search for: gk20a_volt_oclass

...veau_device *device) > device->oclass[NVDEV_ENGINE_GR ] = gk20a_graph_oclass; > device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; > device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass; > + device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; > break; > case 0xf0: > device->cname = "GK110"; > diff --git a/nvkm/include/subdev/volt.h b/nvkm/include/subdev/volt.h > index 820b62ffd75b..67db5e58880d 100644 > --- a/nvkm/include/subdev/volt.h > +++ b/nvkm/include/subdev/volt.h > @@ -52,6 +52,7 @...

...ve0_identify(struct nouveau_device *device) device->oclass[NVDEV_ENGINE_GR ] = gk20a_graph_oclass; device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass; + device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; break; case 0xf0: device->cname = "GK110"; diff --git a/nvkm/include/subdev/volt.h b/nvkm/include/subdev/volt.h index 820b62ffd75b..67db5e58880d 100644 --- a/nvkm/include/subdev/volt.h +++ b/nvkm/include/subdev/volt.h @@ -52,6 +52,7 @@ int _nouveau_volt_init(struct nouveau_ob...

...ve0_identify(struct nouveau_device *device) device->oclass[NVDEV_ENGINE_GR ] = gk20a_graph_oclass; device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass; + device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; break; case 0xf0: device->cname = "GK110"; diff --git a/nvkm/include/subdev/volt.h b/nvkm/include/subdev/volt.h index 820b62ffd75b..67db5e58880d 100644 --- a/nvkm/include/subdev/volt.h +++ b/nvkm/include/subdev/volt.h @@ -52,6 +52,7 @@ int _nouveau_volt_init(struct nouveau_ob...

Some Tegra drivers might be complied as kernel modules, and they need the fuse information for initialization. One example is the GK20A Nouveau driver. It needs the GPU speedo value to calculate frequency-voltage table. So export the tegra_sku_info. Signed-off-by: Vince Hsu <vinceh at nvidia.com> --- v2: add more description why we need this patch drivers/soc/tegra/fuse/fuse-tegra.c | 1

...device->oclass[NVDEV_ENGINE_GR ] = gk20a_graph_oclass; >> device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; >> device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass; >> + device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; >> break; >> case 0xf0: >> device->cname = "GK110"; >> diff --git a/nvkm/include/subdev/volt.h b/nvkm/include/subdev/volt.h >> index 820b62ffd75b..67db5e58880d 100644 >> --- a/nvkm/include/subdev/volt.h >> +++ b/nvkm/...

...oclass[NVDEV_ENGINE_GR ] = gk20a_graph_oclass; >>> device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; >>> device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass; >>> + device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; >>> break; >>> case 0xf0: >>> device->cname = "GK110"; >>> diff --git a/nvkm/include/subdev/volt.h b/nvkm/include/subdev/volt.h >>> index 820b62ffd75b..67db5e58880d 100644 >>> --- a/nvkm/include/subdev/vol...

Signed-off-by: Vince Hsu <vinceh at nvidia.com> --- nvkm/subdev/volt/base.c | 67 ++++++++++++++++++++++++++++--------------------- 1 file changed, 38 insertions(+), 29 deletions(-) diff --git a/nvkm/subdev/volt/base.c b/nvkm/subdev/volt/base.c index 32794a999106..26ccd8df193f 100644 --- a/nvkm/subdev/volt/base.c +++ b/nvkm/subdev/volt/base.c @@ -101,6 +101,41 @@ nouveau_volt_set_id(struct

Some Tegra drivers might be compiled as kernel modules, and they need the fuse information for initialization. One example is the GK20A Nouveau driver. It needs the GPU speedo value to calculate frequency-voltage table. So export the tegra_sku_info. Signed-off-by: Vince Hsu <vinceh at nvidia.com> Acked-by: Alexandre Courbot <acourbot at nvidia.com> Acked-by: Thierry Reding <treding

Some Tegra drivers might be compiled as kernel modules, and they need the fuse information for initialization. One example is the GK20A Nouveau driver. It needs the GPU speedo value to calculate frequency-voltage table. So export the tegra_sku_info. Signed-off-by: Vince Hsu <vinceh at nvidia.com> Acked-by: Alexandre Courbot <acourbot at nvidia.com> Acked-by: Thierry Reding <treding

..._ENGINE_GR ] = gk20a_graph_oclass; >>>> device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; >>>> device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass; >>>> + device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; >>>> break; >>>> case 0xf0: >>>> device->cname = "GK110"; >>>> diff --git a/nvkm/include/subdev/volt.h b/nvkm/include/subdev/volt.h >>>> index 820b62ffd75b..67db5e58880d 100644 >>>> --- a...

