Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <jaeg...@kernel.org>:

On Wed,  4 Dec 2024 11:31:13 +0800 you wrote:
> In SSR mode, the segment selected for allocation might be the same as
> the target segment of the GC triggered by ioctl, resulting in the GC
> moving the CURSEG_I(sbi, type)->segno.
> Thread A                              Thread B or Thread A
> - f2fs_ioc_gc_range
>  - __f2fs_ioc_gc_range(.victim_segno=segno#N)
>   - f2fs_gc
>    - __get_victim
>     - f2fs_get_victim
>     : segno#N is valid, return segno#N as source segment of GC
>                                       - f2fs_allocate_data_block
>                                               - need_new_seg
>                                               - get_ssr_segment
>                                               - f2fs_get_victim
>                                               : get segno #N as destination 
> segment
>                                               - change_curseg
> 
> [...]

Here is the summary with links:
  - [f2fs-dev,V2] f2fs: The GC triggered by ioctl also needs to mark the segno 
as victim
    https://git.kernel.org/jaegeuk/f2fs/c/e9a844f6e487

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html




_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Reply via email to