Re: [f2fs-dev] [PATCH 4/4] f2fs: Run f2fs_write_end_io() asynchronously

On 6/13/26 05:03, Bart Van Assche wrote:
> On 6/12/26 5:05 AM, Chao Yu wrote:
>> Actually, I'm worry about whether there is potential performance regression
>> caused by enabling this by default, also a little bit concern about the affection for PC|laptop scenario which is used by individual user.
> 
> According to Gemini F2FS is primarily used on PC's and laptops to
> increase random write IOPS for rotating media (hard disks). I don't
> expect the performance impact of this patch series to be measurable for
> rotating media because the time required to queue and schedule work is
> very small compared to the I/O completion time for these devices.

Hmm, I suspect there is potential non rotating media user, scheduling
can be the bottleneck there, maybe.

> 
> I ran a new set of measurements on a UFS 4 test setup with zoned
> storage (ZUFS). On that test setup I see that the variation between test
> runs is much larger than the variation in test results with or without
> this patch series. In other words, I don't think that we have to worry
> about the performance impact of this patch series.

Well, it needs to be verified for each kernel version you backported, on ZUFS
or non ZUFS, right? I guess Randall or Leo may have time to assist to setup the
performance test. :)

Thanks,

> 
> Bart.



_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel