Skip to content

Commit 9c4ae4d

Browse files
authored
Merge pull request #4206 from martin-frbg/issue4201-2
Work around miscompilation of zdot_thunderx2t99 by the current NVIDIA HPC compiler
2 parents 3bb70b8 + fc8894d commit 9c4ae4d

File tree

1 file changed

+6
-1
lines changed

1 file changed

+6
-1
lines changed

kernel/arm64/zdot_thunderx2t99.c

Lines changed: 6 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -24,7 +24,12 @@ CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
2424
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
2525
USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
2626
*****************************************************************************/
27-
27+
#ifdef __NVCOMPILER
28+
#define NVCOMPVERS ( __NVCOMPILER_MAJOR__ * 100 + __NVCOMPILER_MINOR__ )
29+
#if (NVCOMPVERS < 2309)
30+
#pragma opt 1
31+
#endif
32+
#endif
2833

2934
#include "common.h"
3035

0 commit comments

Comments
 (0)