[5/5] drm/msm/A6xx: Add support for using system cache(llc)
diff mbox

Message ID 1521789591-28628-6-git-send-email-smasetty@codeaurora.org
State Not Applicable, archived
Headers show

Commit Message

Sharat Masetty March 23, 2018, 7:19 a.m. UTC
The last level system cache can be partitioned to 32
different slices of which GPU has two slices preallocated.
The "gpu" slice is used for caching GPU buffers and
the "gpuhtw" slice is used for caching the GPU SMMU
pagetables.  This patch talks to the core system cache
driver to acquire the slice handles, configure the SCID's
to those slices and activates and deactivates the slices
upon GPU power collapse and restore.

Some support from the IOMMU driver is also needed to
make use of the system cache. IOMMU_UPSTREAM_HINT is
a buffer protection flag which enables caching GPU data
buffers in the system cache with memory attributes such
as outer cacheable, read-allocate, write-allocate for buffers.
The GPU then has the ability to override a few cacheability
parameters which it does to override write-allocate to
write-no-allocate as the GPU hardware does not benefit much
from it.
Similarly DOMAIN_ATTR_USE_UPSTREAM_HINT is another domain level
attribute used by the IOMMU driver to set the right attributes
to cache the hardware pagetables into the system cache.

Signed-off-by: Sharat Masetty <smasetty@codeaurora.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 162 +++++++++++++++++++++++++++++++++-
 drivers/gpu/drm/msm/adreno/a6xx_gpu.h |   9 ++
 drivers/gpu/drm/msm/msm_iommu.c       |  13 +++
 drivers/gpu/drm/msm/msm_mmu.h         |   3 +
 4 files changed, 186 insertions(+), 1 deletion(-)

Comments

Jordan Crouse April 3, 2018, 9:24 p.m. UTC | #1
On Fri, Mar 23, 2018 at 12:49:51PM +0530, Sharat Masetty wrote:
> The last level system cache can be partitioned to 32
> different slices of which GPU has two slices preallocated.
> The "gpu" slice is used for caching GPU buffers and
> the "gpuhtw" slice is used for caching the GPU SMMU
> pagetables.  This patch talks to the core system cache
> driver to acquire the slice handles, configure the SCID's
> to those slices and activates and deactivates the slices
> upon GPU power collapse and restore.
> 
> Some support from the IOMMU driver is also needed to
> make use of the system cache. IOMMU_UPSTREAM_HINT is
> a buffer protection flag which enables caching GPU data
> buffers in the system cache with memory attributes such
> as outer cacheable, read-allocate, write-allocate for buffers.
> The GPU then has the ability to override a few cacheability
> parameters which it does to override write-allocate to
> write-no-allocate as the GPU hardware does not benefit much
> from it.
> Similarly DOMAIN_ATTR_USE_UPSTREAM_HINT is another domain level
> attribute used by the IOMMU driver to set the right attributes
> to cache the hardware pagetables into the system cache.

This has a dependency on the LLCC driver and the API to that may change (it is
under review now).  When it does, this will have to naturally change as well but
that'll be a minor tweek and won't affect the functionality of this driver so
pending those changes..

