diff mbox series

[v4,2/2] virtiofs: use GFP_NOFS when enqueuing request through kworker

Message ID 20240831093750.1593871-3-houtao@huaweicloud.com (mailing list archive)
State New
Headers show
Series virtiofs: fix the warning for kernel direct IO | expand

Commit Message

Hou Tao Aug. 31, 2024, 9:37 a.m. UTC
From: Hou Tao <houtao1@huawei.com>

When invoking virtio_fs_enqueue_req() through kworker, both the
allocation of the sg array and the bounce buffer still use GFP_ATOMIC.
Considering the size of the sg array may be greater than PAGE_SIZE, use
GFP_NOFS instead of GFP_ATOMIC to lower the possibility of memory
allocation failure and to avoid unnecessarily depleting the atomic
reserves. GFP_NOFS is not passed to virtio_fs_enqueue_req() directly,
GFP_KERNEL and memalloc_nofs_{save|restore} helpers are used instead.

It may seem OK to pass GFP_NOFS to virtio_fs_enqueue_req() as well when
queuing the request for the first time, but this is not the case. The
reason is that fuse_request_queue_background() may call
->queue_request_and_unlock() while holding fc->bg_lock, which is a
spin-lock. Therefore, still use GFP_ATOMIC for it.

Signed-off-by: Hou Tao <houtao1@huawei.com>
---
 fs/fuse/virtio_fs.c | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

Comments

Jingbo Xu Sept. 3, 2024, 9:34 a.m. UTC | #1
On 8/31/24 5:37 PM, Hou Tao wrote:
> From: Hou Tao <houtao1@huawei.com>
> 
> When invoking virtio_fs_enqueue_req() through kworker, both the
> allocation of the sg array and the bounce buffer still use GFP_ATOMIC.
> Considering the size of the sg array may be greater than PAGE_SIZE, use
> GFP_NOFS instead of GFP_ATOMIC to lower the possibility of memory
> allocation failure and to avoid unnecessarily depleting the atomic
> reserves. GFP_NOFS is not passed to virtio_fs_enqueue_req() directly,
> GFP_KERNEL and memalloc_nofs_{save|restore} helpers are used instead.
> 
> It may seem OK to pass GFP_NOFS to virtio_fs_enqueue_req() as well when
> queuing the request for the first time, but this is not the case. The
> reason is that fuse_request_queue_background() may call
> ->queue_request_and_unlock() while holding fc->bg_lock, which is a
> spin-lock. Therefore, still use GFP_ATOMIC for it.

Actually, .wake_pending_and_unlock() is called under fiq->lock and
GFP_ATOMIC is requisite.


