fa.linux.kernel Google Group

Re: i686 quirk for AMD Geode

2009-11-08T18:23:32Z

Right, but what if is_geode_lx() is called befor the x86.id change takes effect?
Maybe something like this?
--- a/arch/x86/include/asm/geode.h 2009-11-08 19:13:43.531117343 +0100
+++ b/arch/x86/include/asm/geode.h 2009-11-08 19:19:42.130618023 +0100
@@ -177,7 +177,7 @@
static inline int is_geode_lx(void)

Re: [git patches] rt2800 drivers updates

2009-11-08T18:15:01Z

The current branch is now rt2800-v3.1, changes:
- fixed build problem reported by Hauke Mehrtens
- added "rt2x00: Remove deprecated ieee80211_rx_status->qual usage" patch
from Ivo and "rt2x00: Update copyright statements." from you
It may take some time to get it to kernel.org mirrors though..

Re: i686 quirk for AMD Geode

2009-11-08T18:13:36Z

This is on my TODO list, but better to discuss it in the GCC mailing list,
the kernel should use -march=geode and GCC should generate the best code
for the AMD Geode, actually the i686 one

Re: i686 quirk for AMD Geode

2009-11-08T18:10:41Z

Good. So there's really no point in breaking kernel.
..fix gcc to genereta code at least as good as for i686 and you are
done...?
Pavel

Re: i686 quirk for AMD Geode

2009-11-08T18:05:07Z

True, but also remove the duplicate function is_geode in the NAND driver
and use the identical one defined in geode.h:
--- a/drivers/mtd/nand/cs553x_nand .c 2009-11-08 18:58:14.835043214 +0100
+++ b/drivers/mtd/nand/cs553x_nand .c 2009-11-08 19:00:07.914117831 +0100
@@ -30,6 +30,7 @@
#include <asm/msr.h>

Re: [PATCH -tip] x86,apic: Use PAGE_SIZE instead of numbers

2009-11-08T18:04:07Z

Not in the next 10-20 years or so. (and any prediction beyond that is
pointless i suspect)
Ingo

Re: [PATCH] trivial: fix checking socket() in net tstamp example

2009-11-08T17:56:03Z

Hi Patrick,
By the way, I tried igb hardware timestamp but HWTSTAMP_FILTER_ALL works
almost like HWTSTAMP_FILTER_PTP_*. Isn't it supposed to timestamp all
of the incoming packets? Maybe there is something wrong with my test
setup?
Thanks,
Ali

[PATCH] drivers/media/video: correct initialization of audio_mode

2009-11-08T17:49:19Z

This initialization of the value of audio_mode is the one used if nothing
matches in the subsequent switch. The variable audio_mode is subsequently
assigned to constants such as TUNER_AUDIO_MONO and TUNER_AUDIO_STEREO.
TUNER_AUDIO_STEREO has the same value as V4L2_TUNER_MODE_STEREO, so it
would seem better to use that value here.

Re: i686 quirk for AMD Geode

2009-11-08T17:47:19Z

Gcc 4.4 already knows about it, just sucks at optimizing:
gcc -c -O3 -march=geode ./dry.c -o dry1.o
gcc -DPASS2 -O3 -march=geode ./dry.c dry1.o -o dry2
Dhrystone Benchmark, Version C, Version 2.2
Program compiled without 'register' attribute
Using times(), HZ=100
Trying 5000000 runs through Dhrystone:

Re: mmotm 2009-11-01-10-01 uploaded

2009-11-08T17:43:59Z

OK. yes, weird.
thanks.
---
~Randy

Re: i686 quirk for AMD Geode

2009-11-08T17:37:28Z

..
Teach gcc that geodelx exists? No need to break kernel for that... and
you probably can gain even bigger gains.
Pavel

Re: [PATCH -tip] x86,apic: Use PAGE_SIZE instead of numbers

2009-11-08T17:37:02Z

[Maciej W. Rozycki - Sun, Nov 08, 2009 at 05:28:05PM +0000]
Yeah, as far as I know. I didn't find any sign if we going to
change PAGE_SHIFT.
-- Cyrill

Re: i686 quirk for AMD Geode

2009-11-08T17:35:54Z

Hi!
It is not.
NOP should be fast, so this is bad idea.
Pavel

Re: [PATCH 4/6] hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events

2009-11-08T17:32:30Z

When we originally had this discussion on LKML, one of the use-cases
cited was [link]. I can't see why such
need should be restricted to a given CPU only, rather than a subset of
CPUs (say 'x' is a variable normally read/written-to in the interrupt
path, and if the said interrupt is has a cpu affinity to a subset of

Re: [PATCH 5/7] hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events

2009-11-08T17:31:35Z

So, will the breakpoint values be stored in per-cpu variables and be
restored when a cpu (from the list of for_each_possible_cpu) eventually
turns online due to cpu-hotplug or resume from hibernate (as originally
done by load_debug_registers())?
A few more observations....
int reserve_bp_slot(struct perf_event *bp)