linux-kernel - Re: [PATCH 0/9] Minor cleanups and improvements to swap freeing code

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: <94a5ebb9bd4c8076f23489e014462ac6d169cae0.camel@linux.intel.com>
Date: Fri, 14 Mar 2025 13:27:55 -0700
From: Tim Chen <tim.c.chen@...ux.intel.com>
To: Kemeng Shi <shikemeng@...weicloud.com>, akpm@...ux-foundation.org
Cc: kasong@...cent.com, linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/9] Minor cleanups and improvements to swap freeing code

On Fri, 2025-03-14 at 05:05 +0800, Kemeng Shi wrote:
> Hi All,
> This series contains some cleanups and improvements which are made
> during learning swapfile. Here is a summary of the changes:

Nice work.

> 1. Function nameing improvments.
> -Use "put" instead of "free" to name functions which only do actual free
> when count drops to zero.
> -Use "entry" to name function only frees one swap slot. Use "entries" to
> name function could may free multi swap slots within one cluster. Use
> "_nr" suffix to name function which could free multi swap slots spanning
> cross multi clusters.

Will be nice to add some comments in the code about functions with _nr 
crossing the cluster boundaries and those without stay within a cluster. 

> 2. Eliminate the need to set swap slot to intermediate SWAP_HAS_CACHE
> value before do actual free by using swap_entry_range_free()
> 3. Add helpers swap_entries_put_map() and swap_entries_put_cache() as a
> general-purpose routine to free swap entries within a single cluster
> which will try batch-remove first and fallback to put eatch entry
> indvidually with cluster lock acquired/released only once. By using 
> these helpers, we could remove repeated code, levarage batch-remove in
> more cases and aoivd to acquire/release cluster lock for each single
> swap entry.

Wonder if the batching shows up in any swap performance improvement?

Tim

> 
> Kemeng Shi (9):
>   mm: swap: rename __swap_[entry/entries]_free[_locked] to
>     swap_[entry/entries]_put[_locked]
>   mm: swap: factor out the actual swap entry freeing logic to new helper
>   mm: swap: use __swap_entry_free() to free swap entry in
>     swap_entry_put_locked()
>   mm: swap: remove unneeded VM_BUG_ON(*map != SWAP_HAS_CACHE) in
>     swap_entry_range_free()
>   mm: swap: use swap_entries_free() drop last 1 flag in
>     swap_entries_put_nr()
>   mm: swap: drop last SWAP_MAP_SHMEM flag in batch in
>     swap_entries_put_nr()
>   mm: swap: free each cluster individually in swap_entries_put_map_nr()
>   mm: swap: factor out helper to drop cache of entries within a single
>     cluster
>   mm: swap: replace cluster_swap_free_nr() with
>     swap_entries_put_[map/cache]()
> 
>  mm/swapfile.c | 173 +++++++++++++++++++++++++-------------------------
>  1 file changed, 86 insertions(+), 87 deletions(-)
>