From: Sun YangKai <sunk67188@gmail.com>
To: linux-btrfs <linux-btrfs@vger.kernel.org>
Subject: Re: [PATCH v2 0/3] btrfs: search_tree ioctl performance improvements and cleanups
Date: Tue, 12 Aug 2025 20:05:04 +0800 [thread overview]
Message-ID: <12722055.O9o76ZdvQC@saltykitkat> (raw)
In-Reply-To: <20250726135214.16000-1-sunk67188@gmail.com>
> This series optimizes the search_tree ioctl path used by tools like
> compsize and cleans up related code:
>
> Patch 1: Narrow loop variable scope
>
> Patch 2: Early exit for out-of-range keys
>
> Replace continue with early exit when keys exceed max_key
>
> Provide measurable performance improvements:
> Cold cache: 34.61s → 30.40s (about 12% improvement)
> Hot cache: 14.19s → 10.57s (about 25% improvement)
>
> Patch 3: Simplify key range checking
>
> Replace key_in_sk() helper with direct comparisons
>
> Add ASSERT for min_key validation (safe due to forward search)
>
> Maintain equivalent functionality with cleaner implementation
>
> These changes optimize a critical path for filesystem analysis tools while
> improving code maintainability. The performance gains are particularly
> noticeable when scanning large filesystems.
>
> Thanks,
> Sun YangKai
>
> ---
>
> Changes since v1:
>
> * Replace the WARN_ON with ASSERT, since the condition is a runtime error.
> Suggested by David Sterba.
>
> ---
>
> Sun YangKai (3):
> btrfs: narrow loop variable scope in copy_to_sk()
> btrfs: early exit the searching process in search_tree ioctl
> btrfs: replace key_in_sk() with a simple btrfs_key compare
>
> fs/btrfs/ioctl.c | 55 +++++++++++++++++-------------------------------
> 1 file changed, 19 insertions(+), 36 deletions(-)
Politely ping. Is there anything blocking this?
next prev parent reply other threads:[~2025-08-12 12:05 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-12 4:31 [PATCH 0/3] btrfs: search_tree ioctl performance improvements and cleanups Sun YangKai
2025-06-12 4:31 ` [PATCH 1/3] btrfs: narrow loop variable scope in copy_to_sk() Sun YangKai
2025-06-12 4:31 ` [PATCH 2/3] btrfs: early exit the searching process in search_tree ioctl Sun YangKai
2025-06-12 4:31 ` [PATCH 3/3] btrfs: replace key_in_sk() with a simple btrfs_key compare Sun YangKai
2025-06-19 13:43 ` David Sterba
2025-06-19 14:31 ` Sun YangKai
2025-06-19 13:51 ` [PATCH 0/3] btrfs: search_tree ioctl performance improvements and cleanups David Sterba
2025-07-26 13:51 ` [PATCH v2 " Sun YangKai
2025-07-26 13:51 ` [PATCH v2 1/3] btrfs: narrow loop variable scope in copy_to_sk() Sun YangKai
2025-07-26 13:51 ` [PATCH v2 2/3] btrfs: early exit the searching process in search_tree ioctl Sun YangKai
2025-07-26 13:51 ` [PATCH v2 3/3] btrfs: replace key_in_sk() with a simple btrfs_key compare Sun YangKai
2025-08-12 12:05 ` Sun YangKai [this message]
2025-09-09 10:54 ` [PATCH v2 0/3] btrfs: search_tree ioctl performance improvements and cleanups Sun YangKai
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=12722055.O9o76ZdvQC@saltykitkat \
--to=sunk67188@gmail.com \
--cc=linux-btrfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox