From: Mikko Perttunen mperttunen@nvidia.com
------------- Merging notes -------------
This series should be applied on top of the Host1x context isolation series.
------------- Changes in v2 -------------
Rebased on top of v5.19-rc3.
Updated Host1x device tree schema based on Rob's comments.
No changes to node names done, as that is a change that should be done uniformly in a separate series for all Tegra generations, if deemed appropriate.
------------- Original message -------------
Hi all,
This series adds support for Host1x and VIC on the recently released Tegra234 (Orin) SoC. It's split into the following parts:
* Device tree binding updates * Cleanup in host1x driver * Add programming of new registers and old registers that now need to be programmed to a non-reset value * Tegra234 device data and headers * Rewrite of the job opcode sequence, and related patches to support MLOCKs on Tegra186+.
The rewrite of the job opcode sequence brings Tegra186, Tegra194 and Tegra234 support to a 'full-featured' status that is necessary to support all host1x features in the future. This should not have any impact on older SoCs.
This series should be applied on top of the Host1x context isolation series.
Tested on Jetson AGX Xavier and Jetson AGX Orin.
Thanks, Mikko
Mikko Perttunen (13): dt-bindings: Add bindings for Tegra234 Host1x and VIC dt-bindings: Add headers for Host1x and VIC on Tegra234 arm64: tegra: Add Host1x and VIC on Tegra234 gpu: host1x: Deduplicate hardware headers gpu: host1x: Simplify register mapping and add common aperture gpu: host1x: Program virtualization tables gpu: host1x: Allow reset to be missing gpu: host1x: Program interrupt destinations on Tegra234 gpu: host1x: Tegra234 device data and headers gpu: host1x: Rewrite job opcode sequence gpu: host1x: Add MLOCK release code on Tegra234 gpu: host1x: Use RESTART_W to skip timed out jobs on Tegra186+ drm/tegra: vic: Add Tegra234 support
.../display/tegra/nvidia,tegra124-vic.yaml | 1 + .../display/tegra/nvidia,tegra20-host1x.yaml | 110 +++++++++-- arch/arm64/boot/dts/nvidia/tegra234.dtsi | 46 +++++ drivers/gpu/drm/tegra/drm.c | 1 + drivers/gpu/drm/tegra/vic.c | 12 ++ drivers/gpu/host1x/Makefile | 3 +- drivers/gpu/host1x/cdma.c | 19 +- drivers/gpu/host1x/dev.c | 112 +++++++---- drivers/gpu/host1x/dev.h | 11 ++ drivers/gpu/host1x/hw/cdma_hw.c | 34 ++++ drivers/gpu/host1x/hw/channel_hw.c | 144 ++++++++------ drivers/gpu/host1x/hw/host1x01_hardware.h | 114 +---------- drivers/gpu/host1x/hw/host1x02_hardware.h | 113 +---------- drivers/gpu/host1x/hw/host1x04_hardware.h | 113 +---------- drivers/gpu/host1x/hw/host1x05_hardware.h | 113 +---------- drivers/gpu/host1x/hw/host1x06_hardware.h | 128 +------------ drivers/gpu/host1x/hw/host1x07_hardware.h | 128 +------------ drivers/gpu/host1x/hw/host1x08.c | 33 ++++ drivers/gpu/host1x/hw/host1x08.h | 15 ++ drivers/gpu/host1x/hw/host1x08_hardware.h | 21 ++ drivers/gpu/host1x/hw/hw_host1x08_channel.h | 11 ++ drivers/gpu/host1x/hw/hw_host1x08_common.h | 11 ++ .../gpu/host1x/hw/hw_host1x08_hypervisor.h | 9 + drivers/gpu/host1x/hw/hw_host1x08_uclass.h | 181 ++++++++++++++++++ drivers/gpu/host1x/hw/hw_host1x08_vm.h | 36 ++++ drivers/gpu/host1x/hw/intr_hw.c | 11 ++ drivers/gpu/host1x/hw/opcodes.h | 150 +++++++++++++++ include/dt-bindings/clock/tegra234-clock.h | 4 + include/dt-bindings/memory/tegra234-mc.h | 5 + .../dt-bindings/power/tegra234-powergate.h | 1 + include/dt-bindings/reset/tegra234-reset.h | 1 + 31 files changed, 877 insertions(+), 814 deletions(-) create mode 100644 drivers/gpu/host1x/hw/host1x08.c create mode 100644 drivers/gpu/host1x/hw/host1x08.h create mode 100644 drivers/gpu/host1x/hw/host1x08_hardware.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_channel.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_common.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_uclass.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_vm.h create mode 100644 drivers/gpu/host1x/hw/opcodes.h
From: Mikko Perttunen mperttunen@nvidia.com
Update VIC and Host1x bindings for changes in Tegra234.
Namely, - New compatible strings - Sharded syncpoint interrupts - Optional reset.
Also, fix the order of descriptions for VM/hypervisor register apertures -- while the reg-names specification was correct, the descriptions for these were switched.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- v2: * Add back 'required' for resets/reset-names on older SoCs * Simplified reg descriptions * Updated commit message --- .../display/tegra/nvidia,tegra124-vic.yaml | 1 + .../display/tegra/nvidia,tegra20-host1x.yaml | 110 +++++++++++++++--- 2 files changed, 95 insertions(+), 16 deletions(-)
diff --git a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml index 37bb5ddc1963..7200095ef19e 100644 --- a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml +++ b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml @@ -21,6 +21,7 @@ properties: - nvidia,tegra210-vic - nvidia,tegra186-vic - nvidia,tegra194-vic + - nvidia,tegra234-vic
- items: - const: nvidia,tegra132-vic diff --git a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml index 0adeb03b9e3a..5fe25e0a8d48 100644 --- a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml +++ b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml @@ -24,6 +24,7 @@ properties: - nvidia,tegra210-host1x - nvidia,tegra186-host1x - nvidia,tegra194-host1x + - nvidia,tegra234-host1x
- items: - const: nvidia,tegra132-host1x @@ -31,23 +32,19 @@ properties:
reg: minItems: 1 - maxItems: 2 + maxItems: 3
reg-names: minItems: 1 - maxItems: 2 + maxItems: 3
interrupts: - items: - - description: host1x syncpoint interrupt - - description: host1x general interrupt minItems: 1 + maxItems: 9
interrupt-names: - items: - - const: syncpt - - const: host1x minItems: 1 + maxItems: 9
'#address-cells': description: The number of cells used to represent physical base addresses @@ -110,13 +107,35 @@ required: - reg - clocks - clock-names - - resets - - reset-names
additionalProperties: type: object
allOf: + - if: + properties: + compatible: + contains: + enum: + - nvidia,tegra20-host1x + - nvidia,tegra30-host1x + - nvidia,tegra114-host1x + - nvidia,tegra124-host1x + - nvidia,tegra210-host1x + then: + properties: + interrupts: + items: + - description: host1x syncpoint interrupt + - description: host1x general interrupt + + interrupt-names: + items: + - const: syncpt + - const: host1x + required: + - resets + - reset-names - if: properties: compatible: @@ -133,10 +152,8 @@ allOf:
reg: items: - - description: physical base address and length of the register - region assigned to the VM - - description: physical base address and length of the register - region used by the hypervisor + - description: region used by the hypervisor + - description: region assigned to the virtual machine
resets: maxItems: 1 @@ -144,6 +161,67 @@ allOf: reset-names: maxItems: 1
+ interrupts: + items: + - description: host1x syncpoint interrupt + - description: host1x general interrupt + + interrupt-names: + items: + - const: syncpt + - const: host1x + + iommu-map: + description: Specification of stream IDs available for memory context device + use. Should be a mapping of IDs 0..n to IOMMU entries corresponding to + usable stream IDs. + + required: + - reg-names + - if: + properties: + compatible: + contains: + enum: + - nvidia,tegra234-host1x + then: + properties: + reg-names: + items: + - const: common + - const: hypervisor + - const: vm + + reg: + items: + - description: region used by host1x server + - description: region used by the hypervisor + - description: region assigned to the virtual machine + + interrupts: + items: + - description: host1x syncpoint interrupt 0 + - description: host1x syncpoint interrupt 1 + - description: host1x syncpoint interrupt 2 + - description: host1x syncpoint interrupt 3 + - description: host1x syncpoint interrupt 4 + - description: host1x syncpoint interrupt 5 + - description: host1x syncpoint interrupt 6 + - description: host1x syncpoint interrupt 7 + - description: host1x general interrupt + + interrupt-names: + items: + - const: syncpt0 + - const: syncpt1 + - const: syncpt2 + - const: syncpt3 + - const: syncpt4 + - const: syncpt5 + - const: syncpt6 + - const: syncpt7 + - const: host1x + iommu-map: description: Specification of stream IDs available for memory context device use. Should be a mapping of IDs 0..n to IOMMU entries corresponding to @@ -160,8 +238,8 @@ examples: host1x@50000000 { compatible = "nvidia,tegra20-host1x"; reg = <0x50000000 0x00024000>; - interrupts = <0 65 0x04 /* mpcore syncpt */ - 0 67 0x04>; /* mpcore general */ + interrupts = <0 65 0x04>, /* mpcore syncpt */ + <0 67 0x04>; /* mpcore general */ interrupt-names = "syncpt", "host1x"; clocks = <&tegra_car TEGRA20_CLK_HOST1X>; clock-names = "host1x";
On 22/06/2022 13:37, Mikko Perttunen wrote:
From: Mikko Perttunen mperttunen@nvidia.com
Update VIC and Host1x bindings for changes in Tegra234.
