| Message ID | 20220315061031.21642-1-moudy.ho@mediatek.com |
|---|---|
| Headers | show |
| Series | Add mutex support for MDP | expand |
Hi, Moudy: On Tue, 2022-03-15 at 14:10 +0800, Moudy Ho wrote: > Due to hardware limitations, MDP3 is necessary to enable MUTEX in > each frame for SOF triggering and cooperate with CMDQ control to > reduce the amount of interrupts generated(also, reduce frame > latency). > > In response to the above situation, a new interface > "mtk_mutex_enable_by_cmdq" has been added to achieve the purpose. > > Signed-off-by: Moudy Ho <moudy.ho@mediatek.com> > --- > drivers/soc/mediatek/mtk-mutex.c | 42 > +++++++++++++++++++++++++- > include/linux/soc/mediatek/mtk-mutex.h | 2 ++ > 2 files changed, 43 insertions(+), 1 deletion(-) > > diff --git a/drivers/soc/mediatek/mtk-mutex.c > b/drivers/soc/mediatek/mtk-mutex.c > index 88fb4fc8f216..0d85111cfbe2 100644 > --- a/drivers/soc/mediatek/mtk-mutex.c > +++ b/drivers/soc/mediatek/mtk-mutex.c > @@ -7,10 +7,16 @@ > #include <linux/iopoll.h> > #include <linux/module.h> > #include <linux/of_device.h> > +#include <linux/of_address.h> > #include <linux/platform_device.h> > #include <linux/regmap.h> > #include <linux/soc/mediatek/mtk-mmsys.h> > #include <linux/soc/mediatek/mtk-mutex.h> > +#include <linux/soc/mediatek/mtk-cmdq.h> > + > +#define MTK_MUTEX_ENABLE BIT(0) This is not related to cmdq. Separate this to another patch and apply to mtk_mutex_enable() also. > +#define MTK_MUTEX_MOD_MASK 0x07FFFFFF Useless, remove. > +#define MTK_MUTEX_SOF_MASK 0x00000007 Ditto. > > #define MT2701_MUTEX0_MOD0 0x2c > #define MT2701_MUTEX0_SOF0 0x30 > @@ -167,6 +173,7 @@ struct mtk_mutex_data { > const unsigned int mutex_sof_reg; > const unsigned long long *mutex_table_mod; > const bool no_clk; > + const bool has_gce_client_reg; > }; > > struct mtk_mutex_ctx { > @@ -175,6 +182,8 @@ struct mtk_mutex_ctx { > void __iomem *regs; > struct mtk_mutex mutex[10]; > const struct mtk_mutex_data *data; > + phys_addr_t addr; > + struct cmdq_client_reg cmdq_reg; > }; > > static const unsigned int mt2701_mutex_mod[DDP_COMPONENT_ID_MAX] = { > @@ -357,6 +366,7 @@ static const struct mtk_mutex_data > mt8183_mutex_driver_data = { > .mutex_sof_reg = MT8183_MUTEX0_SOF0, > .mutex_table_mod = mt8183_mutex_table_mod, > .no_clk = true, > + .has_gce_client_reg = true, > }; > > static const struct mtk_mutex_data mt8186_mutex_driver_data = { > @@ -639,6 +649,22 @@ void mtk_mutex_enable(struct mtk_mutex *mutex) > } > EXPORT_SYMBOL_GPL(mtk_mutex_enable); > > +void mtk_mutex_enable_by_cmdq(struct mtk_mutex *mutex, > + struct cmdq_pkt *pkt) > +{ > +#if IS_REACHABLE(CONFIG_MTK_CMDQ) > + struct mtk_mutex_ctx *mtx = container_of(mutex, struct > mtk_mutex_ctx, > + mutex[mutex->id]); > + > + WARN_ON(&mtx->mutex[mutex->id] != mutex); > + > + cmdq_pkt_write_mask(pkt, mtx->cmdq_reg.subsys, > + mtx->aaddr + DISP_REG_MUTEX_EN(mutex->id), > + MTK_MUTEX_ENABLE, MTK_MUTEX_ENABLE); #else /* show some error message here */ > +#endif > +} > +EXPORT_SYMBOL_GPL(mtk_mutex_enable_by_cmdq); > + > void mtk_mutex_disable(struct mtk_mutex *mutex) > { > struct mtk_mutex_ctx *mtx = container_of(mutex, struct > mtk_mutex_ctx, > @@ -677,7 +703,7 @@ static int mtk_mutex_probe(struct platform_device > *pdev) > { > struct device *dev = &pdev->dev; > struct mtk_mutex_ctx *mtx; > - struct resource *regs; > + struct resource *regs, addr; > int i; > > mtx = devm_kzalloc(dev, sizeof(*mtx), GFP_KERNEL); > @@ -698,6 +724,20 @@ static int mtk_mutex_probe(struct > platform_device *pdev) > } > } > > + if (of_address_to_resource(dev->of_node, 