[PATCH v2 00/11] Improve stability and add bug fixes of Aspeed video engine driver

[PATCH v2 00/11] Improve stability and add bug fixes of Aspeed video engine driver

[Jae Hyun Yoo]

This patch series improves stability of Aspeed video engine driver by fixing
clock control and irq handling logic in the driver. Also, it adds a couple of
bug fixes and a workaroud for a silicon bug.

Changes since v1:
- Removed spinlock handling code from 0001 patch.
- Added 4 more patches.

Jae Hyun Yoo (11):
  media: aspeed: fix a kernel warning on clk control
  media: aspeed: refine clock control logic
  media: aspeed: change irq to threaded irq
  media: aspeed: remove IRQF_SHARED flag
  media: aspeed: reduce noisy log printing outs
  media: aspeed: remove checking of VE_INTERRUPT_CAPTURE_COMPLETE
  media: aspeed: refine interrupt handling logic
  media: aspeed: remove source buffer allocation before mode detection
  media: aspeed: use different delays for triggering VE H/W reset
  media: aspeed: fix an incorrect timeout checking in mode detection
  media: aspeed: add a workaround to fix a silicon bug

 drivers/media/platform/aspeed-video.c | 140 +++++++++++++++-----------
 1 file changed, 80 insertions(+), 60 deletions(-)

-- 
2.21.0

[PATCH v2 01/11] media: aspeed: fix a kernel warning on clk control

[Jae Hyun Yoo]

Video engine clock control can be double disabled and eventually
it causes a kernel warning with stack dump printing out like below:

[  515.540498] ------------[ cut here ]------------
[  515.545174] WARNING: CPU: 0 PID: 1310 at drivers/clk/clk.c:684 clk_core_unprepare+0x13c/0x170
[  515.553806] vclk-gate already unprepared
[  515.557841] CPU: 0 PID: 1310 Comm: obmc-ikvm Tainted: G        W         5.0.6-df66fbc97853fbba90a0bfa44de32f3d5f7602b4 #1
[  515.568973] Hardware name: Generic DT based system
[  515.573777] Backtrace:
[  515.576272] [<80107cdc>] (dump_backtrace) from [<80107f10>] (show_stack+0x20/0x24)
[  515.583930]  r7:803a5614 r6:00000009 r5:00000000 r4:9d88fe1c
[  515.589712] [<80107ef0>] (show_stack) from [<80690184>] (dump_stack+0x20/0x28)
[  515.597053] [<80690164>] (dump_stack) from [<80116044>] (__warn.part.3+0xb4/0xdc)
[  515.604557] [<80115f90>] (__warn.part.3) from [<801160d8>] (warn_slowpath_fmt+0x6c/0x90)
[  515.612734]  r6:000002ac r5:8080befc r4:80a07008
[  515.617463] [<80116070>] (warn_slowpath_fmt) from [<803a5614>] (clk_core_unprepare+0x13c/0x170)
[  515.626167]  r3:8080cdf4 r2:8080bfc0
[  515.629834]  r7:98d682a8 r6:9d8a9200 r5:9e5151a0 r4:97abd620
[  515.635530] [<803a54d8>] (clk_core_unprepare) from [<803a76a4>] (clk_unprepare+0x34/0x3c)
[  515.643812]  r5:9e5151a0 r4:97abd620
[  515.647529] [<803a7670>] (clk_unprepare) from [<804f36ec>] (aspeed_video_off+0x38/0x50)
[  515.655539]  r5:9e5151a0 r4:9e504000
[  515.659242] [<804f36b4>] (aspeed_video_off) from [<804f4358>] (aspeed_video_release+0x90/0x114)
[  515.668036]  r5:9e5044b0 r4:9e504000
[  515.671643] [<804f42c8>] (aspeed_video_release) from [<804d302c>] (v4l2_release+0xd4/0xe8)
[  515.679999]  r7:98d682a8 r6:9d087810 r5:9d8a9200 r4:9e504318
[  515.685695] [<804d2f58>] (v4l2_release) from [<80236454>] (__fput+0x98/0x1c4)
[  515.692914]  r5:9e51b608 r4:9d8a9200
[  515.696597] [<802363bc>] (__fput) from [<802365e8>] (____fput+0x18/0x1c)
[  515.703315]  r9:80a0700c r8:801011e4 r7:00000000 r6:80a64b9c r5:9d8e35a0 r4:9d8e38dc
[  515.711167] [<802365d0>] (____fput) from [<80131ca4>] (task_work_run+0x7c/0xa0)
[  515.718596] [<80131c28>] (task_work_run) from [<80106884>] (do_work_pending+0x4a8/0x578)
[  515.726777]  r7:801011e4 r6:80a07008 r5:9d88ffb0 r4:ffffe000
[  515.732466] [<801063dc>] (do_work_pending) from [<8010106c>] (slow_work_pending+0xc/0x20)
[  515.740727] Exception stack(0x9d88ffb0 to 0x9d88fff8)
[  515.745840] ffa0:                                     00000000 76f18094 00000000 00000000
[  515.754122] ffc0: 00000007 00176778 7eda4c20 00000006 00000000 00000000 48e20fa4 00000000
[  515.762386] ffe0: 00000002 7eda4b08 00000000 48f91efc 80000010 00000007
[  515.769097]  r10:00000000 r9:9d88e000 r8:801011e4 r7:00000006 r6:7eda4c20 r5:00176778
[  515.777006]  r4:00000007
[  515.779558] ---[ end trace 12c04aadef8afbbb ]---
[  515.784176] ------------[ cut here ]------------
[  515.788817] WARNING: CPU: 0 PID: 1310 at drivers/clk/clk.c:825 clk_core_disable+0x18c/0x204
[  515.797161] eclk-gate already disabled
[  515.800916] CPU: 0 PID: 1310 Comm: obmc-ikvm Tainted: G        W         5.0.6-df66fbc97853fbba90a0bfa44de32f3d5f7602b4 #1
[  515.811945] Hardware name: Generic DT based system
[  515.816730] Backtrace:
[  515.819210] [<80107cdc>] (dump_backtrace) from [<80107f10>] (show_stack+0x20/0x24)
[  515.826782]  r7:803a5900 r6:00000009 r5:00000000 r4:9d88fe04
[  515.832454] [<80107ef0>] (show_stack) from [<80690184>] (dump_stack+0x20/0x28)
[  515.839687] [<80690164>] (dump_stack) from [<80116044>] (__warn.part.3+0xb4/0xdc)
[  515.847170] [<80115f90>] (__warn.part.3) from [<801160d8>] (warn_slowpath_fmt+0x6c/0x90)
[  515.855247]  r6:00000339 r5:8080befc r4:80a07008
[  515.859868] [<80116070>] (warn_slowpath_fmt) from [<803a5900>] (clk_core_disable+0x18c/0x204)
[  515.868385]  r3:8080cdd0 r2:8080c00c
[  515.871957]  r7:98d682a8 r6:9d8a9200 r5:97abd560 r4:97abd560
[  515.877615] [<803a5774>] (clk_core_disable) from [<803a59a0>] (clk_core_disable_lock+0x28/0x34)
[  515.886301]  r7:98d682a8 r6:9d8a9200 r5:97abd560 r4:a0000013
[  515.891960] [<803a5978>] (clk_core_disable_lock) from [<803a7714>] (clk_disable+0x2c/0x30)
[  515.900216]  r5:9e5151a0 r4:9e515f60
[  515.903816] [<803a76e8>] (clk_disable) from [<804f36f8>] (aspeed_video_off+0x44/0x50)
[  515.911656] [<804f36b4>] (aspeed_video_off) from [<804f4358>] (aspeed_video_release+0x90/0x114)
[  515.920341]  r5:9e5044b0 r4:9e504000
[  515.923921] [<804f42c8>] (aspeed_video_release) from [<804d302c>] (v4l2_release+0xd4/0xe8)
[  515.932184]  r7:98d682a8 r6:9d087810 r5:9d8a9200 r4:9e504318
[  515.937851] [<804d2f58>] (v4l2_release) from [<80236454>] (__fput+0x98/0x1c4)
[  515.944980]  r5:9e51b608 r4:9d8a9200
[  515.948559] [<802363bc>] (__fput) from [<802365e8>] (____fput+0x18/0x1c)
[  515.955257]  r9:80a0700c r8:801011e4 r7:00000000 r6:80a64b9c r5:9d8e35a0 r4:9d8e38dc
[  515.963008] [<802365d0>] (____fput) from [<80131ca4>] (task_work_run+0x7c/0xa0)
[  515.970333] [<80131c28>] (task_work_run) from [<80106884>] (do_work_pending+0x4a8/0x578)
[  515.978421]  r7:801011e4 r6:80a07008 r5:9d88ffb0 r4:ffffe000
[  515.984086] [<801063dc>] (do_work_pending) from [<8010106c>] (slow_work_pending+0xc/0x20)
[  515.992247] Exception stack(0x9d88ffb0 to 0x9d88fff8)
[  515.997296] ffa0:                                     00000000 76f18094 00000000 00000000
[  516.005473] ffc0: 00000007 00176778 7eda4c20 00000006 00000000 00000000 48e20fa4 00000000
[  516.013642] ffe0: 00000002 7eda4b08 00000000 48f91efc 80000010 00000007
[  516.020257]  r10:00000000 r9:9d88e000 r8:801011e4 r7:00000006 r6:7eda4c20 r5:00176778
[  516.028072]  r4:00000007
[  516.030606] ---[ end trace 12c04aadef8afbbc ]---