Namely,
- New compatible strings
- Sharded syncpoint interrupts
- Optional reset.
Also, fix the order of descriptions for VM/hypervisor register apertures -- while the reg-names specification was correct, the descriptions for these were switched.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com
v2:
- Add back 'required' for resets/reset-names on older SoCs
- Simplified reg descriptions
- Updated commit message
.../display/tegra/nvidia,tegra124-vic.yaml | 1 + .../display/tegra/nvidia,tegra20-host1x.yaml | 110 +++++++++++++++--- 2 files changed, 95 insertions(+), 16 deletions(-)
diff --git a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml index 37bb5ddc1963..7200095ef19e 100644 --- a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml +++ b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra124-vic.yaml @@ -21,6 +21,7 @@ properties: - nvidia,tegra210-vic - nvidia,tegra186-vic - nvidia,tegra194-vic
- nvidia,tegra234-vic - items: - const: nvidia,tegra132-vicdiff --git a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml index 0adeb03b9e3a..5fe25e0a8d48 100644 --- a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml +++ b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.yaml @@ -24,6 +24,7 @@ properties: - nvidia,tegra210-host1x - nvidia,tegra186-host1x - nvidia,tegra194-host1x
- nvidia,tegra234-host1x - items: - const: nvidia,tegra132-host1x@@ -31,23 +32,19 @@ properties:
reg: minItems: 1
- maxItems: 2
maxItems: 3
reg-names: minItems: 1
- maxItems: 2
maxItems: 3
interrupts:
- items:
- description: host1x syncpoint interrupt- description: host1x general interrupt
maxItems: 9
interrupt-names:
- items:
- const: syncpt- const: host1x
maxItems: 9
'#address-cells': description: The number of cells used to represent physical base addresses
@@ -110,13 +107,35 @@ required:
- reg
- clocks
- clock-names
- resets
- reset-names
additionalProperties: type: object
allOf:
- if:
properties:compatible:contains:enum:- nvidia,tegra20-host1x- nvidia,tegra30-host1x- nvidia,tegra114-host1x- nvidia,tegra124-host1x- nvidia,tegra210-host1x- then:
properties:interrupts:items:- description: host1x syncpoint interrupt- description: host1x general interruptinterrupt-names:items:- const: syncpt- const: host1xrequired:- resets- reset-names
- if: properties: compatible:
@@ -133,10 +152,8 @@ allOf:
reg: items:
- description: physical base address and length of the registerregion assigned to the VM- description: physical base address and length of the registerregion used by the hypervisor
- description: region used by the hypervisor- description: region assigned to the virtual machine resets: maxItems: 1@@ -144,6 +161,67 @@ allOf: reset-names: maxItems: 1
interrupts:items:- description: host1x syncpoint interrupt- description: host1x general interruptinterrupt-names:items:- const: syncpt- const: host1xiommu-map:description: Specification of stream IDs available for memory context deviceuse. Should be a mapping of IDs 0..n to IOMMU entries corresponding tousable stream IDs.required:- reg-names
- if:
properties:compatible:contains:enum:- nvidia,tegra234-host1x- then:
properties:reg-names:items:- const: common- const: hypervisor- const: vmreg:items:- description: region used by host1x server- description: region used by the hypervisor- description: region assigned to the virtual machineinterrupts:items:- description: host1x syncpoint interrupt 0- description: host1x syncpoint interrupt 1- description: host1x syncpoint interrupt 2- description: host1x syncpoint interrupt 3- description: host1x syncpoint interrupt 4- description: host1x syncpoint interrupt 5- description: host1x syncpoint interrupt 6- description: host1x syncpoint interrupt 7- description: host1x general interruptinterrupt-names:items:- const: syncpt0- const: syncpt1- const: syncpt2- const: syncpt3- const: syncpt4- const: syncpt5- const: syncpt6- const: syncpt7- const: host1xiommu-map: description: Specification of stream IDs available for memory context device use. Should be a mapping of IDs 0..n to IOMMU entries corresponding to@@ -160,8 +238,8 @@ examples: host1x@50000000 { compatible = "nvidia,tegra20-host1x"; reg = <0x50000000 0x00024000>;
interrupts = <0 65 0x04 /* mpcore syncpt */0 67 0x04>; /* mpcore general */
interrupts = <0 65 0x04>, /* mpcore syncpt */<0 67 0x04>; /* mpcore general */
Split this part to separate patch.
Rest looks good but because of dependency cannot be tested by automation.
