Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 4 Oct 2019 15:13:38 +0100
From:      tech-lists <tech-lists@zyxst.net>
To:        freebsd-hackers@freebsd.org
Subject:   CPUTYPE?= in make.conf
Message-ID:  <20191004141338.GA72749@bastion.zyxst.net>

next in thread | raw e-mail | index | archive | help

--OgqxwSJOaUobr8KG
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hi,

I hope this is the right place to ask about this; if not please suggest whe=
re?

I have a large poudriere system running 12.1-STABLE #0 r352869. It compiles
ports in poudriere jails for many other smaller systems, some on same arch
(amd64) but different cpu types.

The host cputype is:

# llvm-tblgen --version
LLVM (http://llvm.org/):
  LLVM version 8.0.1
  Optimized build.
  Default target: x86_64-unknown-freebsd12.1
  Host CPU: sandybridge

The actual cpu is Intel(R) Xeon(R) CPU E5-2690 0 @ 2.90GHz (x2)

If I make a make.conf for a poudriere jail and define CPUTYPE?=3Dbtver1, se=
veral
ports fail to compile, like python36 and lzo2.

If I do the same and define it as CPUTYPE?=3Dhaswell, *everything* fails ap=
art
=66rom building the jail itself.

If I do the same thing on a different host, where the host cpu is: Intel(R)
Core(TM) i7-4770K CPU @ 3.50GHz

no failures happen. llvm-tblgen here reports haswell. The freebsd version is
the same.

What I'm asking is, is this a hardware failure on the failing system? Or is=
 it
a capabilities issue, in that the one having the problem hasn't got the rig=
ht
instruction set?

The Xenon differs from the i7 in that it doesn't have these cpu instructions
available:

AVX AVX2 BMI1 BMI2 F16C FMA3=20
Supervisor Mode Execution Protection
Advanced Bit manipulation
Enhanced REP MOVSB/STOSB
INVPCID
MOVBE
RD/WR FSGSBASE instructions
RDRAND

I'm hoping for a performance boost if ports are compiled to the specific
cputype of the target machine. if I don't specify cputype at all on the
failing system, the ports will compile. Is it an instruction set issue or is
the Xenon actually failing in some way?

thanks,
--=20
J.

--OgqxwSJOaUobr8KG
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=BAiD
-----END PGP SIGNATURE-----

--OgqxwSJOaUobr8KG--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20191004141338.GA72749>