Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 5 Mar 2017 21:43:03 -0500
From:      Pedro Giffuni <pfg@FreeBSD.org>
To:        John Baldwin <jhb@freebsd.org>
Cc:        Slawa Olhovchenkov <slw@zxy.spb.ru>, src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org
Subject:   Re: svn commit: r314669 - head/sys/i386/conf
Message-ID:  <3351E9C7-CC12-472A-9E81-4973710E9C8C@FreeBSD.org>
In-Reply-To: <1537596.qhqdTsTLLF@ralph.baldwin.cx>
References:  <201703041504.v24F4HMh023937@repo.freebsd.org> <7873439.f6BlOXHt6g@ralph.baldwin.cx> <c5fc59db-3d9a-92f9-0e10-e9002b38b9f8@FreeBSD.org> <1537596.qhqdTsTLLF@ralph.baldwin.cx>

next in thread | previous in thread | raw e-mail | index | archive | help

> Il giorno 05 mar 2017, alle ore 18:24, John Baldwin <jhb@freebsd.org> =
ha scritto:
>=20
> On Saturday, March 04, 2017 08:14:11 PM Pedro Giffuni wrote:
>>=20
>> On 3/4/2017 5:51 PM, John Baldwin wrote:
>>> On Saturday, March 04, 2017 03:49:52 PM Pedro Giffuni wrote:
>>>>> Il giorno 04 mar 2017, alle ore 14:43, John Baldwin =
<jhb@freebsd.org> ha scritto:
>>>>>=20
>>>>> On Saturday, March 04, 2017 10:52:46 AM Pedro Giffuni wrote:
>>>>>> On 03/04/17 10:32, Slawa Olhovchenkov wrote:
>>>>>>> On Sat, Mar 04, 2017 at 03:04:17PM +0000, Pedro F. Giffuni =
wrote:
>>>>>>>=20
>>>>>>>> Author: pfg
>>>>>>>> Date: Sat Mar  4 15:04:17 2017
>>>>>>>> New Revision: 314669
>>>>>>>> URL: https://svnweb.freebsd.org/changeset/base/314669
>>>>>>>>=20
>>>>>>>> Log:
>>>>>>>>  Drop i486 from the default i386 GENERIC kernel configuration.
>>>>>>>>=20
>>>>>>>>  80486 production was stopped by Intel on September 2007. =
Dropping the 486
>>>>>>>>  configuration option from the GENERIC kernel improves =
performance
>>>>>>>>  slightly.
>>>>>>>>=20
>>>>>>>>  Removing I486_CPU is consistent at this time: we don't support =
any
>>>>>>>>  processor without a FPU and the PC-98 arch, which frequently =
involved i486
>>>>>>>>  CPUs, is also gone so we don't test such platforms anymore.
>>>>>>> What is realy mean?
>>>>>> This means we don't do work-arounds that would be required for =
raw 486.
>>>>>> Instead we will use the 586 instructions by default.
>>>>> This doesn't change that.  The kernel already has runtime tests in =
place
>>>>> for new things on 486 and later via cpuid.
>>>>>=20
>>>> Hmm ..then I am wondering if I effectively changed anything?
>>> The only change is a 486 now panics on boot when it used to work =
fine. :-/
>>>=20
>>> Nothing for other CPUs has changed.
>>=20
>> Not much has been lost then.
>> FWIW, I have a "Pentium overdrive" somewhere in the basement which =
could=20
>> theoretically boot FreeBSD 12 but last I remember just rebuilding a=20=

>> kernel was painful and the memory and HD limitations really make it a =
no-go.
>=20
> So I would rather support 486 in GENERIC or not support it at all.  It =
doesn't
> cost anything for it to be in GENERIC, so if we have it, I think we =
should ship
> it.  Also, the original justification for this commit of a 4% =
performance gain
> doesn't seem to have any basis in fact.  The one gain I can think of =
is
> de-cluttering some things like identcpu.c and initcpu.c which can only =
happen if
> we remove code entirely, not from removing an option in GENERIC.
>=20

OK, that=E2=80=99s reasonable. I will revert the change.

Pedro.





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3351E9C7-CC12-472A-9E81-4973710E9C8C>