Re: [Mesa-dev] [PATCH 3/5] clover: Report default values for half and double fp configs v2

2014-06-26 Thread Francisco Jerez 
Tom Stellard  writes:

> From: Matt Arsenault 
>
> v2:
>   -Fix indentation
> ---
>  src/gallium/state_trackers/clover/api/device.cpp  | 11 +--
>  src/gallium/state_trackers/clover/core/device.cpp | 24 
> +++
>  src/gallium/state_trackers/clover/core/device.hpp |  3 +++
>  3 files changed, 36 insertions(+), 2 deletions(-)
>
> diff --git a/src/gallium/state_trackers/clover/api/device.cpp 
> b/src/gallium/state_trackers/clover/api/device.cpp
> index 97b2cf9..7006702 100644
> --- a/src/gallium/state_trackers/clover/api/device.cpp
> +++ b/src/gallium/state_trackers/clover/api/device.cpp
> @@ -201,8 +201,15 @@ clGetDeviceInfo(cl_device_id d_dev, cl_device_info param,
>break;
>  
> case CL_DEVICE_SINGLE_FP_CONFIG:
> -  buf.as_scalar() =
> - CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
> +  buf.as_scalar() = dev.single_fp_config();
> +  break;
> +
> +   case CL_DEVICE_DOUBLE_FP_CONFIG:
> +  buf.as_scalar() = dev.double_fp_config();
> +  break;
> +
> +   case CL_DEVICE_HALF_FP_CONFIG:
> +  buf.as_scalar() = dev.half_fp_config();
>break;


We need some mechanism to find out if the device supports half
precision.  Until then can you drop this case statement and
device::half_fp_config()?

>  
> case CL_DEVICE_GLOBAL_MEM_CACHE_TYPE:
> diff --git a/src/gallium/state_trackers/clover/core/device.cpp 
> b/src/gallium/state_trackers/clover/core/device.cpp
> index b6078db..bc6b761 100644
> --- a/src/gallium/state_trackers/clover/core/device.cpp
> +++ b/src/gallium/state_trackers/clover/core/device.cpp
>[...]
> +cl_device_fp_config
> +device::double_fp_config() const {
> +   // TODO: Get these from somewhere. This is the "mandated minimum double
> +   // precision floating-point capability"
> +   return CL_FP_FMA
> +   | CL_FP_ROUND_TO_NEAREST
> +   | CL_FP_ROUND_TO_ZERO
> +   | CL_FP_ROUND_TO_INF
> +   | CL_FP_INF_NAN
> +   | CL_FP_DENORM;

This should probably return 0 if the device doesn't support doubles.

>[...]


pgpfiOJbzsgmI.pgp
Description: PGP signature
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 3/5] clover: Report default values for half and double fp configs v2

2014-06-25 Thread Tom Stellard 
From: Matt Arsenault 

v2:
  -Fix indentation
---
 src/gallium/state_trackers/clover/api/device.cpp  | 11 +--
 src/gallium/state_trackers/clover/core/device.cpp | 24 +++
 src/gallium/state_trackers/clover/core/device.hpp |  3 +++
 3 files changed, 36 insertions(+), 2 deletions(-)

diff --git a/src/gallium/state_trackers/clover/api/device.cpp 
b/src/gallium/state_trackers/clover/api/device.cpp
index 97b2cf9..7006702 100644
--- a/src/gallium/state_trackers/clover/api/device.cpp
+++ b/src/gallium/state_trackers/clover/api/device.cpp
@@ -201,8 +201,15 @@ clGetDeviceInfo(cl_device_id d_dev, cl_device_info param,
   break;
 
case CL_DEVICE_SINGLE_FP_CONFIG:
-  buf.as_scalar() =
- CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
+  buf.as_scalar() = dev.single_fp_config();
+  break;
+
+   case CL_DEVICE_DOUBLE_FP_CONFIG:
+  buf.as_scalar() = dev.double_fp_config();
+  break;
+
+   case CL_DEVICE_HALF_FP_CONFIG:
+  buf.as_scalar() = dev.half_fp_config();
   break;
 
case CL_DEVICE_GLOBAL_MEM_CACHE_TYPE:
diff --git a/src/gallium/state_trackers/clover/core/device.cpp 
b/src/gallium/state_trackers/clover/core/device.cpp
index b6078db..bc6b761 100644
--- a/src/gallium/state_trackers/clover/core/device.cpp
+++ b/src/gallium/state_trackers/clover/core/device.cpp
@@ -169,6 +169,30 @@ device::max_compute_units() const {
   PIPE_COMPUTE_CAP_MAX_COMPUTE_UNITS)[0];
 }
 