To prevent this issue, this commit adds clock status checking
logic into the Aspeed video engine driver.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
Reviewed-by: Eddie James <eajames@linux.ibm.com>
---
v1 -> v2:
 Removed spinlock handling code. Actual fix for removing the kernel warning is
 blocking double disabling of video clocks. Refactoring of protection mechanism
 of this driver will be submitted using a seprate patch series.

 drivers/media/platform/aspeed-video.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index 8144fe36ad48..562d7c0adc78 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -187,6 +187,7 @@ enum {
 	VIDEO_STREAMING,
 	VIDEO_FRAME_INPRG,
 	VIDEO_STOPPED,
+	VIDEO_CLOCKS_ON,
 };
 
 struct aspeed_video_addr {
@@ -483,19 +484,29 @@ static void aspeed_video_enable_mode_detect(struct aspeed_video *video)
 
 static void aspeed_video_off(struct aspeed_video *video)
 {
+	if (!test_bit(VIDEO_CLOCKS_ON, &video->flags))
+		return;
+
 	/* Disable interrupts */
 	aspeed_video_write(video, VE_INTERRUPT_CTRL, 0);
 
 	/* Turn off the relevant clocks */
 	clk_disable_unprepare(video->vclk);
 	clk_disable_unprepare(video->eclk);
+
+	clear_bit(VIDEO_CLOCKS_ON, &video->flags);
 }
 
 static void aspeed_video_on(struct aspeed_video *video)
 {
+	if (test_bit(VIDEO_CLOCKS_ON, &video->flags))
+		return;
+
 	/* Turn on the relevant clocks */
 	clk_prepare_enable(video->eclk);
 	clk_prepare_enable(video->vclk);
+
+	set_bit(VIDEO_CLOCKS_ON, &video->flags);
 }
 
 static void aspeed_video_bufs_done(struct aspeed_video *video,
-- 
2.21.0

[PATCH v2 02/11] media: aspeed: refine clock control logic

[Jae Hyun Yoo]

Currently, this driver calls clk_prepare and clk_unprepare from
interrupt context too but these should be called from sleepable
context only. To fix this issue, this commit splits out
clk_enable/disable and clk_prepare/unprepare, and it places
clk_prepare/unprepare calls into the module probe/remove function.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
Reviewed-by: Eddie James <eajames@linux.ibm.com>
---
v1 -> v2:
 None.

 drivers/media/platform/aspeed-video.c | 38 ++++++++++++++++++++-------
 1 file changed, 29 insertions(+), 9 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index 562d7c0adc78..7982ce634936 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -491,8 +491,8 @@ static void aspeed_video_off(struct aspeed_video *video)
 	aspeed_video_write(video, VE_INTERRUPT_CTRL, 0);
 
 	/* Turn off the relevant clocks */
-	clk_disable_unprepare(video->vclk);
-	clk_disable_unprepare(video->eclk);
+	clk_disable(video->vclk);
+	clk_disable(video->eclk);
 
 	clear_bit(VIDEO_CLOCKS_ON, &video->flags);
 }
@@ -503,8 +503,8 @@ static void aspeed_video_on(struct aspeed_video *video)
 		return;
 
 	/* Turn on the relevant clocks */
-	clk_prepare_enable(video->eclk);
-	clk_prepare_enable(video->vclk);
+	clk_enable(video->eclk);
+	clk_enable(video->vclk);
 
 	set_bit(VIDEO_CLOCKS_ON, &video->flags);
 }
@@ -1613,31 +1613,46 @@ static int aspeed_video_init(struct aspeed_video *video)
 		return PTR_ERR(video->eclk);
 	}
 
+	rc = clk_prepare(video->eclk);
+	if (rc)
+		return rc;
+
 	video->vclk = devm_clk_get(dev, "vclk");
 	if (IS_ERR(video->vclk)) {
 		dev_err(dev, "Unable to get VCLK\n");
-		return PTR_ERR(video->vclk);
+		rc = PTR_ERR(video->vclk);
+		goto err_unprepare_eclk;
 	}
 
+	rc = clk_prepare(video->vclk);
+	if (rc)
+		goto err_unprepare_eclk;
+
 	of_reserved_mem_device_init(dev);
 
 	rc = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32));
 	if (rc) {
 		dev_err(dev, "Failed to set DMA mask\n");
-		of_reserved_mem_device_release(dev);
-		return rc;
+		goto err_release_reserved_mem;
 	}
 
 	if (!aspeed_video_alloc_buf(video, &video->jpeg,
 				    VE_JPEG_HEADER_SIZE)) {
 		dev_err(dev, "Failed to allocate DMA for JPEG header\n");
-		of_reserved_mem_device_release(dev);
-		return rc;
+		goto err_release_reserved_mem;
 	}
 
 	aspeed_video_init_jpeg_table(video->jpeg.virt, video->yuv420);
 
 	return 0;
+
+err_release_reserved_mem:
+	of_reserved_mem_device_release(dev);
+	clk_unprepare(video->vclk);
+err_unprepare_eclk:
+	clk_unprepare(video->eclk);
+
+	return rc;
 }
 
 static int aspeed_video_probe(struct platform_device *pdev)
@@ -1681,6 +1696,11 @@ static int aspeed_video_remove(struct platform_device *pdev)
 	struct v4l2_device *v4l2_dev = dev_get_drvdata(dev);
 	struct aspeed_video *video = to_aspeed_video(v4l2_dev);
 
