On Mon, 23 Apr 2001, Byeong-ryeol Kim wrote:
> On Sun, 22 Apr 2001, Alan Cox wrote:
>
> > > > Using /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o
> > > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > > symbol rwsem_up_write_wake
> > > > /lib/modules/2.4.3-a
On Sun, 22 Apr 2001, Alan Cox wrote:
> > > Using /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > symbol rwsem_up_write_wake
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > sy
On 22 Apr 2001, Jes Sorensen wrote:
> > "Alan" == Alan Cox <[EMAIL PROTECTED]> writes:
> Alan> The recommended compilers for non x86 are different too - eg you
> Alan> need 2.96 gcc for IA64, you need 2.95 not egcs for mips and so
> Alan> on.
>
> In principle you just need 2.7.2.3 for m68k, b
Alan Cox wrote:
>
> > > Using /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > symbol rwsem_up_write_wake
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > symbol rwsem_down_wri
Hello Alan , To whom is this attributed ? Tia , JimL
On Sun, 22 Apr 2001, Alan Cox wrote:
> o Hopefully fix bugtraq reported netfilter ftp
> flaw
++
| James W. Laferriere | System Techniques |
On 04.22 Dieter Nützel wrote:
> > My belief however is that several million people have gcc 2.96-69+, about 50
> > are likely to have random cvs snapshots and none of them are going to build
> > kernels with them anyway, as they wont work __builtin_expect or otherwise.
> >
> > Alan
>
> I will no
Alan Cox wrote:
> > > Using /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > symbol rwsem_up_write_wake
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > symbol rwsem_down_write
On 2001.04.22 11:48 Alan Cox wrote:
> > > Using /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > symbol rwsem_up_write_wake
> > > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > > sym
> > Using /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o
> > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > symbol rwsem_up_write_wake
> > /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o: unresolved
> > symbol rwsem_down_write_failed
>
> Same thing wit
> In principle you just need 2.7.2.3 for m68k, but someone decided to
> raise the bar for all architectures by putting a check in a common
> header file.
I suspect you would find that some of the problems with the initialisers
in structures were common to 2.7.2 across all platforms, but I may be
> "Roman" == Roman Zippel <[EMAIL PROTECTED]> writes:
Roman> Hi, Jes Sorensen wrote:
>> In principle you just need 2.7.2.3 for m68k, but someone decided to
>> raise the bar for all architectures by putting a check in a common
>> header file.
Roman> IIRC 2.7.2.3 has problems with labeled ini
Hi,
Jes Sorensen wrote:
> In principle you just need 2.7.2.3 for m68k, but someone decided to
> raise the bar for all architectures by putting a check in a common
> header file.
IIRC 2.7.2.3 has problems with labeled initializers for structures,
which makes 2.7.2.3 unusable for all archs under
On 2001.04.22 09:25 John Cavan wrote:
> Alan Cox wrote:
> > 2.4.3-ac12
> > o Further semaphore fixes (David Howells)
>
> Getting unresolved symbols in some modules (notably, for me, microcode.o
> and radeon.o)...
>
> Using /lib/modules/2.4.3-ac12/kernel/drivers/cha
> My belief however is that several million people have gcc 2.96-69+, about 50
> are likely to have random cvs snapshots and none of them are going to build
> kernels with them anyway, as they wont work __builtin_expect or otherwise.
>
> Alan
I will not add fuel to the fire, but isn't 2.4.XX the
In case everyone missed my original patch =P
http://marc.theaimsgroup.com/?l=linux-kernel&m=98791931115515&w=2
Jes Sorensen wrote:
>
> > "Alan" == Alan Cox <[EMAIL PROTECTED]> writes:
>
> Alan> The recommended compilers for non x86 are different too - eg you
> Alan> need 2.96 gcc for IA64
Alan Cox wrote:
> 2.4.3-ac12
> o Further semaphore fixes (David Howells)
Getting unresolved symbols in some modules (notably, for me, microcode.o
and radeon.o)...
Using /lib/modules/2.4.3-ac12/kernel/drivers/char/drm/radeon.o
/lib/modules/2.4.3-ac12/kernel/drivers/c
> "Alan" == Alan Cox <[EMAIL PROTECTED]> writes:
Alan> The recommended compilers for non x86 are different too - eg you
Alan> need 2.96 gcc for IA64, you need 2.95 not egcs for mips and so
Alan> on.
In principle you just need 2.7.2.3 for m68k, but someone decided to
raise the bar for all arc
f5ibh wrote:
> Alan,
>
>
>>> /usr/src/linux-2.4.3-ac12/lib/lib.a(rwsem.o): In function
>>> `rwsem_up_write_wake':rwsem.o(.text+0x3c6): undefined reference to
>>> `__builtin_expect'
>>
>> Add a
>>
>> #define __builtin_expect
>
>
> I had the same problem here, adding #define __builtin_expect
> Are you being deliberately obtuse? 2.97+ snapshots do all support
> builtin_expect, which is what we were discussing.
I think we are having different conversations here.
The only valid inputs to the question are
Recommended
---
egcs-1.1.2 (miscom
Alan,
>> /usr/src/linux-2.4.3-ac12/lib/lib.a(rwsem.o): In function
>> `rwsem_up_write_wake':rwsem.o(.text+0x3c6): undefined reference to
>> `__builtin_expect'
>
>Add a
>
>#define __builtin_expect
I had the same problem here, adding #define __builtin_expect in ../lib/rwsem.c
solved the problem.
>There are no gcc 2.97 snapshots that compile the kernel correctly because
>they have the broken bitfield packing ABI change.
Oh right. I didn't know about that particular nicety.
>My belief however is that several million people have gcc 2.96-69+, about 50
>are likely to have random cvs snaps
21 matches
Mail list logo
