thanks
Giacomo
--
_____
Giacomo Mulas
_
INAF - Osservatorio Astronomico di Cagliari
via della scienza 5 - 09047 Selargius (CA)
tel. +39 070 71180247
mob. : +
ance, best regards
Giacomo Mulas
#include
#include
#include
void saxpy(float a, float* x, float* y, int sz) {
#pragma omp target teams distribute parallel for simd \
num_teams(3) map(to:x[0:sz]) map(tofrom:y[0:sz])
for (int i = 0; i < sz; i++) {
if (omp_is_initial_device()) {
ance, best regards
Giacomo Mulas
#include
#include
#include
void saxpy(float a, float* x, float* y, int sz) {
#pragma omp target teams distribute parallel for simd \
num_teams(3) map(to:x[0:sz]) map(tofrom:y[0:sz])
for (int i = 0; i < sz; i++) {
if (omp_is_initial_device()) {
ance, best regards
Giacomo Mulas
#include
#include
#include
void saxpy(float a, float* x, float* y, int sz) {
#pragma omp target teams distribute parallel for simd \
num_teams(3) map(to:x[0:sz]) map(tofrom:y[0:sz])
for (int i = 0; i < sz; i++) {
if (omp_is_initial_device()) {
load-nvptx does not appear to install them anywhere, nor
to depend on any other separate package that does. This effectively renders the
package unusable.
Thanks in advance, best regards
Giacomo Mulas
*** Reporter, please consider answering these questions, where appropriate ***
* What led
he default compiler fails.
Thanks in advance, best regards
Giacomo Mulas
-- System Information:
Debian Release: trixie/sid
APT prefers unstable
APT policy: (401, 'unstable'), (10, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Kernel: Linux 6.7
el free to reassign the bug, if appropriate.
Thanks in advance, best regards
Giacomo Mulas
-- System Information:
Debian Release: trixie/sid
APT prefers unstable
APT policy: (401, 'unstable'), (10, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Ke
in the support libraries, that problem
would still remain regardless of what you do on nvptx-tools, wouldn't it?
Thanks, bye
Giacomo
--
_________
Giacomo Mulas
_
INAF
same error message I get with gcc-12.
Is there something wrong in how GCC 12 handles nvptx code generation
options?
bye
Giacomo
--
_____
Giacomo Mulas
_
INAF - Osservato
s dropped"
ok, but I am puzzled by this: if the issue is with nvptx-tools and not with
gcc-12-offload-nvptx, why does gcc-11-offload-nvptx work, producing working
executables that target sm_35 if I compile with gcc-11 on the same laptop?
Thanks, bye
Giacomo
--
_____
of 8
Hello world! I am thread 3 out of 8
Would it be possible to change the default -misa of gcc 12 to sm_35,
to enable gpu offloading to nvidia to work with gcc-12? And/or, is there
some undocumented, or poorly documented, way to actually specify on the
command line the requested cuda level arch
:i386 (--configure):
dependency problems - leaving unconfigured
If only one arch is installed, then things work. But this is obviously a
serious issue for almost any amd64 installation that also wants to be able
to have a fairly complete i386 runtime.
Best regards
Giacomo Mulas
-- System
the (now default) 4.9 version of gfortran. I just searched
"by hand" all such mod files on my sid system and filed bug reports for
their packages, but this could (should?) be made automatic whenever a new
version of gfortran is made the default.
Thanks in advance, bye
Giacomo Mula
at they did not need to waste the same time I wasted to
track and solve the same problem...
Bye
Giacomo
--
_________
Giacomo Mulas
_
OSSERVATORIO ASTRONOMICO DI CAGLIARI
Str. 54,
14 matches
Mail list logo