interrupt-names = "syncpt", "host1x"; clocks = <&tegra_car TEGRA20_CLK_HOST1X>; clock-names = "host1x";
Best regards, Krzysztof
From: Mikko Perttunen mperttunen@nvidia.com
Add clock, memory controller, powergate and reset dt-binding headers for Host1x and VIC on Tegra234.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- include/dt-bindings/clock/tegra234-clock.h | 4 ++++ include/dt-bindings/memory/tegra234-mc.h | 5 +++++ include/dt-bindings/power/tegra234-powergate.h | 1 + include/dt-bindings/reset/tegra234-reset.h | 1 + 4 files changed, 11 insertions(+)
diff --git a/include/dt-bindings/clock/tegra234-clock.h b/include/dt-bindings/clock/tegra234-clock.h index bd4c3086a2da..6e4e5cc75631 100644 --- a/include/dt-bindings/clock/tegra234-clock.h +++ b/include/dt-bindings/clock/tegra234-clock.h @@ -38,6 +38,8 @@ * throughput and memory controller power. */ #define TEGRA234_CLK_EMC 31U +/** @brief output of mux controlled by CLK_RST_CONTROLLER_CLK_SOURCE_HOST1X */ +#define TEGRA234_CLK_HOST1X 46U /** @brief output of gate CLK_ENB_FUSE */ #define TEGRA234_CLK_FUSE 40U /** @brief output of mux controlled by CLK_RST_CONTROLLER_CLK_SOURCE_I2C1 */ @@ -132,6 +134,8 @@ #define TEGRA234_CLK_UARTA 155U /** @brief output of gate CLK_ENB_PEX1_CORE_6 */ #define TEGRA234_CLK_PEX1_C6_CORE 161U +/** @brief output of mux controlled by CLK_RST_CONTROLLER_CLK_SOURCE_VIC */ +#define TEGRA234_CLK_VIC 167U /** @brief output of gate CLK_ENB_PEX2_CORE_7 */ #define TEGRA234_CLK_PEX2_C7_CORE 171U /** @brief output of gate CLK_ENB_PEX2_CORE_8 */ diff --git a/include/dt-bindings/memory/tegra234-mc.h b/include/dt-bindings/memory/tegra234-mc.h index e3b0e9da295d..73fdd18523a9 100644 --- a/include/dt-bindings/memory/tegra234-mc.h +++ b/include/dt-bindings/memory/tegra234-mc.h @@ -26,6 +26,8 @@ #define TEGRA234_SID_PCIE8 0x09 #define TEGRA234_SID_PCIE10 0x0b #define TEGRA234_SID_BPMP 0x10 +#define TEGRA234_SID_HOST1X 0x27 +#define TEGRA234_SID_VIC 0x34
/* * memory client IDs @@ -33,6 +35,7 @@
/* High-definition audio (HDA) read clients */ #define TEGRA234_MEMORY_CLIENT_HDAR 0x15 +#define TEGRA234_MEMORY_CLIENT_HOST1XDMAR 0x16 /* PCIE6 read clients */ #define TEGRA234_MEMORY_CLIENT_PCIE6AR 0x28 /* PCIE6 write clients */ @@ -65,6 +68,8 @@ #define TEGRA234_MEMORY_CLIENT_SDMMCRAB 0x63 /* sdmmcd memory write client */ #define TEGRA234_MEMORY_CLIENT_SDMMCWAB 0x67 +#define TEGRA234_MEMORY_CLIENT_VICSRD 0x6c +#define TEGRA234_MEMORY_CLIENT_VICSWR 0x6d /* BPMP read client */ #define TEGRA234_MEMORY_CLIENT_BPMPR 0x93 /* BPMP write client */ diff --git a/include/dt-bindings/power/tegra234-powergate.h b/include/dt-bindings/power/tegra234-powergate.h index f610eee9bce8..c3f7e380d2c6 100644 --- a/include/dt-bindings/power/tegra234-powergate.h +++ b/include/dt-bindings/power/tegra234-powergate.h @@ -18,5 +18,6 @@ #define TEGRA234_POWER_DOMAIN_MGBEA 17U #define TEGRA234_POWER_DOMAIN_MGBEB 18U #define TEGRA234_POWER_DOMAIN_MGBEC 19U +#define TEGRA234_POWER_DOMAIN_VIC 29U
#endif diff --git a/include/dt-bindings/reset/tegra234-reset.h b/include/dt-bindings/reset/tegra234-reset.h index 547ca3b60caa..1971400bf360 100644 --- a/include/dt-bindings/reset/tegra234-reset.h +++ b/include/dt-bindings/reset/tegra234-reset.h @@ -44,6 +44,7 @@ #define TEGRA234_RESET_QSPI1 77U #define TEGRA234_RESET_SDMMC4 85U #define TEGRA234_RESET_UARTA 100U +#define TEGRA234_RESET_VIC 113U #define TEGRA234_RESET_PEX0_CORE_0 116U #define TEGRA234_RESET_PEX0_CORE_1 117U #define TEGRA234_RESET_PEX0_CORE_2 118U
On 22/06/2022 13:37, Mikko Perttunen wrote:
From: Mikko Perttunen mperttunen@nvidia.com
Add clock, memory controller, powergate and reset dt-binding headers for Host1x and VIC on Tegra234.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com
include/dt-bindings/clock/tegra234-clock.h | 4 ++++ include/dt-bindings/memory/tegra234-mc.h | 5 +++++ include/dt-bindings/power/tegra234-powergate.h | 1 + include/dt-bindings/reset/tegra234-reset.h | 1 + 4 files changed, 11 insertions(+)
It's not the fault of this patch but in the past you started encoding some register offsets or values in the bindings. That's not what bindings headers are for. Store here hardware-independent IDs or nothing. For new Tegra SoCs I might start pushing this back...
Anyway, it's not this patch which introduced it, so: Acked-by: Krzysztof Kozlowski krzysztof.kozlowski@linaro.org
Best regards, Krzysztof
From: Mikko Perttunen mperttunen@nvidia.com
Add device tree nodes for Host1x and VIC on Tegra234.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- arch/arm64/boot/dts/nvidia/tegra234.dtsi | 46 ++++++++++++++++++++++++ 1 file changed, 46 insertions(+)
diff --git a/arch/arm64/boot/dts/nvidia/tegra234.dtsi b/arch/arm64/boot/dts/nvidia/tegra234.dtsi index cb3af539e477..cae68e59580c 100644 --- a/arch/arm64/boot/dts/nvidia/tegra234.dtsi +++ b/arch/arm64/boot/dts/nvidia/tegra234.dtsi @@ -454,6 +454,52 @@ misc@100000 { status = "okay"; };
+ host1x@13e00000 { + compatible = "nvidia,tegra234-host1x"; + reg = <0x13e00000 0x10000>, + <0x13e10000 0x10000>, + <0x13e40000 0x10000>; + reg-names = "common", "hypervisor", "vm"; + interrupts = <GIC_SPI 448 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 449 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 450 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 451 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 452 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 453 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 454 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 455 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 263 IRQ_TYPE_LEVEL_HIGH>; + interrupt-names = "syncpt0", "syncpt1", "syncpt2", "syncpt3", "syncpt4", + "syncpt5", "syncpt6", "syncpt7", "host1x"; + clocks = <&bpmp TEGRA234_CLK_HOST1X>; + clock-names = "host1x"; + + #address-cells = <1>; + #size-cells = <1>; + + ranges = <0x15000000 0x15000000 0x01000000>; + interconnects = <&mc TEGRA234_MEMORY_CLIENT_HOST1XDMAR &emc>; + interconnect-names = "dma-mem"; + iommus = <&smmu_niso1 TEGRA234_SID_HOST1X>; + + vic@15340000 { + compatible = "nvidia,tegra234-vic"; + reg = <0x15340000 0x00040000>; + interrupts = <GIC_SPI 206 IRQ_TYPE_LEVEL_HIGH>; + clocks = <&bpmp TEGRA234_CLK_VIC>; + clock-names = "vic"; + resets = <&bpmp TEGRA234_RESET_VIC>; + reset-names = "vic"; + + power-domains = <&bpmp TEGRA234_POWER_DOMAIN_VIC>; + interconnects = <&mc TEGRA234_MEMORY_CLIENT_VICSRD &emc>, + <&mc TEGRA234_MEMORY_CLIENT_VICSWR &emc>; + interconnect-names = "dma-mem", "write"; + iommus = <&smmu_niso1 TEGRA234_SID_VIC>; + dma-coherent; + }; + }; + gpio: gpio@2200000 { compatible = "nvidia,tegra234-gpio"; reg-names = "security", "gpio";
From: Mikko Perttunen mperttunen@nvidia.com
Host1x class information and opcodes are unchanged or backwards compatible across SoCs so let's not duplicate them for each one but have them in a shared header file.