> 
> Signed-off-by: Hou Tao <houtao1@huawei.com>
> ---
>  fs/fuse/virtio_fs.c | 24 +++++++++++++++---------
>  1 file changed, 15 insertions(+), 9 deletions(-)
> 
> diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c
> index 43d66ab5e891..9bc48b3ca384 100644
> --- a/fs/fuse/virtio_fs.c
> +++ b/fs/fuse/virtio_fs.c
> @@ -95,7 +95,8 @@ struct virtio_fs_req_work {
>  };
>  
>  static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
> -				 struct fuse_req *req, bool in_flight);
> +				 struct fuse_req *req, bool in_flight,
> +				 gfp_t gfp);
>  
>  static const struct constant_table dax_param_enums[] = {
>  	{"always",	FUSE_DAX_ALWAYS },
> @@ -439,6 +440,8 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
>  
>  	/* Dispatch pending requests */
>  	while (1) {
> +		unsigned int flags;
> +
>  		spin_lock(&fsvq->lock);
>  		req = list_first_entry_or_null(&fsvq->queued_reqs,
>  					       struct fuse_req, list);
> @@ -449,7 +452,9 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
>  		list_del_init(&req->list);
>  		spin_unlock(&fsvq->lock);
>  
> -		ret = virtio_fs_enqueue_req(fsvq, req, true);
> +		flags = memalloc_nofs_save();
> +		ret = virtio_fs_enqueue_req(fsvq, req, true, GFP_KERNEL);
> +		memalloc_nofs_restore(flags);
>  		if (ret < 0) {
>  			if (ret == -ENOSPC) {
>  				spin_lock(&fsvq->lock);
> @@ -550,7 +555,7 @@ static void virtio_fs_hiprio_dispatch_work(struct work_struct *work)
>  }
>  
>  /* Allocate and copy args into req->argbuf */
> -static int copy_args_to_argbuf(struct fuse_req *req)
> +static int copy_args_to_argbuf(struct fuse_req *req, gfp_t gfp)
>  {
>  	struct fuse_args *args = req->args;
>  	unsigned int offset = 0;
> @@ -564,7 +569,7 @@ static int copy_args_to_argbuf(struct fuse_req *req)
>  	len = fuse_len_args(num_in, (struct fuse_arg *) args->in_args) +
>  	      fuse_len_args(num_out, args->out_args);
>  
> -	req->argbuf = kmalloc(len, GFP_ATOMIC);
> +	req->argbuf = kmalloc(len, gfp);
>  	if (!req->argbuf)
>  		return -ENOMEM;
>  
> @@ -1239,7 +1244,8 @@ static unsigned int sg_init_fuse_args(struct scatterlist *sg,
>  
>  /* Add a request to a virtqueue and kick the device */
>  static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
> -				 struct fuse_req *req, bool in_flight)
> +				 struct fuse_req *req, bool in_flight,
> +				 gfp_t gfp)
>  {
>  	/* requests need at least 4 elements */
>  	struct scatterlist *stack_sgs[6];
> @@ -1260,8 +1266,8 @@ static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
>  	/* Does the sglist fit on the stack? */
>  	total_sgs = sg_count_fuse_req(req);
>  	if (total_sgs > ARRAY_SIZE(stack_sgs)) {
> -		sgs = kmalloc_array(total_sgs, sizeof(sgs[0]), GFP_ATOMIC);
> -		sg = kmalloc_array(total_sgs, sizeof(sg[0]), GFP_ATOMIC);
> +		sgs = kmalloc_array(total_sgs, sizeof(sgs[0]), gfp);
> +		sg = kmalloc_array(total_sgs, sizeof(sg[0]), gfp);
>  		if (!sgs || !sg) {
>  			ret = -ENOMEM;
>  			goto out;
> @@ -1269,7 +1275,7 @@ static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
>  	}
>  
>  	/* Use a bounce buffer since stack args cannot be mapped */
> -	ret = copy_args_to_argbuf(req);
> +	ret = copy_args_to_argbuf(req, gfp);
>  	if (ret < 0)
>  		goto out;
>  
> @@ -1367,7 +1373,7 @@ __releases(fiq->lock)
>  		 queue_id);
>  
>  	fsvq = &fs->vqs[queue_id];
> -	ret = virtio_fs_enqueue_req(fsvq, req, false);
> +	ret = virtio_fs_enqueue_req(fsvq, req, false, GFP_ATOMIC);
>  	if (ret < 0) {
>  		if (ret == -ENOSPC) {
>  			/*

LGTM.

Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Hou Tao Sept. 4, 2024, 3:53 a.m. UTC | #2
On 9/3/2024 5:34 PM, Jingbo Xu wrote:
>
> On 8/31/24 5:37 PM, Hou Tao wrote:
>> From: Hou Tao <houtao1@huawei.com>
>>
>> When invoking virtio_fs_enqueue_req() through kworker, both the
>> allocation of the sg array and the bounce buffer still use GFP_ATOMIC.
>> Considering the size of the sg array may be greater than PAGE_SIZE, use
>> GFP_NOFS instead of GFP_ATOMIC to lower the possibility of memory
>> allocation failure and to avoid unnecessarily depleting the atomic
>> reserves. GFP_NOFS is not passed to virtio_fs_enqueue_req() directly,
>> GFP_KERNEL and memalloc_nofs_{save|restore} helpers are used instead.
>>
>> It may seem OK to pass GFP_NOFS to virtio_fs_enqueue_req() as well when
>> queuing the request for the first time, but this is not the case. The
>> reason is that fuse_request_queue_background() may call
>> ->queue_request_and_unlock() while holding fc->bg_lock, which is a
>> spin-lock. Therefore, still use GFP_ATOMIC for it.
> Actually, .wake_pending_and_unlock() is called under fiq->lock and
> GFP_ATOMIC is requisite.

