[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20161101084641.2a66qov3itzfbmnt@pd.tnic>
Date: Tue, 1 Nov 2016 09:46:41 +0100
From: Borislav Petkov <bp@...en8.de>
To: He Chen <he.chen@...ux.intel.com>
Cc: Paolo Bonzini <pbonzini@...hat.com>,
"Luc, Piotr" <Piotr.Luc@...el.com>,
"kvm@...r.kernel.org" <kvm@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"tglx@...utronix.de" <tglx@...utronix.de>,
"x86@...nel.org" <x86@...nel.org>, "hpa@...or.com" <hpa@...or.com>,
"mingo@...hat.com" <mingo@...hat.com>,
"Kang, Luwei" <luwei.kang@...el.com>,
"rkrcmar@...hat.com" <rkrcmar@...hat.com>
Subject: Re: [PATCH] x86/cpuid: expose AVX512_4VNNIW and AVX512_4FMAPS
features to kvm guest
On Tue, Nov 01, 2016 at 03:48:50PM +0800, He Chen wrote:
> Before sending a patch, let me check if my understanding is right...
> I will add a helper in scattered.c like:
>
> unsigned int get_scattered_cpuid_features(unsigned int level,
> unsigned int sub_leaf, enum cpuid_regs reg)
> {
> u32 val = 0;
> const struct cpuid_bit *cb;
>
> for (cb = cpuid_bits; cb->feature; cb++) {
Right, we can improve that by keeping cpuid_bit.level sorted so that you
can break out early if the level is exceeded. For that please sort it
and add a comment stating that the leaf should be kept sorted ontop of
it.
And then you do something like this:
u32 get_scattered_cpuid_leaf(unsigned int level, unsigned int sub_leaf,
enum cpuid_regs reg)
{
u32 cpuid_val = 0;
for (cb = cpuid_bits; cb->feature; cb++) {
if (level < cb->level)
continue;
if (level > cb->level)
break;
if (cb->reg == reg && sub_leaf == cb->sub_leaf) {
if (cpu_has(cb->feature))
cpuid_val |= BIT(cb->bit);
}
}
return cpuid_val;
}
Completely untested, of course.
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
Powered by blists - more mailing lists