At the same time, add opcode functions for acquire/release_mlock.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/hw/host1x01_hardware.h | 114 +--------------- drivers/gpu/host1x/hw/host1x02_hardware.h | 113 +--------------- drivers/gpu/host1x/hw/host1x04_hardware.h | 113 +--------------- drivers/gpu/host1x/hw/host1x05_hardware.h | 113 +--------------- drivers/gpu/host1x/hw/host1x06_hardware.h | 128 +----------------- drivers/gpu/host1x/hw/host1x07_hardware.h | 128 +----------------- drivers/gpu/host1x/hw/opcodes.h | 150 ++++++++++++++++++++++ 7 files changed, 156 insertions(+), 703 deletions(-) create mode 100644 drivers/gpu/host1x/hw/opcodes.h
diff --git a/drivers/gpu/host1x/hw/host1x01_hardware.h b/drivers/gpu/host1x/hw/host1x01_hardware.h index fe59df1d3dc3..cb93d7c1808c 100644 --- a/drivers/gpu/host1x/hw/host1x01_hardware.h +++ b/drivers/gpu/host1x/hw/host1x01_hardware.h @@ -15,118 +15,6 @@ #include "hw_host1x01_sync.h" #include "hw_host1x01_uclass.h"
-static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h"
#endif diff --git a/drivers/gpu/host1x/hw/host1x02_hardware.h b/drivers/gpu/host1x/hw/host1x02_hardware.h index af60d7fb016d..2d1282b9bc33 100644 --- a/drivers/gpu/host1x/hw/host1x02_hardware.h +++ b/drivers/gpu/host1x/hw/host1x02_hardware.h @@ -15,117 +15,6 @@ #include "hw_host1x02_sync.h" #include "hw_host1x02_uclass.h"
-static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h"
#endif diff --git a/drivers/gpu/host1x/hw/host1x04_hardware.h b/drivers/gpu/host1x/hw/host1x04_hardware.h index 4f9bcddf27e3..84d244e8af30 100644 --- a/drivers/gpu/host1x/hw/host1x04_hardware.h +++ b/drivers/gpu/host1x/hw/host1x04_hardware.h @@ -15,117 +15,6 @@ #include "hw_host1x04_sync.h" #include "hw_host1x04_uclass.h"
-static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h"
#endif diff --git a/drivers/gpu/host1x/hw/host1x05_hardware.h b/drivers/gpu/host1x/hw/host1x05_hardware.h index af3ab4b7f010..1dcde6ec7909 100644 --- a/drivers/gpu/host1x/hw/host1x05_hardware.h +++ b/drivers/gpu/host1x/hw/host1x05_hardware.h @@ -15,117 +15,6 @@ #include "hw_host1x05_sync.h" #include "hw_host1x05_uclass.h"
-static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h"
#endif diff --git a/drivers/gpu/host1x/hw/host1x06_hardware.h b/drivers/gpu/host1x/hw/host1x06_hardware.h index 5d515745eee7..c05cfa7e3090 100644 --- a/drivers/gpu/host1x/hw/host1x06_hardware.h +++ b/drivers/gpu/host1x/hw/host1x06_hardware.h @@ -16,132 +16,6 @@ #include "hw_host1x06_vm.h" #include "hw_host1x06_hypervisor.h"
-static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -static inline u32 host1x_opcode_setstreamid(unsigned streamid) -{ - return (7 << 28) | streamid; -} - -static inline u32 host1x_opcode_setpayload(unsigned payload) -{ - return (9 << 28) | payload; -} - -static inline u32 host1x_opcode_gather_wide(unsigned count) -{ - return (12 << 28) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h"
#endif diff --git a/drivers/gpu/host1x/hw/host1x07_hardware.h b/drivers/gpu/host1x/hw/host1x07_hardware.h index 82c0cc9bb0b5..d67364e03956 100644 --- a/drivers/gpu/host1x/hw/host1x07_hardware.h +++ b/drivers/gpu/host1x/hw/host1x07_hardware.h @@ -16,132 +16,6 @@ #include "hw_host1x07_vm.h" #include "hw_host1x07_hypervisor.h"
-static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -static inline u32 host1x_opcode_setstreamid(unsigned streamid) -{ - return (7 << 28) | streamid; -} - -static inline u32 host1x_opcode_setpayload(unsigned payload) -{ - return (9 << 28) | payload; -} - -static inline u32 host1x_opcode_gather_wide(unsigned count) -{ - return (12 << 28) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h"
#endif diff --git a/drivers/gpu/host1x/hw/opcodes.h b/drivers/gpu/host1x/hw/opcodes.h new file mode 100644 index 000000000000..649614499b04 --- /dev/null +++ b/drivers/gpu/host1x/hw/opcodes.h @@ -0,0 +1,150 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x opcodes + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef __HOST1X_OPCODES_H +#define __HOST1X_OPCODES_H + +#include <linux/types.h> + +static inline u32 host1x_class_host_wait_syncpt( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_wait_syncpt_indx_f(indx) + | host1x_uclass_wait_syncpt_thresh_f(threshold); +} + +static inline u32 host1x_class_host_load_syncpt_base( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_load_syncpt_base_base_indx_f(indx) + | host1x_uclass_load_syncpt_base_value_f(threshold); +} + +static inline u32 host1x_class_host_wait_syncpt_base( + unsigned indx, unsigned base_indx, unsigned offset) +{ + return host1x_uclass_wait_syncpt_base_indx_f(indx) + | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_wait_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt_base( + unsigned base_indx, unsigned offset) +{ + return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_incr_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt( + unsigned cond, unsigned indx) +{ + return host1x_uclass_incr_syncpt_cond_f(cond) + | host1x_uclass_incr_syncpt_indx_f(indx); +} + +static inline u32 host1x_class_host_indoff_reg_write( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indbe_f(0xf) + | host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +static inline u32 host1x_class_host_indoff_reg_read( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset) + | host1x_uclass_indoff_rwn_read_v(); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +static inline u32 host1x_opcode_setclass( + unsigned class_id, unsigned offset, unsigned mask) +{ + return (0 << 28) | (offset << 16) | (class_id << 6) | mask; +} + +static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) +{ + return (1 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) +{ + return (2 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) +{ + return (3 << 28) | (offset << 16) | mask; +} + +static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) +{ + return (4 << 28) | (offset << 16) | value; +} + +static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) +{ + return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), + host1x_class_host_incr_syncpt(cond, indx)); +} + +static inline u32 host1x_opcode_restart(unsigned address) +{ + return (5 << 28) | (address >> 4); +} + +static inline u32 host1x_opcode_gather(unsigned count) +{ + return (6 << 28) | count; +} + +static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | count; +} + +static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; +} + +static inline u32 host1x_opcode_setstreamid(unsigned streamid) +{ + return (7 << 28) | streamid; +} + +static inline u32 host1x_opcode_setpayload(unsigned payload) +{ + return (9 << 28) | payload; +} + +static inline u32 host1x_opcode_gather_wide(unsigned count) +{ + return (12 << 28) | count; +} + +static inline u32 host1x_opcode_acquire_mlock(unsigned mlock) +{ + return (14 << 28) | (0 << 24) | mlock; +} + +static inline u32 host1x_opcode_release_mlock(unsigned mlock) +{ + return (14 << 28) | (1 << 24) | mlock; +} + +#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) + +#endif
From: Mikko Perttunen mperttunen@nvidia.com
Refactor 'regs' property loading using devm_platform_ioremap_* and add loading of the 'common' region found on Tegra234.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/dev.c | 46 +++++++++++++++++----------------------- drivers/gpu/host1x/dev.h | 3 +++ 2 files changed, 22 insertions(+), 27 deletions(-)
diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index 89cc79a48eab..8c6ce8014c09 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -40,6 +40,11 @@ #include "hw/host1x06.h" #include "hw/host1x07.h"
+void host1x_common_writel(struct host1x *host1x, u32 v, u32 r) +{ + writel(v, host1x->common_regs + r); +} + void host1x_hypervisor_writel(struct host1x *host1x, u32 v, u32 r) { writel(v, host1x->hv_regs + r); @@ -412,7 +417,6 @@ static int host1x_get_resets(struct host1x *host) static int host1x_probe(struct platform_device *pdev) { struct host1x *host; - struct resource *regs, *hv_regs = NULL; int syncpt_irq; int err;
@@ -423,25 +427,23 @@ static int host1x_probe(struct platform_device *pdev) host->info = of_device_get_match_data(&pdev->dev);
if (host->info->has_hypervisor) { - regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "vm"); - if (!regs) { - dev_err(&pdev->dev, "failed to get vm registers\n"); - return -ENXIO; - } + host->regs = devm_platform_ioremap_resource_byname(pdev, "vm"); + if (IS_ERR(host->regs)) + return PTR_ERR(host->regs); + + host->hv_regs = devm_platform_ioremap_resource_byname(pdev, "hypervisor"); + if (IS_ERR(host->hv_regs)) + return PTR_ERR(host->hv_regs);
- hv_regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, - "hypervisor"); - if (!hv_regs) { - dev_err(&pdev->dev, - "failed to get hypervisor registers\n"); - return -ENXIO; + if (host->info->has_common) { + host->common_regs = devm_platform_ioremap_resource_byname(pdev, "common"); + if (IS_ERR(host->common_regs)) + return PTR_ERR(host->common_regs); } } else { - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!regs) { - dev_err(&pdev->dev, "failed to get registers\n"); - return -ENXIO; - } + host->regs = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(host->regs)) + return PTR_ERR(host->regs); }
syncpt_irq = platform_get_irq(pdev, 0); @@ -456,16 +458,6 @@ static int host1x_probe(struct platform_device *pdev) /* set common host1x device data */ platform_set_drvdata(pdev, host);
- host->regs = devm_ioremap_resource(&pdev->dev, regs); - if (IS_ERR(host->regs)) - return PTR_ERR(host->regs); - - if (host->info->has_hypervisor) { - host->hv_regs = devm_ioremap_resource(&pdev->dev, hv_regs); - if (IS_ERR(host->hv_regs)) - return PTR_ERR(host->hv_regs); - } - host->dev->dma_parms = &host->dma_parms; dma_set_max_seg_size(host->dev, UINT_MAX);
diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h index 7552a4554534..85edcc6e0fc7 100644 --- a/drivers/gpu/host1x/dev.h +++ b/drivers/gpu/host1x/dev.h @@ -100,6 +100,7 @@ struct host1x_info { u64 dma_mask; /* mask of addressable memory */ bool has_wide_gather; /* supports GATHER_W opcode */ bool has_hypervisor; /* has hypervisor registers */ + bool has_common; /* has common registers separate from hypervisor */ unsigned int num_sid_entries; const struct host1x_sid_entry *sid_table; /* @@ -115,6 +116,7 @@ struct host1x {
void __iomem *regs; void __iomem *hv_regs; /* hypervisor region */ + void __iomem *common_regs; struct host1x_syncpt *syncpt; struct host1x_syncpt_base *bases; struct device *dev; @@ -156,6 +158,7 @@ struct host1x { struct host1x_bo_cache cache; };
+void host1x_common_writel(struct host1x *host1x, u32 v, u32 r); void host1x_hypervisor_writel(struct host1x *host1x, u32 r, u32 v); u32 host1x_hypervisor_readl(struct host1x *host1x, u32 r); void host1x_sync_writel(struct host1x *host1x, u32 r, u32 v);
From: Mikko Perttunen mperttunen@nvidia.com
Program virtualization tables specifying which VMs have access to which Host1x hardware resources. Programming these has become mandatory in Tegra234.