Er, but virtio_fs_wake_pending_and_unlock() unlocks fiq->lock before
queuing the request.
>
>
>> Signed-off-by: Hou Tao <houtao1@huawei.com>
>> ---
>>  fs/fuse/virtio_fs.c | 24 +++++++++++++++---------
>>  1 file changed, 15 insertions(+), 9 deletions(-)
>>
>> diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c
>> index 43d66ab5e891..9bc48b3ca384 100644
>> --- a/fs/fuse/virtio_fs.c
>> +++ b/fs/fuse/virtio_fs.c
>> @@ -95,7 +95,8 @@ struct virtio_fs_req_work {
>>  };
>>  
>>  static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
>> -				 struct fuse_req *req, bool in_flight);
>> +				 struct fuse_req *req, bool in_flight,
>> +				 gfp_t gfp);
>>  
>>  static const struct constant_table dax_param_enums[] = {
>>  	{"always",	FUSE_DAX_ALWAYS },
>> @@ -439,6 +440,8 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
>>  
>>  	/* Dispatch pending requests */
>>  	while (1) {
>> +		unsigned int flags;
>> +
>>  		spin_lock(&fsvq->lock);
>>  		req = list_first_entry_or_null(&fsvq->queued_reqs,
>>  					       struct fuse_req, list);
>> @@ -449,7 +452,9 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
>>  		list_del_init(&req->list);
>>  		spin_unlock(&fsvq->lock);
>>  
>> -		ret = virtio_fs_enqueue_req(fsvq, req, true);
>> +		flags = memalloc_nofs_save();
>> +		ret = virtio_fs_enqueue_req(fsvq, req, true, GFP_KERNEL);
>> +		memalloc_nofs_restore(flags);
>>  		if (ret < 0) {
>>  			if (ret == -ENOSPC) {
>>  				spin_lock(&fsvq->lock);
>> @@ -550,7 +555,7 @@ static void virtio_fs_hiprio_dispatch_work(struct work_struct *work)
>>  }
>>  
>>  /* Allocate and copy args into req->argbuf */
>> -static int copy_args_to_argbuf(struct fuse_req *req)
>> +static int copy_args_to_argbuf(struct fuse_req *req, gfp_t gfp)
>>  {
>>  	struct fuse_args *args = req->args;
>>  	unsigned int offset = 0;
>> @@ -564,7 +569,7 @@ static int copy_args_to_argbuf(struct fuse_req *req)
>>  	len = fuse_len_args(num_in, (struct fuse_arg *) args->in_args) +
>>  	      fuse_len_args(num_out, args->out_args);
>>  
>> -	req->argbuf = kmalloc(len, GFP_ATOMIC);
>> +	req->argbuf = kmalloc(len, gfp);
>>  	if (!req->argbuf)
>>  		return -ENOMEM;
>>  
>> @@ -1239,7 +1244,8 @@ static unsigned int sg_init_fuse_args(struct scatterlist *sg,
>>  
>>  /* Add a request to a virtqueue and kick the device */
>>  static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
>> -				 struct fuse_req *req, bool in_flight)
>> +				 struct fuse_req *req, bool in_flight,
>> +				 gfp_t gfp)
>>  {
>>  	/* requests need at least 4 elements */
>>  	struct scatterlist *stack_sgs[6];
>> @@ -1260,8 +1266,8 @@ static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
>>  	/* Does the sglist fit on the stack? */
>>  	total_sgs = sg_count_fuse_req(req);
>>  	if (total_sgs > ARRAY_SIZE(stack_sgs)) {
>> -		sgs = kmalloc_array(total_sgs, sizeof(sgs[0]), GFP_ATOMIC);
>> -		sg = kmalloc_array(total_sgs, sizeof(sg[0]), GFP_ATOMIC);
>> +		sgs = kmalloc_array(total_sgs, sizeof(sgs[0]), gfp);
>> +		sg = kmalloc_array(total_sgs, sizeof(sg[0]), gfp);
>>  		if (!sgs || !sg) {
>>  			ret = -ENOMEM;
>>  			goto out;
>> @@ -1269,7 +1275,7 @@ static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
>>  	}
>>  
>>  	/* Use a bounce buffer since stack args cannot be mapped */
>> -	ret = copy_args_to_argbuf(req);
>> +	ret = copy_args_to_argbuf(req, gfp);
>>  	if (ret < 0)
>>  		goto out;
>>  
>> @@ -1367,7 +1373,7 @@ __releases(fiq->lock)
>>  		 queue_id);
>>  
>>  	fsvq = &fs->vqs[queue_id];
>> -	ret = virtio_fs_enqueue_req(fsvq, req, false);
>> +	ret = virtio_fs_enqueue_req(fsvq, req, false, GFP_ATOMIC);
>>  	if (ret < 0) {
>>  		if (ret == -ENOSPC) {
>>  			/*
> LGTM.
>
> Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com>

Thanks for the review.
>
>
Jingbo Xu Sept. 4, 2024, 12:12 p.m. UTC | #3
On 9/4/24 11:53 AM, Hou Tao wrote:
> 
> 
> On 9/3/2024 5:34 PM, Jingbo Xu wrote:
>>
>> On 8/31/24 5:37 PM, Hou Tao wrote:
>>> From: Hou Tao <houtao1@huawei.com>
>>>
>>> When invoking virtio_fs_enqueue_req() through kworker, both the
>>> allocation of the sg array and the bounce buffer still use GFP_ATOMIC.
>>> Considering the size of the sg array may be greater than PAGE_SIZE, use
>>> GFP_NOFS instead of GFP_ATOMIC to lower the possibility of memory
>>> allocation failure and to avoid unnecessarily depleting the atomic
>>> reserves. GFP_NOFS is not passed to virtio_fs_enqueue_req() directly,
>>> GFP_KERNEL and memalloc_nofs_{save|restore} helpers are used instead.
>>>
>>> It may seem OK to pass GFP_NOFS to virtio_fs_enqueue_req() as well when
>>> queuing the request for the first time, but this is not the case. The
>>> reason is that fuse_request_queue_background() may call
>>> ->queue_request_and_unlock() while holding fc->bg_lock, which is a
>>> spin-lock. Therefore, still use GFP_ATOMIC for it.
>> Actually, .wake_pending_and_unlock() is called under fiq->lock and
>> GFP_ATOMIC is requisite.
> 
> Er, but virtio_fs_wake_pending_and_unlock() unlocks fiq->lock before
> queuing the request.

