Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-23 Thread Pavel Machek
On Fri 2008-01-18 01:27:06, H. Peter Anvin wrote: > Kyle McMartin wrote: > >On Fri, Jan 18, 2008 at 06:53:53AM +0100, Andi Kleen > >wrote: > >>One problem that we had in the past is that some > >>simulators > >>only implement the absolutely minimum feature set and > >>you > >>might have well bro

Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-18 Thread Andi Kleen
On Fri, Jan 18, 2008 at 08:56:43AM -0500, H. Peter Anvin wrote: > Andi Kleen wrote: > >>Simulators can be fixed, > > > >They could, but why? I don't know of a good reason to require CLFLUSH. > > Well, simulators are generally expected to follow the architecture, not > vice versa. I would tend t

Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-18 Thread H. Peter Anvin
Andi Kleen wrote: Simulators can be fixed, They could, but why? I don't know of a good reason to require CLFLUSH. Well, simulators are generally expected to follow the architecture, not vice versa. I would tend to agree with the coupling that recent versions of Bochs appeared to have made

Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-17 Thread Andi Kleen
> Simulators can be fixed, They could, but why? I don't know of a good reason to require CLFLUSH. > I just verified that Bochs 2.3.0 lacks this CPUID bit whereas the > current version, 2.3.6, enables CLFLUSH iff SSE2 is enabled. Qemu 0.9.0 > has CLFLUSH. Andi, do you happen to know of any sp

Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-17 Thread H. Peter Anvin
Kyle McMartin wrote: On Fri, Jan 18, 2008 at 06:53:53AM +0100, Andi Kleen wrote: One problem that we had in the past is that some simulators only implement the absolutely minimum feature set and you might have well broken one of these with this. Yeah, true. Please ignore the patch folks. chee

Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-17 Thread Kyle McMartin
On Fri, Jan 18, 2008 at 06:53:53AM +0100, Andi Kleen wrote: > One problem that we had in the past is that some simulators > only implement the absolutely minimum feature set and you > might have well broken one of these with this. Yeah, true. Please ignore the patch folks. cheers, Kyle -- To unsu

Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-17 Thread Andi Kleen
Kyle McMartin <[EMAIL PROTECTED]> writes: > Hopefully nobody will be stupid enough to implement a cpu without > it. Frankly, it seems safe enough given we already require SSE2. > > This means the compiler can optimise away "if (!cpu_has_clflush)" > blocks. The original required CPUID bit set for

Re: [PATCH] x86: make clflush a required feature on x86_64

2008-01-17 Thread H. Peter Anvin
Kyle McMartin wrote: Hopefully nobody will be stupid enough to implement a cpu without it. Frankly, it seems safe enough given we already require SSE2. This means the compiler can optimise away "if (!cpu_has_clflush)" blocks. Signed-off-by: Kyle McMartin <[EMAIL PROTECTED]> --- include/asm-x86

[PATCH] x86: make clflush a required feature on x86_64

2008-01-17 Thread Kyle McMartin
Hopefully nobody will be stupid enough to implement a cpu without it. Frankly, it seems safe enough given we already require SSE2. This means the compiler can optimise away "if (!cpu_has_clflush)" blocks. Signed-off-by: Kyle McMartin <[EMAIL PROTECTED]> --- include/asm-x86/cpufeature_64.h |