[<prev] [next>] [day] [month] [year] [list]
Message-ID: <aGznjhBVCkifc6BD@infradead.org>
Date: Tue, 8 Jul 2025 02:40:30 -0700
From: Christoph Hellwig <hch@...radead.org>
To: "ywen.chen" <ywen.chen@...mail.com>
Cc: Christoph Hellwig <hch@...radead.org>,
Eric Biggers <ebiggers@...nel.org>, brauner <brauner@...nel.org>,
tytso <tytso@....edu>, linux-kernel <linux-kernel@...r.kernel.org>,
linux-f2fs-devel <linux-f2fs-devel@...ts.sourceforge.net>,
"adilger.kernel" <adilger.kernel@...ger.ca>,
viro <viro@...iv.linux.org.uk>,
linux-fsdevel <linux-fsdevel@...r.kernel.org>,
jaegeuk <jaegeuk@...nel.org>,
linux-ext4 <linux-ext4@...r.kernel.org>
Subject: Re: 回复: [PATCH v3 1/2] libfs: reduce the number of memory
allocations in generic_ci_match
On Tue, Jul 08, 2025 at 10:36:07AM +0800, ywen.chen wrote:
> 1. Use the latest version of the fsck.f2fs tool to correct
> the file system.
> 2. Use a relatively new version of the kernel. (For example,
> linear search cannot be turned off in v6.6)
>
>
> The performance gain of this commit is very obvious in scenarios
> where linear search is not turned off. In scenarios where linear
> search is turned off, no performance problems will be introduced
> either.
Turning off hashed lookups was a stupid idea and should not last. Hashed
directory lookups are the one file system improvement that significantly
improved lookup performance. So don't work around it, because otherwise
you will be creating more optimizations to work around the lack of
hashed lookups.
Powered by blists - more mailing lists