Reviewed-by: Jordan Crouse <jcrouse@codeaurora.org>

> Signed-off-by: Sharat Masetty <smasetty@codeaurora.org>
> ---
>  drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 162 +++++++++++++++++++++++++++++++++-
>  drivers/gpu/drm/msm/adreno/a6xx_gpu.h |   9 ++
>  drivers/gpu/drm/msm/msm_iommu.c       |  13 +++
>  drivers/gpu/drm/msm/msm_mmu.h         |   3 +
>  4 files changed, 186 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> index bd50674..e4554eb 100644
> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> @@ -13,6 +13,7 @@
>  
>  #include <linux/qcom_scm.h>
>  #include <linux/soc/qcom/mdt_loader.h>
> +#include <linux/soc/qcom/llcc-qcom.h>
>  
>  #include "msm_gem.h"
>  #include "msm_mmu.h"
> @@ -913,6 +914,154 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
>  	~0
>  };
>  
> +#define A6XX_LLC_NUM_GPU_SCIDS		5
> +#define A6XX_GPU_LLC_SCID_NUM_BITS	5
> +
> +#define A6XX_GPU_LLC_SCID_MASK \
> +	((1 << (A6XX_LLC_NUM_GPU_SCIDS * A6XX_GPU_LLC_SCID_NUM_BITS)) - 1)
> +
> +#define A6XX_GPUHTW_LLC_SCID_SHIFT	25
> +#define A6XX_GPUHTW_LLC_SCID_MASK \
> +	(((1 << A6XX_GPU_LLC_SCID_NUM_BITS) - 1) << A6XX_GPUHTW_LLC_SCID_SHIFT)
> +
> +static inline void a6xx_gpu_cx_rmw(struct a6xx_llc *llc,
> +	u32 reg, u32 mask, u32 or)
> +{
> +	msm_rmw(llc->mmio + (reg << 2), mask, or);
> +}
> +
> +static void a6xx_llc_deactivate(struct msm_gpu *gpu)
> +{
> +	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
> +	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
> +	struct a6xx_llc *llc = &a6xx_gpu->llc;
> +
> +	llcc_slice_deactivate(llc->gpu_llc_slice);
> +	llcc_slice_deactivate(llc->gpuhtw_llc_slice);
> +}
> +
> +static void a6xx_llc_activate(struct msm_gpu *gpu)
> +{
> +	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
> +	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
> +	struct a6xx_llc *llc = &a6xx_gpu->llc;
> +
> +	if (!llc->mmio)
> +		return;
> +
> +	if (llc->gpu_llc_slice)
> +		if (!llcc_slice_activate(llc->gpu_llc_slice))
> +			/* Program the sub-cache ID for all GPU blocks */
> +			a6xx_gpu_cx_rmw(llc,
> +				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
> +				A6XX_GPU_LLC_SCID_MASK,
> +				(llc->cntl1_regval &
> +					A6XX_GPU_LLC_SCID_MASK));
> +
> +	if (llc->gpuhtw_llc_slice)
> +		if (!llcc_slice_activate(llc->gpuhtw_llc_slice))
> +			/* Program the sub-cache ID for GPU pagetables */
> +			a6xx_gpu_cx_rmw(llc,
> +				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
> +				A6XX_GPUHTW_LLC_SCID_MASK,
> +				(llc->cntl1_regval &
> +					A6XX_GPUHTW_LLC_SCID_MASK));
> +
> +	/* Program cacheability overrides */
> +	a6xx_gpu_cx_rmw(llc, REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_0, 0xF,
> +		llc->cntl0_regval);
> +}
> +
> +void a6xx_llc_slices_destroy(struct a6xx_llc *llc)
> +{
> +	if (llc->mmio) {
> +		iounmap(llc->mmio);
> +		llc->mmio = NULL;
> +	}
> +
> +	llcc_slice_putd(llc->gpu_llc_slice);
> +	llc->gpu_llc_slice = NULL;
> +
> +	llcc_slice_putd(llc->gpuhtw_llc_slice);
> +	llc->gpuhtw_llc_slice = NULL;
> +}
> +
> +static int a6xx_llc_slices_init(struct platform_device *pdev,
> +		struct a6xx_llc *llc)
> +{
> +	int i;
> +
> +	/* Get the system cache slice descriptor for GPU and GPUHTWs */
> +	llc->gpu_llc_slice = llcc_slice_getd(&pdev->dev, "gpu");
> +	if (IS_ERR(llc->gpu_llc_slice))
> +		llc->gpu_llc_slice = NULL;
> +
> +	llc->gpuhtw_llc_slice = llcc_slice_getd(&pdev->dev, "gpuhtw");
> +	if (IS_ERR(llc->gpuhtw_llc_slice))
> +		llc->gpuhtw_llc_slice = NULL;
> +
> +	if (llc->gpu_llc_slice == NULL && llc->gpuhtw_llc_slice == NULL)
> +		return -1;
> +
> +	/* Map registers */
> +	llc->mmio = msm_ioremap(pdev, "cx_mem", "gpu_cx");
> +	if (IS_ERR(llc->mmio)) {
> +		llc->mmio = NULL;
> +		a6xx_llc_slices_destroy(llc);
> +		return -1;
> +	}
> +
> +	/*
> +	 * Setup GPU system cache CNTL0 and CNTL1 register values.
> +	 * These values will be programmed everytime GPU comes out
> +	 * of power collapse as these are non-retention registers.
> +	 */
> +
> +	/*
> +	 * CNTL0 provides options to override the settings for the
> +	 * read and write allocation policies for the LLC. These
> +	 * overrides are global for all memory transactions from
> +	 * the GPU.
> +	 *
> +	 * 0x3: read-no-alloc-overridden = 0
> +	 *      read-no-alloc = 0 - Allocate lines on read miss
> +	 *      write-no-alloc-overridden = 1
> +	 *      write-no-alloc = 1 - Do not allocates lines on write miss
> +	 */
> +	llc->cntl0_regval = 0x03;
> +
> +	/*
> +	 * CNTL1 is used to specify SCID for (CP, TP, VFD, CCU and UBWC
> +	 * FLAG cache) GPU blocks. This value will be passed along with
> +	 * the address for any memory transaction from GPU to identify
> +	 * the sub-cache for that transaction.
> +	 *
> +	 * Currently there is only one SCID allocated for all GPU blocks
> +	 * Hence set same SCID for all the blocks.
> +	 */
> +
> +	if (llc->gpu_llc_slice) {
> +		u32 gpu_scid = llcc_get_slice_id(llc->gpu_llc_slice);
> +
> +		for (i = 0; i < A6XX_LLC_NUM_GPU_SCIDS; i++)
> +			llc->cntl1_regval |=
> +				gpu_scid << (A6XX_GPU_LLC_SCID_NUM_BITS * i);
> +	}
> +
> +	/*
> +	 * Set SCID for GPU IOMMU. This will be used to access
> +	 * page tables that are cached in LLC.
> +	 */
> +	if (llc->gpuhtw_llc_slice) {
> +		u32 gpuhtw_scid = llcc_get_slice_id(llc->gpuhtw_llc_slice);
> +
> +		llc->cntl1_regval |=
> +			gpuhtw_scid << A6XX_GPUHTW_LLC_SCID_SHIFT;
