It is possible that mtk_vcodec_enc_pw_on fails, and in that scenario
the PM counter is not incremented, and subsequent call to
mtk_vcodec_enc_pw_off decrements the counter, leading to a PM imbalance.
Fix by bailing out of venc_if_encode in the case when mtk_vcodec_enc_pw_on
fails.
Fixes: 4e855a6efa54 ("[media] vcodec: mediatek: Add Mediatek V4L2 Video Encoder Driver")
Signed-off-by: Eugen Hristev <[email protected]>
Reviewed-by: AngeloGioacchino Del Regno <[email protected]>
---
Changes in v2:
- collect R-b
.../platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.c | 4 +++-
.../platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.h | 2 +-
drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c | 5 ++++-
3 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.c b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.c
index a22b7dfc656e..1a2b14a3e219 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.c
@@ -58,13 +58,15 @@ int mtk_vcodec_init_enc_clk(struct mtk_vcodec_enc_dev *mtkdev)
return 0;
}
-void mtk_vcodec_enc_pw_on(struct mtk_vcodec_pm *pm)
+int mtk_vcodec_enc_pw_on(struct mtk_vcodec_pm *pm)
{
int ret;
ret = pm_runtime_resume_and_get(pm->dev);
if (ret)
dev_err(pm->dev, "pm_runtime_resume_and_get fail: %d", ret);
+
+ return ret;
}
void mtk_vcodec_enc_pw_off(struct mtk_vcodec_pm *pm)
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.h b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.h
index 157ea08ba9e3..2e28f25e36cc 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.h
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_pm.h
@@ -10,7 +10,7 @@
#include "mtk_vcodec_enc_drv.h"
int mtk_vcodec_init_enc_clk(struct mtk_vcodec_enc_dev *dev);
-void mtk_vcodec_enc_pw_on(struct mtk_vcodec_pm *pm);
+int mtk_vcodec_enc_pw_on(struct mtk_vcodec_pm *pm);
void mtk_vcodec_enc_pw_off(struct mtk_vcodec_pm *pm);
void mtk_vcodec_enc_clock_on(struct mtk_vcodec_pm *pm);
void mtk_vcodec_enc_clock_off(struct mtk_vcodec_pm *pm);
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c b/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c
index c402a686f3cb..e83747b8d69a 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c
@@ -64,7 +64,9 @@ int venc_if_encode(struct mtk_vcodec_enc_ctx *ctx,
ctx->dev->curr_ctx = ctx;
spin_unlock_irqrestore(&ctx->dev->irqlock, flags);
- mtk_vcodec_enc_pw_on(&ctx->dev->pm);
+ ret = mtk_vcodec_enc_pw_on(&ctx->dev->pm);
+ if (ret)
+ goto venc_if_encode_pw_on_err;
mtk_vcodec_enc_clock_on(&ctx->dev->pm);
ret = ctx->enc_if->encode(ctx->drv_handle, opt, frm_buf,
bs_buf, result);
@@ -75,6 +77,7 @@ int venc_if_encode(struct mtk_vcodec_enc_ctx *ctx,
ctx->dev->curr_ctx = NULL;
spin_unlock_irqrestore(&ctx->dev->irqlock, flags);
+venc_if_encode_pw_on_err:
mtk_venc_unlock(ctx);
return ret;
}
--
2.34.1
As IOMMUs are supported, dma-ranges is not mentioned but
additionalProperties=false, thus we have an error when adding dma-ranges.
Add dma-ranges as a possible property because this may be present.
Signed-off-by: Eugen Hristev <[email protected]>
---
Changes in v2:
- none.
.../devicetree/bindings/media/mediatek,vcodec-encoder.yaml | 2 ++
1 file changed, 2 insertions(+)
diff --git a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
index a2051b31fa29..403530de5624 100644
--- a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
+++ b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
@@ -38,6 +38,8 @@ properties:
minItems: 1
maxItems: 5
+ dma-ranges: true
+
assigned-clocks: true
assigned-clock-parents: true
--
2.34.1
Looking at the binding it makes sense that the `-vp8` compatible has
the `venc_lt_sel` while the other bindings have the `venc_sel` as name for
the clock.
This was also mentioned in the txt version of the binding before the
conversion:
`
clock-names: avc encoder must contain "venc_sel", vp8 encoder must
contain "venc_lt_sel", decoder must contain "vcodecpll", "univpll_d2",
`
So it is easier to check for compatible that includes vp8, since that's
just one, to have the requirement for the clock name property as
`venc_lt_sel`, rather than for all the others, some of which are missing,
thus for them, the requirement is wrongly `venc_lt_sel`.
Reordered the if/then/else to match `-vp8` and have all the rest of
the compatibles using the other clock name (`venc_sel`).
Signed-off-by: Eugen Hristev <[email protected]>
---
Changes in v2:
- new patch.
