From 0ec053539253ad93b840e208b6a7d740734a2f71 Mon Sep 17 00:00:00 2001
From: Andreas Kloeckner <inform@tiker.net>
Date: Thu, 8 Jul 2010 20:21:34 -0400
Subject: [PATCH] Only use 'go-faster' options in matrix multiply example on
Nvidia.
---
examples/matrix-multiply.py | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/examples/matrix-multiply.py b/examples/matrix-multiply.py
index 142591f1..c8fab12e 100644
--- a/examples/matrix-multiply.py
+++ b/examples/matrix-multiply.py
@@ -161,8 +161,12 @@ h_c = numpy.empty((c_height, c_width)).astype(numpy.float32)
kernel_params = {"block_size": block_size,
"w_a":a_width, "h_a":a_height, "w_b":b_width}
+if "NVIDIA" in queue.device.vendor:
+ options = "-cl-mad-enable -cl-fast-relaxed-math"
+else:
+ options = None
prg = cl.Program(ctx, KERNEL_CODE % kernel_params,
- ).build(options="-cl-mad-enable -cl-fast-relaxed-math")
+ ).build(options=options)
kernel = prg.matrixMul
#print prg.binaries[0]
--
GitLab