+	aspeed_video_off(video);
+
+	clk_unprepare(video->vclk);
+	clk_unprepare(video->eclk);
+
 	video_unregister_device(&video->vdev);
 
 	vb2_queue_release(&video->queue);
-- 
2.21.0

[PATCH v2 03/11] media: aspeed: change irq to threaded irq

[Jae Hyun Yoo]

Differently from other Aspeed drivers, this driver calls clock
control APIs in interrupt context. Since ECLK is coupled with a
reset bit in clk-aspeed module, aspeed_clk_enable will make 10ms of
busy waiting delay for triggering the reset and it will eventually
disturb other drivers' interrupt handling. To fix this issue, this
commit changes this driver's irq to threaded irq so that the delay
can be happened in a thread context.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
Reviewed-by: Eddie James <eajames@linux.ibm.com>
---
v1 -> v2:
 None.

 drivers/media/platform/aspeed-video.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index 7982ce634936..f7db8969c8f2 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -1600,8 +1600,9 @@ static int aspeed_video_init(struct aspeed_video *video)
 		return -ENODEV;
 	}
 
-	rc = devm_request_irq(dev, irq, aspeed_video_irq, IRQF_SHARED,
-			      DEVICE_NAME, video);
+	rc = devm_request_threaded_irq(dev, irq, NULL, aspeed_video_irq,
+				       IRQF_ONESHOT | IRQF_SHARED, DEVICE_NAME,
+				       video);
 	if (rc < 0) {
 		dev_err(dev, "Unable to request IRQ %d\n", irq);
 		return rc;
-- 
2.21.0

[PATCH v2 04/11] media: aspeed: remove IRQF_SHARED flag

[Jae Hyun Yoo]

Video Engine has a dedicated interrupt line so this driver doesn't
need to use IRQF_SHARED flag so remove it. Also, it'd be good for
following what Thomas recommended in the IRQF_ONESHOT support
patch like below:

"Note that for now IRQF_ONESHOT cannot be used with IRQF_SHARED to
avoid complex accounting mechanisms."

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
Reviewed-by: Eddie James <eajames@linux.ibm.com>
---
v1 -> v2:
 None.

 drivers/media/platform/aspeed-video.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index f7db8969c8f2..d1b541409544 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -1601,8 +1601,7 @@ static int aspeed_video_init(struct aspeed_video *video)
 	}
 
 	rc = devm_request_threaded_irq(dev, irq, NULL, aspeed_video_irq,
-				       IRQF_ONESHOT | IRQF_SHARED, DEVICE_NAME,
-				       video);
+				       IRQF_ONESHOT, DEVICE_NAME, video);
 	if (rc < 0) {
 		dev_err(dev, "Unable to request IRQ %d\n", irq);
 		return rc;
-- 
2.21.0

[PATCH v2 05/11] media: aspeed: reduce noisy log printing outs

[Jae Hyun Yoo]

Currently, this driver prints out too much log messages when a
mode change happens, video turned off by screen saver and etc.
Actually, all cases are reported to user space properly. Also,
these are not critical errors but recoverable things, so this
commit changes the log level of some noisy printing outs.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
Reviewed-by: Eddie James <eajames@linux.ibm.com>
---
v1 -> v2:
 None.

 drivers/media/platform/aspeed-video.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index d1b541409544..92abdfc79e76 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -441,7 +441,7 @@ static int aspeed_video_start_frame(struct aspeed_video *video)
 
 	if (!(seq_ctrl & VE_SEQ_CTRL_COMP_BUSY) ||
 	    !(seq_ctrl & VE_SEQ_CTRL_CAP_BUSY)) {
-		dev_err(video->dev, "Engine busy; don't start frame\n");
+		dev_dbg(video->dev, "Engine busy; don't start frame\n");
 		return -EBUSY;
 	}
 
@@ -769,7 +769,7 @@ static void aspeed_video_get_resolution(struct aspeed_video *video)
 						      res_check(video),
 						      MODE_DETECT_TIMEOUT);
 		if (!rc) {
-			dev_err(video->dev, "Timed out; first mode detect\n");
+			dev_dbg(video->dev, "Timed out; first mode detect\n");
 			clear_bit(VIDEO_RES_DETECT, &video->flags);
 			return;
 		}
@@ -787,7 +787,7 @@ static void aspeed_video_get_resolution(struct aspeed_video *video)
 						      MODE_DETECT_TIMEOUT);
 		clear_bit(VIDEO_RES_DETECT, &video->flags);
 		if (!rc) {
-			dev_err(video->dev, "Timed out; second mode detect\n");
+			dev_dbg(video->dev, "Timed out; second mode detect\n");
 			return;
 		}
 
@@ -821,7 +821,7 @@ static void aspeed_video_get_resolution(struct aspeed_video *video)
 	} while (invalid_resolution && (tries++ < INVALID_RESOLUTION_RETRIES));
 
 	if (invalid_resolution) {
-		dev_err(video->dev, "Invalid resolution detected\n");
+		dev_dbg(video->dev, "Invalid resolution detected\n");
 		return;
 	}
 
@@ -1456,7 +1456,7 @@ static void aspeed_video_stop_streaming(struct vb2_queue *q)
 				!test_bit(VIDEO_FRAME_INPRG, &video->flags),
 				STOP_TIMEOUT);
 	if (!rc) {
-		dev_err(video->dev, "Timed out when stopping streaming\n");
+		dev_dbg(video->dev, "Timed out when stopping streaming\n");
 
 		/*
 		 * Need to force stop any DMA and try and get HW into a good
-- 
2.21.0

[PATCH v2 06/11] media: aspeed: remove checking of VE_INTERRUPT_CAPTURE_COMPLETE

[Jae Hyun Yoo]

VE_INTERRUPT_CAPTURE_COMPLETE and VE_INTERRUPT_COMP_COMPLETE are
not set at the same time but the current interrupt handling
mechanism of this driver doesn't clear the interrupt flag until
both two are set, and this behavior causes unnecessary interrupt
handler calls. In fact, this driver provides JPEG format only so
taking care of the VE_INTERRUPT_COMP_COMPLETE is enough for getting
compressed image frame so this commit gets rid of the
VE_INTERRUPT_CAPTURE_COMPLETE checking logic to simplify the logic.
Handling of VE_INTERRUPT_CAPTURE_COMPLETE could be added back later
when it's actually needed.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
Reviewed-by: Eddie James <eajames@linux.ibm.com>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
---
v1 -> v2:
 None.

