lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <7a6d2c91-2221-ce07-995b-9f648f9ba2d3@fastmail.fm>
Date:   Sat, 16 Sep 2023 13:15:06 +0200
From:   Bernd Schubert <bernd.schubert@...tmail.fm>
To:     Kemeng Shi <shikemeng@...weicloud.com>, miklos@...redi.hu,
        linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/2] fuse: move FR_WAITING set from
 fuse_request_queue_background to fuse_simple_background



On 9/13/23 11:22, Kemeng Shi wrote:
> Current way to set FR_WAITING in fuse_simple_background:
> fuse_simple_background
> 	if (args->force)
> 		fuse_request_alloc
> 		/* need to increase num_waiting before request is queued */
> 	else
> 		fuse_get_req
> 			atomic_inc(&fc->num_waiting);
> 			__set_bit(FR_WAITING, &req->flags);
> 
> 	fuse_request_queue_background
> 		if (!test_bit(FR_WAITING, &req->flags)
> 			__set_bit(FR_WAITING, &req->flags);
> 			atomic_inc(&fc->num_waiting);
> 
> We only need to increase num_waiting for force allocated reqeust in
> fuse_request_queue_background. Simply increase num_waiting in force block
> to remove unnecessary test_bit(FR_WAITING).
> This patch also makes it more intuitive to remove FR_WAITING usage in next
> commit.
> 
> Signed-off-by: Kemeng Shi <shikemeng@...weicloud.com>
> ---
>   fs/fuse/dev.c | 8 ++++----
>   1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
> index 1a8f82f478cb..59e1357d4880 100644
> --- a/fs/fuse/dev.c
> +++ b/fs/fuse/dev.c
> @@ -528,10 +528,6 @@ static bool fuse_request_queue_background(struct fuse_req *req)
>   	bool queued = false;
>   
>   	WARN_ON(!test_bit(FR_BACKGROUND, &req->flags));
> -	if (!test_bit(FR_WAITING, &req->flags)) {
> -		__set_bit(FR_WAITING, &req->flags);
> -		atomic_inc(&fc->num_waiting);
> -	}
>   	__set_bit(FR_ISREPLY, &req->flags);
>   	spin_lock(&fc->bg_lock);
>   	if (likely(fc->connected)) {
> @@ -553,10 +549,14 @@ int fuse_simple_background(struct fuse_mount *fm, struct fuse_args *args,
>   	struct fuse_req *req;
>   
>   	if (args->force) {
> +		struct fuse_conn *fc = fm->fc;
> +
>   		WARN_ON(!args->nocreds);
>   		req = fuse_request_alloc(fm, gfp_flags);
>   		if (!req)
>   			return -ENOMEM;
> +		atomic_inc(&fc->num_waiting);
> +		__set_bit(FR_WAITING, &req->flags);
>   		__set_bit(FR_BACKGROUND, &req->flags);
>   	} else {
>   		WARN_ON(args->nocreds);

Looks good to me.
Reviewed-by: Bernd Schubert <bschubert@....com>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