Simplifying ARMv8 build parameters
ARMv8 builds were a bit mixed up, with ThunderX2 code in ARMv8 mode (which is not right because TX2 is ARMv8.1) as well as requiring a few redundancies in the defines, making it harder to maintain and understand what core has what. A few other minor issues were also fixed. Tests were made on the following cores: A53, A57, A72, Falkor, ThunderX, ThunderX2, and XGene. Tests were: OpenBLAS/test, OpenBLAS/benchmark, BLAS-Tester. A summary: * Removed TX2 code from ARMv8 build, to make sure it is compatible with all ARMv8 cores, not just v8.1. Also, the TX2 code has actually harmed performance on big cores. * Commoned up ARMv8 architectures' defines in params.h, to make sure that all will benefit from ARMv8 settings, in addition to their own. * Adding a few more cores, using ARMv8's include strategy, to benefit from compiler optimisations using mtune. Also updated cache information from the manuals, making sure we set good conservative values by default. Removed Vulcan, as it's an alias to TX2. * Auto-detecting most of those cores, but also updating the forced compilation in getarch.c, to make sure the parameters are the same whether compiled natively or forced arch. Benefits: * ARMv8 build is now guaranteed to work on all ARMv8 cores * Improved performance for ARMv8 builds on some cores (A72, Falkor, ThunderX1 and 2: up to 11%) over current develop * Improved performance for *all* cores comparing to develop branch before TX2's patch (9% ~ 36%) * ThunderX1 builds are 14% faster than ARMv8 on TX1, 9% faster than current develop's branch and 8% faster than deveop before tx2 patches Issues: * Regression from current develop branch for A53 (-12%) and A57 (-3%) with ARMv8 builds, but still faster than before TX2's commit (+15% and +24% respectively). This can be improved with a simplification of TX2's code, to be done in future patches. At least the code is guaranteed to be ARMv8.0 now. Comments: * CortexA57 builds are unchanged on A57 hardware from develop's branch, which makes sense, as it's untouched. * CortexA72 builds improve over A57 on A72 hardware, even if they're using the same includes due to new compiler tunning in the makefile.
This commit is contained in:
Renato Golin
@@ -2543,49 +2543,9 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#define SYMV_P 16
|
||||
#endif
|
||||
|
||||
|
||||
#if defined(CORTEXA57)
|
||||
#define SNUMOPT 2
|
||||
#define DNUMOPT 2
|
||||
|
||||
#define GEMM_DEFAULT_OFFSET_A 0
|
||||
#define GEMM_DEFAULT_OFFSET_B 0
|
||||
#define GEMM_DEFAULT_ALIGN 0x03fffUL
|
||||
|
||||
#define SGEMM_DEFAULT_UNROLL_M 16
|
||||
#define SGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define DGEMM_DEFAULT_UNROLL_M 8
|
||||
#define DGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define CGEMM_DEFAULT_UNROLL_M 8
|
||||
#define CGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define ZGEMM_DEFAULT_UNROLL_M 4
|
||||
#define ZGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define SGEMM_DEFAULT_P 512
|
||||
#define DGEMM_DEFAULT_P 256
|
||||
#define CGEMM_DEFAULT_P 256
|
||||
#define ZGEMM_DEFAULT_P 128
|
||||
|
||||
#define SGEMM_DEFAULT_Q 1024
|
||||
#define DGEMM_DEFAULT_Q 512
|
||||
#define CGEMM_DEFAULT_Q 512
|
||||
#define ZGEMM_DEFAULT_Q 512
|
||||
|
||||
#define SGEMM_DEFAULT_R 4096
|
||||
#define DGEMM_DEFAULT_R 4096
|
||||
#define CGEMM_DEFAULT_R 4096
|
||||
#define ZGEMM_DEFAULT_R 2048
|
||||
|
||||
|
||||
#define SYMV_P 16
|
||||
#endif
|
||||
|
||||
// Common ARMv8 parameters
|
||||
#if defined(ARMV8)
|
||||
|
||||
#if defined(OS_DARWIN) && defined(CROSS)
|
||||
#define SNUMOPT 2
|
||||
#define DNUMOPT 2
|
||||
|
||||
@@ -2593,6 +2553,11 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#define GEMM_DEFAULT_OFFSET_B 0
|
||||
#define GEMM_DEFAULT_ALIGN 0x03fffUL
|
||||
|
||||
#define SYMV_P 16
|
||||
|
||||
// Darwin / Cross
|
||||