 drivers/media/platform/aspeed-video.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index 92abdfc79e76..1cba582918cc 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -463,8 +463,7 @@ static int aspeed_video_start_frame(struct aspeed_video *video)
 	aspeed_video_write(video, VE_COMP_ADDR, addr);
 
 	aspeed_video_update(video, VE_INTERRUPT_CTRL, 0,
-			    VE_INTERRUPT_COMP_COMPLETE |
-			    VE_INTERRUPT_CAPTURE_COMPLETE);
+			    VE_INTERRUPT_COMP_COMPLETE);
 
 	aspeed_video_update(video, VE_SEQ_CTRL, 0,
 			    VE_SEQ_CTRL_TRIG_CAPTURE | VE_SEQ_CTRL_TRIG_COMP);
@@ -568,8 +567,7 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
 		}
 	}
 
-	if ((sts & VE_INTERRUPT_COMP_COMPLETE) &&
-	    (sts & VE_INTERRUPT_CAPTURE_COMPLETE)) {
+	if (sts & VE_INTERRUPT_COMP_COMPLETE) {
 		struct aspeed_video_buffer *buf;
 		u32 frame_size = aspeed_video_read(video,
 						   VE_OFFSET_COMP_STREAM);
@@ -598,11 +596,9 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
 				    VE_SEQ_CTRL_FORCE_IDLE |
 				    VE_SEQ_CTRL_TRIG_COMP, 0);
 		aspeed_video_update(video, VE_INTERRUPT_CTRL,
-				    VE_INTERRUPT_COMP_COMPLETE |
-				    VE_INTERRUPT_CAPTURE_COMPLETE, 0);
+				    VE_INTERRUPT_COMP_COMPLETE, 0);
 		aspeed_video_write(video, VE_INTERRUPT_STATUS,
-				   VE_INTERRUPT_COMP_COMPLETE |
-				   VE_INTERRUPT_CAPTURE_COMPLETE);
+				   VE_INTERRUPT_COMP_COMPLETE);
 
 		if (test_bit(VIDEO_STREAMING, &video->flags) && buf)
 			aspeed_video_start_frame(video);
-- 
2.21.0

[PATCH v2 07/11] media: aspeed: refine interrupt handling logic

[Jae Hyun Yoo]

There are cases that interrupt bits are cleared by a 500ms delayed
work which causes unnecessary irq calls. Also, the current
interrupt handler returns IRQ_HANDLED always but it should return
IRQ_NONE if there is any unhandled interrupt. So this commit
refines the interrupt handling logic to fix these issues.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Reviewed-by: Eddie James <eajames@linux.ibm.com>
---
v1 -> v2:
 None.

 drivers/media/platform/aspeed-video.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index 1cba582918cc..c0b889141b8f 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -488,6 +488,7 @@ static void aspeed_video_off(struct aspeed_video *video)
 
 	/* Disable interrupts */
 	aspeed_video_write(video, VE_INTERRUPT_CTRL, 0);
+	aspeed_video_write(video, VE_INTERRUPT_STATUS, 0xffffffff);
 
 	/* Turn off the relevant clocks */
 	clk_disable(video->vclk);
@@ -554,7 +555,7 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
 					    VE_INTERRUPT_MODE_DETECT, 0);
 			aspeed_video_write(video, VE_INTERRUPT_STATUS,
 					   VE_INTERRUPT_MODE_DETECT);
-
+			sts &= ~VE_INTERRUPT_MODE_DETECT;
 			set_bit(VIDEO_MODE_DETECT_DONE, &video->flags);
 			wake_up_interruptible_all(&video->wait);
 		} else {
@@ -599,12 +600,12 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
 				    VE_INTERRUPT_COMP_COMPLETE, 0);
 		aspeed_video_write(video, VE_INTERRUPT_STATUS,
 				   VE_INTERRUPT_COMP_COMPLETE);
-
+		sts &= ~VE_INTERRUPT_COMP_COMPLETE;
 		if (test_bit(VIDEO_STREAMING, &video->flags) && buf)
 			aspeed_video_start_frame(video);
 	}
 
-	return IRQ_HANDLED;
+	return sts ? IRQ_NONE : IRQ_HANDLED;
 }
 
 static void aspeed_video_check_and_set_polarity(struct aspeed_video *video)
-- 
2.21.0

[PATCH v2 08/11] media: aspeed: remove source buffer allocation before mode detection

[Jae Hyun Yoo]

Mode detection doesn't require source buffer allocation so this
commit removes that.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
---
v1 -> v2:
 New.

 drivers/media/platform/aspeed-video.c | 21 ---------------------
 1 file changed, 21 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index c0b889141b8f..4647ed2e9e63 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -731,27 +731,6 @@ static void aspeed_video_get_resolution(struct aspeed_video *video)
 	det->height = MIN_HEIGHT;
 	video->v4l2_input_status = V4L2_IN_ST_NO_SIGNAL;
 
-	/*
-	 * Since we need max buffer size for detection, free the second source
-	 * buffer first.
-	 */
-	if (video->srcs[1].size)
-		aspeed_video_free_buf(video, &video->srcs[1]);
-
-	if (video->srcs[0].size < VE_MAX_SRC_BUFFER_SIZE) {
-		if (video->srcs[0].size)
-			aspeed_video_free_buf(video, &video->srcs[0]);
-
-		if (!aspeed_video_alloc_buf(video, &video->srcs[0],
-					    VE_MAX_SRC_BUFFER_SIZE)) {
-			dev_err(video->dev,
-				"Failed to allocate source buffers\n");
-			return;
-		}
-	}
-
-	aspeed_video_write(video, VE_SRC0_ADDR, video->srcs[0].dma);
-
 	do {
 		if (tries) {
 			set_current_state(TASK_INTERRUPTIBLE);
-- 
2.21.0

Re: [PATCH v2 08/11] media: aspeed: remove source buffer allocation before mode detection

[Eddie James]

On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
> Mode detection doesn't require source buffer allocation so this
> commit removes that.


Reviewed-by: Eddie James <eajames@linux.ibm.com>


>
> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
> ---
> v1 -> v2:
>   New.
>
>   drivers/media/platform/aspeed-video.c | 21 ---------------------
>   1 file changed, 21 deletions(-)
>
> diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
> index c0b889141b8f..4647ed2e9e63 100644
> --- a/drivers/media/platform/aspeed-video.c
> +++ b/drivers/media/platform/aspeed-video.c
> @@ -731,27 +731,6 @@ static void aspeed_video_get_resolution(struct aspeed_video *video)
>   	det->height = MIN_HEIGHT;
>   	video->v4l2_input_status = V4L2_IN_ST_NO_SIGNAL;
>   
> -	/*
> -	 * Since we need max buffer size for detection, free the second source
> -	 * buffer first.
> -	 */
> -	if (video->srcs[1].size)
> -		aspeed_video_free_buf(video, &video->srcs[1]);
> -
> -	if (video->srcs[0].size < VE_MAX_SRC_BUFFER_SIZE) {
> -		if (video->srcs[0].size)
> -			aspeed_video_free_buf(video, &video->srcs[0]);
> -
> -		if (!aspeed_video_alloc_buf(video, &video->srcs[0],
> -					    VE_MAX_SRC_BUFFER_SIZE)) {
> -			dev_err(video->dev,
> -				"Failed to allocate source buffers\n");
> -			return;
> -		}
> -	}
> -
> -	aspeed_video_write(video, VE_SRC0_ADDR, video->srcs[0].dma);
> -
>   	do {
>   		if (tries) {
>   			set_current_state(TASK_INTERRUPTIBLE);