.../bindings/media/mediatek,vcodec-encoder.yaml | 11 ++++-------
1 file changed, 4 insertions(+), 7 deletions(-)
diff --git a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
index 403530de5624..5166c1c6934e 100644
--- a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
+++ b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
@@ -111,10 +111,7 @@ allOf:
properties:
compatible:
enum:
- - mediatek,mt8173-vcodec-enc
- - mediatek,mt8188-vcodec-enc
- - mediatek,mt8192-vcodec-enc
- - mediatek,mt8195-vcodec-enc
+ - mediatek,mt8173-vcodec-enc-vp8
then:
properties:
@@ -124,8 +121,8 @@ allOf:
maxItems: 1
clock-names:
items:
- - const: venc_sel
- else: # for vp8 hw encoder
+ - const: venc_lt_sel
+ else:
properties:
clock:
items:
@@ -133,7 +130,7 @@ allOf:
maxItems: 1
clock-names:
items:
- - const: venc_lt_sel
+ - const: venc_sel
additionalProperties: false
--
2.34.1
Clock name should be `venc_sel` as per binding.
Fix the warning message :
arch/arm64/boot/dts/mediatek/mt8192-asurada-hayato-r1.dtb: vcodec@17020000: clock-names:0: 'venc_sel' was expected
from schema $id: http://devicetree.org/schemas/media/mediatek,vcodec-encoder.yaml#
Fixes: aa8f3711fc87 ("arm64: dts: mt8192: Add H264 venc device node")
Signed-off-by: Eugen Hristev <[email protected]>
---
Changes in v2:
- new patch.
arch/arm64/boot/dts/mediatek/mt8192.dtsi | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm64/boot/dts/mediatek/mt8192.dtsi b/arch/arm64/boot/dts/mediatek/mt8192.dtsi
index 69f4cded5dbb..f1fc14e53f8c 100644
--- a/arch/arm64/boot/dts/mediatek/mt8192.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8192.dtsi
@@ -1770,7 +1770,7 @@ vcodec_enc: vcodec@17020000 {
mediatek,scp = <&scp>;
power-domains = <&spm MT8192_POWER_DOMAIN_VENC>;
clocks = <&vencsys CLK_VENC_SET1_VENC>;
- clock-names = "venc-set1";
+ clock-names = "venc_sel";
assigned-clocks = <&topckgen CLK_TOP_VENC_SEL>;
assigned-clock-parents = <&topckgen CLK_TOP_UNIVPLL_D4>;
};
--
2.34.1
Add compatible for the mt8186 encoder which currently works in the same
way as mt8183.
Signed-off-by: Eugen Hristev <[email protected]>
---
Changes in v2:
- new patch.
.../media/mediatek,vcodec-encoder.yaml | 20 +++++++++++--------
1 file changed, 12 insertions(+), 8 deletions(-)
diff --git a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
index 5166c1c6934e..dfabc2faea6c 100644
--- a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
+++ b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
@@ -16,14 +16,18 @@ description: |+
properties:
compatible:
- enum:
- - mediatek,mt8173-vcodec-enc-vp8
- - mediatek,mt8173-vcodec-enc
- - mediatek,mt8183-vcodec-enc
- - mediatek,mt8188-vcodec-enc
- - mediatek,mt8192-vcodec-enc
- - mediatek,mt8195-vcodec-enc
-
+ oneOf:
+ - items:
+ - enum:
+ - mediatek,mt8173-vcodec-enc-vp8
+ - mediatek,mt8173-vcodec-enc
+ - mediatek,mt8183-vcodec-enc
+ - mediatek,mt8188-vcodec-enc
+ - mediatek,mt8192-vcodec-enc
+ - mediatek,mt8195-vcodec-enc
+ - items:
+ - const: mediatek,mt8186-vcodec-enc
+ - const: mediatek,mt8183-vcodec-enc
reg:
maxItems: 1
--
2.34.1
The larb clock is in fact a subsys clock, so it must be prefixed by
'subsys-' to be correctly identified in the driver.
Fixes: d9e43c1e7a38 ("arm64: dts: mt8186: Add power domains controller")
Signed-off-by: Eugen Hristev <[email protected]>
---
Changes in v2:
- none.
arch/arm64/boot/dts/mediatek/mt8186.dtsi | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm64/boot/dts/mediatek/mt8186.dtsi b/arch/arm64/boot/dts/mediatek/mt8186.dtsi
index df0c04f2ba1d..66ead3f23336 100644
--- a/arch/arm64/boot/dts/mediatek/mt8186.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8186.dtsi
@@ -1061,7 +1061,7 @@ power-domain@MT8186_POWER_DOMAIN_VENC {
reg = <MT8186_POWER_DOMAIN_VENC>;
clocks = <&topckgen CLK_TOP_VENC>,
<&vencsys CLK_VENC_CKE1_VENC>;
- clock-names = "venc0", "larb";
+ clock-names = "venc0", "subsys-larb";
mediatek,infracfg = <&infracfg_ao>;
#power-domain-cells = <0>;
};
--
2.34.1
From: Kyrie Wu <[email protected]>
Add video encoder node.
