Package: gcc-12-offload-nvptx
Version: 12.4.0-2
Severity: grave
Justification: renders package unusable
Dear Maintainer,
Any attempt to compile code that offloads to nvptx-none fails with the
following error:
gmulas@capitanata:~/mintest$ gcc -fopenmp -O3 -ggdb -fcf-protection=check
-foffload=default -foffload=nvptx-none="-O3 -ggdb -fopt-info -lm -latomic
-mgomp" -o mintest mintest.c
mintest.c:5:9: optimized: basic block part vectorized using 16 byte vectors
collect2: fatal error: ld terminated with signal 11 [Errore di segmentazione],
core dumped
compilation terminated.
nvptx mkoffload: fatal error: x86_64-linux-gnu-accel-nvptx-none-gcc-12 returned
1 exit status
compilation terminated.
lto-wrapper: fatal error:
/usr/libexec/gcc/x86_64-linux-gnu/12//accel/nvptx-none/mkoffload returned 1
exit status
compilation terminated.
/usr/bin/ld: error: lto-wrapper failed
collect2: error: ld returned 1 exit status
This occurs with any gcc (or g++) version installable on debian sid, so I
suspect it is connected to some update on the nvptx-tools side, but since it
is reproduced using gcc I am initially reporting the bug on
gcc-12-offload-nvptx. If/when you can trace it to some other package, of
course feel free to move the bug where it belongs. This is rather critical,
since it makes it impossible to compile any code that offloads to nvptx. I
attach a minimal test code that reproduces it. It compiles and runs
flawlessly on a debian stable machine.
Thanks in advance, best regards
Giacomo Mulas
#include <stdio.h>
#include <stdlib.h>
#include <omp.h>
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()) {
printf("Running on host\n");
} else {
int nthreads= omp_get_num_threads();
int nteams= omp_get_num_teams();
printf("Running on device with %d teams (fixed) in total and %d
threads in each team\n",nteams,nthreads);
}
printf("Thread %d %i\n", omp_get_thread_num(), i );
y[i] = a * x[i] + y[i];
}
}
int main(int argc, char** argv) {
float a = 2.0;
int sz = 16;
float *x = (float *) calloc( sz, sizeof *x );
float *y = (float *) calloc( sz, sizeof *y );
//Set values
int num_devices = omp_get_num_devices();
printf("Number of available devices %d\n", num_devices);
saxpy( a, x, y, sz );
return 0;
}
-- System Information:
Debian Release: trixie/sid
APT prefers unstable
APT policy: (401, 'unstable'), (10, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Kernel: Linux 6.11.9-amd64 (SMP w/12 CPU threads; PREEMPT)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE,
TAINT_UNSIGNED_MODULE
Locale: LANG=it_IT.UTF-8, LC_CTYPE=it_IT.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled
Versions of packages gcc-12-offload-nvptx depends on:
ii gcc-12 12.4.0-2
ii gcc-12-base 12.4.0-2
ii libc6 2.40-3
ii libc6-dev 2.40-3
ii libgmp10 2:6.3.0+dfsg-2+b2
ii libgomp-plugin-nvptx1 14.2.0-8
ii libmpc3 1.3.1-1+b3
ii libmpfr6 4.2.1-1+b2
ii libzstd1 1.5.6+dfsg-1+b1
ii nvptx-tools 0.20240810-1
ii zlib1g 1:1.3.dfsg+really1.3.1-1+b1
gcc-12-offload-nvptx recommends no packages.
gcc-12-offload-nvptx suggests no packages.
-- no debconf information