[PATCH v2 09/11] media: aspeed: use different delays for triggering VE H/W reset

[Jae Hyun Yoo]

In case of watchdog timeout detected while doing mode detection,
it's better triggering video engine hardware reset immediately so
this commit fixes code for the case. Other than the case, it will
trigger video engine hardware reset after RESOLUTION_CHANGE_DELAY.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
---
v1 -> v2:
 New.

 drivers/media/platform/aspeed-video.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index 4647ed2e9e63..67f476bf0a03 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -522,7 +522,7 @@ static void aspeed_video_bufs_done(struct aspeed_video *video,
 	spin_unlock_irqrestore(&video->lock, flags);
 }
 
-static void aspeed_video_irq_res_change(struct aspeed_video *video)
+static void aspeed_video_irq_res_change(struct aspeed_video *video, ulong delay)
 {
 	dev_dbg(video->dev, "Resolution changed; resetting\n");
 
@@ -532,7 +532,7 @@ static void aspeed_video_irq_res_change(struct aspeed_video *video)
 	aspeed_video_off(video);
 	aspeed_video_bufs_done(video, VB2_BUF_STATE_ERROR);
 
-	schedule_delayed_work(&video->res_work, RESOLUTION_CHANGE_DELAY);
+	schedule_delayed_work(&video->res_work, delay);
 }
 
 static irqreturn_t aspeed_video_irq(int irq, void *arg)
@@ -545,7 +545,7 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
 	 * re-initialize
 	 */
 	if (sts & VE_INTERRUPT_MODE_DETECT_WD) {
-		aspeed_video_irq_res_change(video);
+		aspeed_video_irq_res_change(video, 0);
 		return IRQ_HANDLED;
 	}
 
@@ -563,7 +563,8 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
 			 * Signal acquired while NOT doing resolution
 			 * detection; reset the engine and re-initialize
 			 */
-			aspeed_video_irq_res_change(video);
+			aspeed_video_irq_res_change(video,
+						    RESOLUTION_CHANGE_DELAY);
 			return IRQ_HANDLED;
 		}
 	}
-- 
2.21.0

Re: [PATCH v2 09/11] media: aspeed: use different delays for triggering VE H/W reset

[Eddie James]

On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
> In case of watchdog timeout detected while doing mode detection,
> it's better triggering video engine hardware reset immediately so
> this commit fixes code for the case. Other than the case, it will
> trigger video engine hardware reset after RESOLUTION_CHANGE_DELAY.


Reviewed-by: Eddie James <eajames@linux.ibm.com>


>
> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
> ---
> v1 -> v2:
>   New.
>
>   drivers/media/platform/aspeed-video.c | 9 +++++----
>   1 file changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
> index 4647ed2e9e63..67f476bf0a03 100644
> --- a/drivers/media/platform/aspeed-video.c
> +++ b/drivers/media/platform/aspeed-video.c
> @@ -522,7 +522,7 @@ static void aspeed_video_bufs_done(struct aspeed_video *video,
>   	spin_unlock_irqrestore(&video->lock, flags);
>   }
>   
> -static void aspeed_video_irq_res_change(struct aspeed_video *video)
> +static void aspeed_video_irq_res_change(struct aspeed_video *video, ulong delay)
>   {
>   	dev_dbg(video->dev, "Resolution changed; resetting\n");
>   
> @@ -532,7 +532,7 @@ static void aspeed_video_irq_res_change(struct aspeed_video *video)
>   	aspeed_video_off(video);
>   	aspeed_video_bufs_done(video, VB2_BUF_STATE_ERROR);
>   
> -	schedule_delayed_work(&video->res_work, RESOLUTION_CHANGE_DELAY);
> +	schedule_delayed_work(&video->res_work, delay);
>   }
>   
>   static irqreturn_t aspeed_video_irq(int irq, void *arg)
> @@ -545,7 +545,7 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
>   	 * re-initialize
>   	 */
>   	if (sts & VE_INTERRUPT_MODE_DETECT_WD) {
> -		aspeed_video_irq_res_change(video);
> +		aspeed_video_irq_res_change(video, 0);
>   		return IRQ_HANDLED;
>   	}
>   
> @@ -563,7 +563,8 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
>   			 * Signal acquired while NOT doing resolution
>   			 * detection; reset the engine and re-initialize
>   			 */
> -			aspeed_video_irq_res_change(video);
> +			aspeed_video_irq_res_change(video,
> +						    RESOLUTION_CHANGE_DELAY);
>   			return IRQ_HANDLED;
>   		}
>   	}

[PATCH v2 10/11] media: aspeed: fix an incorrect timeout checking in mode detection

[Jae Hyun Yoo]

There is an incorrect timeout checking in mode detection logic so
it misses resolution detecting chances. This commit fixes the bug.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
---
v1 -> v2:
 New.

 drivers/media/platform/aspeed-video.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index 67f476bf0a03..b05b073b63bc 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -735,7 +735,7 @@ static void aspeed_video_get_resolution(struct aspeed_video *video)
 	do {
 		if (tries) {
 			set_current_state(TASK_INTERRUPTIBLE);
-			if (schedule_timeout(INVALID_RESOLUTION_DELAY))
+			if (!schedule_timeout(INVALID_RESOLUTION_DELAY))
 				return;
 		}
 
-- 
2.21.0

Re: [PATCH v2 10/11] media: aspeed: fix an incorrect timeout checking in mode detection

[Eddie James]

On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
> There is an incorrect timeout checking in mode detection logic so
> it misses resolution detecting chances. This commit fixes the bug.
>
> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
> ---
> v1 -> v2:
>   New.
>
>   drivers/media/platform/aspeed-video.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
> index 67f476bf0a03..b05b073b63bc 100644
> --- a/drivers/media/platform/aspeed-video.c
> +++ b/drivers/media/platform/aspeed-video.c
> @@ -735,7 +735,7 @@ static void aspeed_video_get_resolution(struct aspeed_video *video)
>   	do {
>   		if (tries) {
>   			set_current_state(TASK_INTERRUPTIBLE);
> -			if (schedule_timeout(INVALID_RESOLUTION_DELAY))
> +			if (!schedule_timeout(INVALID_RESOLUTION_DELAY))
>   				return;


schedule_timeout returns 0 when the timer has expired otherwise the 
remaining time in  jiffies will be returned. So if it was interrupted (timer did not expire 
and it returns non-zero) then we should return, rather than keep trying. 
So I think it was correct before. Thanks, Eddie

>   		}
>   

Re: [PATCH v2 10/11] media: aspeed: fix an incorrect timeout checking in mode detection

[Jae Hyun Yoo]

On 5/29/2019 7:03 AM, Eddie James wrote:
> 
> On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
>> There is an incorrect timeout checking in mode detection logic so
>> it misses resolution detecting chances. This commit fixes the bug.
>>
>> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
>> ---
>> v1 -> v2:
>>   New.
>>
>>   drivers/media/platform/aspeed-video.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/media/platform/aspeed-video.c 
>> b/drivers/media/platform/aspeed-video.c
>> index 67f476bf0a03..b05b073b63bc 100644
>> --- a/drivers/media/platform/aspeed-video.c
>> +++ b/drivers/media/platform/aspeed-video.c
>> @@ -735,7 +735,7 @@ static void aspeed_video_get_resolution(struct 
>> aspeed_video *video)
>>       do {
>>           if (tries) {
>>               set_current_state(TASK_INTERRUPTIBLE);
>> -            if (schedule_timeout(INVALID_RESOLUTION_DELAY))
>> +            if (!schedule_timeout(INVALID_RESOLUTION_DELAY))
>>                   return;
> 
> 
> schedule_timeout returns 0 when the timer has expired otherwise the 
> remaining time in  jiffies will be returned. So if it was interrupted 
> (timer did not expire and it returns non-zero) then we should return, 
> rather than keep trying. So I think it was correct before. Thanks, Eddie

I thought that there is an explicit waking up case of this waiting
before the delay expires but I checked code again that there isn't. So
yes, you are right. Will drop this change from this series.

