svn commit: r387283 - head/devel/sdl20

Tijl Coosemans tijl at FreeBSD.org
Mon May 25 09:13:54 UTC 2015


On Mon, 25 May 2015 10:51:29 +0200 Mathieu Arnold <mat at FreeBSD.org> wrote:
> +--On 25 mai 2015 10:37:25 +0200 Tijl Coosemans <tijl at FreeBSD.org> wrote:
> | On Sun, 24 May 2015 20:01:47 +0200 Jan Beich <jbeich at FreeBSD.org> wrote:
> |> Tijl Coosemans <tijl at FreeBSD.org> writes:
> |>> +.if defined(MACHINE_CPU) && ${MACHINE_CPU:Mmmx}
> |>> +CONFIGURE_ARGS+=	--enable-mmx
> |>> +.else
> |>> +CONFIGURE_ARGS+=	--disable-mmx
> |>> +.endif
> |> 
> |> Why not use smth like the following instead ?
> |> 
> |>   OPTIONS_DEFAULT_i386=	${MACHINE_CPU:tu:MMMX}
> | 
> | Because an option doesn't make much sense for CPU features.  Either
> | the target CPU supports MMX or it doesn't.  It's not a user choice.
> 
> Well, yes and no, I'm building packages on a:
> 
> CPU: Intel(R) Xeon(R) CPU E5-2430L 0 @ 2.00GHz (2000.04-MHz K8-class CPU)
>   Origin = "GenuineIntel"  Id = 0x206d7  Family = 0x6  Model = 0x2d
> Stepping = 7
> 
> Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
> 
> Features2=0x1fbee3ff<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,DCA,SSE4.1,SSE4.2,x2APIC,POPCNT,TSCDLT,AESNI,XSAVE,OSXSAVE,AVX>
>   AMD Features=0x2c100800<SYSCALL,NX,Page1GB,RDTSCP,LM>
>   AMD Features2=0x1<LAHF>
>   VT-x: PAT,HLT,MTF,PAUSE,EPT,UG,VPID
>   TSC: P-state invariant, performance statistics
> 
> but my i386 packages have to work on a 
> 
> CPU: Geode(TM) Integrated Processor by AMD PCS (498.05-MHz 586-class CPU)
>   Origin = "AuthenticAMD"  Id = 0x5a2  Stepping = 2
>   Features=0x88a93d<FPU,DE,PSE,TSC,MSR,CX8,SEP,PGE,CMOV,CLFLUSH,MMX>
>   AMD Features=0xc0400000<MMX+,3DNow!+,3DNow!>
> 
> 
> True, they all support MMX instructions, but I'd say it's more by chance
> than anything else :-)

MACHINE_CPU is defined in /usr/share/mk/bsd.cpu.mk based on CPUTYPE
which you can set in make.conf.  The host CPU doesn't matter.


More information about the svn-ports-all mailing list