+cl_device_fp_config
+device::single_fp_config() const {
+   // TODO: Get these from somewhere.
+   return CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
+}
+
+cl_device_fp_config
+device::double_fp_config() const {
+   // TODO: Get these from somewhere. This is the "mandated minimum double
+   // precision floating-point capability"
+   return CL_FP_FMA
+   | CL_FP_ROUND_TO_NEAREST
+   | CL_FP_ROUND_TO_ZERO
+   | CL_FP_ROUND_TO_INF
+   | CL_FP_INF_NAN
+   | CL_FP_DENORM;
+}
+
+cl_device_fp_config
+device::half_fp_config() const {
+   // TODO: Get these from somewhere.
+   return CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
+}
+
 std::vector
 device::max_block_size() const {
auto v = get_compute_param(pipe, PIPE_COMPUTE_CAP_MAX_BLOCK_SIZE);
diff --git a/src/gallium/state_trackers/clover/core/device.hpp 
b/src/gallium/state_trackers/clover/core/device.hpp
index 731c31e..16831ab 100644
--- a/src/gallium/state_trackers/clover/core/device.hpp
+++ b/src/gallium/state_trackers/clover/core/device.hpp
@@ -63,6 +63,9 @@ namespace clover {
   cl_ulong max_mem_alloc_size() const;
   cl_uint max_clock_frequency() const;
   cl_uint max_compute_units() const;
+  cl_device_fp_config single_fp_config() const;
+  cl_device_fp_config double_fp_config() const;
+  cl_device_fp_config half_fp_config() const;
 
   std::vector max_block_size() const;
   std::string device_name() const;
-- 
1.8.1.5

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 3/5] clover: Report default values for half and double fp configs

2014-06-21 Thread Francisco Jerez 
Tom Stellard  writes:

> From: Matt Arsenault 
>
> ---
>  src/gallium/state_trackers/clover/api/device.cpp  | 11 +--
>  src/gallium/state_trackers/clover/core/device.cpp | 24 
> +++
>  src/gallium/state_trackers/clover/core/device.hpp |  3 +++
>  3 files changed, 36 insertions(+), 2 deletions(-)
>
> diff --git a/src/gallium/state_trackers/clover/api/device.cpp 
> b/src/gallium/state_trackers/clover/api/device.cpp
> index 1bc2692..dc8e22c 100644
> --- a/src/gallium/state_trackers/clover/api/device.cpp
> +++ b/src/gallium/state_trackers/clover/api/device.cpp
> @@ -201,8 +201,15 @@ clGetDeviceInfo(cl_device_id d_dev, cl_device_info param,
>break;
>  
> case CL_DEVICE_SINGLE_FP_CONFIG:
> -  buf.as_scalar() =
> - CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
> +  buf.as_scalar() = dev.single_fp_config();
> +  break;
> +
> +   case CL_DEVICE_DOUBLE_FP_CONFIG:
> +  buf.as_scalar() = dev.double_fp_config();
> +  break;
> +
> +   case CL_DEVICE_HALF_FP_CONFIG:
> +  buf.as_scalar() = dev.half_fp_config();
>break;
>  
> case CL_DEVICE_GLOBAL_MEM_CACHE_TYPE:
> diff --git a/src/gallium/state_trackers/clover/core/device.cpp 
> b/src/gallium/state_trackers/clover/core/device.cpp
> index bc3e3e6..6d52dd4 100644
> --- a/src/gallium/state_trackers/clover/core/device.cpp
> +++ b/src/gallium/state_trackers/clover/core/device.cpp
> @@ -163,6 +163,30 @@ device::max_clock_frequency() const {
>
> PIPE_COMPUTE_CAP_MAX_CLOCK_FREQUENCY)[0];
>  }
>  
> +cl_device_fp_config
> +device::single_fp_config() const {
> +   // TODO: Get these from somewhere.
> +   return CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
> +}
> +
> +cl_device_fp_config
> +device::double_fp_config() const {
> +   // TODO: Get these from somewhere. This is the "mandated minimum double
> +   // precision floating-point capability"
> +return CL_FP_FMA

Weird indentation here, otherwise looks good to me.

