All of lore.kernel.org
 help / color / mirror / Atom feed
From: Brian Gerst <bgerst@didntduck.org>
To: Adrian Bunk <bunk@kernel.org>
Cc: Sam Ravnborg <sam@ravnborg.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 03/10] x86: arch/x86/Kconfig.cpu unification
Date: Tue, 06 Nov 2007 02:10:27 -0500	[thread overview]
Message-ID: <47301363.9030303@didntduck.org> (raw)
In-Reply-To: <20071106025253.GF3444@stusta.de>

Adrian Bunk wrote:
> On Tue, Nov 06, 2007 at 03:46:30AM +0100, Sam Ravnborg wrote:
>> On Mon, Nov 05, 2007 at 07:53:39PM -0500, Brian Gerst wrote:
>>> Sam Ravnborg wrote:
>>>> Move all CPU definitions to Kconfig.cpu
>>> This patch causes build failure on x86_64:
>>>
>>> lib/rwsem.c: In function ‘__init_rwsem’:
>>> lib/rwsem.c:24: error: ‘struct rw_semaphore’ has no member named ‘count’
>>> lib/rwsem.c:24: error: ‘RWSEM_UNLOCKED_VALUE’ undeclared (first use in
>>> this function)
>>> lib/rwsem.c:24: error: (Each undeclared identifier is reported only once
>>> lib/rwsem.c:24: error: for each function it appears in.)
>>> lib/rwsem.c: In function ‘__rwsem_do_wake’:
>>> lib/rwsem.c:64: error: implicit declaration of function
>>> ‘rwsem_atomic_update’
>>> lib/rwsem.c:64: error: ‘RWSEM_ACTIVE_BIAS’ undeclared (first use in this
>>> function)
>>> lib/rwsem.c:66: error: ‘RWSEM_ACTIVE_MASK’ undeclared (first use in this
>>> function)
>>> lib/rwsem.c:115: error: ‘RWSEM_WAITING_BIAS’ undeclared (first use in
>>> this function)
>>> lib/rwsem.c:120: error: implicit declaration of function ‘rwsem_atomic_add’
>>> lib/rwsem.c: In function ‘rwsem_down_failed_common’:
>>> lib/rwsem.c:169: error: ‘RWSEM_ACTIVE_MASK’ undeclared (first use in
>>> this function)
>>> lib/rwsem.c: In function ‘rwsem_down_read_failed’:
>>> lib/rwsem.c:197: error: ‘RWSEM_WAITING_BIAS’ undeclared (first use in
>>> this function)
>>> lib/rwsem.c:197: error: ‘RWSEM_ACTIVE_BIAS’ undeclared (first use in
>>> this function)
>>> lib/rwsem.c: In function ‘rwsem_down_write_failed’:
>>> lib/rwsem.c:210: error: ‘RWSEM_ACTIVE_BIAS’ undeclared (first use in
>>> this function)
>>>
>>> git bisect gives:
>>> 55432d383871d127685d78f94e0ad60307dcc0e9 is first bad commit
>>> commit 55432d383871d127685d78f94e0ad60307dcc0e9
>>> Author: Sam Ravnborg <sam@ravnborg.org>
>>> Date:   Thu Nov 1 21:54:39 2007 +0100
>>>
>>>     x86: arch/x86/Kconfig.cpu unification
>> Seems that a default value fail to change value when we shift
>> from 32 to 64 bit. I have it reproduced here.
>> Will look at it tonight.
> 
> The problem is that you source arch/x86/Kconfig.cpu from 
> arch/x86/Kconfig.x86_64 but didn't give RWSEM_XCHGADD_ALGORITHM 
> the required dependency on X86_32.
> 
>> 	Sam
> 
> cu
> Adrian
> 

The quick fix is to move the RWSEM_XCHGADD_ALGORITHM and
RWSEM_XCHGADD_ALGORITHM defines from Kconfig.cpu to Kconfig.i386.

Looking at this closer, the 64-bit code has been using the generic
spinlock-based implementation, while the 32-bit code uses an optimized
version on all but true 386 cpus.  The 64-bit code should be adapted to
use the optimized version at some point.

  reply	other threads:[~2007-11-06  7:10 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-03 23:51 [PATCH 01/10] x86: unification of cfufreq/Kconfig Sam Ravnborg
2007-11-03 23:51 ` [PATCH 02/10] x86: start unification of arch/x86/Kconfig.* Sam Ravnborg
2007-11-03 23:51   ` [PATCH 03/10] x86: arch/x86/Kconfig.cpu unification Sam Ravnborg
2007-11-03 23:51     ` [PATCH 04/10] x86: add X86_32 dependency to i386 specific symbols in Kconfig.i386 Sam Ravnborg
2007-11-03 23:51       ` [PATCH 05/10] x86: add X86_64 dependency to x86_64 specific symbols in Kconig.x86_64 Sam Ravnborg
2007-11-03 23:51         ` [PATCH 06/10] x86: copy x86_64 specific Kconfig symbols to Kconifg.i386 Sam Ravnborg
2007-11-03 23:51           ` [PATCH 07/10] x86: add remaning bits from x86_64 to Kconfig.i386 Sam Ravnborg
2007-11-03 23:51             ` [PATCH 08/10] x86: combine all config options with prompts in Kconfig Sam Ravnborg
2007-11-03 23:51               ` [PATCH 09/10] x86: select i386 or x86_64 at config time Sam Ravnborg
2007-11-03 23:51                 ` [PATCH 10/10] x86: enable make ARCH=x86 Sam Ravnborg
2007-11-06  0:53     ` [PATCH 03/10] x86: arch/x86/Kconfig.cpu unification Brian Gerst
2007-11-06  2:46       ` Sam Ravnborg
2007-11-06  2:52         ` Adrian Bunk
2007-11-06  7:10           ` Brian Gerst [this message]
2007-11-04  1:44   ` [PATCH 02/10] x86: start unification of arch/x86/Kconfig.* Adrian Bunk
2007-11-04 18:07     ` Sam Ravnborg
2007-11-04  1:28 ` [PATCH 01/10] x86: unification of cfufreq/Kconfig Adrian Bunk
2007-11-04  8:35   ` Sam Ravnborg
2007-11-06  7:38 ` Dave Jones
2007-11-06  8:13   ` Sam Ravnborg
2007-11-06 10:49     ` Adrian Bunk

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=47301363.9030303@didntduck.org \
    --to=bgerst@didntduck.org \
    --cc=bunk@kernel.org \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=sam@ravnborg.org \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.