#if defined(OS_DARWIN) && defined(CROSS)
|
||||
|
||||
#define SGEMM_DEFAULT_UNROLL_M 2
|
||||
#define SGEMM_DEFAULT_UNROLL_N 2
|
||||
|
||||
@@ -2620,15 +2585,11 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#define CGEMM_DEFAULT_R 4096
|
||||
#define ZGEMM_DEFAULT_R 4096
|
||||
|
||||
#define SYMV_P 16
|
||||
#else
|
||||
#else // Linux / Native
|
||||
|
||||
#define SNUMOPT 2
|
||||
#define DNUMOPT 2
|
||||
|
||||
#define GEMM_DEFAULT_OFFSET_A 0
|
||||
#define GEMM_DEFAULT_OFFSET_B 0
|
||||
#define GEMM_DEFAULT_ALIGN 0x03fffUL
|
||||
#if defined(CORTEXA53) || defined(CORTEXA57) || \
|
||||
defined(CORTEXA72) || defined(CORTEXA73) || \
|
||||
defined(FALKOR)
|
||||
|
||||
#define SGEMM_DEFAULT_UNROLL_M 16
|
||||
#define SGEMM_DEFAULT_UNROLL_N 4
|
||||
@@ -2642,33 +2603,22 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#define ZGEMM_DEFAULT_UNROLL_M 4
|
||||
#define ZGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define SGEMM_DEFAULT_P 128
|
||||
#define DGEMM_DEFAULT_P 160
|
||||
#define CGEMM_DEFAULT_P 128
|
||||
#define SGEMM_DEFAULT_P 512
|
||||
#define DGEMM_DEFAULT_P 256
|
||||
#define CGEMM_DEFAULT_P 256
|
||||
#define ZGEMM_DEFAULT_P 128
|
||||
|
||||
#define SGEMM_DEFAULT_Q 352
|
||||
#define DGEMM_DEFAULT_Q 128
|
||||
#define CGEMM_DEFAULT_Q 224
|
||||
#define ZGEMM_DEFAULT_Q 112
|
||||
#define SGEMM_DEFAULT_Q 1024
|
||||
#define DGEMM_DEFAULT_Q 512
|
||||
#define CGEMM_DEFAULT_Q 512
|
||||
#define ZGEMM_DEFAULT_Q 512
|
||||
|
||||
#define SGEMM_DEFAULT_R 4096
|
||||
#define DGEMM_DEFAULT_R 4096
|
||||
#define CGEMM_DEFAULT_R 4096
|
||||
#define ZGEMM_DEFAULT_R 4096
|
||||
#define ZGEMM_DEFAULT_R 2048
|
||||
|
||||
#define SYMV_P 16
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
||||
#if defined(THUNDERX)
|
||||
#define SNUMOPT 2
|
||||
#define DNUMOPT 2
|
||||
|
||||
#define GEMM_DEFAULT_OFFSET_A 0
|
||||
#define GEMM_DEFAULT_OFFSET_B 0
|
||||
#define GEMM_DEFAULT_ALIGN 0x03fffUL
|
||||
#elif defined(THUNDERX)
|
||||
|
||||
#define SGEMM_DEFAULT_UNROLL_M 4
|
||||
#define SGEMM_DEFAULT_UNROLL_N 4
|
||||
@@ -2697,17 +2647,7 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#define CGEMM_DEFAULT_R 4096
|
||||
#define ZGEMM_DEFAULT_R 4096
|
||||
|
||||
|
||||
#define SYMV_P 16
|
||||
#endif
|
||||
|
||||
#if defined(THUNDERX2T99) || defined(VULCAN)
|
||||
#define SNUMOPT 2
|
||||
#define DNUMOPT 2
|
||||
|
||||
#define GEMM_DEFAULT_OFFSET_A 0
|
||||
#define GEMM_DEFAULT_OFFSET_B 0
|
||||
#define GEMM_DEFAULT_ALIGN 0x03fffUL
|
||||
#elif defined(THUNDERX2T99)
|
||||
|
||||
#define SGEMM_DEFAULT_UNROLL_M 16
|
||||
#define SGEMM_DEFAULT_UNROLL_N 4
|
||||
@@ -2736,8 +2676,40 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#define CGEMM_DEFAULT_R 4096
|
||||
#define ZGEMM_DEFAULT_R 4096
|
||||
|
||||
#define SYMV_P 16
|
||||
#endif
|
||||
#else // Other/undetected ARMv8 cores
|
||||
|
||||
#define SGEMM_DEFAULT_UNROLL_M 16
|
||||
#define SGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define DGEMM_DEFAULT_UNROLL_M 8
|
||||
#define DGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define CGEMM_DEFAULT_UNROLL_M 8
|
||||
#define CGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define ZGEMM_DEFAULT_UNROLL_M 4
|
||||
#define ZGEMM_DEFAULT_UNROLL_N 4
|
||||
|
||||
#define SGEMM_DEFAULT_P 128
|
||||
#define DGEMM_DEFAULT_P 160
|
||||
#define CGEMM_DEFAULT_P 128
|
||||
#define ZGEMM_DEFAULT_P 128
|
||||
|
||||
#define SGEMM_DEFAULT_Q 352
|
||||
#define DGEMM_DEFAULT_Q 128
|
||||
#define CGEMM_DEFAULT_Q 224
|
||||
#define ZGEMM_DEFAULT_Q 112
|
||||
|
||||
#define SGEMM_DEFAULT_R 4096
|
||||
#define DGEMM_DEFAULT_R 4096
|
||||
#define CGEMM_DEFAULT_R 4096
|
||||
#define ZGEMM_DEFAULT_R 4096
|
||||
|
||||
#endif // Cores
|
||||
|
||||
#endif // Linux / Darwin
|
||||
|
||||
#endif // ARMv8
|
||||
|
||||
#if defined(ARMV5)
|
||||
#define SNUMOPT 2
|
||||
|
||||
Reference in New Issue
Block a user