[patch,libgfortran] PR51119 - MATMUL slow for large matrices

[Thomas Koenig]

Hi Jerry,

> With these changes, OK for trunk?

Just going over this with a fine comb...

One thing just struck me:   The loop variables should be index_type, so

       const index_type m = xcount, n = ycount, k = count;

[...]

    index_type a_dim1, a_offset, b_dim1, b_offset, c_dim1, c_offset, i1, i2,
	      i3, i4, i5, i6;

       /* Local variables */
       GFC_REAL_4 t1[65536], /* was [256][256] */
		 f11, f12, f21, f22, f31, f32, f41, f42,
		 f13, f14, f23, f24, f33, f34, f43, f44;
       index_type i, j, l, ii, jj, ll;
       index_type isec, jsec, lsec, uisec, ujsec, ulsec;

I agree that we should do the tuning of the inline limit
separately.

When we do that, we should think about -Os.  With the buffering, we have
much more memory usage in the library function.  If -Os is in force,
we should also consider raising the limit for inlining.

Since I was involved in the development, I would like to give others a
few days to raise more comments.  If there are none, OK to commit with
the above change within a few days. Of course, somebody else might also
OK this patch :-)

Regards

	Thomas