Thanks for the review!

Regards,
Jae

[PATCH v2 11/11] media: aspeed: add a workaround to fix a silicon bug

[Jae Hyun Yoo]

AST2500 silicon revision A1 and A2 have a silicon bug which causes
extremly long capturing time on specific resolutions (1680 width).
To fix the bug, this commit adjusts the capturing window register
setting to 1728 if detected width is 1680. The compression window
register setting will be kept as the original width so output
result will be the same.

Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
---
v1 -> v2:
 New.

 drivers/media/platform/aspeed-video.c | 26 +++++++++++++++++++-------
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
index b05b073b63bc..f93989f532d6 100644
--- a/drivers/media/platform/aspeed-video.c
+++ b/drivers/media/platform/aspeed-video.c
@@ -824,8 +824,27 @@ static void aspeed_video_set_resolution(struct aspeed_video *video)
 	struct v4l2_bt_timings *act = &video->active_timings;
 	unsigned int size = act->width * act->height;
 
+	/* Set capture/compression frame sizes */
 	aspeed_video_calc_compressed_size(video, size);
 
+	if (video->active_timings.width == 1680) {
+		/*
+		 * This is a workaround to fix a silicon bug on A1 and A2
+		 * revisions. Since it doesn't break capturing operation on A0
+		 * revision, use it for all revisions without checking the
+		 * revision ID.
+		 */
+		aspeed_video_write(video, VE_CAP_WINDOW,
+				   1728 << 16 | act->height);
+		size += (1728 - 1680) * video->active_timings.height;
+	} else {
+		aspeed_video_write(video, VE_CAP_WINDOW,
+				   act->width << 16 | act->height);
+	}
+	aspeed_video_write(video, VE_COMP_WINDOW,
+			   act->width << 16 | act->height);
+	aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 4);
+
 	/* Don't use direct mode below 1024 x 768 (irqs don't fire) */
 	if (size < DIRECT_FETCH_THRESHOLD) {
 		aspeed_video_write(video, VE_TGS_0,
@@ -842,13 +861,6 @@ static void aspeed_video_set_resolution(struct aspeed_video *video)
 		aspeed_video_update(video, VE_CTRL, 0, VE_CTRL_DIRECT_FETCH);
 	}
 
-	/* Set capture/compression frame sizes */
-	aspeed_video_write(video, VE_CAP_WINDOW,
-			   act->width << 16 | act->height);
-	aspeed_video_write(video, VE_COMP_WINDOW,
-			   act->width << 16 | act->height);
-	aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 4);
-
 	size *= 4;
 
 	if (size == video->srcs[0].size / 2) {
-- 
2.21.0

Re: [PATCH v2 11/11] media: aspeed: add a workaround to fix a silicon bug

[Eddie James]

On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
> AST2500 silicon revision A1 and A2 have a silicon bug which causes
> extremly long capturing time on specific resolutions (1680 width).
> To fix the bug, this commit adjusts the capturing window register
> setting to 1728 if detected width is 1680. The compression window
> register setting will be kept as the original width so output
> result will be the same.


This is a bit curious, why 1728 in particular? And what is the behavior 
of the VE when the capture window is larger than the actual source 
resolution?

Thanks,

Eddie


>
> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
> ---
> v1 -> v2:
>   New.
>
>   drivers/media/platform/aspeed-video.c | 26 +++++++++++++++++++-------
>   1 file changed, 19 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
> index b05b073b63bc..f93989f532d6 100644
> --- a/drivers/media/platform/aspeed-video.c
> +++ b/drivers/media/platform/aspeed-video.c
> @@ -824,8 +824,27 @@ static void aspeed_video_set_resolution(struct aspeed_video *video)
>   	struct v4l2_bt_timings *act = &video->active_timings;
>   	unsigned int size = act->width * act->height;
>   
> +	/* Set capture/compression frame sizes */
>   	aspeed_video_calc_compressed_size(video, size);
>   
> +	if (video->active_timings.width == 1680) {
> +		/*
> +		 * This is a workaround to fix a silicon bug on A1 and A2
> +		 * revisions. Since it doesn't break capturing operation on A0
> +		 * revision, use it for all revisions without checking the
> +		 * revision ID.
> +		 */
> +		aspeed_video_write(video, VE_CAP_WINDOW,
> +				   1728 << 16 | act->height);
> +		size += (1728 - 1680) * video->active_timings.height;
> +	} else {
> +		aspeed_video_write(video, VE_CAP_WINDOW,
> +				   act->width << 16 | act->height);
> +	}
> +	aspeed_video_write(video, VE_COMP_WINDOW,
> +			   act->width << 16 | act->height);
> +	aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 4);
> +
>   	/* Don't use direct mode below 1024 x 768 (irqs don't fire) */
>   	if (size < DIRECT_FETCH_THRESHOLD) {
>   		aspeed_video_write(video, VE_TGS_0,
> @@ -842,13 +861,6 @@ static void aspeed_video_set_resolution(struct aspeed_video *video)
>   		aspeed_video_update(video, VE_CTRL, 0, VE_CTRL_DIRECT_FETCH);
>   	}
>   
> -	/* Set capture/compression frame sizes */
> -	aspeed_video_write(video, VE_CAP_WINDOW,
> -			   act->width << 16 | act->height);
> -	aspeed_video_write(video, VE_COMP_WINDOW,
> -			   act->width << 16 | act->height);
> -	aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 4);
> -
>   	size *= 4;
>   
>   	if (size == video->srcs[0].size / 2) {

Re: [PATCH v2 11/11] media: aspeed: add a workaround to fix a silicon bug

[Jae Hyun Yoo]

On 5/29/2019 7:07 AM, Eddie James wrote:
> 
> On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
>> AST2500 silicon revision A1 and A2 have a silicon bug which causes
>> extremly long capturing time on specific resolutions (1680 width).
>> To fix the bug, this commit adjusts the capturing window register
>> setting to 1728 if detected width is 1680. The compression window
>> register setting will be kept as the original width so output
>> result will be the same.
> 
> 
> This is a bit curious, why 1728 in particular? And what is the behavior 
> of the VE when the capture window is larger than the actual source 
> resolution?

