[PATCH v11 06/12] media: mediatek: jpeg: fix decoding buffer number setting timing issue

[Kyrie Wu (吴晗)]

On Tue, 2025-12-16 at 16:37 -0500, Nicolas Dufresne wrote:
> Hi,
> 
> Le mardi 02 décembre 2025 à 17:47 +0800, Kyrie Wu a écrit :
> > move decoding buffer increase code into spinlock
> > protecting aera for multi-core
> 
> I don't understand this text, please rework.
Dear Nicolas,

The dst buf parameters only need to set when the power
set succussed and protect the setting by spinlock.

I will rework the commit message in the coming version.

Thanks.

Regards,
Kyrie.
> 
> > 
> > Fixes: dedc21500334 ("media: mtk-jpegdec: add jpeg decode worker
> > interface")
> > 
> 
> Drop that blank line.
> 
> > Signed-off-by: Kyrie Wu <kyrie.wu at mediatek.com>
> > ---
> >  drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c   | 9 +++--
> > ----
> >  drivers/media/platform/mediatek/jpeg/mtk_jpeg_dec_hw.c | 1 +
> >  drivers/media/platform/mediatek/jpeg/mtk_jpeg_enc_hw.c | 1 +
> >  3 files changed, 5 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> > b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> > index e45d7e82b8a0..5ffaee4dcd19 100644
> > --- a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> > +++ b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> > @@ -1746,7 +1746,6 @@ static void mtk_jpegdec_worker(struct
> > work_struct *work)
> >  
> >  	v4l2_m2m_buf_copy_metadata(src_buf, dst_buf, true);
> >  	jpeg_src_buf = mtk_jpeg_vb2_to_srcbuf(&src_buf->vb2_buf);
> > -	jpeg_dst_buf = mtk_jpeg_vb2_to_srcbuf(&dst_buf->vb2_buf);
> >  
> >  	if (mtk_jpeg_check_resolution_change(ctx,
> >  					     &jpeg_src_buf->dec_param)) 
> > {
> > @@ -1755,11 +1754,6 @@ static void mtk_jpegdec_worker(struct
> > work_struct *work)
> >  		goto getbuf_fail;
> >  	}
> >  
> > -	jpeg_src_buf->curr_ctx = ctx;
> > -	jpeg_src_buf->frame_num = ctx->total_frame_num;
> > -	jpeg_dst_buf->curr_ctx = ctx;
> > -	jpeg_dst_buf->frame_num = ctx->total_frame_num;
> > -
> >  	mtk_jpegdec_set_hw_param(ctx, hw_id, src_buf, dst_buf);
> >  	ret = pm_runtime_resume_and_get(comp_jpeg[hw_id]->dev);
> >  	if (ret < 0) {
> > @@ -1784,6 +1778,9 @@ static void mtk_jpegdec_worker(struct
> > work_struct *work)
> >  			     
> > msecs_to_jiffies(MTK_JPEG_HW_TIMEOUT_MSEC));
> >  
> >  	spin_lock_irqsave(&comp_jpeg[hw_id]->hw_lock, flags);
> > +	jpeg_dst_buf = mtk_jpeg_vb2_to_srcbuf(&dst_buf->vb2_buf);
> > +	jpeg_dst_buf->curr_ctx = ctx;
> > +	jpeg_dst_buf->frame_num = ctx->total_frame_num;
> >  	ctx->total_frame_num++;
> >  	mtk_jpeg_dec_reset(comp_jpeg[hw_id]->reg_base);
> >  	mtk_jpeg_dec_set_config(comp_jpeg[hw_id]->reg_base,
> > diff --git a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_dec_hw.c
> > b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_dec_hw.c
> > index 5f1557dafad6..b3142dc9be85 100644
> > --- a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_dec_hw.c
> > +++ b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_dec_hw.c
> > @@ -513,6 +513,7 @@ static void mtk_jpegdec_put_buf(struct
> > mtk_jpegdec_comp_dev *jpeg)
> >  				v4l2_m2m_buf_done(&tmp_dst_done_buf->b,
> >  						  VB2_BUF_STATE_DONE);
> >  				ctx->last_done_frame_num++;
> > +				break;
> >  			}
> >  		}
> >  	}
> > diff --git a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_enc_hw.c
> > b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_enc_hw.c
> > index 785db5ba4770..82c971936c4d 100644
> > --- a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_enc_hw.c
> > +++ b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_enc_hw.c
> > @@ -242,6 +242,7 @@ static void mtk_jpegenc_put_buf(struct
> > mtk_jpegenc_comp_dev *jpeg)
> >  				v4l2_m2m_buf_done(&tmp_dst_done_buf->b,
> >  						  VB2_BUF_STATE_DONE);
> >  				ctx->last_done_frame_num++;
> > +				break;
> >  			}
> >  		}
> >  	}
> 
> Normally I'd seek for the context in the code. But I think I'll wait
> for a
> proper commit message, and do a proper review then.
> 
> Nicolas