Received-SPF: pass (google.com: domain of linux-kernel+bounces-207351-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Date: Sun, 09 Jun 2024 04:47:51 -0700
From: "H. Peter Anvin" <hpa@zytor.com>
To: Borislav Petkov <bp@alien8.de>,
        Linus Torvalds <torvalds@linux-foundation.org>
CC: Ingo Molnar <mingo@kernel.org>, Thomas Gleixner <tglx@linutronix.de>,
        Rasmus Villemoes <linux@rasmusvillemoes.dk>,
        Josh Poimboeuf <jpoimboe@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        the arch/x86 maintainers <x86@kernel.org>,
        linux-arch <linux-arch@vger.kernel.org>
Subject: Re: [PATCH] x86: add 'runtime constant' infrastructure
User-Agent: K-9 Mail for Android
In-Reply-To: <20240609112240.GBZmWQgNQXguD_8Nc8@fat_crate.local>
References: <20240608193504.429644-2-torvalds@linux-foundation.org> <20240609112240.GBZmWQgNQXguD_8Nc8@fat_crate.local>
Message-ID: <00BA183C-4EE6-46AE-AEC8-94B612222373@zytor.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable

On June 9, 2024 4:22:40 AM PDT, Borislav Petkov <bp@alien8=2Ede> wrote:
>On Sat, Jun 08, 2024 at 12:35:05PM -0700, Linus Torvalds wrote:
>> Ingo / Peter / Borislav - I enabled this for 32-bit x86 too, because it
>> was literally trivial (had to remove a "q" from "movq")=2E  I did a
>> test-build and it looks find, but I didn't actually try to boot it=2E=
=20
>
>Will do once you have your final version=2E I still have an Atom, 32-bit
>only laptop lying around here=2E
>
>> +#define runtime_const_ptr(sym) ({				\
>> +	typeof(sym) __ret;					\
>> +	asm("mov %1,%0\n1:\n"					\
>> +		"=2Epushsection runtime_ptr_" #sym ",\"a\"\n\t"	\
>> +		"=2Elong 1b - %c2 - =2E\n\t"			\
>> +		"=2Epopsection"					\
>> +		:"=3Dr" (__ret)					\
>> +		:"i" ((unsigned long)0x0123456789abcdefull),	\
>> +		 "i" (sizeof(long)));				\
>> +	__ret; })
>
>You might wanna use asm symbolic names for the operands so that it is
>more readable:
>
>#define runtime_const_ptr(sym) ({                                        =
       \
>        typeof(sym) __ret;                                               =
       \
>        asm("mov %[constant] ,%[__ret]\n1:\n"                            =
       \
>                "=2Epushsection runtime_ptr_" #sym ",\"a\"\n\t"          =
         \
>                "=2Elong 1b - %c[sizeoflong] - =2E\n\t"                  =
           \
>                "=2Epopsection"                                          =
         \
>                : [__ret] "=3Dr" (__ret)                                 =
         \
>                : [constant] "i" ((unsigned long)0x0123456789abcdefull), =
       \
>                  [sizeoflong] "i" (sizeof(long)));                      =
       \
>        __ret; })
>
>For example=2E
>
>> +// The 'typeof' will create at _least_ a 32-bit type, but
>> +// will happily also take a bigger type and the 'shrl' will
>> +// clear the upper bits
>
>Can we pls use the multiline comments, like you do below in the same
>file=2E
>
>Otherwise, it looks ok to me and it boots in a guest=2E
>
>I'll take the final version for a spin on real hw in a couple of days,
>once the review dust settles=2E
>
>Thx=2E
>

So the biggest difference versus what I had in progress was that I had the=
 idea of basically doing "ro_after_init" behavior by doing memory reference=
s until alternatives are run=2E

I don't know if that was overthinking the problem=2E=2E=2E