For an example, if resolution is 1680x1050, capturing operation takes
very long time because VE has the silicon bug. So this patch adjusts
the 'Capture Window' register slightly larger than 1680 to avoid the
issue. As a result, source buffer will copy 1728x1050 frames from the
original screen buffer but the image is still has valid information.
As the next step in compression phase, it will set the 'Compression
Window' register as '1680x1050' so it will compress using the original
image resolution which is a cropped image from the '1728x1050' source
buffer.

You can compare results using these shell commands in Ubuntu GUI
desktop.

$ xrandr --newmode "1680x1050_60.00"  146.25  1680 1784 1960 1240  1050 
1053 1059 1089 -hsync +vsync
$ xrandr --addmode VGA-1 1680x1050_60.00
$ xrandr --output VGA-1 --mode 1680x1050_60.00

I'm also curious about why that is 1728. Actually, this workaround was
provided from the chip vendor, Aspeed, and they use this in their SDK
code too. Let's check it to Ryan.


Hi Ryan,

Can you please explain why that is 1728 in particular.

Thanks,
Jae

> 
> Thanks,
> 
> Eddie
> 
> 
>>
>> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
>> ---
>> v1 -> v2:
>>   New.
>>
>>   drivers/media/platform/aspeed-video.c | 26 +++++++++++++++++++-------
>>   1 file changed, 19 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/media/platform/aspeed-video.c 
>> b/drivers/media/platform/aspeed-video.c
>> index b05b073b63bc..f93989f532d6 100644
>> --- a/drivers/media/platform/aspeed-video.c
>> +++ b/drivers/media/platform/aspeed-video.c
>> @@ -824,8 +824,27 @@ static void aspeed_video_set_resolution(struct 
>> aspeed_video *video)
>>       struct v4l2_bt_timings *act = &video->active_timings;
>>       unsigned int size = act->width * act->height;
>> +    /* Set capture/compression frame sizes */
>>       aspeed_video_calc_compressed_size(video, size);
>> +    if (video->active_timings.width == 1680) {
>> +        /*
>> +         * This is a workaround to fix a silicon bug on A1 and A2
>> +         * revisions. Since it doesn't break capturing operation on A0
>> +         * revision, use it for all revisions without checking the
>> +         * revision ID.
>> +         */
>> +        aspeed_video_write(video, VE_CAP_WINDOW,
>> +                   1728 << 16 | act->height);
>> +        size += (1728 - 1680) * video->active_timings.height;
>> +    } else {
>> +        aspeed_video_write(video, VE_CAP_WINDOW,
>> +                   act->width << 16 | act->height);
>> +    }
>> +    aspeed_video_write(video, VE_COMP_WINDOW,
>> +               act->width << 16 | act->height);
>> +    aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 4);
>> +
>>       /* Don't use direct mode below 1024 x 768 (irqs don't fire) */
>>       if (size < DIRECT_FETCH_THRESHOLD) {
>>           aspeed_video_write(video, VE_TGS_0,
>> @@ -842,13 +861,6 @@ static void aspeed_video_set_resolution(struct 
>> aspeed_video *video)
>>           aspeed_video_update(video, VE_CTRL, 0, VE_CTRL_DIRECT_FETCH);
>>       }
>> -    /* Set capture/compression frame sizes */
>> -    aspeed_video_write(video, VE_CAP_WINDOW,
>> -               act->width << 16 | act->height);
>> -    aspeed_video_write(video, VE_COMP_WINDOW,
>> -               act->width << 16 | act->height);
>> -    aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 4);
>> -
>>       size *= 4;
>>       if (size == video->srcs[0].size / 2) {
> 
> 

RE: [PATCH v2 11/11] media: aspeed: add a workaround to fix a silicon bug

[Ryan Chen]

> 
> On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
>> AST2500 silicon revision A1 and A2 have a silicon bug which causes 
>> extremly long capturing time on specific resolutions (1680 width).
>> To fix the bug, this commit adjusts the capturing window register 
>> setting to 1728 if detected width is 1680. The compression window 
>> register setting will be kept as the original width so output result 
>> will be the same.
> 
> 
> This is a bit curious, why 1728 in particular? And what is the 
> behavior of the VE when the capture window is larger than the actual 
> source resolution?

>For an example, if resolution is 1680x1050, capturing operation takes very long time because VE has the silicon bug. So this patch adjusts the 'Capture Window' register slightly larger than >1680 to avoid the issue. As a result, source buffer will copy 1728x1050 frames from the original screen buffer but the image is still has valid information.
>As the next step in compression phase, it will set the 'Compression Window' register as '1680x1050' so it will compress using the original image resolution which is a cropped image from the >'1728x1050' source buffer.

>You can compare results using these shell commands in Ubuntu GUI desktop.

>$ xrandr --newmode "1680x1050_60.00"  146.25  1680 1784 1960 1240  1050
>1053 1059 1089 -hsync +vsync
>$ xrandr --addmode VGA-1 1680x1050_60.00 $ xrandr --output VGA-1 --mode 1680x1050_60.00

>I'm also curious about why that is 1728. Actually, this workaround was provided from the chip vendor, Aspeed, and they use this in their SDK code too. Let's check it to Ryan.


>Hi Ryan,

>Can you please explain why that is 1728 in particular.

>Thanks,
>Jae

That have two factor, one is data too huge cause the memory bandwidth is too busy.  
The other is vga resolution width is not align to 32 or 64 pixels (32bpp or 16bpp).
Those will cause engine read latency time too long, maybe engine will hange.

Ryan

> 
> Thanks,
> 
> Eddie
> 
> 
>>
>> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
>> ---
>> v1 -> v2:
>>   New.
>>
>>   drivers/media/platform/aspeed-video.c | 26 
>> +++++++++++++++++++-------
>>   1 file changed, 19 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/media/platform/aspeed-video.c
>> b/drivers/media/platform/aspeed-video.c
>> index b05b073b63bc..f93989f532d6 100644
>> --- a/drivers/media/platform/aspeed-video.c
>> +++ b/drivers/media/platform/aspeed-video.c
>> @@ -824,8 +824,27 @@ static void aspeed_video_set_resolution(struct
>> aspeed_video *video)
>>       struct v4l2_bt_timings *act = &video->active_timings;
>>       unsigned int size = act->width * act->height;
>> +    /* Set capture/compression frame sizes */
>>       aspeed_video_calc_compressed_size(video, size);
>> +    if (video->active_timings.width == 1680) {
>> +        /*
>> +         * This is a workaround to fix a silicon bug on A1 and A2
>> +         * revisions. Since it doesn't break capturing operation on 
>> +A0
>> +         * revision, use it for all revisions without checking the
>> +         * revision ID.
>> +         */
>> +        aspeed_video_write(video, VE_CAP_WINDOW,
>> +                   1728 << 16 | act->height);
>> +        size += (1728 - 1680) * video->active_timings.height;
>> +    } else {
>> +        aspeed_video_write(video, VE_CAP_WINDOW,
>> +                   act->width << 16 | act->height);
>> +    }
>> +    aspeed_video_write(video, VE_COMP_WINDOW,
>> +               act->width << 16 | act->height);
>> +    aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 
>> +4);
>> +
>>       /* Don't use direct mode below 1024 x 768 (irqs don't fire) */
>>       if (size < DIRECT_FETCH_THRESHOLD) {
>>           aspeed_video_write(video, VE_TGS_0, @@ -842,13 +861,6 @@ 
>> static void aspeed_video_set_resolution(struct
>> aspeed_video *video)
>>           aspeed_video_update(video, VE_CTRL, 0, 
>> VE_CTRL_DIRECT_FETCH);
>>       }
>> -    /* Set capture/compression frame sizes */
>> -    aspeed_video_write(video, VE_CAP_WINDOW,
>> -               act->width << 16 | act->height);
>> -    aspeed_video_write(video, VE_COMP_WINDOW,
>> -               act->width << 16 | act->height);
>> -    aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width * 
>> 4);
>> -
>>       size *= 4;
>>       if (size == video->srcs[0].size / 2) {
> 
> 