> +   | CL_FP_ROUND_TO_NEAREST
> +   | CL_FP_ROUND_TO_ZERO
> +   | CL_FP_ROUND_TO_INF
> +   | CL_FP_INF_NAN
> +   | CL_FP_DENORM;
> +}
> +
> +cl_device_fp_config
> +device::half_fp_config() const {
> +   // TODO: Get these from somewhere.
> +   return CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
> +}
> +
>  std::vector
>  device::max_block_size() const {
> auto v = get_compute_param(pipe, 
> PIPE_COMPUTE_CAP_MAX_BLOCK_SIZE);
> diff --git a/src/gallium/state_trackers/clover/core/device.hpp 
> b/src/gallium/state_trackers/clover/core/device.hpp
> index 3662c6b..380029e 100644
> --- a/src/gallium/state_trackers/clover/core/device.hpp
> +++ b/src/gallium/state_trackers/clover/core/device.hpp
> @@ -62,6 +62,9 @@ namespace clover {
>size_t max_threads_per_block() const;
>cl_ulong max_mem_alloc_size() const;
>cl_uint max_clock_frequency() const;
> +  cl_device_fp_config single_fp_config() const;
> +  cl_device_fp_config double_fp_config() const;
> +  cl_device_fp_config half_fp_config() const;
>  
>std::vector max_block_size() const;
>std::string device_name() const;
> -- 
> 1.8.1.4


pgpBM336cJ0NR.pgp
Description: PGP signature
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 3/5] clover: Report default values for half and double fp configs

2014-06-17 Thread Tom Stellard 
From: Matt Arsenault 

---
 src/gallium/state_trackers/clover/api/device.cpp  | 11 +--
 src/gallium/state_trackers/clover/core/device.cpp | 24 +++
 src/gallium/state_trackers/clover/core/device.hpp |  3 +++
 3 files changed, 36 insertions(+), 2 deletions(-)

diff --git a/src/gallium/state_trackers/clover/api/device.cpp 
b/src/gallium/state_trackers/clover/api/device.cpp
index 1bc2692..dc8e22c 100644
--- a/src/gallium/state_trackers/clover/api/device.cpp
+++ b/src/gallium/state_trackers/clover/api/device.cpp
@@ -201,8 +201,15 @@ clGetDeviceInfo(cl_device_id d_dev, cl_device_info param,
   break;
 
case CL_DEVICE_SINGLE_FP_CONFIG:
-  buf.as_scalar() =
- CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
+  buf.as_scalar() = dev.single_fp_config();
+  break;
+
+   case CL_DEVICE_DOUBLE_FP_CONFIG:
+  buf.as_scalar() = dev.double_fp_config();
+  break;
+
+   case CL_DEVICE_HALF_FP_CONFIG:
+  buf.as_scalar() = dev.half_fp_config();
   break;
 
case CL_DEVICE_GLOBAL_MEM_CACHE_TYPE:
diff --git a/src/gallium/state_trackers/clover/core/device.cpp 
b/src/gallium/state_trackers/clover/core/device.cpp
index bc3e3e6..6d52dd4 100644
--- a/src/gallium/state_trackers/clover/core/device.cpp
+++ b/src/gallium/state_trackers/clover/core/device.cpp
@@ -163,6 +163,30 @@ device::max_clock_frequency() const {
   PIPE_COMPUTE_CAP_MAX_CLOCK_FREQUENCY)[0];
 }
 
+cl_device_fp_config
+device::single_fp_config() const {
+   // TODO: Get these from somewhere.
+   return CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
+}
+
+cl_device_fp_config
+device::double_fp_config() const {
+   // TODO: Get these from somewhere. This is the "mandated minimum double
+   // precision floating-point capability"
+return CL_FP_FMA
+   | CL_FP_ROUND_TO_NEAREST
+   | CL_FP_ROUND_TO_ZERO
+   | CL_FP_ROUND_TO_INF
+   | CL_FP_INF_NAN
+   | CL_FP_DENORM;
+}
+
+cl_device_fp_config
+device::half_fp_config() const {
+   // TODO: Get these from somewhere.
+   return CL_FP_DENORM | CL_FP_INF_NAN | CL_FP_ROUND_TO_NEAREST;
+}
+
 std::vector
 device::max_block_size() const {
auto v = get_compute_param(pipe, PIPE_COMPUTE_CAP_MAX_BLOCK_SIZE);
diff --git a/src/gallium/state_trackers/clover/core/device.hpp 
b/src/gallium/state_trackers/clover/core/device.hpp
index 3662c6b..380029e 100644
--- a/src/gallium/state_trackers/clover/core/device.hpp
+++ b/src/gallium/state_trackers/clover/core/device.hpp
@@ -62,6 +62,9 @@ namespace clover {
   size_t max_threads_per_block() const;
   cl_ulong max_mem_alloc_size() const;
   cl_uint max_clock_frequency() const;
+  cl_device_fp_config single_fp_config() const;
+  cl_device_fp_config double_fp_config() const;
+  cl_device_fp_config half_fp_config() const;
 
   std::vector max_block_size() const;
   std::string device_name() const;
-- 
1.8.1.4

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev