[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <7bf64b12-ea59-470c-ac31-8226be4eb496@zytor.com>
Date: Wed, 12 Jun 2024 19:56:28 -0700
From: Xin Li <xin@...or.com>
To: Borislav Petkov <bp@...en8.de>
Cc: linux-kernel@...r.kernel.org, linux-perf-users@...r.kernel.org,
tglx@...utronix.de, mingo@...hat.com, dave.hansen@...ux.intel.com,
x86@...nel.org, hpa@...or.com, will@...nel.org, peterz@...radead.org,
akpm@...ux-foundation.org, acme@...nel.org, namhyung@...nel.org
Subject: Re: [RESEND v1 1/3] x86/cpufeatures: Add {required,disabled} feature
configs
On 6/12/2024 6:33 AM, Borislav Petkov wrote:
> On Thu, May 09, 2024 at 01:53:38PM -0700, Xin Li (Intel) wrote:
>> diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu
>> index 2a7279d80460..719302d37053 100644
>> --- a/arch/x86/Kconfig.cpu
>> +++ b/arch/x86/Kconfig.cpu
>> @@ -358,6 +358,10 @@ config X86_P6_NOP
>> depends on X86_64
>> depends on (MCORE2 || MPENTIUM4 || MPSC)
>>
>> +config X86_REQUIRED_FEATURE_NOPL
>
> Can we keep the X86_{REQUIRED,DISABLED}_ prefixes solely in
> arch/x86/Kconfig.cpufeatures and not spill them out into the rest of the tree?
Good point, let me try.
>
> This way there will be no confusion between X86_FEATURE_, X86_REQUIRED_FEATURE_,
> X86_DISABLED_FEATURE_ and so on, and which one I am supposed to use where...
Currently X86_DISABLED_FEATURE_ macros are defined only in:
arch/x86/Kconfig.cpufeatures. But X86_REQUIRED_FEATURE_ macros are not.
>> diff --git a/arch/x86/Kconfig.cpufeatures b/arch/x86/Kconfig.cpufeatures
>> new file mode 100644
>> index 000000000000..326a8410ff06
>> --- /dev/null
>> +++ b/arch/x86/Kconfig.cpufeatures
>> @@ -0,0 +1,153 @@
>> +# SPDX-License-Identifier: GPL-2.0
>> +#
>> +# x86 feature bits (see arch/x86/include/asm/cpufeatures.h) that are
>> +# either REQUIRED to be enabled, or DISABLED (always ignored) for this
>> +# particular compile-time configuration. The tests for these features
>> +# are turned into compile-time constants via the generated
>> +# <asm/featuremasks.h>.
>> +#
>> +# The naming of these variables *must* match asm/cpufeatures.h.
>
> I presume they must match X86_FEATURE_<name>?
Right, let me add this.
>
> And REQUIRED and DISABLED is manipulated in by the script?
>
> I guess I'll see later.
Yep!
Thanks!
Xin
Powered by blists - more mailing lists