For now, since the driver does not operate as a Host1x hypervisor, we basically allow access to everything to everyone.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/dev.c | 21 ++++++++++++++++++--- drivers/gpu/host1x/dev.h | 8 ++++++++ 2 files changed, 26 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index 8c6ce8014c09..5a57165ab0f6 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -217,7 +217,7 @@ static const struct of_device_id host1x_of_match[] = { }; MODULE_DEVICE_TABLE(of, host1x_of_match);
-static void host1x_setup_sid_table(struct host1x *host) +static void host1x_setup_virtualization_tables(struct host1x *host) { const struct host1x_info *info = host->info; unsigned int i; @@ -231,6 +231,21 @@ static void host1x_setup_sid_table(struct host1x *host) host1x_hypervisor_writel(host, entry->offset, entry->base); host1x_hypervisor_writel(host, entry->limit, entry->base + 4); } + + for (i = 0; i < info->streamid_vm_table.count; i++) { + /* Allow access to all stream IDs to all VMs. */ + host1x_hypervisor_writel(host, 0xff, info->streamid_vm_table.base + 4 * i); + } + + for (i = 0; i < info->classid_vm_table.count; i++) { + /* Allow access to all classes to all VMs. */ + host1x_hypervisor_writel(host, 0xff, info->classid_vm_table.base + 4 * i); + } + + for (i = 0; i < info->mmio_vm_table.count; i++) { + /* Use VM1 (that's us) as originator VMID for engine MMIO accesses. */ + host1x_hypervisor_writel(host, 0x1, info->mmio_vm_table.base + 4 * i); + } }
static bool host1x_wants_iommu(struct host1x *host1x) @@ -602,7 +617,7 @@ static int __maybe_unused host1x_runtime_suspend(struct device *dev) return 0;
resume_host1x: - host1x_setup_sid_table(host); + host1x_setup_virtualization_tables(host); host1x_syncpt_restore(host); host1x_intr_start(host);
@@ -632,7 +647,7 @@ static int __maybe_unused host1x_runtime_resume(struct device *dev) goto disable_clk; }
- host1x_setup_sid_table(host); + host1x_setup_virtualization_tables(host); host1x_syncpt_restore(host); host1x_intr_start(host);
diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h index 85edcc6e0fc7..920e5548cfbc 100644 --- a/drivers/gpu/host1x/dev.h +++ b/drivers/gpu/host1x/dev.h @@ -90,6 +90,11 @@ struct host1x_sid_entry { unsigned int limit; };
+struct host1x_table_desc { + unsigned int base; + unsigned int count; +}; + struct host1x_info { unsigned int nb_channels; /* host1x: number of channels supported */ unsigned int nb_pts; /* host1x: number of syncpoints supported */ @@ -103,6 +108,9 @@ struct host1x_info { bool has_common; /* has common registers separate from hypervisor */ unsigned int num_sid_entries; const struct host1x_sid_entry *sid_table; + struct host1x_table_desc streamid_vm_table; + struct host1x_table_desc classid_vm_table; + struct host1x_table_desc mmio_vm_table; /* * On T20-T148, the boot chain may setup DC to increment syncpoints * 26/27 on VBLANK. As such we cannot use these syncpoints until
From: Mikko Perttunen mperttunen@nvidia.com
Host1x on Tegra234 does not have a software-controllable reset line. As such, don't bail out if we don't find one in the device tree.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/dev.c | 3 --- 1 file changed, 3 deletions(-)
diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index 5a57165ab0f6..c28939850bf6 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -423,9 +423,6 @@ static int host1x_get_resets(struct host1x *host) return err; }
- if (WARN_ON(!host->resets[1].rstc)) - return -ENOENT; - return 0; }
From: Mikko Perttunen mperttunen@nvidia.com
On Tegra234, each Host1x VM has 8 interrupt lines. Each syncpoint can be configured with which interrupt line should be used for threshold interrupt, allowing for load balancing.