> +	}
> +
> +	return 0;
> +}
> +
>  static int a6xx_pm_resume(struct msm_gpu *gpu)
>  {
>  	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
> @@ -923,6 +1072,9 @@ static int a6xx_pm_resume(struct msm_gpu *gpu)
>  
>  	gpu->needs_hw_init = true;
>  
> +	/* Activate LLC slices */
> +	a6xx_llc_activate(gpu);
> +
>  	return ret;
>  }
>  
> @@ -931,6 +1083,9 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
>  	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
>  	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
>  
> +	/* Deactivate LLC slices */
> +	a6xx_llc_deactivate(gpu);
> +
>  	/*
>  	 * Make sure the GMU is idle before continuing (because some transitions
>  	 * may use VBIF
> @@ -993,6 +1148,8 @@ static void a6xx_destroy(struct msm_gpu *gpu)
>  		drm_gem_object_unreference_unlocked(a6xx_gpu->sqe_bo);
>  	}
>  
> +	a6xx_llc_slices_destroy(&a6xx_gpu->llc);
> +
>  	a6xx_gmu_remove(a6xx_gpu);
>  
>  	adreno_gpu_cleanup(adreno_gpu);
> @@ -1040,7 +1197,10 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
>  	adreno_gpu->registers = a6xx_registers;
>  	adreno_gpu->reg_offsets = a6xx_register_offsets;
>  
> -	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4, 0);
> +	ret = a6xx_llc_slices_init(pdev, &a6xx_gpu->llc);
> +
> +	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4,
> +			ret ? 0 : MMU_FEATURE_USE_SYSTEM_CACHE);
>  	if (ret) {
>  		a6xx_destroy(&(a6xx_gpu->base.base));
>  		return ERR_PTR(ret);
> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
> index 21ab701..392c426 100644
> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
> @@ -21,6 +21,14 @@
>  
>  extern bool hang_debug;
>  
> +struct a6xx_llc {
> +	void __iomem *mmio;
> +	void *gpu_llc_slice;
> +	void *gpuhtw_llc_slice;
> +	u32 cntl0_regval;
> +	u32 cntl1_regval;
> +};
> +
>  struct a6xx_gpu {
>  	struct adreno_gpu base;
>  
> @@ -46,6 +54,7 @@ struct a6xx_gpu {
>  	uint64_t scratch_iova;
>  
>  	struct a6xx_gmu gmu;
> +	struct a6xx_llc llc;
>  };
>  
>  #define to_a6xx_gpu(x) container_of(x, struct a6xx_gpu, base)
> diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
> index 1ab629b..6c03eda 100644
> --- a/drivers/gpu/drm/msm/msm_iommu.c
> +++ b/drivers/gpu/drm/msm/msm_iommu.c
> @@ -39,6 +39,16 @@ static int msm_iommu_attach(struct msm_mmu *mmu, const char * const *names,
>  {
>  	struct msm_iommu *iommu = to_msm_iommu(mmu);
>  	int ret;
> +	int gpu_htw_llc = 1;
> +
> +	/*
> +	 * This allows GPU to set the bus attributes required
> +	 * to use system cache on behalf of the iommu page table
> +	 * walker.
> +	 */
> +	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
> +		iommu_domain_set_attr(iommu->domain,
> +				DOMAIN_ATTR_USE_UPSTREAM_HINT, &gpu_htw_llc);
>  
>  	pm_runtime_get_suppliers(mmu->dev);
>  	ret = iommu_attach_device(iommu->domain, mmu->dev);
> @@ -63,6 +73,9 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova,
>  	struct msm_iommu *iommu = to_msm_iommu(mmu);
>  	size_t ret;
>  
> +	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
> +		prot |= IOMMU_USE_UPSTREAM_HINT;
> +
>  	pm_runtime_get_suppliers(mmu->dev);
>  	ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot);
>  	pm_runtime_put_suppliers(mmu->dev);
> diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h
> index 85df78d..257bdea 100644
> --- a/drivers/gpu/drm/msm/msm_mmu.h
> +++ b/drivers/gpu/drm/msm/msm_mmu.h
> @@ -30,6 +30,9 @@ struct msm_mmu_funcs {
>  	void (*destroy)(struct msm_mmu *mmu);
>  };
>  
> +/* MMU features */
> +#define MMU_FEATURE_USE_SYSTEM_CACHE (1 << 0)
> +
>  struct msm_mmu {
>  	const struct msm_mmu_funcs *funcs;
>  	struct device *dev;
> -- 
> 1.9.1
> 
> _______________________________________________
> Freedreno mailing list
> Freedreno@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/freedreno
Sharat Masetty April 5, 2018, 5:25 a.m. UTC | #2
On 4/4/2018 2:54 AM, Jordan Crouse wrote:
> On Fri, Mar 23, 2018 at 12:49:51PM +0530, Sharat Masetty wrote:
>> The last level system cache can be partitioned to 32
>> different slices of which GPU has two slices preallocated.
>> The "gpu" slice is used for caching GPU buffers and
>> the "gpuhtw" slice is used for caching the GPU SMMU
>> pagetables.  This patch talks to the core system cache
>> driver to acquire the slice handles, configure the SCID's
>> to those slices and activates and deactivates the slices
>> upon GPU power collapse and restore.
>>
>> Some support from the IOMMU driver is also needed to
>> make use of the system cache. IOMMU_UPSTREAM_HINT is
>> a buffer protection flag which enables caching GPU data
>> buffers in the system cache with memory attributes such
>> as outer cacheable, read-allocate, write-allocate for buffers.
>> The GPU then has the ability to override a few cacheability
>> parameters which it does to override write-allocate to
>> write-no-allocate as the GPU hardware does not benefit much
>> from it.
>> Similarly DOMAIN_ATTR_USE_UPSTREAM_HINT is another domain level
>> attribute used by the IOMMU driver to set the right attributes
>> to cache the hardware pagetables into the system cache.
> 
> This has a dependency on the LLCC driver and the API to that may change (it is
> under review now).  When it does, this will have to naturally change as well but
> that'll be a minor tweek and won't affect the functionality of this driver so
> pending those changes..

