Re: [f2fs-dev] [PATCH 1/5] f2fs: change default compression algorithm

[Jaegeuk Kim]

On 03/11, Chao Yu wrote:
> On 2020/3/11 0:15, Eric Biggers wrote:
> > On Tue, Mar 10, 2020 at 08:50:05PM +0800, Chao Yu wrote:
> >> Use LZ4 as default compression algorithm, as compared to LZO, it shows
> >> almost the same compression ratio and much better decompression speed.
> >>
> >> Signed-off-by: Chao Yu <yuchao0@xxxxxxxxxx>
> >> ---
> >>  fs/f2fs/super.c | 2 +-
> >>  1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> >> index db3a63f7c769..ebffe7aa08ee 100644
> >> --- a/fs/f2fs/super.c
> >> +++ b/fs/f2fs/super.c
> >> @@ -1577,7 +1577,7 @@ static void default_options(struct f2fs_sb_info *sbi)
> >>  	F2FS_OPTION(sbi).test_dummy_encryption = false;
> >>  	F2FS_OPTION(sbi).s_resuid = make_kuid(&init_user_ns, F2FS_DEF_RESUID);
> >>  	F2FS_OPTION(sbi).s_resgid = make_kgid(&init_user_ns, F2FS_DEF_RESGID);
> >> -	F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZO;
> >> +	F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZ4;
> >>  	F2FS_OPTION(sbi).compress_log_size = MIN_COMPRESS_LOG_SIZE;
> >>  	F2FS_OPTION(sbi).compress_ext_cnt = 0;
> >>  	F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_ON;
> > 
> > This makes sense, but it's unclear to me why comparing the different compression
> > algorithms is happening just now, after support for both LZO and LZ4 was already
> > merged into mainline and now has to be supported forever.  During review months
> > ago, multiple people suggested that LZ4 is better than LZO, so there's not much
> > reason to support LZO at all.
> 
> Agreed,
> 
> Jaegeuk, thoughts?

Supporting LZO or whatever algorithms looks fine to me as long as kernel
supports without huge maintenance effort. I think it'd be good to provide
a way for users to compare several algorithms in whatever their ways.

And, unfortunately, I don't think we can remove LZO at this point, even for -rc.

Thanks,

> 
> Let me remove lzo if you have no objection on this.
> 
> Thanks,
> 
> > 
> > - Eric
> > .
> >