# Fortran 2023 standard **URL:** \ **Category:** Help\ **Created:** [February 5, 2023, 9:42pm UTC](https://fortran-lang.discourse.group/t/fortran-2023-standard/5137 "2023-02-05T21:42:09Z")\ **Posts on this page:** 1\ **Showing post:** 50
**Author:** ![RonShepard](https://avatars.discourse-cdn.com/v4/letter/r/a3d4f5/32.png) [@RonShepard](https://fortran-lang.discourse.group/u/RonShepard)\ **Post date:** [July 1, 2023, 5:44am UTC](https://fortran-lang.discourse.group/t/fortran-2023-standard/5137/50 "2023-07-01T05:44:49Z")
> [@certik](#): > > I didn’t know you can use `-framework Accelerate` with gfortran. Here is a thread from last year that discusses some errors in some of the real32 routines in the Accelerate framework. I think examples were given with mkl and OpenBLAS as the discussion progressed. > [@How many BLAS libraries have this error?](https://fortran-lang.discourse.group/t/how-many-blas-libraries-have-this-error/4454): > > There is a fairly well-known error in many BLAS libraries. Here is some code that demonstrates it. program test\_blas implicit none real :: dot, snorm, x(3) = [1.0, 2.0, 3.0], y(3) = [3.0, 0.0, 4.0] character(\*), parameter :: cfmt='(a,i0,2(a,f0.3))' block integer, parameter :: wp=kind(1.e0) real(wp), external :: sdot, snrm2 dot = sdot(3,x,1,y,1) snorm = snrm2(3,y,1) write(\*,cfmt) 'wp kind=', wp, ' sdot=', dot, ' snrm2=', snorm end block block integ… I have not tested a recent Xcode install, which is where Accelerate lives, so I’m not sure if this has been corrected or not by Apple since that thread was posted. I did receive an email that they received my bug report and that they were looking at it. --- _[View the full topic](https://fortran-lang.discourse.group/t/fortran-2023-standard/5137)._