Thanks for the review Jordan. Vivek will also submit the SMMU changes 
for the UPSTREAM_HINT support to the mailing list soon. So once the 
dependencies are sorted out, I will review and submit a fresh patch set 
if needed.
> 
> Reviewed-by: Jordan Crouse <jcrouse@codeaurora.org> >
>> Signed-off-by: Sharat Masetty <smasetty@codeaurora.org>
>> ---
>>   drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 162 +++++++++++++++++++++++++++++++++-
>>   drivers/gpu/drm/msm/adreno/a6xx_gpu.h |   9 ++
>>   drivers/gpu/drm/msm/msm_iommu.c       |  13 +++
>>   drivers/gpu/drm/msm/msm_mmu.h         |   3 +
>>   4 files changed, 186 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>> index bd50674..e4554eb 100644
>> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>> @@ -13,6 +13,7 @@
>>   
>>   #include <linux/qcom_scm.h>
>>   #include <linux/soc/qcom/mdt_loader.h>
>> +#include <linux/soc/qcom/llcc-qcom.h>
>>   
>>   #include "msm_gem.h"
>>   #include "msm_mmu.h"
>> @@ -913,6 +914,154 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
>>   	~0
>>   };
>>   
>> +#define A6XX_LLC_NUM_GPU_SCIDS		5
>> +#define A6XX_GPU_LLC_SCID_NUM_BITS	5
>> +
>> +#define A6XX_GPU_LLC_SCID_MASK \
>> +	((1 << (A6XX_LLC_NUM_GPU_SCIDS * A6XX_GPU_LLC_SCID_NUM_BITS)) - 1)
>> +
>> +#define A6XX_GPUHTW_LLC_SCID_SHIFT	25
>> +#define A6XX_GPUHTW_LLC_SCID_MASK \
>> +	(((1 << A6XX_GPU_LLC_SCID_NUM_BITS) - 1) << A6XX_GPUHTW_LLC_SCID_SHIFT)
>> +
>> +static inline void a6xx_gpu_cx_rmw(struct a6xx_llc *llc,
>> +	u32 reg, u32 mask, u32 or)
>> +{
>> +	msm_rmw(llc->mmio + (reg << 2), mask, or);
>> +}
>> +
>> +static void a6xx_llc_deactivate(struct msm_gpu *gpu)
>> +{
>> +	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
>> +	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
>> +	struct a6xx_llc *llc = &a6xx_gpu->llc;
>> +
>> +	llcc_slice_deactivate(llc->gpu_llc_slice);
>> +	llcc_slice_deactivate(llc->gpuhtw_llc_slice);
>> +}
>> +
>> +static void a6xx_llc_activate(struct msm_gpu *gpu)
>> +{
>> +	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
>> +	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
>> +	struct a6xx_llc *llc = &a6xx_gpu->llc;
>> +
>> +	if (!llc->mmio)
>> +		return;
>> +
>> +	if (llc->gpu_llc_slice)
>> +		if (!llcc_slice_activate(llc->gpu_llc_slice))
>> +			/* Program the sub-cache ID for all GPU blocks */
>> +			a6xx_gpu_cx_rmw(llc,
>> +				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
>> +				A6XX_GPU_LLC_SCID_MASK,
>> +				(llc->cntl1_regval &
>> +					A6XX_GPU_LLC_SCID_MASK));
>> +
>> +	if (llc->gpuhtw_llc_slice)
>> +		if (!llcc_slice_activate(llc->gpuhtw_llc_slice))
>> +			/* Program the sub-cache ID for GPU pagetables */
>> +			a6xx_gpu_cx_rmw(llc,
>> +				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
>> +				A6XX_GPUHTW_LLC_SCID_MASK,
>> +				(llc->cntl1_regval &
>> +					A6XX_GPUHTW_LLC_SCID_MASK));
>> +
>> +	/* Program cacheability overrides */
>> +	a6xx_gpu_cx_rmw(llc, REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_0, 0xF,
>> +		llc->cntl0_regval);
>> +}
>> +
>> +void a6xx_llc_slices_destroy(struct a6xx_llc *llc)
>> +{
>> +	if (llc->mmio) {
>> +		iounmap(llc->mmio);
>> +		llc->mmio = NULL;
>> +	}
>> +
>> +	llcc_slice_putd(llc->gpu_llc_slice);
>> +	llc->gpu_llc_slice = NULL;
>> +
>> +	llcc_slice_putd(llc->gpuhtw_llc_slice);
>> +	llc->gpuhtw_llc_slice = NULL;
>> +}
>> +
>> +static int a6xx_llc_slices_init(struct platform_device *pdev,
>> +		struct a6xx_llc *llc)
>> +{
>> +	int i;
>> +
>> +	/* Get the system cache slice descriptor for GPU and GPUHTWs */
>> +	llc->gpu_llc_slice = llcc_slice_getd(&pdev->dev, "gpu");
>> +	if (IS_ERR(llc->gpu_llc_slice))
>> +		llc->gpu_llc_slice = NULL;
>> +
>> +	llc->gpuhtw_llc_slice = llcc_slice_getd(&pdev->dev, "gpuhtw");
>> +	if (IS_ERR(llc->gpuhtw_llc_slice))
>> +		llc->gpuhtw_llc_slice = NULL;
>> +
>> +	if (llc->gpu_llc_slice == NULL && llc->gpuhtw_llc_slice == NULL)
>> +		return -1;
>> +
>> +	/* Map registers */
>> +	llc->mmio = msm_ioremap(pdev, "cx_mem", "gpu_cx");
>> +	if (IS_ERR(llc->mmio)) {
>> +		llc->mmio = NULL;
>> +		a6xx_llc_slices_destroy(llc);
>> +		return -1;
>> +	}
>> +
>> +	/*
>> +	 * Setup GPU system cache CNTL0 and CNTL1 register values.
>> +	 * These values will be programmed everytime GPU comes out
>> +	 * of power collapse as these are non-retention registers.
>> +	 */
>> +
>> +	/*
>> +	 * CNTL0 provides options to override the settings for the
>> +	 * read and write allocation policies for the LLC. These
>> +	 * overrides are global for all memory transactions from
>> +	 * the GPU.
>> +	 *
>> +	 * 0x3: read-no-alloc-overridden = 0
>> +	 *      read-no-alloc = 0 - Allocate lines on read miss
>> +	 *      write-no-alloc-overridden = 1
>> +	 *      write-no-alloc = 1 - Do not allocates lines on write miss
>> +	 */
>> +	llc->cntl0_regval = 0x03;
>> +
>> +	/*
>> +	 * CNTL1 is used to specify SCID for (CP, TP, VFD, CCU and UBWC
>> +	 * FLAG cache) GPU blocks. This value will be passed along with
>> +	 * the address for any memory transaction from GPU to identify
>> +	 * the sub-cache for that transaction.
>> +	 *
>> +	 * Currently there is only one SCID allocated for all GPU blocks
>> +	 * Hence set same SCID for all the blocks.
>> +	 */
>> +
>> +	if (llc->gpu_llc_slice) {
>> +		u32 gpu_scid = llcc_get_slice_id(llc->gpu_llc_slice);
>> +
>> +		for (i = 0; i < A6XX_LLC_NUM_GPU_SCIDS; i++)
>> +			llc->cntl1_regval |=
>> +				gpu_scid << (A6XX_GPU_LLC_SCID_NUM_BITS * i);
>> +	}
>> +
>> +	/*
>> +	 * Set SCID for GPU IOMMU. This will be used to access
>> +	 * page tables that are cached in LLC.
>> +	 */
>> +	if (llc->gpuhtw_llc_slice) {
>> +		u32 gpuhtw_scid = llcc_get_slice_id(llc->gpuhtw_llc_slice);
>> +
>> +		llc->cntl1_regval |=
>> +			gpuhtw_scid << A6XX_GPUHTW_LLC_SCID_SHIFT;
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>>   static int a6xx_pm_resume(struct msm_gpu *gpu)
>>   {
>>   	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
>> @@ -923,6 +1072,9 @@ static int a6xx_pm_resume(struct msm_gpu *gpu)
>>   
>>   	gpu->needs_hw_init = true;
>>   
>> +	/* Activate LLC slices */
>> +	a6xx_llc_activate(gpu);
>> +
>>   	return ret;
>>   }
>>   
>> @@ -931,6 +1083,9 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
>>   	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
>>   	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
>>   
>> +	/* Deactivate LLC slices */
>> +	a6xx_llc_deactivate(gpu);
>> +
>>   	/*
>>   	 * Make sure the GMU is idle before continuing (because some transitions
>>   	 * may use VBIF
>> @@ -993,6 +1148,8 @@ static void a6xx_destroy(struct msm_gpu *gpu)
>>   		drm_gem_object_unreference_unlocked(a6xx_gpu->sqe_bo);
>>   	}
>>   
>> +	a6xx_llc_slices_destroy(&a6xx_gpu->llc);
>> +
>>   	a6xx_gmu_remove(a6xx_gpu);
>>   
>>   	adreno_gpu_cleanup(adreno_gpu);
>> @@ -1040,7 +1197,10 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
>>   	adreno_gpu->registers = a6xx_registers;
>>   	adreno_gpu->reg_offsets = a6xx_register_offsets;
>>   
>> -	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4, 0);
>> +	ret = a6xx_llc_slices_init(pdev, &a6xx_gpu->llc);
>> +
>> +	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4,
>> +			ret ? 0 : MMU_FEATURE_USE_SYSTEM_CACHE);
>>   	if (ret) {
>>   		a6xx_destroy(&(a6xx_gpu->base.base));
>>   		return ERR_PTR(ret);
>> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
>> index 21ab701..392c426 100644
>> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
>> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
>> @@ -21,6 +21,14 @@
>>   
>>   extern bool hang_debug;
>>   
>> +struct a6xx_llc {
>> +	void __iomem *mmio;
>> +	void *gpu_llc_slice;
>> +	void *gpuhtw_llc_slice;
>> +	u32 cntl0_regval;
>> +	u32 cntl1_regval;
>> +};
>> +
>>   struct a6xx_gpu {
>>   	struct adreno_gpu base;
>>   
>> @@ -46,6 +54,7 @@ struct a6xx_gpu {
>>   	uint64_t scratch_iova;
>>   
>>   	struct a6xx_gmu gmu;
>> +	struct a6xx_llc llc;
>>   };
>>   
>>   #define to_a6xx_gpu(x) container_of(x, struct a6xx_gpu, base)
>> diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
>> index 1ab629b..6c03eda 100644
>> --- a/drivers/gpu/drm/msm/msm_iommu.c
>> +++ b/drivers/gpu/drm/msm/msm_iommu.c
>> @@ -39,6 +39,16 @@ static int msm_iommu_attach(struct msm_mmu *mmu, const char * const *names,
>>   {
>>   	struct msm_iommu *iommu = to_msm_iommu(mmu);
>>   	int ret;
>> +	int gpu_htw_llc = 1;
>> +
>> +	/*
>> +	 * This allows GPU to set the bus attributes required
>> +	 * to use system cache on behalf of the iommu page table
>> +	 * walker.
>> +	 */
>> +	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
>> +		iommu_domain_set_attr(iommu->domain,
>> +				DOMAIN_ATTR_USE_UPSTREAM_HINT, &gpu_htw_llc);
>>   
>>   	pm_runtime_get_suppliers(mmu->dev);
>>   	ret = iommu_attach_device(iommu->domain, mmu->dev);
>> @@ -63,6 +73,9 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova,
>>   	struct msm_iommu *iommu = to_msm_iommu(mmu);
>>   	size_t ret;
>>   
>> +	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
>> +		prot |= IOMMU_USE_UPSTREAM_HINT;
>> +
>>   	pm_runtime_get_suppliers(mmu->dev);
>>   	ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot);
>>   	pm_runtime_put_suppliers(mmu->dev);
>> diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h
>> index 85df78d..257bdea 100644
>> --- a/drivers/gpu/drm/msm/msm_mmu.h
>> +++ b/drivers/gpu/drm/msm/msm_mmu.h
>> @@ -30,6 +30,9 @@ struct msm_mmu_funcs {
>>   	void (*destroy)(struct msm_mmu *mmu);
>>   };
>>   
>> +/* MMU features */
>> +#define MMU_FEATURE_USE_SYSTEM_CACHE (1 << 0)
>> +
>>   struct msm_mmu {
>>   	const struct msm_mmu_funcs *funcs;
>>   	struct device *dev;
>> -- 
>> 1.9.1
>>
>> _______________________________________________
>> Freedreno mailing list
>> Freedreno@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/freedreno
>
Vivek Gautam April 5, 2018, 7:04 a.m. UTC | #3
Hi Sharat,