...+181,7 @@ nve0_identify(struct nouveau_device *device) > device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; > device->oclass[NVDEV_ENGINE_PERFMON] = &gk20a_perfmon_oclass; > device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; > + device->oclass[NVDEV_SUBDEV_DVFS ] = gk20a_dvfs_oclass; > break; > case 0xf0: > device->cname = "GK110"; > diff --git a/nvkm/include/core/device.h b/nvkm/include/core/device.h > index 2ec2e50d3676..162fa6...

...; device->oclass[NVDEV_ENGINE_COPY2 ] = >>> &nve0_copy2_oclass; >>> device->oclass[NVDEV_ENGINE_PERFMON] = >>> &gk20a_perfmon_oclass; >>> device->oclass[NVDEV_SUBDEV_VOLT ] = >>> &gk20a_volt_oclass; >>> + device->oclass[NVDEV_SUBDEV_DVFS ] = gk20a_dvfs_oclass; >>> break; >>> case 0xf0: >>> device->cname = "GK110"; >>> diff --git a/nvkm/include/core/device.h b/nvkm/include/co...

Hi, This is a try to have some simple DVFS (Dynamic Voltage and Frequency Scaling) support for GK20A. Instead of relying on other existing frequency scaling framework, we create a simple subdev in Nouveau for the same purpose. That's because we don't want to make the DVFS implementation for GK20A far more than enough in the beginning and hinder the implementation for dGPU in the future.

...ice.h> #include <engine/dmaobj.h> @@ -180,6 +181,7 @@ nve0_identify(struct nouveau_device *device) device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; device->oclass[NVDEV_ENGINE_PERFMON] = &gk20a_perfmon_oclass; device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; + device->oclass[NVDEV_SUBDEV_DVFS ] = gk20a_dvfs_oclass; break; case 0xf0: device->cname = "GK110"; diff --git a/nvkm/include/core/device.h b/nvkm/include/core/device.h index 2ec2e50d3676..162fa6a3cfcc 100644 --- a/nvkm/include/core/device.h +++ b/nvkm/include/core/devi...

...identify(struct nouveau_device *device) >> device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; >> device->oclass[NVDEV_ENGINE_PERFMON] = &gk20a_perfmon_oclass; >> device->oclass[NVDEV_SUBDEV_VOLT ] = &gk20a_volt_oclass; >> + device->oclass[NVDEV_SUBDEV_DVFS ] = gk20a_dvfs_oclass; >> break; >> case 0xf0: >> device->cname = "GK110"; >> diff --git a/nvkm/include/core/device.h b/nvkm/include/core/device.h >>...

...ice->oclass[NVDEV_ENGINE_COPY2 ] = >>>> &nve0_copy2_oclass; >>>> device->oclass[NVDEV_ENGINE_PERFMON] = >>>> &gk20a_perfmon_oclass; >>>> device->oclass[NVDEV_SUBDEV_VOLT ] = >>>> &gk20a_volt_oclass; >>>> + device->oclass[NVDEV_SUBDEV_DVFS ] = gk20a_dvfs_oclass; >>>> break; >>>> case 0xf0: >>>> device->cname = "GK110"; >>>> diff --git a/nvkm/include/core/dev...

There might be some callers of nouveau_clock_astate(), and they are from inetrrupt context. So we must ensure that this function can be atomic in that condition. This patch adds one parameter which is subsequently passed to nouveau_pstate_calc(). Therefore we can choose whether we want to wait for the pstate work's completion or not. Signed-off-by: Vince Hsu <vinceh at nvidia.com> ---