For now, to keep backwards compatibility, just set all syncpoints to the first interrupt.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/hw/intr_hw.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
diff --git a/drivers/gpu/host1x/hw/intr_hw.c b/drivers/gpu/host1x/hw/intr_hw.c index f56375ee6e71..9acccdb139e6 100644 --- a/drivers/gpu/host1x/hw/intr_hw.c +++ b/drivers/gpu/host1x/hw/intr_hw.c @@ -76,6 +76,17 @@ static void intr_hw_init(struct host1x *host, u32 cpm) /* update host clocks per usec */ host1x_sync_writel(host, cpm, HOST1X_SYNC_USEC_CLK); #endif +#if HOST1X_HW >= 8 + u32 id; + + /* + * Program threshold interrupt destination among 8 lines per VM, + * per syncpoint. For now, just direct all to the first interrupt + * line. + */ + for (id = 0; id < host->info->nb_pts; id++) + host1x_sync_writel(host, 0, HOST1X_SYNC_SYNCPT_INTR_DEST(id)); +#endif }
static int
From: Mikko Perttunen mperttunen@nvidia.com
Add device data and chip headers for Tegra234.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/Makefile | 3 +- drivers/gpu/host1x/dev.c | 42 ++++ drivers/gpu/host1x/hw/host1x08.c | 33 ++++ drivers/gpu/host1x/hw/host1x08.h | 15 ++ drivers/gpu/host1x/hw/host1x08_hardware.h | 21 ++ drivers/gpu/host1x/hw/hw_host1x08_channel.h | 11 ++ drivers/gpu/host1x/hw/hw_host1x08_common.h | 4 + .../gpu/host1x/hw/hw_host1x08_hypervisor.h | 9 + drivers/gpu/host1x/hw/hw_host1x08_uclass.h | 181 ++++++++++++++++++ drivers/gpu/host1x/hw/hw_host1x08_vm.h | 36 ++++ 10 files changed, 354 insertions(+), 1 deletion(-) create mode 100644 drivers/gpu/host1x/hw/host1x08.c create mode 100644 drivers/gpu/host1x/hw/host1x08.h create mode 100644 drivers/gpu/host1x/hw/host1x08_hardware.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_channel.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_common.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_uclass.h create mode 100644 drivers/gpu/host1x/hw/hw_host1x08_vm.h
diff --git a/drivers/gpu/host1x/Makefile b/drivers/gpu/host1x/Makefile index 8a65e13d113a..a35ec85f8d65 100644 --- a/drivers/gpu/host1x/Makefile +++ b/drivers/gpu/host1x/Makefile @@ -16,7 +16,8 @@ host1x-y = \ hw/host1x04.o \ hw/host1x05.o \ hw/host1x06.o \ - hw/host1x07.o + hw/host1x07.o \ + hw/host1x08.o
obj-$(CONFIG_TEGRA_HOST1X) += host1x.o obj-$(CONFIG_TEGRA_HOST1X_CONTEXT_BUS) += context_bus.o diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index c28939850bf6..0cd3f97e7e49 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -39,6 +39,7 @@ #include "hw/host1x05.h" #include "hw/host1x06.h" #include "hw/host1x07.h" +#include "hw/host1x08.h"
void host1x_common_writel(struct host1x *host1x, u32 v, u32 r) { @@ -205,7 +206,48 @@ static const struct host1x_info host1x07_info = { .reserve_vblank_syncpts = false, };
+/* + * Tegra234 has two stream ID protection tables, one for setting stream IDs + * through the channel path via SETSTREAMID, and one for setting them via + * MMIO. We program each engine's data stream ID in the channel path table + * and firmware stream ID in the MMIO path table. + */ +static const struct host1x_sid_entry tegra234_sid_table[] = { + { + /* VIC channel */ + .base = 0x17b8, + .offset = 0x30, + .limit = 0x30 + }, + { + /* VIC MMIO */ + .base = 0x1688, + .offset = 0x34, + .limit = 0x34 + }, +}; + +static const struct host1x_info host1x08_info = { + .nb_channels = 63, + .nb_pts = 1024, + .nb_mlocks = 24, + .nb_bases = 0, + .init = host1x08_init, + .sync_offset = 0x0, + .dma_mask = DMA_BIT_MASK(40), + .has_wide_gather = true, + .has_hypervisor = true, + .has_common = true, + .num_sid_entries = ARRAY_SIZE(tegra234_sid_table), + .sid_table = tegra234_sid_table, + .streamid_vm_table = { 0x1004, 128 }, + .classid_vm_table = { 0x1404, 25 }, + .mmio_vm_table = { 0x1504, 25 }, + .reserve_vblank_syncpts = false, +}; + static const struct of_device_id host1x_of_match[] = { + { .compatible = "nvidia,tegra234-host1x", .data = &host1x08_info, }, { .compatible = "nvidia,tegra194-host1x", .data = &host1x07_info, }, { .compatible = "nvidia,tegra186-host1x", .data = &host1x06_info, }, { .compatible = "nvidia,tegra210-host1x", .data = &host1x05_info, }, diff --git a/drivers/gpu/host1x/hw/host1x08.c b/drivers/gpu/host1x/hw/host1x08.c new file mode 100644 index 000000000000..754890c34c74 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra234 SoCs + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x08.h" +#include "host1x08_hardware.h" + +/* include code */ +#define HOST1X_HW 8 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x08_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x08.h b/drivers/gpu/host1x/hw/host1x08.h new file mode 100644 index 000000000000..a6bad56e44cf --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra234 SoCs + * + * Copyright (c) 2018 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X08_H +#define HOST1X_HOST1X08_H + +struct host1x; + +int host1x08_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x08_hardware.h b/drivers/gpu/host1x/hw/host1x08_hardware.h new file mode 100644 index 000000000000..936243060bff --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08_hardware.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra234 + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X08_HARDWARE_H +#define __HOST1X_HOST1X08_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x08_uclass.h" +#include "hw_host1x08_vm.h" +#include "hw_host1x08_hypervisor.h" +#include "hw_host1x08_common.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_channel.h b/drivers/gpu/host1x/hw/hw_host1x08_channel.h new file mode 100644 index 000000000000..c9272d2ab14a --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_channel.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef HOST1X_HW_HOST1X08_CHANNEL_H +#define HOST1X_HW_HOST1X08_CHANNEL_H + +#define HOST1X_CHANNEL_SMMU_STREAMID 0x084 + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_common.h b/drivers/gpu/host1x/hw/hw_host1x08_common.h new file mode 100644 index 000000000000..4df28440b86b --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_common.h @@ -0,0 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ diff --git a/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h b/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h new file mode 100644 index 000000000000..22964324c914 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h @@ -0,0 +1,9 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_HV_SYNCPT_PROT_EN 0x1724 +#define HOST1X_HV_SYNCPT_PROT_EN_CH_EN BIT(1) +#define HOST1X_HV_CH_MLOCK_EN(x) (0x1700 + (x * 4)) +#define HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(x) (0x1710 + (x * 4)) diff --git a/drivers/gpu/host1x/hw/hw_host1x08_uclass.h b/drivers/gpu/host1x/hw/hw_host1x08_uclass.h new file mode 100644 index 000000000000..724cccd71aa1 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X08_UCLASS_H +#define HOST1X_HW_HOST1X08_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 10; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_vm.h b/drivers/gpu/host1x/hw/hw_host1x08_vm.h new file mode 100644 index 000000000000..1455a4670bf8 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_vm.