The matrix size of LAPACKE_sge_trans in lapacke_sgeqrt_work.c seems to be not correct #766
Description
Description
When LAPACKE_sgeqrt is executed with row major, the following code is executed to convert the result obtained with column major to row major.
lapack/LAPACKE/src/lapacke_sgeqrt_work.c
Lines 82 to 83 in 77fac7e
Here, ldt * MIN(m,n) should be the number of elements in t_t (= the number of elements in t), but this expression takes more values when ldt > nbnb > n.
This causes a segmentation fault when LAPACKE_sgeqrt is executed with row major and ldt > n.
Here is an example that causes the problem:
#include <stdio.h> #include <time.h> #include <lapacke.h> int main(int argc, char *argv[]) { srandom(42); const int N = 16; const int NB = 4; const int LDA = N; const int LDT = 100 * N; // set the value so that LDT > N holds. // allocate A and T float *A = (float *)malloc(N * LDA * sizeof(float)); float *T = (float *)malloc(N * LDT * sizeof(float)); // initialize A for (int i = 0; i < N; i++) { for (int j = 0; j < N; j++) { A[i * LDA + j] = (float)random() / (float)RAND_MAX; } } // call sgeqrt LAPACKE_sgeqrt(LAPACK_ROW_MAJOR, N, N, NB, A, LDA, T, LDT); // print results for (int i = 0; i < N; i++) { for (int j = 0; j < N; j++) { printf("% 8f ", A[i * LDA + j]); } printf("\n"); } printf("\n"); for (int i = 0; i < NB; i++) { for (int j = 0; j < N; j++) { printf("% 8f ", T[i * LDT + j]); } printf("\n"); } // free memory free(A); free(T); return 0; }Shouldn't this line look like this?
LAPACKE_sge_trans(LAPACK_COL_MAJOR, nb, MIN(m,n), t_t, ldt_t, t, ldt );I'm thinking that other LAPACKE_*geqrt functions may contain similar problems.
Checklist
- I've included a minimal example to reproduce the issue
- I'd be willing to make a PR to solve this issue