On 3/23/2018 12:49 PM, Sharat Masetty wrote:
> The last level system cache can be partitioned to 32
> different slices of which GPU has two slices preallocated.
> The "gpu" slice is used for caching GPU buffers and
> the "gpuhtw" slice is used for caching the GPU SMMU
> pagetables.  This patch talks to the core system cache
> driver to acquire the slice handles, configure the SCID's
> to those slices and activates and deactivates the slices
> upon GPU power collapse and restore.
>
> Some support from the IOMMU driver is also needed to
> make use of the system cache. IOMMU_UPSTREAM_HINT is
> a buffer protection flag which enables caching GPU data
> buffers in the system cache with memory attributes such
> as outer cacheable, read-allocate, write-allocate for buffers.
> The GPU then has the ability to override a few cacheability
> parameters which it does to override write-allocate to
> write-no-allocate as the GPU hardware does not benefit much
> from it.
> Similarly DOMAIN_ATTR_USE_UPSTREAM_HINT is another domain level
> attribute used by the IOMMU driver to set the right attributes
> to cache the hardware pagetables into the system cache.
>
> Signed-off-by: Sharat Masetty <smasetty@codeaurora.org>
> ---

Couple of minor nits. Please see comments inline below.

>   drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 162 +++++++++++++++++++++++++++++++++-
>   drivers/gpu/drm/msm/adreno/a6xx_gpu.h |   9 ++
>   drivers/gpu/drm/msm/msm_iommu.c       |  13 +++
>   drivers/gpu/drm/msm/msm_mmu.h         |   3 +
>   4 files changed, 186 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> index bd50674..e4554eb 100644
> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> @@ -13,6 +13,7 @@
>   
>   #include <linux/qcom_scm.h>
>   #include <linux/soc/qcom/mdt_loader.h>
> +#include <linux/soc/qcom/llcc-qcom.h>
>   
>   #include "msm_gem.h"
>   #include "msm_mmu.h"
> @@ -913,6 +914,154 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
>   	~0
>   };
>   
> +#define A6XX_LLC_NUM_GPU_SCIDS		5
> +#define A6XX_GPU_LLC_SCID_NUM_BITS	5
> +
> +#define A6XX_GPU_LLC_SCID_MASK \
> +	((1 << (A6XX_LLC_NUM_GPU_SCIDS * A6XX_GPU_LLC_SCID_NUM_BITS)) - 1)
> +
> +#define A6XX_GPUHTW_LLC_SCID_SHIFT	25
> +#define A6XX_GPUHTW_LLC_SCID_MASK \
> +	(((1 << A6XX_GPU_LLC_SCID_NUM_BITS) - 1) << A6XX_GPUHTW_LLC_SCID_SHIFT)
> +
> +static inline void a6xx_gpu_cx_rmw(struct a6xx_llc *llc,
> +	u32 reg, u32 mask, u32 or)
> +{
> +	msm_rmw(llc->mmio + (reg << 2), mask, or);
> +}
> +
> +static void a6xx_llc_deactivate(struct msm_gpu *gpu)
> +{
> +	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
> +	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
> +	struct a6xx_llc *llc = &a6xx_gpu->llc;
> +
> +	llcc_slice_deactivate(llc->gpu_llc_slice);
> +	llcc_slice_deactivate(llc->gpuhtw_llc_slice);
> +}
> +
> +static void a6xx_llc_activate(struct msm_gpu *gpu)
> +{
> +	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
> +	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
> +	struct a6xx_llc *llc = &a6xx_gpu->llc;
> +
> +	if (!llc->mmio)
> +		return;
> +
> +	if (llc->gpu_llc_slice)
> +		if (!llcc_slice_activate(llc->gpu_llc_slice))
> +			/* Program the sub-cache ID for all GPU blocks */
> +			a6xx_gpu_cx_rmw(llc,
> +				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
> +				A6XX_GPU_LLC_SCID_MASK,
> +				(llc->cntl1_regval &
> +					A6XX_GPU_LLC_SCID_MASK));
> +
> +	if (llc->gpuhtw_llc_slice)
> +		if (!llcc_slice_activate(llc->gpuhtw_llc_slice))
> +			/* Program the sub-cache ID for GPU pagetables */
> +			a6xx_gpu_cx_rmw(llc,
> +				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
> +				A6XX_GPUHTW_LLC_SCID_MASK,
> +				(llc->cntl1_regval &
> +					A6XX_GPUHTW_LLC_SCID_MASK));
> +
> +	/* Program cacheability overrides */
> +	a6xx_gpu_cx_rmw(llc, REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_0, 0xF,
> +		llc->cntl0_regval);
> +}
> +
> +void a6xx_llc_slices_destroy(struct a6xx_llc *llc)