Signed-off-by: Kyrie Wu <[email protected]>
Signed-off-by: Allen-KH Cheng <[email protected]>
Reviewed-by: Hsin-Yi Wang <[email protected]>
[[email protected]: minor cleanup]
Signed-off-by: Eugen Hristev <[email protected]>
---
Changes in v2:
- change node name
- change compatible to include 8186
- change props order
- change clock name to cope with binding
arch/arm64/boot/dts/mediatek/mt8186.dtsi | 24 ++++++++++++++++++++++++
1 file changed, 24 insertions(+)
diff --git a/arch/arm64/boot/dts/mediatek/mt8186.dtsi b/arch/arm64/boot/dts/mediatek/mt8186.dtsi
index 66ead3f23336..ebd07bf3d9d2 100644
--- a/arch/arm64/boot/dts/mediatek/mt8186.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8186.dtsi
@@ -1993,6 +1993,30 @@ larb7: smi@17010000 {
power-domains = <&spm MT8186_POWER_DOMAIN_VENC>;
};
+ venc: video-encoder@17020000 {
+ compatible = "mediatek,mt8186-vcodec-enc", "mediatek,mt8183-vcodec-enc";
+ reg = <0 0x17020000 0 0x2000>;
+ #address-cells = <2>;
+ #size-cells = <2>;
+ interrupts = <GIC_SPI 243 IRQ_TYPE_LEVEL_HIGH 0>;
+ iommus = <&iommu_mm IOMMU_PORT_L7_VENC_RCPU>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_REC>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_BSDMA>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_SV_COMV>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_RD_COMV>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_CUR_LUMA>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_CUR_CHROMA>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_REF_LUMA>,
+ <&iommu_mm IOMMU_PORT_L7_VENC_REF_CHROMA>;
+ dma-ranges = <0x1 0x0 0x1 0x0 0x1 0x0>;
+ clocks = <&vencsys CLK_VENC_CKE1_VENC>;
+ clock-names = "venc_sel";
+ assigned-clocks = <&topckgen CLK_TOP_VENC>;
+ assigned-clock-parents = <&topckgen CLK_TOP_UNIVPLL_D3>;
+ power-domains = <&spm MT8186_POWER_DOMAIN_VENC>;
+ mediatek,scp = <&scp>;
+ };
+
camsys: clock-controller@1a000000 {
compatible = "mediatek,mt8186-camsys";
reg = <0 0x1a000000 0 0x1000>;
--
2.34.1
On Fri, 15 Dec 2023 12:45:46 +0200, Eugen Hristev wrote:
> As IOMMUs are supported, dma-ranges is not mentioned but
> additionalProperties=false, thus we have an error when adding dma-ranges.
> Add dma-ranges as a possible property because this may be present.
>
> Signed-off-by: Eugen Hristev <[email protected]>
> ---
> Changes in v2:
> - none.
> .../devicetree/bindings/media/mediatek,vcodec-encoder.yaml | 2 ++
> 1 file changed, 2 insertions(+)
>
Acked-by: Rob Herring <[email protected]>
On Fri, 15 Dec 2023 12:45:47 +0200, Eugen Hristev wrote:
> Looking at the binding it makes sense that the `-vp8` compatible has
> the `venc_lt_sel` while the other bindings have the `venc_sel` as name for
> the clock.
> This was also mentioned in the txt version of the binding before the
> conversion:
> `
> clock-names: avc encoder must contain "venc_sel", vp8 encoder must
> contain "venc_lt_sel", decoder must contain "vcodecpll", "univpll_d2",
> `
>
> So it is easier to check for compatible that includes vp8, since that's
> just one, to have the requirement for the clock name property as
> `venc_lt_sel`, rather than for all the others, some of which are missing,
> thus for them, the requirement is wrongly `venc_lt_sel`.
>
> Reordered the if/then/else to match `-vp8` and have all the rest of
> the compatibles using the other clock name (`venc_sel`).
>
> Signed-off-by: Eugen Hristev <[email protected]>
> ---
> Changes in v2:
> - new patch.
> .../bindings/media/mediatek,vcodec-encoder.yaml | 11 ++++-------
> 1 file changed, 4 insertions(+), 7 deletions(-)
>
Reviewed-by: Rob Herring <[email protected]>
On Fri, 15 Dec 2023 12:45:49 +0200, Eugen Hristev wrote:
> Add compatible for the mt8186 encoder which currently works in the same
> way as mt8183.
>
> Signed-off-by: Eugen Hristev <[email protected]>
> ---
> Changes in v2:
> - new patch.
> .../media/mediatek,vcodec-encoder.yaml | 20 +++++++++++--------
> 1 file changed, 12 insertions(+), 8 deletions(-)
>
Reviewed-by: Rob Herring <[email protected]>