Message ID | 20210617124709.670936-1-libaokun1@huawei.com |
---|---|
State | Superseded |
Headers | show |
Series | [-next,v4] media: staging: atomisp: use list_splice_init in atomisp_compat_css20.c | expand |
Please ignore this email. I have forgot the version information. I'm about to send a patch v5 with the version information. Best Regards. 在 2021/6/17 20:47, Baokun Li 写道: > Using list_splice_init() instead of entire while-loops > in atomisp_compat_css20.c. > > Reported-by: Hulk Robot <hulkci@huawei.com> > Signed-off-by: Baokun Li <libaokun1@huawei.com> > --- > .../media/atomisp/pci/atomisp_compat_css20.c | 35 +++---------------- > 1 file changed, 5 insertions(+), 30 deletions(-) > > diff --git a/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c b/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c > index f60198bb8a1a..7831bdac96ff 100644 > --- a/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c > +++ b/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c > @@ -2144,42 +2144,17 @@ void atomisp_css_stop(struct atomisp_sub_device *asd, > } > > /* move stats buffers to free queue list */ > - while (!list_empty(&asd->s3a_stats_in_css)) { > - s3a_buf = list_entry(asd->s3a_stats_in_css.next, > - struct atomisp_s3a_buf, list); > - list_del(&s3a_buf->list); > - list_add_tail(&s3a_buf->list, &asd->s3a_stats); > - } > - while (!list_empty(&asd->s3a_stats_ready)) { > - s3a_buf = list_entry(asd->s3a_stats_ready.next, > - struct atomisp_s3a_buf, list); > - list_del(&s3a_buf->list); > - list_add_tail(&s3a_buf->list, &asd->s3a_stats); > - } > + list_splice_init(&asd->s3a_stats_in_css, &asd->s3a_stats); > + list_splice_init(&asd->s3a_stats_ready, &asd->s3a_stats); > > spin_lock_irqsave(&asd->dis_stats_lock, irqflags); > - while (!list_empty(&asd->dis_stats_in_css)) { > - dis_buf = list_entry(asd->dis_stats_in_css.next, > - struct atomisp_dis_buf, list); > - list_del(&dis_buf->list); > - list_add_tail(&dis_buf->list, &asd->dis_stats); > - } > + list_splice_init(&asd->dis_stats_in_css, &asd->dis_stats); > asd->params.dis_proj_data_valid = false; > spin_unlock_irqrestore(&asd->dis_stats_lock, irqflags); > > for (i = 0; i < ATOMISP_METADATA_TYPE_NUM; i++) { > - while (!list_empty(&asd->metadata_in_css[i])) { > - md_buf = list_entry(asd->metadata_in_css[i].next, > - struct atomisp_metadata_buf, list); > - list_del(&md_buf->list); > - list_add_tail(&md_buf->list, &asd->metadata[i]); > - } > - while (!list_empty(&asd->metadata_ready[i])) { > - md_buf = list_entry(asd->metadata_ready[i].next, > - struct atomisp_metadata_buf, list); > - list_del(&md_buf->list); > - list_add_tail(&md_buf->list, &asd->metadata[i]); > - } > + list_splice_init(&asd->metadata_in_css[i], &asd->metadata[i]); > + list_splice_init(&asd->metadata_ready[i], &asd->metadata[i]); > } > > atomisp_flush_params_queue(&asd->video_out_capture);
On Thu, Jun 17, 2021 at 08:47:09PM +0800, Baokun Li wrote: > Using list_splice_init() instead of entire while-loops > in atomisp_compat_css20.c. > > Reported-by: Hulk Robot <hulkci@huawei.com> > Signed-off-by: Baokun Li <libaokun1@huawei.com> > --- You missed changelog. And I believe you missed the changes. So, what's the point?
diff --git a/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c b/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c index f60198bb8a1a..7831bdac96ff 100644 --- a/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c +++ b/drivers/staging/media/atomisp/pci/atomisp_compat_css20.c @@ -2144,42 +2144,17 @@ void atomisp_css_stop(struct atomisp_sub_device *asd, } /* move stats buffers to free queue list */ - while (!list_empty(&asd->s3a_stats_in_css)) { - s3a_buf = list_entry(asd->s3a_stats_in_css.next, - struct atomisp_s3a_buf, list); - list_del(&s3a_buf->list); - list_add_tail(&s3a_buf->list, &asd->s3a_stats); - } - while (!list_empty(&asd->s3a_stats_ready)) { - s3a_buf = list_entry(asd->s3a_stats_ready.next, - struct atomisp_s3a_buf, list); - list_del(&s3a_buf->list); - list_add_tail(&s3a_buf->list, &asd->s3a_stats); - } + list_splice_init(&asd->s3a_stats_in_css, &asd->s3a_stats); + list_splice_init(&asd->s3a_stats_ready, &asd->s3a_stats); spin_lock_irqsave(&asd->dis_stats_lock, irqflags); - while (!list_empty(&asd->dis_stats_in_css)) { - dis_buf = list_entry(asd->dis_stats_in_css.next, - struct atomisp_dis_buf, list); - list_del(&dis_buf->list); - list_add_tail(&dis_buf->list, &asd->dis_stats); - } + list_splice_init(&asd->dis_stats_in_css, &asd->dis_stats); asd->params.dis_proj_data_valid = false; spin_unlock_irqrestore(&asd->dis_stats_lock, irqflags); for (i = 0; i < ATOMISP_METADATA_TYPE_NUM; i++) { - while (!list_empty(&asd->metadata_in_css[i])) { - md_buf = list_entry(asd->metadata_in_css[i].next, - struct atomisp_metadata_buf, list); - list_del(&md_buf->list); - list_add_tail(&md_buf->list, &asd->metadata[i]); - } - while (!list_empty(&asd->metadata_ready[i])) { - md_buf = list_entry(asd->metadata_ready[i].next, - struct atomisp_metadata_buf, list); - list_del(&md_buf->list); - list_add_tail(&md_buf->list, &asd->metadata[i]); - } + list_splice_init(&asd->metadata_in_css[i], &asd->metadata[i]); + list_splice_init(&asd->metadata_ready[i], &asd->metadata[i]); } atomisp_flush_params_queue(&asd->video_out_capture);
Using list_splice_init() instead of entire while-loops in atomisp_compat_css20.c. Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Baokun Li <libaokun1@huawei.com> --- .../media/atomisp/pci/atomisp_compat_css20.c | 35 +++---------------- 1 file changed, 5 insertions(+), 30 deletions(-)