static?

> +{
> +	if (llc->mmio) {
> +		iounmap(llc->mmio);
> +		llc->mmio = NULL;
> +	}
> +
> +	llcc_slice_putd(llc->gpu_llc_slice);
> +	llc->gpu_llc_slice = NULL;
> +
> +	llcc_slice_putd(llc->gpuhtw_llc_slice);
> +	llc->gpuhtw_llc_slice = NULL;
> +}
> +
> +static int a6xx_llc_slices_init(struct platform_device *pdev,
> +		struct a6xx_llc *llc)
> +{
> +	int i;
> +
> +	/* Get the system cache slice descriptor for GPU and GPUHTWs */
> +	llc->gpu_llc_slice = llcc_slice_getd(&pdev->dev, "gpu");
> +	if (IS_ERR(llc->gpu_llc_slice))
> +		llc->gpu_llc_slice = NULL;
> +
> +	llc->gpuhtw_llc_slice = llcc_slice_getd(&pdev->dev, "gpuhtw");
> +	if (IS_ERR(llc->gpuhtw_llc_slice))
> +		llc->gpuhtw_llc_slice = NULL;
> +
> +	if (llc->gpu_llc_slice == NULL && llc->gpuhtw_llc_slice == NULL)
> +		return -1;
> +
> +	/* Map registers */
> +	llc->mmio = msm_ioremap(pdev, "cx_mem", "gpu_cx");
> +	if (IS_ERR(llc->mmio)) {
> +		llc->mmio = NULL;
> +		a6xx_llc_slices_destroy(llc);
> +		return -1;
> +	}

If you move this ioremap to start of the function, then you wouldn't need to
call a6xx_llcc_slices_destroy().

regards
Vivek

> +
> +	/*
> +	 * Setup GPU system cache CNTL0 and CNTL1 register values.
> +	 * These values will be programmed everytime GPU comes out
> +	 * of power collapse as these are non-retention registers.
> +	 */
> +
> +	/*
> +	 * CNTL0 provides options to override the settings for the
> +	 * read and write allocation policies for the LLC. These
> +	 * overrides are global for all memory transactions from
> +	 * the GPU.
> +	 *
> +	 * 0x3: read-no-alloc-overridden = 0
> +	 *      read-no-alloc = 0 - Allocate lines on read miss
> +	 *      write-no-alloc-overridden = 1
> +	 *      write-no-alloc = 1 - Do not allocates lines on write miss
> +	 */
> +	llc->cntl0_regval = 0x03;
> +
> +	/*
> +	 * CNTL1 is used to specify SCID for (CP, TP, VFD, CCU and UBWC
> +	 * FLAG cache) GPU blocks. This value will be passed along with
> +	 * the address for any memory transaction from GPU to identify
> +	 * the sub-cache for that transaction.
> +	 *
> +	 * Currently there is only one SCID allocated for all GPU blocks
> +	 * Hence set same SCID for all the blocks.
> +	 */
> +
> +	if (llc->gpu_llc_slice) {
> +		u32 gpu_scid = llcc_get_slice_id(llc->gpu_llc_slice);
> +
> +		for (i = 0; i < A6XX_LLC_NUM_GPU_SCIDS; i++)
> +			llc->cntl1_regval |=
> +				gpu_scid << (A6XX_GPU_LLC_SCID_NUM_BITS * i);
> +	}
> +
> +	/*
> +	 * Set SCID for GPU IOMMU. This will be used to access
> +	 * page tables that are cached in LLC.
> +	 */
> +	if (llc->gpuhtw_llc_slice) {
> +		u32 gpuhtw_scid = llcc_get_slice_id(llc->gpuhtw_llc_slice);
> +
> +		llc->cntl1_regval |=
> +			gpuhtw_scid << A6XX_GPUHTW_LLC_SCID_SHIFT;
> +	}
> +
> +	return 0;
> +}
> +
>   static int a6xx_pm_resume(struct msm_gpu *gpu)
>   {
>   	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
> @@ -923,6 +1072,9 @@ static int a6xx_pm_resume(struct msm_gpu *gpu)
>   
>   	gpu->needs_hw_init = true;
>   
> +	/* Activate LLC slices */
> +	a6xx_llc_activate(gpu);
> +
>   	return ret;
>   }
>   
> @@ -931,6 +1083,9 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
>   	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
>   	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
>   
> +	/* Deactivate LLC slices */
> +	a6xx_llc_deactivate(gpu);
> +
>   	/*
>   	 * Make sure the GMU is idle before continuing (because some transitions
>   	 * may use VBIF
> @@ -993,6 +1148,8 @@ static void a6xx_destroy(struct msm_gpu *gpu)
>   		drm_gem_object_unreference_unlocked(a6xx_gpu->sqe_bo);
>   	}
>   
> +	a6xx_llc_slices_destroy(&a6xx_gpu->llc);
> +
>   	a6xx_gmu_remove(a6xx_gpu);
>   
>   	adreno_gpu_cleanup(adreno_gpu);
> @@ -1040,7 +1197,10 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
>   	adreno_gpu->registers = a6xx_registers;
>   	adreno_gpu->reg_offsets = a6xx_register_offsets;
>   
> -	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4, 0);
> +	ret = a6xx_llc_slices_init(pdev, &a6xx_gpu->llc);
> +
> +	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4,
> +			ret ? 0 : MMU_FEATURE_USE_SYSTEM_CACHE);
>   	if (ret) {
>   		a6xx_destroy(&(a6xx_gpu->base.base));
>   		return ERR_PTR(ret);
> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
> index 21ab701..392c426 100644
> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
> @@ -21,6 +21,14 @@
>   
>   extern bool hang_debug;
>   
> +struct a6xx_llc {
> +	void __iomem *mmio;
> +	void *gpu_llc_slice;
> +	void *gpuhtw_llc_slice;
> +	u32 cntl0_regval;
> +	u32 cntl1_regval;
> +};
> +
>   struct a6xx_gpu {
>   	struct adreno_gpu base;
>   
> @@ -46,6 +54,7 @@ struct a6xx_gpu {
>   	uint64_t scratch_iova;
>   
>   	struct a6xx_gmu gmu;
> +	struct a6xx_llc llc;
>   };
>   
>   #define to_a6xx_gpu(x) container_of(x, struct a6xx_gpu, base)
> diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
> index 1ab629b..6c03eda 100644
> --- a/drivers/gpu/drm/msm/msm_iommu.c
> +++ b/drivers/gpu/drm/msm/msm_iommu.c
> @@ -39,6 +39,16 @@ static int msm_iommu_attach(struct msm_mmu *mmu, const char * const *names,
>   {
>   	struct msm_iommu *iommu = to_msm_iommu(mmu);
>   	int ret;
> +	int gpu_htw_llc = 1;
> +
> +	/*
> +	 * This allows GPU to set the bus attributes required
> +	 * to use system cache on behalf of the iommu page table
> +	 * walker.
> +	 */
> +	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
> +		iommu_domain_set_attr(iommu->domain,
> +				DOMAIN_ATTR_USE_UPSTREAM_HINT, &gpu_htw_llc);
>   
>   	pm_runtime_get_suppliers(mmu->dev);
>   	ret = iommu_attach_device(iommu->domain, mmu->dev);
> @@ -63,6 +73,9 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova,
>   	struct msm_iommu *iommu = to_msm_iommu(mmu);
>   	size_t ret;
>   
> +	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
> +		prot |= IOMMU_USE_UPSTREAM_HINT;
> +
>   	pm_runtime_get_suppliers(mmu->dev);
>   	ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot);
>   	pm_runtime_put_suppliers(mmu->dev);
> diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h
> index 85df78d..257bdea 100644
> --- a/drivers/gpu/drm/msm/msm_mmu.h
> +++ b/drivers/gpu/drm/msm/msm_mmu.h
> @@ -30,6 +30,9 @@ struct msm_mmu_funcs {
>   	void (*destroy)(struct msm_mmu *mmu);
>   };
>   
> +/* MMU features */
> +#define MMU_FEATURE_USE_SYSTEM_CACHE (1 << 0)
> +
>   struct msm_mmu {
>   	const struct msm_mmu_funcs *funcs;
>   	struct device *dev;