Re: [PATCH v2 11/11] media: aspeed: add a workaround to fix a silicon bug

[Jae Hyun Yoo]

Hi Ryan,

On 5/31/2019 9:32 AM, Ryan Chen wrote:
>>
>> On 5/24/19 6:17 PM, Jae Hyun Yoo wrote:
>>> AST2500 silicon revision A1 and A2 have a silicon bug which causes
>>> extremly long capturing time on specific resolutions (1680 width).
>>> To fix the bug, this commit adjusts the capturing window register
>>> setting to 1728 if detected width is 1680. The compression window
>>> register setting will be kept as the original width so output result
>>> will be the same.
>>
>>
>> This is a bit curious, why 1728 in particular? And what is the
>> behavior of the VE when the capture window is larger than the actual
>> source resolution?
> 
>> For an example, if resolution is 1680x1050, capturing operation takes very long time because VE has the silicon bug. So this patch adjusts the 'Capture Window' register slightly larger than >1680 to avoid the issue. As a result, source buffer will copy 1728x1050 frames from the original screen buffer but the image is still has valid information.
>> As the next step in compression phase, it will set the 'Compression Window' register as '1680x1050' so it will compress using the original image resolution which is a cropped image from the >'1728x1050' source buffer.
> 
>> You can compare results using these shell commands in Ubuntu GUI desktop.
> 
>> $ xrandr --newmode "1680x1050_60.00"  146.25  1680 1784 1960 1240  1050
>> 1053 1059 1089 -hsync +vsync
>> $ xrandr --addmode VGA-1 1680x1050_60.00 $ xrandr --output VGA-1 --mode 1680x1050_60.00
> 
>> I'm also curious about why that is 1728. Actually, this workaround was provided from the chip vendor, Aspeed, and they use this in their SDK code too. Let's check it to Ryan.
> 
> 
>> Hi Ryan,
> 
>> Can you please explain why that is 1728 in particular.
> 
>> Thanks,
>> Jae
> 
> That have two factor, one is data too huge cause the memory bandwidth is too busy.
> The other is vga resolution width is not align to 32 or 64 pixels (32bpp or 16bpp).
> Those will cause engine read latency time too long, maybe engine will hange.

Thanks a lot for the detail. I'll add it as an additional comment in
code.

Thanks,
Jae

> 
> Ryan
> 
>>
>> Thanks,
>>
>> Eddie
>>
>>
>>>
>>> Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
>>> ---
>>> v1 -> v2:
>>>    New.
>>>
>>>    drivers/media/platform/aspeed-video.c | 26
>>> +++++++++++++++++++-------
>>>    1 file changed, 19 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/drivers/media/platform/aspeed-video.c
>>> b/drivers/media/platform/aspeed-video.c
>>> index b05b073b63bc..f93989f532d6 100644
>>> --- a/drivers/media/platform/aspeed-video.c
>>> +++ b/drivers/media/platform/aspeed-video.c
>>> @@ -824,8 +824,27 @@ static void aspeed_video_set_resolution(struct
>>> aspeed_video *video)
>>>        struct v4l2_bt_timings *act = &video->active_timings;
>>>        unsigned int size = act->width * act->height;
>>> +    /* Set capture/compression frame sizes */
>>>        aspeed_video_calc_compressed_size(video, size);
>>> +    if (video->active_timings.width == 1680) {
>>> +        /*
>>> +         * This is a workaround to fix a silicon bug on A1 and A2
>>> +         * revisions. Since it doesn't break capturing operation on
>>> +A0
>>> +         * revision, use it for all revisions without checking the
>>> +         * revision ID.
>>> +         */
>>> +        aspeed_video_write(video, VE_CAP_WINDOW,
>>> +                   1728 << 16 | act->height);
>>> +        size += (1728 - 1680) * video->active_timings.height;
>>> +    } else {
>>> +        aspeed_video_write(video, VE_CAP_WINDOW,
>>> +                   act->width << 16 | act->height);
>>> +    }
>>> +    aspeed_video_write(video, VE_COMP_WINDOW,
>>> +               act->width << 16 | act->height);
>>> +    aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width *
>>> +4);
>>> +
>>>        /* Don't use direct mode below 1024 x 768 (irqs don't fire) */
>>>        if (size < DIRECT_FETCH_THRESHOLD) {
>>>            aspeed_video_write(video, VE_TGS_0, @@ -842,13 +861,6 @@
>>> static void aspeed_video_set_resolution(struct
>>> aspeed_video *video)
>>>            aspeed_video_update(video, VE_CTRL, 0,
>>> VE_CTRL_DIRECT_FETCH);
>>>        }
>>> -    /* Set capture/compression frame sizes */
>>> -    aspeed_video_write(video, VE_CAP_WINDOW,
>>> -               act->width << 16 | act->height);
>>> -    aspeed_video_write(video, VE_COMP_WINDOW,
>>> -               act->width << 16 | act->height);
>>> -    aspeed_video_write(video, VE_SRC_SCANLINE_OFFSET, act->width *
>>> 4);
>>> -
>>>        size *= 4;
>>>        if (size == video->srcs[0].size / 2) {
>>
>>

Re: [PATCH v2 00/11] Improve stability and add bug fixes of Aspeed video engine driver

[Hans Verkuil]

Eddie,

Can you review the last 4 new patches? If all is OK, then I can make a pull
request for this series.

Regards,

	Hans

On 5/25/19 1:17 AM, Jae Hyun Yoo wrote:
> This patch series improves stability of Aspeed video engine driver by fixing
> clock control and irq handling logic in the driver. Also, it adds a couple of
> bug fixes and a workaroud for a silicon bug.
> 
> Changes since v1:
> - Removed spinlock handling code from 0001 patch.
> - Added 4 more patches.
> 
> Jae Hyun Yoo (11):
>   media: aspeed: fix a kernel warning on clk control
>   media: aspeed: refine clock control logic
>   media: aspeed: change irq to threaded irq
>   media: aspeed: remove IRQF_SHARED flag
>   media: aspeed: reduce noisy log printing outs
>   media: aspeed: remove checking of VE_INTERRUPT_CAPTURE_COMPLETE
>   media: aspeed: refine interrupt handling logic
>   media: aspeed: remove source buffer allocation before mode detection
>   media: aspeed: use different delays for triggering VE H/W reset
>   media: aspeed: fix an incorrect timeout checking in mode detection
>   media: aspeed: add a workaround to fix a silicon bug
> 
>  drivers/media/platform/aspeed-video.c | 140 +++++++++++++++-----------
>  1 file changed, 80 insertions(+), 60 deletions(-)
>