linux-kernel - Re: [PATCH 1/3] btrfs: get rid of path allocation in btrfs_del_inode

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: <20250415155637.GG16750@suse.cz>
Date: Tue, 15 Apr 2025 17:56:37 +0200
From: David Sterba <dsterba@...e.cz>
To: Sun YangKai <sunk67188@...il.com>
Cc: frank.li@...o.com, clm@...com, dsterba@...e.com, josef@...icpanda.com,
	linux-btrfs@...r.kernel.org, linux-kernel@...r.kernel.org,
	neelx@...e.com
Subject: Re: [PATCH 1/3] btrfs: get rid of path allocation in
 btrfs_del_inode_extref()

On Tue, Apr 15, 2025 at 10:45:06PM +0800, Sun YangKai wrote:
> It seems a nice try to reduce path allocation and improve performance.
> 
> But it also seems make the code less maintainable. I would prefer to  have a 
> comment saying something like the @path argument is just for reuse the 
> btrfs_path allocation and only a released or empty btrfs_path should be used 
> here.

Yes, this should be there, though we use the pattern of passing existing
path to functions so this within what'd consider OK.

> Also, although the path passed is released, it seems the bit flags are still 
> passed, which makes the behavior of the functions a little different. But it 
> seems fine since those bit flags are never set in this code path.

Also a good point, the path should be in a pristine state, as if it were
just allocated. Releasing paths in other functions may want to keep the
bits but in this case we're crossing a function boundary and the same
assumptions may not be the same.

Release resets the ->nodes, so what's left is from ->slots until the the
end of the structure. And a helper for that would be desirable rather
than opencoding that.