(struct device *dev,
return;
}
- snprintf(buf, len, "supplier:%s", dev_name(sup));
+ snprintf(buf, len, "supplier:%s:%s", dev_bus_name(sup), dev_name(sup));
sysfs_remove_link(&con->kobj, buf);
- snprintf(buf, len, "consumer:%s", dev_name(con));
+ snprintf(buf, len, "consumer:%s:%s", dev_bus_name(con), dev_name(con));
sysfs_remove_link(&sup->kobj, buf);
kfree(buf);
}
@@ -737,8 +741,9 @@ struct device_link *device_link_add(struct device *consumer,
link->link_dev.class = &devlink_class;
device_set_pm_not_required(&link->link_dev);
- dev_set_name(&link->link_dev, "%s--%s",
- dev_name(supplier), dev_name(consumer));
+ dev_set_name(&link->link_dev, "%s:%s--%s:%s",
+ dev_bus_name(supplier), dev_name(supplier),
+ dev_bus_name(consumer), dev_name(consumer));
if (device_register(&link->link_dev)) {
put_device(consumer);
put_device(supplier);
@@ -1808,9 +1813,7 @@ const char *dev_driver_string(const struct device *dev)
* never change once they are set, so they don't need special care.
*/
drv = READ_ONCE(dev->driver);
- return drv ? drv->name :
- (dev->bus ? dev->bus->name :
- (dev->class ? dev->class->name : ""));
+ return drv ? drv->name : dev_bus_name(dev);
}
EXPORT_SYMBOL(dev_driver_string);
diff --git a/include/linux/device.h b/include/linux/device.h
index 89bb8b84173e..1779f90eeb4c 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -609,6 +609,18 @@ static inline const char *dev_name(const struct device *dev)
return kobject_name(&dev->kobj);
}
+/**
+ * dev_bus_name - Return a device's bus/class name, if at all possible
+ * @dev: struct device to get the bus/class name of
+ *
+ * Will return the name of the bus/class the device is attached to. If it is
+ * not attached to a bus/class, an empty string will be returned.
+ */
+static inline const char *dev_bus_name(const struct device *dev)
+{
+ return dev->bus ? dev->bus->name : (dev->class ? dev->class->name : "");
+}
+
__printf(2, 3) int dev_set_name(struct device *dev, const char *name, ...);
#ifdef CONFIG_NUMA
--
2.30.0.284.gd98b1dd5eaa7-goog
Corrected to <stable(a)vger.kernel.org>.
- Sedat -
---------- Forwarded message ---------
From: Sedat Dilek <sedat.dilek(a)gmail.com>
Date: Sun, Jan 10, 2021 at 2:18 AM
Subject: depmod fixes for linux-stable releases
To: Sasha Levin <sashal(a)kernel.org>, Greg Kroah-Hartman
<gregkh(a)linuxfoundation.org>
Cc: Linus Torvalds <torvalds(a)linux-foundation.org>,
<linux-kernel(a)vger.kernel.org>, <linux-stable(a)vger.kernel.org>
Hi,
I was CCed on the "depmod: handle the case of /sbin/depmod without
/sbin in PATH" changes to linux-stable releases.
Do you mind also pushing...?
commit 436e980e2ed526832de822cbf13c317a458b78e1
kbuild: don't hardcode depmod path
That was the origin for the depmod follow-up.
Thanks.
Regards,
- Sedat -
[1] https://git.kernel.org/linus/436e980e2ed526832de822cbf13c317a458b78e1
For all PCI functions on the netxen_nic adapter, interrupt
mode (INTx or MSI) configuration is dependent on what has
been configured by the PCI function zero in the shared
interrupt register, as these adapters do not support mixed
mode interrupts among the functions of a given adapter.
Logic for setting MSI/MSI-x interrupt mode in the shared interrupt
register based on PCI function id zero check is not appropriate for
all family of netxen adapters, as for some of the netxen family
adapters PCI function zero is not really meant to be probed/loaded
in the host but rather just act as a management function on the device,
which caused all the other PCI functions on the adapter to always use
legacy interrupt (INTx) mode instead of choosing MSI/MSI-x interrupt mode.
This patch replaces that check with port number so that for all
type of adapters driver attempts for MSI/MSI-x interrupt modes.
Fixes: b37eb210c076 ("netxen_nic: Avoid mixed mode interrupts")
Signed-off-by: Manish Chopra <manishc(a)marvell.com>
Signed-off-by: Igor Russkikh <irusskikh(a)marvell.com>
---
drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c | 7 +------
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index f218477..d258e0c 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -564,11 +564,6 @@ static int netxen_set_features(struct net_device *dev,
.ndo_set_features = netxen_set_features,
};
-static inline bool netxen_function_zero(struct pci_dev *pdev)
-{
- return (PCI_FUNC(pdev->devfn) == 0) ? true : false;
-}
-
static inline void netxen_set_interrupt_mode(struct netxen_adapter *adapter,
u32 mode)
{
@@ -664,7 +659,7 @@ static int netxen_setup_intr(struct netxen_adapter *adapter)
netxen_initialize_interrupt_registers(adapter);
netxen_set_msix_bit(pdev, 0);
- if (netxen_function_zero(pdev)) {
+ if (adapter->portnum == 0) {
if (!netxen_setup_msi_interrupts(adapter, num_msix))
netxen_set_interrupt_mode(adapter, NETXEN_MSI_MODE);
else
--
1.8.3.1