--
To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch
diff mbox

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index bd50674..e4554eb 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -13,6 +13,7 @@ 
 
 #include <linux/qcom_scm.h>
 #include <linux/soc/qcom/mdt_loader.h>
+#include <linux/soc/qcom/llcc-qcom.h>
 
 #include "msm_gem.h"
 #include "msm_mmu.h"
@@ -913,6 +914,154 @@  static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
 	~0
 };
 
+#define A6XX_LLC_NUM_GPU_SCIDS		5
+#define A6XX_GPU_LLC_SCID_NUM_BITS	5
+
+#define A6XX_GPU_LLC_SCID_MASK \
+	((1 << (A6XX_LLC_NUM_GPU_SCIDS * A6XX_GPU_LLC_SCID_NUM_BITS)) - 1)
+
+#define A6XX_GPUHTW_LLC_SCID_SHIFT	25
+#define A6XX_GPUHTW_LLC_SCID_MASK \
+	(((1 << A6XX_GPU_LLC_SCID_NUM_BITS) - 1) << A6XX_GPUHTW_LLC_SCID_SHIFT)
+
+static inline void a6xx_gpu_cx_rmw(struct a6xx_llc *llc,
+	u32 reg, u32 mask, u32 or)
+{
+	msm_rmw(llc->mmio + (reg << 2), mask, or);
+}
+
+static void a6xx_llc_deactivate(struct msm_gpu *gpu)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+	struct a6xx_llc *llc = &a6xx_gpu->llc;
+
+	llcc_slice_deactivate(llc->gpu_llc_slice);
+	llcc_slice_deactivate(llc->gpuhtw_llc_slice);
+}
+
+static void a6xx_llc_activate(struct msm_gpu *gpu)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+	struct a6xx_llc *llc = &a6xx_gpu->llc;
+
+	if (!llc->mmio)
+		return;
+
+	if (llc->gpu_llc_slice)
+		if (!llcc_slice_activate(llc->gpu_llc_slice))
+			/* Program the sub-cache ID for all GPU blocks */
+			a6xx_gpu_cx_rmw(llc,
+				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
+				A6XX_GPU_LLC_SCID_MASK,
+				(llc->cntl1_regval &
+					A6XX_GPU_LLC_SCID_MASK));
+
+	if (llc->gpuhtw_llc_slice)
+		if (!llcc_slice_activate(llc->gpuhtw_llc_slice))
+			/* Program the sub-cache ID for GPU pagetables */
+			a6xx_gpu_cx_rmw(llc,
+				REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_1,
+				A6XX_GPUHTW_LLC_SCID_MASK,
+				(llc->cntl1_regval &
+					A6XX_GPUHTW_LLC_SCID_MASK));
+
+	/* Program cacheability overrides */
+	a6xx_gpu_cx_rmw(llc, REG_A6XX_GPU_CX_MISC_SYSTEM_CACHE_CNTL_0, 0xF,
+		llc->cntl0_regval);
+}
+
+void a6xx_llc_slices_destroy(struct a6xx_llc *llc)
+{
+	if (llc->mmio) {
+		iounmap(llc->mmio);
+		llc->mmio = NULL;
+	}
+
+	llcc_slice_putd(llc->gpu_llc_slice);
+	llc->gpu_llc_slice = NULL;
+
+	llcc_slice_putd(llc->gpuhtw_llc_slice);
+	llc->gpuhtw_llc_slice = NULL;
+}
+
+static int a6xx_llc_slices_init(struct platform_device *pdev,
+		struct a6xx_llc *llc)
+{
+	int i;
+
+	/* Get the system cache slice descriptor for GPU and GPUHTWs */
+	llc->gpu_llc_slice = llcc_slice_getd(&pdev->dev, "gpu");
+	if (IS_ERR(llc->gpu_llc_slice))
+		llc->gpu_llc_slice = NULL;
+
+	llc->gpuhtw_llc_slice = llcc_slice_getd(&pdev->dev, "gpuhtw");
+	if (IS_ERR(llc->gpuhtw_llc_slice))
+		llc->gpuhtw_llc_slice = NULL;
+
+	if (llc->gpu_llc_slice == NULL && llc->gpuhtw_llc_slice == NULL)
+		return -1;
+
+	/* Map registers */
+	llc->mmio = msm_ioremap(pdev, "cx_mem", "gpu_cx");
+	if (IS_ERR(llc->mmio)) {
+		llc->mmio = NULL;
+		a6xx_llc_slices_destroy(llc);
+		return -1;
+	}
+
+	/*
+	 * Setup GPU system cache CNTL0 and CNTL1 register values.
+	 * These values will be programmed everytime GPU comes out
+	 * of power collapse as these are non-retention registers.
+	 */
+
+	/*
+	 * CNTL0 provides options to override the settings for the
+	 * read and write allocation policies for the LLC. These
+	 * overrides are global for all memory transactions from
+	 * the GPU.
+	 *
+	 * 0x3: read-no-alloc-overridden = 0
+	 *      read-no-alloc = 0 - Allocate lines on read miss
+	 *      write-no-alloc-overridden = 1
+	 *      write-no-alloc = 1 - Do not allocates lines on write miss
+	 */
+	llc->cntl0_regval = 0x03;
+
+	/*
+	 * CNTL1 is used to specify SCID for (CP, TP, VFD, CCU and UBWC
+	 * FLAG cache) GPU blocks. This value will be passed along with
+	 * the address for any memory transaction from GPU to identify
+	 * the sub-cache for that transaction.
+	 *
+	 * Currently there is only one SCID allocated for all GPU blocks
+	 * Hence set same SCID for all the blocks.
+	 */
+
+	if (llc->gpu_llc_slice) {
+		u32 gpu_scid = llcc_get_slice_id(llc->gpu_llc_slice);
+
+		for (i = 0; i < A6XX_LLC_NUM_GPU_SCIDS; i++)
+			llc->cntl1_regval |=
+				gpu_scid << (A6XX_GPU_LLC_SCID_NUM_BITS * i);
+	}
+
+	/*
+	 * Set SCID for GPU IOMMU. This will be used to access
+	 * page tables that are cached in LLC.
+	 */
+	if (llc->gpuhtw_llc_slice) {
+		u32 gpuhtw_scid = llcc_get_slice_id(llc->gpuhtw_llc_slice);
+
+		llc->cntl1_regval |=
+			gpuhtw_scid << A6XX_GPUHTW_LLC_SCID_SHIFT;
+	}
+
+	return 0;
+}
+
 static int a6xx_pm_resume(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -923,6 +1072,9 @@  static int a6xx_pm_resume(struct msm_gpu *gpu)
 