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_CHANNEL_DMASTART 0x0000 +#define HOST1X_CHANNEL_DMASTART_HI 0x0004 +#define HOST1X_CHANNEL_DMAPUT 0x0008 +#define HOST1X_CHANNEL_DMAPUT_HI 0x000c +#define HOST1X_CHANNEL_DMAGET 0x0010 +#define HOST1X_CHANNEL_DMAGET_HI 0x0014 +#define HOST1X_CHANNEL_DMAEND 0x0018 +#define HOST1X_CHANNEL_DMAEND_HI 0x001c +#define HOST1X_CHANNEL_DMACTRL 0x0020 +#define HOST1X_CHANNEL_DMACTRL_DMASTOP BIT(0) +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST BIT(1) +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET BIT(2) +#define HOST1X_CHANNEL_CMDFIFO_STAT 0x0024 +#define HOST1X_CHANNEL_CMDFIFO_STAT_EMPTY BIT(13) +#define HOST1X_CHANNEL_CMDFIFO_RDATA 0x0028 +#define HOST1X_CHANNEL_CMDP_OFFSET 0x0030 +#define HOST1X_CHANNEL_CMDP_CLASS 0x0034 +#define HOST1X_CHANNEL_CHANNELSTAT 0x0038 +#define HOST1X_CHANNEL_CMDPROC_STOP 0x0048 +#define HOST1X_CHANNEL_TEARDOWN 0x004c +#define HOST1X_CHANNEL_SMMU_STREAMID 0x0084 + +#define HOST1X_SYNC_SYNCPT_CPU_INCR(x) (0x6400 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(x) (0x6600 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INTR_DEST(x) (0x6684 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(x) (0x770c + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(x) (0x7790 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT(x) (0x8080 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INT_THRESH(x) (0xa088 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP(x) (0xb090 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP_CH(v) (((v) & 0x3f) << 8)
From: Mikko Perttunen mperttunen@nvidia.com
For new (Tegra186+) SoCs, use a new ('full-featured') job opcode sequence that is compatible with virtualization. In particular, the Host1x hardware in Tegra234 is more strict regarding the sequence, requiring ACQUIRE_MLOCK-SETCLASS-SETSTREAMID opcodes to occur in that sequence without gaps (except for SETPAYLOAD), so let's do it properly in one go now.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/hw/channel_hw.c | 144 +++++++++++++++++------------ 1 file changed, 85 insertions(+), 59 deletions(-)
diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c index f84caf06621a..4eb7fb2e4f0a 100644 --- a/drivers/gpu/host1x/hw/channel_hw.c +++ b/drivers/gpu/host1x/hw/channel_hw.c @@ -47,10 +47,41 @@ static void trace_write_gather(struct host1x_cdma *cdma, struct host1x_bo *bo, } }
-static void submit_wait(struct host1x_cdma *cdma, u32 id, u32 threshold, +static void submit_wait(struct host1x_job *job, u32 id, u32 threshold, u32 next_class) { -#if HOST1X_HW >= 2 + struct host1x_cdma *cdma = &job->channel->cdma; + +#if HOST1X_HW >= 6 + u32 stream_id; + + /* + * If a memory context has been set, use it. Otherwise + * (if context isolation is disabled) use the engine's + * firmware stream ID. + */ + if (job->memory_context) + stream_id = job->memory_context->stream_id; + else + stream_id = job->engine_fallback_streamid; + + host1x_cdma_push_wide(cdma, + host1x_opcode_setclass( + HOST1X_CLASS_HOST1X, + HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32, + /* WAIT_SYNCPT_32 is at SYNCPT_PAYLOAD_32+2 */ + BIT(0) | BIT(2) + ), + threshold, + id, + HOST1X_OPCODE_NOP + ); + host1x_cdma_push_wide(&job->channel->cdma, + host1x_opcode_setclass(job->class, 0, 0), + host1x_opcode_setpayload(stream_id), + host1x_opcode_setstreamid(job->engine_streamid_offset / 4), + HOST1X_OPCODE_NOP); +#elif HOST1X_HW >= 2 host1x_cdma_push_wide(cdma, host1x_opcode_setclass( HOST1X_CLASS_HOST1X, @@ -97,7 +128,7 @@ static void submit_gathers(struct host1x_job *job, u32 job_syncpt_base) else threshold = cmd->wait.threshold;
- submit_wait(cdma, cmd->wait.id, threshold, cmd->wait.next_class); + submit_wait(job, cmd->wait.id, threshold, cmd->wait.next_class); } else { struct host1x_job_gather *g = &cmd->gather;
@@ -180,42 +211,70 @@ static void host1x_enable_gather_filter(struct host1x_channel *ch) #endif }
-static void host1x_channel_program_engine_streamid(struct host1x_job *job) +static void channel_program_cdma(struct host1x_job *job) { + struct host1x_cdma *cdma = &job->channel->cdma; + struct host1x_syncpt *sp = job->syncpt; + #if HOST1X_HW >= 6 u32 fence;
- if (!job->memory_context) - return; + /* Enter engine class with invalid stream ID. */ + host1x_cdma_push_wide(cdma, + host1x_opcode_acquire_mlock(job->class), + host1x_opcode_setclass(job->class, 0, 0), + host1x_opcode_setpayload(0), + host1x_opcode_setstreamid(job->engine_streamid_offset / 4));
- fence = host1x_syncpt_incr_max(job->syncpt, 1); + /* Before switching stream ID to real stream ID, ensure engine is idle. */ + fence = host1x_syncpt_incr_max(sp, 1); + host1x_cdma_push(&job->channel->cdma, + host1x_opcode_nonincr(HOST1X_UCLASS_INCR_SYNCPT, 1), + HOST1X_UCLASS_INCR_SYNCPT_INDX_F(job->syncpt->id) | + HOST1X_UCLASS_INCR_SYNCPT_COND_F(4)); + submit_wait(job, job->syncpt->id, fence, job->class);
- /* First, increment a syncpoint on OP_DONE condition.. */ + /* Submit work. */ + job->syncpt_end = host1x_syncpt_incr_max(sp, job->syncpt_incrs); + submit_gathers(job, job->syncpt_end - job->syncpt_incrs);
+ /* Before releasing MLOCK, ensure engine is idle again. */ + fence = host1x_syncpt_incr_max(sp, 1); host1x_cdma_push(&job->channel->cdma, host1x_opcode_nonincr(HOST1X_UCLASS_INCR_SYNCPT, 1), HOST1X_UCLASS_INCR_SYNCPT_INDX_F(job->syncpt->id) | - HOST1X_UCLASS_INCR_SYNCPT_COND_F(1)); + HOST1X_UCLASS_INCR_SYNCPT_COND_F(4)); + submit_wait(job, job->syncpt->id, fence, job->class);
- /* Wait for syncpoint to increment */ + /* Release MLOCK. */ + host1x_cdma_push(cdma, + HOST1X_OPCODE_NOP, host1x_opcode_release_mlock(job->class)); +#else + if (job->serialize) { + /* + * Force serialization by inserting a host wait for the + * previous job to finish before this one can commence. + */ + host1x_cdma_push(cdma, + host1x_opcode_setclass(HOST1X_CLASS_HOST1X, + host1x_uclass_wait_syncpt_r(), 1), + host1x_class_host_wait_syncpt(job->syncpt->id, + host1x_syncpt_read_max(sp))); + }
- host1x_cdma_push(&job->channel->cdma, - host1x_opcode_setclass(HOST1X_CLASS_HOST1X, - host1x_uclass_wait_syncpt_r(), 1), - host1x_class_host_wait_syncpt(job->syncpt->id, fence)); + /* Synchronize base register to allow using it for relative waiting */ + if (sp->base) + synchronize_syncpt_base(job);
- /* - * Now that we know the engine is idle, return to class and - * change stream ID. - */ + /* add a setclass for modules that require it */ + if (job->class) + host1x_cdma_push(cdma, + host1x_opcode_setclass(job->class, 0, 0), + HOST1X_OPCODE_NOP);
- host1x_cdma_push(&job->channel->cdma, - host1x_opcode_setclass(job->class, 0, 0), - HOST1X_OPCODE_NOP); + job->syncpt_end = host1x_syncpt_incr_max(sp, job->syncpt_incrs);
- host1x_cdma_push(&job->channel->cdma, - host1x_opcode_setpayload(job->memory_context->stream_id), - host1x_opcode_setstreamid(job->engine_streamid_offset / 4)); + submit_gathers(job, job->syncpt_end - job->syncpt_incrs); #endif }
@@ -223,7 +282,6 @@ static int channel_submit(struct host1x_job *job) { struct host1x_channel *ch = job->channel; struct host1x_syncpt *sp = job->syncpt; - u32 user_syncpt_incrs = job->syncpt_incrs; u32 prev_max = 0; u32 syncval; int err; @@ -251,6 +309,7 @@ static int channel_submit(struct host1x_job *job)
host1x_channel_set_streamid(ch); host1x_enable_gather_filter(ch); + host1x_hw_syncpt_assign_to_channel(host, sp, ch);
/* begin a CDMA submit */ err = host1x_cdma_begin(&ch->cdma, job); @@ -259,40 +318,7 @@ static int channel_submit(struct host1x_job *job) goto error; }
