Weslley S. Pereira

> @weslleyspereira Which three letter identifier should I use? If `GQR` is reused (xG_GQR_CS), the code would be tested together with the QR and RQ factorization. I see... Maybe just...

Ah... now I saw your mention to the __three__-letter `PATH`. I thought it could be the `SRNAMT` id, sorry. I would suggest `gQR`, but the cases don't matter. This takes...

Hi @haldaas. I am trying to solve this issue. Can you please review #540? Thanks.

#515 closed the CBLAS part. We still need to change `LAPACKE/include/lapack.h` to solve this issue.

Hi @eshpc. As far as I know, nobody is working on that, and it would certainly be good to have these ifdefs on LAPACKE too. Thanks!

The double precision tests still show: 39 failing tests. See https://github.com/Reference-LAPACK/lapack/actions/runs/2806513051 My personal machine found 16 tests failing for the CS decomposition in double precision: ```ctest Command: "/usr/bin/cmake" "-DTEST=/home/weslleyp/storage/lapack/build_RelWithDebInfo/bin/xeigtstd" "-DINPUT=/home/weslleyp/storage/lapack/TESTING/csd.in"...

> my tests in #697 showed you'd need to restore dorbdb6.f to fix the 39 errors. Thanks, @martin-frbg! I completely bypassed your comment: https://github.com/Reference-LAPACK/lapack/issues/695#issuecomment-1203569363. I will apply those changes too.

@martin-frbg, your minimal fix makes my tests pass indeed. Thanks! However, I understand that setting X1 and X2 to this code is important for dorbdb5. See: https://github.com/Reference-LAPACK/lapack/blob/3381a0ec8249be684428421bd30ed388de93fc47/SRC/dorbdb5.f#L218-L226 The point here,...

I have rebased with the current master ( that includes #702 ), and then all tests pass in my machine. I am now running the tests on Github Actions. Thanks...

I will merge this PR since the failing tests in double precision were fixed by #702. __[Edit:] Actually, can someone review my changes so I can merge it? Thanks!__