Re: [PATCH] erofs: don't calculate new start when expanding read length

[Yue Hu <zbestahu@gmail.com>]

On Thu, 25 May 2023 13:56:14 +0800
Gao Xiang <hsiangkao@linux.alibaba.com> wrote:

> On 2023/5/25 22:51, Yue Hu wrote:
> > From: Yue Hu <huyue2@coolpad.com>
> > 
> > We only expand the trailing edge and not the leading edge.  So no need
> > to obtain new start again.  Let's use the existing ->headoffset instead.
> > 
> > Signed-off-by: Yue Hu <huyue2@coolpad.com>
> > ---
> >   fs/erofs/zdata.c | 12 +++++-------
> >   1 file changed, 5 insertions(+), 7 deletions(-)
> > 
> > diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
> > index 874fee35af32..bab8dcb8e848 100644
> > --- a/fs/erofs/zdata.c
> > +++ b/fs/erofs/zdata.c
> > @@ -1828,26 +1828,24 @@ static void z_erofs_pcluster_readmore(struct z_erofs_decompress_frontend *f,
> >   {
> >   	struct inode *inode = f->inode;
> >   	struct erofs_map_blocks *map = &f->map;
> > -	erofs_off_t cur, end;
> > +	erofs_off_t cur, end, headoffset = f->headoffset;  
> 
> 
> That is not quite useful, or could you merge this info the original patch?

Okay, i will do the merge later.

> 
> Thanks,
> Gao Xiang
> 
> >   	int err;
> >   
> >   	if (backmost) {
> >   		if (rac)
> > -			end = f->headoffset + readahead_length(rac) - 1;
> > +			end = headoffset + readahead_length(rac) - 1;
> >   		else
> > -			end = f->headoffset + PAGE_SIZE - 1;
> > +			end = headoffset + PAGE_SIZE - 1;
> >   		map->m_la = end;
> >   		err = z_erofs_map_blocks_iter(inode, map,
> >   					      EROFS_GET_BLOCKS_READMORE);
> >   		if (err)
> >   			return;
> >   
> > -		/* expend ra for the trailing edge if readahead */
> > +		/* expand ra for the trailing edge if readahead */
> >   		if (rac) {
> > -			loff_t newstart = readahead_pos(rac);
> > -
> >   			cur = round_up(map->m_la + map->m_llen, PAGE_SIZE);
> > -			readahead_expand(rac, newstart, cur - newstart);
> > +			readahead_expand(rac, headoffset, cur - headoffset);
> >   			return;
> >   		}
> >   		end = round_up(end, PAGE_SIZE);