Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk

To:	NeilBrown <neilb@xxxxxxx>
Subject:	Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock
From:	Eric Dumazet <eric.dumazet@xxxxxxxxx>
Date:	Tue, 15 Apr 2014 22:13:46 -0700
Cc:	linux-mm@xxxxxxxxx, linux-nfs@xxxxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, xfs@xxxxxxxxxxx, netdev@xxxxxxxxxxxxxxx
Delivered-to:	xfs@xxxxxxxxxxx
Dkim-signature:	v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=message-id:subject:from:to:cc:date:in-reply-to:references :content-type:content-transfer-encoding:mime-version; bh=xD9DIIaID91y+ESCvfXO5o/T2UrOqB3PRddDdod4zWQ=; b=hfAL7yISP8AIC3Dp2qe/J11jhTDeUkI/hm4Nf9b0KKZfVrx5C/VseUQY7feF2HTtdS OtoO5WgO3tZt1aTw7dQXV/z7sa+IE1W61FjqlH0JbBGpqQT/rL9AipASDbE8EsQpwQBn 8GkFbhUeMUcHfocSi1v0UGnVQ5FDXSIUTzKA0mdnAv+AieloZZBwt3aGMwtLf60HmE/E 4Rr8gAHNu4Gciucn6RNohqM5ciX4qeVCgxWpLv0fSx/mjWqDf+xgaezKFkWxYO4u0gVh ZFMtK1aFRDAmnUUWgSbWDGMkONpCR7rFPaAW/CsqtfRn5WIy71l3yccxJSNoGSrAXytx uzdw==
In-reply-to:	<20140416040336.10604.96000.stgit@xxxxxxxxxxxxxx>
References:	<20140416033623.10604.69237.stgit@xxxxxxxxxxxxxx> <20140416040336.10604.96000.stgit@xxxxxxxxxxxxxx>

On Wed, 2014-04-16 at 14:03 +1000, NeilBrown wrote:
> sk_lock can be taken while reclaiming memory (in nfsd for loop-back
> NFS mounts, and presumably in nfs), and memory can be allocated while
> holding sk_lock, at least via:
> 
>  inet_listen -> inet_csk_listen_start ->reqsk_queue_alloc
> 
> So to avoid deadlocks, always set PF_FSTRANS while holding sk_lock.
> 
> This deadlock was found by lockdep.

Wow, this is adding expensive stuff in fast path, only for nfsd :(

BTW, why should the current->flags should be saved on a socket field,
and not a current->save_flags. This really looks a thread property, not
a socket one.

Why nfsd could not have PF_FSTRANS in its current->flags ?

For applications handling millions of sockets, this makes a difference.

<Prev in Thread]	Current Thread	[Next in Thread>
Re: [PATCH 05/19] SUNRPC: track whether a request is coming from a loop-back interface., (continued) Re: [PATCH 05/19] SUNRPC: track whether a request is coming from a loop-back interface., NeilBrown [PATCH 06/19] nfsd: set PF_FSTRANS for nfsd threads., NeilBrown Re: [PATCH 06/19] nfsd: set PF_FSTRANS for nfsd threads., Peter Zijlstra [PATCH 07/19] nfsd and VM: use PF_LESS_THROTTLE to avoid throttle in shrink_inactive_list., NeilBrown [PATCH 08/19] Set PF_FSTRANS while write_cache_pages calls ->writepage, NeilBrown *[PATCH 09/19] XFS: ensure xfs_file__read cannot deadlock in memory allocation.*, NeilBrown* *Re: [PATCH 09/19] XFS: ensure xfs_file__read cannot deadlock in memory allocation.*, Dave Chinner* *Re: [PATCH 09/19] XFS: ensure xfs_file__read cannot deadlock in memory allocation.*, NeilBrown* *Re: [PATCH 09/19] XFS: ensure xfs_file__read cannot deadlock in memory allocation.*, Dave Chinner* [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, NeilBrown Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, Eric Dumazet <= Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, NeilBrown Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, David Miller Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, NeilBrown [PATCH 11/19] FS: set PF_FSTRANS while holding mmap_sem in exec.c, NeilBrown [PATCH 12/19] NET: set PF_FSTRANS while holding rtnl_lock, NeilBrown [PATCH 13/19] MM: set PF_FSTRANS while allocating per-cpu memory to avoid deadlock., NeilBrown Re: [PATCH 13/19] MM: set PF_FSTRANS while allocating per-cpu memory to avoid deadlock., Dave Chinner Re: [PATCH 13/19] MM: set PF_FSTRANS while allocating per-cpu memory to avoid deadlock., NeilBrown Re: [PATCH 13/19] MM: set PF_FSTRANS while allocating per-cpu memory to avoid deadlock., Dave Chinner [PATCH 14/19] driver core: set PF_FSTRANS while holding gdp_mutex, NeilBrown

<Prev in Thread]

Current Thread

[Next in Thread>

Previous by Date:	[PATCH -v2] check: add support for an external file containing tests to exclude, Theodore Ts'o
Next by Date:	Re: [PATCH 04/19] Make effect of PF_FSTRANS to disable __GFP_FS universal., Dave Chinner
Previous by Thread:	[PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, NeilBrown
Next by Thread:	Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, NeilBrown
Indexes:	[Date] [Thread] [Top] [All Lists]

Previous by Date:

[PATCH -v2] check: add support for an external file containing tests to exclude, Theodore Ts'o

Next by Date:

Re: [PATCH 04/19] Make effect of PF_FSTRANS to disable __GFP_FS universal., Dave Chinner

Previous by Thread:

[PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, NeilBrown

Next by Thread:

Re: [PATCH 10/19] NET: set PF_FSTRANS while holding sk_lock, NeilBrown

Indexes:

[Date] [Thread] [Top] [All Lists]