0, &addr) < 0) > + mtx->addr = 0L; > + else > + mtx->addr = addr.start; > + > +#if IS_REACHABLE(CONFIG_MTK_CMDQ) > + if (mtx->data->has_gce_client_reg) { In current upstreamed mt8183.dtsi [1], mutex device node has no mediatek,gce-client-reg property and display maybe work (?). So I think this should be backward-compatible. [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/boot/dts/mediatek/mt8183.dtsi?h=v5.17-rc8#n1510 Regards, CK > + if (cmdq_dev_get_client_reg(dev, &mtx->cmdq_reg, 0)) { > + dev_err(dev, "No mediatek,gce-client-reg!\n"); > + return ret; > + } > + } > +#endif > + > regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); > mtx->regs = devm_ioremap_resource(dev, regs); > if (IS_ERR(mtx->regs)) { > diff --git a/include/linux/soc/mediatek/mtk-mutex.h > b/include/linux/soc/mediatek/mtk-mutex.h > index c8355bb0e6d6..18a4b1dc3773 100644 > --- a/include/linux/soc/mediatek/mtk-mutex.h > +++ b/include/linux/soc/mediatek/mtk-mutex.h > @@ -45,6 +45,8 @@ void mtk_mutex_set_mod(struct mtk_mutex *mutex, > void mtk_mutex_set_sof(struct mtk_mutex *mutex, > enum mtk_mutex_table_index idx); > void mtk_mutex_enable(struct mtk_mutex *mutex); > +void mtk_mutex_enable_by_cmdq(struct mtk_mutex *mutex, > + struct cmdq_pkt *pkt); > void mtk_mutex_disable(struct mtk_mutex *mutex); > void mtk_mutex_remove_comp(struct mtk_mutex *mutex, > enum mtk_ddp_comp_id id);
Change since V12: - Rebase on linux-next - Remove ISP related settings in MMSYS - Removed CMDQ operations previously used by MDP in MMSYS - Move mediatek MUTEX dt-binding path - Add additional property in MUTEX for CMDQ operations Change since V11: - Rebase on v5.17-rc6. [1]: https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux.git/commit/?h=v5.17-next/soc&id=5f9b5b757e44de47ebdc116c14b90e3cc8bc7acb [2]: https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux.git/commit/?h=v5.17-next/soc&id=831785f0e5b919c29e1bc5f9a74e9ebd38289e24 [3]: https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux.git/commit/?h=v5.17-next/soc&id=15f1768365aed810826a61fef4a744437aa5b426 Change since v10: - For some ISP application scenarios, such as preview and recording at the same time. The routing table needs to be discarded, and the calculation result on the SCP side is used to write a suitable mux setting for 1 input and 2 output. - Adjust the GCE client register parsing method to avoid redundant warning logs. Change since v9: - Add API for MDP getting mutex mod. Hi, This patch splits mmsys and mutex settings from Media Data Path 3 (MDP3), and original mailling list list below: https://patchwork.kernel.org/project/linux-mediatek/cover/20211201095031.31606-1-moudy.ho@mediatek.com/ Corresponding settings and interfaces are added for MDP operation in the mutex drivers, which increases the independence of the modules Moudy Ho (6): soc: mediatek: mutex: add common interface to accommodate multiple modules operationg MUTEX soc: mediatek: mutex: add 8183 MUTEX MOD settings for MDP dt-bindings: soc: mediatek: move out common module from display folder dt-bindings: soc: mediatek: add gce-client-reg for MUTEX dts: arm64: mt8183: add GCE client property for Mediatek MUTEX soc: mediatek: mutex: add functions that operate registers by CMDQ .../mediatek/mediatek,mutex.yaml | 18 +- arch/arm64/boot/dts/mediatek/mt8183.dtsi | 1 + drivers/soc/mediatek/mtk-mutex.c | 185 +++++++++++++++++- include/linux/soc/mediatek/mtk-mutex.h | 35 ++++ 4 files changed, 231 insertions(+), 8 deletions(-) rename Documentation/devicetree/bindings/{display => soc}/mediatek/mediatek,mutex.yaml (77%)