Problem with matrix multiplication dimentions

[Eliyaan]

Describe the bug

As discussed here with spytheman : https://discord.com/channels/592103645835821068/763496422166626336/1202923962326323220

When using large matrices (that are not handled by simple loops but by dgemv) it outputs matrix dimention errors. (but maybe I just didn't understand something)

Expected Behavior

The matrix multiplications to work.

Current Behavior

[/home/nopana/NeuralNetworks-V-Module/examples/tst.v:8] 'works': works
[/home/nopana/NeuralNetworks-V-Module/examples/tst.v:12] 'works': works
================ V panic ================
   module: vsl.vlas.internal.blas
 function: dgemv()
  message: blas: bad leading dimension of A
     file: /home/nopana/.vmodules/vsl/vlas/internal/blas/dgemv.v:18
   v hash: 0a612bf
=========================================

Reproduction Steps

import vsl.la 

fn main() {
    // works 
    ma_small := la.Matrix.new[f64](3, 2)
    vec_small := []f64{len:2, init:1.0}
    _ := la.matrix_vector_mul(1.0, ma_small, vec_small)
    dump("works")
    ma_tr_small := la.Matrix.new[f64](1, 2)
    vec_tr_small := []f64{len:1, init:1.0}
    _ := la.matrix_tr_vector_mul(1.0, ma_tr_small, vec_tr_small)
    dump("works")

    // does not
    ma_big := la.Matrix.new[f64](30, 70)
    vec_big := []f64{len:70, init:1.0}
    _ := la.matrix_vector_mul(1.0, ma_big, vec_big)
    dump("works")
    ma_tr_big := la.Matrix.new[f64](10, 20)
    vec_tr_big := []f64{len:10, init:1.0}
    _ := la.matrix_tr_vector_mul(1.0, ma_tr_big, vec_tr_big)
    dump("works")
}

Possible Solution

The checks that dgemv may be wrong (a month ago I did modify them on a separate version and then the matmult did work fine).

Additional Information/Context

No response

V version

V 0.4.4 07016fb

Version used

last vsl commit

Environment details (OS name and version, etc.)

V full version: V 0.4.4 673a2f4.07016fb
OS: linux, Linux version 6.6.13-200.fc39.x86_64 (mockbuild@72f11b2996ed4699b0f705186172808f) (gcc (GCC) 13.2.1 20231205 (Red Hat 13.2.1-6), GNU ld version 2.40-13.fc39) #1 SMP PREEMPT_DYNAMIC Sat Jan 20 18:03:28 UTC 2024
Processor: 4 cpus, 64bit, little endian, Intel(R) Pentium(R) Gold 7505 @ 2.00GHz

getwd: /home/nopana/NeuralNetworks-V-Module
vexe: /home/nopana/v/v
vexe mtime: 2024-02-02 15:38:32

vroot: OK, value: /home/nopana/v
VMODULES: OK, value: /home/nopana/.vmodules
VTMP: OK, value: /tmp/v_1000

Git version: git version 2.43.0
Git vroot status: weekly.2024.05-15-g07016fb3
.git/config present: true

CC version: cc (GCC) 13.2.1 20231205 (Red Hat 13.2.1-6)
thirdparty/tcc status: thirdparty-linux-amd64 12f392c3

[ulises-jeremias]

Solved on master