 	gpu->needs_hw_init = true;
 
+	/* Activate LLC slices */
+	a6xx_llc_activate(gpu);
+
 	return ret;
 }
 
@@ -931,6 +1083,9 @@  static int a6xx_pm_suspend(struct msm_gpu *gpu)
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 
+	/* Deactivate LLC slices */
+	a6xx_llc_deactivate(gpu);
+
 	/*
 	 * Make sure the GMU is idle before continuing (because some transitions
 	 * may use VBIF
@@ -993,6 +1148,8 @@  static void a6xx_destroy(struct msm_gpu *gpu)
 		drm_gem_object_unreference_unlocked(a6xx_gpu->sqe_bo);
 	}
 
+	a6xx_llc_slices_destroy(&a6xx_gpu->llc);
+
 	a6xx_gmu_remove(a6xx_gpu);
 
 	adreno_gpu_cleanup(adreno_gpu);
@@ -1040,7 +1197,10 @@  struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	adreno_gpu->registers = a6xx_registers;
 	adreno_gpu->reg_offsets = a6xx_register_offsets;
 
-	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4, 0);
+	ret = a6xx_llc_slices_init(pdev, &a6xx_gpu->llc);
+
+	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4,
+			ret ? 0 : MMU_FEATURE_USE_SYSTEM_CACHE);
 	if (ret) {
 		a6xx_destroy(&(a6xx_gpu->base.base));
 		return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index 21ab701..392c426 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -21,6 +21,14 @@ 
 
 extern bool hang_debug;
 
+struct a6xx_llc {
+	void __iomem *mmio;
+	void *gpu_llc_slice;
+	void *gpuhtw_llc_slice;
+	u32 cntl0_regval;
+	u32 cntl1_regval;
+};
+
 struct a6xx_gpu {
 	struct adreno_gpu base;
 
@@ -46,6 +54,7 @@  struct a6xx_gpu {
 	uint64_t scratch_iova;
 
 	struct a6xx_gmu gmu;
+	struct a6xx_llc llc;
 };
 
 #define to_a6xx_gpu(x) container_of(x, struct a6xx_gpu, base)
diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
index 1ab629b..6c03eda 100644
--- a/drivers/gpu/drm/msm/msm_iommu.c
+++ b/drivers/gpu/drm/msm/msm_iommu.c
@@ -39,6 +39,16 @@  static int msm_iommu_attach(struct msm_mmu *mmu, const char * const *names,
 {
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 	int ret;
+	int gpu_htw_llc = 1;
+
+	/*
+	 * This allows GPU to set the bus attributes required
+	 * to use system cache on behalf of the iommu page table
+	 * walker.
+	 */
+	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
+		iommu_domain_set_attr(iommu->domain,
+				DOMAIN_ATTR_USE_UPSTREAM_HINT, &gpu_htw_llc);
 
 	pm_runtime_get_suppliers(mmu->dev);
 	ret = iommu_attach_device(iommu->domain, mmu->dev);
@@ -63,6 +73,9 @@  static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova,
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 	size_t ret;
 
+	if (msm_mmu_has_feature(mmu, MMU_FEATURE_USE_SYSTEM_CACHE))
+		prot |= IOMMU_USE_UPSTREAM_HINT;
+
 	pm_runtime_get_suppliers(mmu->dev);
 	ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot);
 	pm_runtime_put_suppliers(mmu->dev);
diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h
index 85df78d..257bdea 100644
--- a/drivers/gpu/drm/msm/msm_mmu.h
+++ b/drivers/gpu/drm/msm/msm_mmu.h
@@ -30,6 +30,9 @@  struct msm_mmu_funcs {
 	void (*destroy)(struct msm_mmu *mmu);
 };
 
+/* MMU features */
+#define MMU_FEATURE_USE_SYSTEM_CACHE (1 << 0)
+
 struct msm_mmu {
 	const struct msm_mmu_funcs *funcs;
 	struct device *dev;