Skip to contentSkip to conversation

Linux v2.6.12-rc6

DISCUSSION Voluspa 6 participants 7 messages Jun 7, 2005

lore
Share & tools
lore

Voluspa

author
#0 lore raw

CC arch/x86_64/kernel/irq.o
CC arch/x86_64/kernel/ptrace.o
arch/x86_64/kernel/ptrace.c: In function `putreg':
arch/x86_64/kernel/ptrace.c:285: error: duplicate case value arch/x86_64/kernel/ptrace.c:280: error: previously used here make[1]: *** [arch/x86_64/kernel/ptrace.o] Error 1
make: *** [arch/x86_64/kernel] Error 2

real    0m5.524s
user    0m4.227s
sys     0m0.724s

Mvh
Mats Johannesson
--

#1 lore raw
Quoted 7 lines · Voluspa

CC arch/x86_64/kernel/irq.o
CC arch/x86_64/kernel/ptrace.o
arch/x86_64/kernel/ptrace.c: In function `putreg':
arch/x86_64/kernel/ptrace.c:285: error: duplicate case value arch/x86_64/kernel/ptrace.c:280: error: previously used here make[1]: *** [arch/x86_64/kernel/ptrace.o] Error 1
make: *** [arch/x86_64/kernel] Error 2

builds fine here - and i cannot see how those case values could be duplicate. Are you sure your build is completely clean?

Ingo

author
#2 lore raw
Quoted 11 lines · On Tue, 7 Jun 2005 08:18:31 +0200 Ingo Molnar wrote:
Quoted 7 lines · Voluspa

CC arch/x86_64/kernel/irq.o
CC arch/x86_64/kernel/ptrace.o
arch/x86_64/kernel/ptrace.c: In function `putreg':
arch/x86_64/kernel/ptrace.c:285: error: duplicate case value arch/x86_64/kernel/ptrace.c:280: error: previously used here make[1]: *** [arch/x86_64/kernel/ptrace.o] Error 1
make: *** [arch/x86_64/kernel] Error 2

Quoted 2 lines

builds fine here - and i cannot see how those case values could be duplicate. Are you sure your build is completely clean?

Ah, sorry about the noise... I've been away from kernel testing too long. I patched a 2.6.11.11 tree without noticing all the rejects (this new machine is fast). But from what I remember, it was decided to do the -rc patches against the latest stable codebase, in this case .11 Shrug.

Mvh
Mats Johannesson
--

#3 lore raw
Quoted 5 lines · Voluspa

Ah, sorry about the noise... I've been away from kernel testing too long. I patched a 2.6.11.11 tree without noticing all the rejects (this new machine is fast). But from what I remember, it was decided to do the -rc patches against the latest stable codebase, in this case .11 Shrug.

I dunno what the change was, patch didn't apply cleanly to 2.6.11, (no idea if bad .bz2, finger trouble), so I download whole thing instead, now running on three x86 boxen.
--Grant.

#4 lore raw
Quoted 9 lines · Voluspa
Quoted 5 lines

Ah, sorry about the noise... I've been away from kernel testing too long. I patched a 2.6.11.11 tree without noticing all the rejects (this new machine is fast). But from what I remember, it was decided to do the -rc patches against the latest stable codebase, in this case .11 Shrug.

I dunno what the change was, patch didn't apply cleanly to 2.6.11, (no idea if bad .bz2, finger trouble), so I download whole thing instead, now running on three x86 boxen.

It was explicitly stated by Linus way back 2.6.8.1 time that subsequent patches are against the base of the previous release, so -rcs are against 2.6.x not 2.6.x.y.. which all makes great sense if development is happening in parallel... I'm not sure I've ever heard anything else stated to oppose this, but apparently some people have..

Dave.

#5 lore raw
Quoted 6 lines · On Tue, 7 Jun 2005 17:44:00 +1000 Dave Airlie wrote:
Quoted 6 lines · Voluspa
Quoted 5 lines

Ah, sorry about the noise... I've been away from kernel testing too long. I patched a 2.6.11.11 tree without noticing all the rejects (this new machine is fast). But from what I remember, it was decided to do the -rc patches against the latest stable codebase, in this case .11 Shrug.

[...]

Quoted 5 lines

It was explicitly stated by Linus way back 2.6.8.1 time that subsequent patches are against the base of the previous release, so -rcs are against 2.6.x not 2.6.x.y.. which all makes great sense if development is happening in parallel... I'm not sure I've ever heard anything else stated to oppose this, but apparently some people have..

Yes, I remember that episode well. But times and policy change, don't they? When the "stable" dot-releases were introduced, there was another discussion, albeit brief, of where the base should lie. And my memory must have really left for an early vacation since the .x.y decision came so easily and quickly to my mind.

Anyway, the -rc6 applied cleanly to a 2.6.11 kernel and my uptime is now 27 minutes.

Issue: The new "conservative" qpufreq governor doesn't work. Stuck on full blast. Loading the "ondemand" and trying to echo that choice into the scaling_governor file doesn't work ("cannot overwrite existing file"). Must reboot since the "conservative" don't want to be unloaded (is in use). Guess I could have unloaded the whole ACPI module system in correct order, but a reboot was quicker --> 20 seconds...

Ask for more info about my setup if someone would like to work on this.

Meta-issue: Anyone not using hotplug/udev/whatever will trip on the kbd-now-being-another-mouse. Should result in plenty of user bugreports around the lists unless the fact is posted somewhere prominently. I took a brief tour with the -rc5 (downloaded a whole kernel) and just shook my head when the mice in X became totally unruly. Didn't know if I should laugh or cry, that's how funny the cursor behaved. I've now bumped up both the internal touchpad and the usb mouse one notch higher in the /dev/input/mouseX list and all is well.

Mvh
Mats Johannesson
--

#6 lore raw
In reply to Voluspa
Quoted 5 lines · Voluspa

Ah, sorry about the noise... I've been away from kernel testing too long. I patched a 2.6.11.11 tree without noticing all the rejects (this new machine is fast). But from what I remember, it was decided to do the -rc patches against the latest stable codebase, in this case .11 Shrug.

Try adding "-s" to the patch command then.

Signature · 1 line
Matthias Andree