- if (job->serialize) { - /* - * Force serialization by inserting a host wait for the - * previous job to finish before this one can commence. - */ - host1x_cdma_push(&ch->cdma, - host1x_opcode_setclass(HOST1X_CLASS_HOST1X, - host1x_uclass_wait_syncpt_r(), 1), - host1x_class_host_wait_syncpt(job->syncpt->id, - host1x_syncpt_read_max(sp))); - } - - /* Synchronize base register to allow using it for relative waiting */ - if (sp->base) - synchronize_syncpt_base(job); - - host1x_hw_syncpt_assign_to_channel(host, sp, ch); - - /* add a setclass for modules that require it */ - if (job->class) - host1x_cdma_push(&ch->cdma, - host1x_opcode_setclass(job->class, 0, 0), - HOST1X_OPCODE_NOP); - - /* - * Ensure engine DMA is idle and set new stream ID. May increment - * syncpt max. - */ - host1x_channel_program_engine_streamid(job); - - syncval = host1x_syncpt_incr_max(sp, user_syncpt_incrs); - job->syncpt_end = syncval; - - submit_gathers(job, syncval - user_syncpt_incrs); + channel_program_cdma(job);
/* end CDMA submit & stash pinned hMems into sync queue */ host1x_cdma_end(&ch->cdma, job);
From: Mikko Perttunen mperttunen@nvidia.com
With the full-featured opcode sequence using MLOCKs, we need to also unlock those MLOCKs in the event of a timeout. However, it turns out that on Tegra186/Tegra194, by default, we don't need to do this; furthermore, on Tegra234 it is much simpler to do; so only implement this on Tegra234 for the time being.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/hw/cdma_hw.c | 34 ++++++++++++++++++++++ drivers/gpu/host1x/hw/hw_host1x08_common.h | 7 +++++ 2 files changed, 41 insertions(+)
diff --git a/drivers/gpu/host1x/hw/cdma_hw.c b/drivers/gpu/host1x/hw/cdma_hw.c index e49cd5b8f735..1b65a10b9dfc 100644 --- a/drivers/gpu/host1x/hw/cdma_hw.c +++ b/drivers/gpu/host1x/hw/cdma_hw.c @@ -238,6 +238,37 @@ static void cdma_resume(struct host1x_cdma *cdma, u32 getptr) cdma_timeout_restart(cdma, getptr); }
+static void timeout_release_mlock(struct host1x_cdma *cdma) +{ +#if HOST1X_HW >= 8 + /* Tegra186 and Tegra194 require a more complicated MLOCK release + * sequence. Furthermore, those chips by default don't enforce MLOCKs, + * so it turns out that if we don't /actually/ need MLOCKs, we can just + * ignore them. + * + * As such, for now just implement this on Tegra234 where things are + * stricter but also easy to implement. + */ + struct host1x_channel *ch = cdma_to_channel(cdma); + struct host1x *host1x = cdma_to_host1x(cdma); + u32 offset; + + switch (ch->client->class) { + case HOST1X_CLASS_VIC: + offset = HOST1X_COMMON_VIC_MLOCK; + break; + case HOST1X_CLASS_NVDEC: + offset = HOST1X_COMMON_NVDEC_MLOCK; + break; + default: + WARN(1, "%s was not updated for class %u", __func__, ch->client->class); + return; + } + + host1x_common_writel(host1x, 0x0, offset); +#endif +} + /* * If this timeout fires, it indicates the current sync_queue entry has * exceeded its TTL and the userctx should be timed out and remaining @@ -288,6 +319,9 @@ static void cdma_timeout_handler(struct work_struct *work) /* stop HW, resetting channel/module */ host1x_hw_cdma_freeze(host1x, cdma);
+ /* release any held MLOCK */ + timeout_release_mlock(cdma); + host1x_cdma_update_sync_queue(cdma, ch->dev); mutex_unlock(&cdma->lock); } diff --git a/drivers/gpu/host1x/hw/hw_host1x08_common.h b/drivers/gpu/host1x/hw/hw_host1x08_common.h index 4df28440b86b..8e0c99150ec2 100644 --- a/drivers/gpu/host1x/hw/hw_host1x08_common.h +++ b/drivers/gpu/host1x/hw/hw_host1x08_common.h @@ -2,3 +2,10 @@ /* * Copyright (c) 2022 NVIDIA Corporation. */ + +#define HOST1X_COMMON_OFA_MLOCK 0x4050 +#define HOST1X_COMMON_NVJPG1_MLOCK 0x4070 +#define HOST1X_COMMON_VIC_MLOCK 0x4078 +#define HOST1X_COMMON_NVENC_MLOCK 0x407c +#define HOST1X_COMMON_NVDEC_MLOCK 0x4080 +#define HOST1X_COMMON_NVJPG_MLOCK 0x4084
From: Mikko Perttunen mperttunen@nvidia.com
When MLOCK enforcement is enabled, the 0-word write currently done is rejected by the hardware outside of an MLOCK region. As such, on these chips, which also have the newer, more convenient RESTART_W opcode, use that instead to skip over the timed out job.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/host1x/cdma.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/host1x/cdma.c b/drivers/gpu/host1x/cdma.c index 765e5aa64eb6..bb1f3c746be4 100644 --- a/drivers/gpu/host1x/cdma.c +++ b/drivers/gpu/host1x/cdma.c @@ -457,9 +457,24 @@ void host1x_cdma_update_sync_queue(struct host1x_cdma *cdma, * to offset 0xbad. This does nothing but * has a easily detected signature in debug * traces. + * + * On systems with MLOCK enforcement enabled, + * the above 0 word writes would fall foul of + * the enforcement. As such, in the first slot + * put a RESTART_W opcode to the beginning + * of the next job. We don't use this for older + * chips since those only support the RESTART + * opcode with inconvenient alignment requirements. */ - mapped[2*slot+0] = 0x1bad0000; - mapped[2*slot+1] = 0x1bad0000; + if (i == 0 && host1x->info->has_wide_gather) { + unsigned int next_job = (job->first_get/8 + job->num_slots) + % HOST1X_PUSHBUFFER_SLOTS; + mapped[2*slot+0] = (0xd << 28) | (next_job * 2); + mapped[2*slot+1] = 0x0; + } else { + mapped[2*slot+0] = 0x1bad0000; + mapped[2*slot+1] = 0x1bad0000; + } }
job->cancelled = true;
From: Mikko Perttunen mperttunen@nvidia.com
Add Tegra234 support for VIC. It is backwards compatible with Tegra194.
Signed-off-by: Mikko Perttunen mperttunen@nvidia.com --- drivers/gpu/drm/tegra/drm.c | 1 + drivers/gpu/drm/tegra/vic.c | 12 ++++++++++++ 2 files changed, 13 insertions(+)
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index 9464f522e257..426ffeb51f72 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -1380,6 +1380,7 @@ static const struct of_device_id host1x_drm_subdevs[] = { { .compatible = "nvidia,tegra194-sor", }, { .compatible = "nvidia,tegra194-vic", }, { .compatible = "nvidia,tegra194-nvdec", }, + { .compatible = "nvidia,tegra234-vic", }, { /* sentinel */ } };
diff --git a/drivers/gpu/drm/tegra/vic.c b/drivers/gpu/drm/tegra/vic.c index c5526bda88d6..c1734f823d04 100644 --- a/drivers/gpu/drm/tegra/vic.c +++ b/drivers/gpu/drm/tegra/vic.c @@ -447,11 +447,20 @@ static const struct vic_config vic_t194_config = { .supports_sid = true, };
+#define NVIDIA_TEGRA_234_VIC_FIRMWARE "nvidia/tegra234/vic.bin" + +static const struct vic_config vic_t234_config = { + .firmware = NVIDIA_TEGRA_234_VIC_FIRMWARE, + .version = 0x23, + .supports_sid = true, +}; + static const struct of_device_id tegra_vic_of_match[] = { { .compatible = "nvidia,tegra124-vic", .data = &vic_t124_config }, { .compatible = "nvidia,tegra210-vic", .data = &vic_t210_config }, { .compatible = "nvidia,tegra186-vic", .data = &vic_t186_config }, { .compatible = "nvidia,tegra194-vic", .data = &vic_t194_config }, + { .compatible = "nvidia,tegra234-vic", .data = &vic_t234_config }, { }, }; MODULE_DEVICE_TABLE(of, tegra_vic_of_match); @@ -590,3 +599,6 @@ MODULE_FIRMWARE(NVIDIA_TEGRA_186_VIC_FIRMWARE); #if IS_ENABLED(CONFIG_ARCH_TEGRA_194_SOC) MODULE_FIRMWARE(NVIDIA_TEGRA_194_VIC_FIRMWARE); #endif +#if IS_ENABLED(CONFIG_ARCH_TEGRA_234_SOC) +MODULE_FIRMWARE(NVIDIA_TEGRA_234_VIC_FIRMWARE); +#endif
On 22/06/2022 13:37, Mikko Perttunen wrote:
From: Mikko Perttunen mperttunen@nvidia.com
Merging notes
This series should be applied on top of the Host1x context isolation series.
The dependency should be first merged into next, so automated tools can process it.
Best regards, Krzysztof
dri-devel@lists.freedesktop.org
-
Krzysztof Kozlowski -
Mikko Perttunen