On Tue, Sep 06, 2022 at 11:43:47PM +0800, Chao Yu wrote:
> > @@ -1768,6 +1745,8 @@ static void f2fs_verify_cluster(struct work_struct *work)
> > void f2fs_decompress_end_io(struct decompress_io_ctx *dic, bool failed,
> > bool in_task)
> > {
> > + int i;
> > +
> > if (!failed && dic->need_verity) {
> > /*
> > * Note that to avoid deadlocks, the verity work can't be done
> > @@ -1777,9 +1756,28 @@ void f2fs_decompress_end_io(struct decompress_io_ctx *dic, bool failed,
> > */
> > INIT_WORK(&dic->verity_work, f2fs_verify_cluster);
> > fsverity_enqueue_verify_work(&dic->verity_work);
> > - } else {
> > - __f2fs_decompress_end_io(dic, failed, in_task);
>
> Will it be possible to clean up __f2fs_decompress_end_io() and
> f2fs_verify_cluster(), they looks almost similar...
>
I feel that it's simpler to keep them separate.
> > +static void f2fs_finish_read_bio(struct bio *bio, bool in_task,
> > + bool fail_compressed)
>
> Not sure, fail_decompress or fail_decompression may looks more readable?
>
I'll add a field 'bool decompression_attempted' to struct bio_post_read_ctx so
that this extra argument won't be needed.
- Eric