Alright, I missed that :(
diff mbox series

Patch

diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c
index 43d66ab5e891..9bc48b3ca384 100644
--- a/fs/fuse/virtio_fs.c
+++ b/fs/fuse/virtio_fs.c
@@ -95,7 +95,8 @@  struct virtio_fs_req_work {
 };
 
 static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
-				 struct fuse_req *req, bool in_flight);
+				 struct fuse_req *req, bool in_flight,
+				 gfp_t gfp);
 
 static const struct constant_table dax_param_enums[] = {
 	{"always",	FUSE_DAX_ALWAYS },
@@ -439,6 +440,8 @@  static void virtio_fs_request_dispatch_work(struct work_struct *work)
 
 	/* Dispatch pending requests */
 	while (1) {
+		unsigned int flags;
+
 		spin_lock(&fsvq->lock);
 		req = list_first_entry_or_null(&fsvq->queued_reqs,
 					       struct fuse_req, list);
@@ -449,7 +452,9 @@  static void virtio_fs_request_dispatch_work(struct work_struct *work)
 		list_del_init(&req->list);
 		spin_unlock(&fsvq->lock);
 
-		ret = virtio_fs_enqueue_req(fsvq, req, true);
+		flags = memalloc_nofs_save();
+		ret = virtio_fs_enqueue_req(fsvq, req, true, GFP_KERNEL);
+		memalloc_nofs_restore(flags);
 		if (ret < 0) {
 			if (ret == -ENOSPC) {
 				spin_lock(&fsvq->lock);
@@ -550,7 +555,7 @@  static void virtio_fs_hiprio_dispatch_work(struct work_struct *work)
 }
 
 /* Allocate and copy args into req->argbuf */
-static int copy_args_to_argbuf(struct fuse_req *req)
+static int copy_args_to_argbuf(struct fuse_req *req, gfp_t gfp)
 {
 	struct fuse_args *args = req->args;
 	unsigned int offset = 0;
@@ -564,7 +569,7 @@  static int copy_args_to_argbuf(struct fuse_req *req)
 	len = fuse_len_args(num_in, (struct fuse_arg *) args->in_args) +
 	      fuse_len_args(num_out, args->out_args);
 
-	req->argbuf = kmalloc(len, GFP_ATOMIC);
+	req->argbuf = kmalloc(len, gfp);
 	if (!req->argbuf)
 		return -ENOMEM;
 
@@ -1239,7 +1244,8 @@  static unsigned int sg_init_fuse_args(struct scatterlist *sg,
 
 /* Add a request to a virtqueue and kick the device */
 static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
-				 struct fuse_req *req, bool in_flight)
+				 struct fuse_req *req, bool in_flight,
+				 gfp_t gfp)
 {
 	/* requests need at least 4 elements */
 	struct scatterlist *stack_sgs[6];
@@ -1260,8 +1266,8 @@  static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
 	/* Does the sglist fit on the stack? */
 	total_sgs = sg_count_fuse_req(req);
 	if (total_sgs > ARRAY_SIZE(stack_sgs)) {
-		sgs = kmalloc_array(total_sgs, sizeof(sgs[0]), GFP_ATOMIC);
-		sg = kmalloc_array(total_sgs, sizeof(sg[0]), GFP_ATOMIC);
+		sgs = kmalloc_array(total_sgs, sizeof(sgs[0]), gfp);
+		sg = kmalloc_array(total_sgs, sizeof(sg[0]), gfp);
 		if (!sgs || !sg) {
 			ret = -ENOMEM;
 			goto out;
@@ -1269,7 +1275,7 @@  static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
 	}
 
 	/* Use a bounce buffer since stack args cannot be mapped */
-	ret = copy_args_to_argbuf(req);
+	ret = copy_args_to_argbuf(req, gfp);
 	if (ret < 0)
 		goto out;
 
@@ -1367,7 +1373,7 @@  __releases(fiq->lock)
 		 queue_id);
 
 	fsvq = &fs->vqs[queue_id];
-	ret = virtio_fs_enqueue_req(fsvq, req, false);
+	ret = virtio_fs_enqueue_req(fsvq, req, false, GFP_ATOMIC);
 	if (ret < 0) {
 		if (ret == -ENOSPC) {
 			/*