Longhaul - Use hardware support

Hardware is stopping bus masters. There is no need to implement this in Linux kernel. I tested it really hard. Even doing md5sum on my files before and after. No corruption. 2 days without even single lockup. No "dma broken" messages. My system: Epia M10000 (CLE266 chipset) with VIA Nehemiah 1GHz.

Changes in Kconfig:
1. ACPI_PROCESSOR - this code need ACPI C3 state working,
2. !X86_UP_APIC - (un)patched "longhaul" isn't APIC compatible,
3. !SMP - (un)patched "longhaul" isn't SMP compatible. First step would be APIC support. Second - per CPU private data.

Changes in longhaul.h:
- clock tables merged - differences are little, "longhaul" will do errata for detected CPU,
- eblcr tables merged - reason above,
- format of voltage tables changed,
- added mobile vrm.

Changes in longhaul.c:
- most important - now C3 state is causing transition,
- code resopnsible for clearing "bus master" bit removed,
- protect bcr2 transition in the same way as longhaul,
- check if Longhaul MSR is present, don't assume that this stepping have this MSR, and other don't,
- voltage scaling added. By default disabled because is untested. Looks like processors on Epia mainboards don't support voltage scaling,
- some FSB scaling compatibility,
- some minor changes.

---

diff -uprN -X linux-2.6.17-git20-vanilla/Documentation/dontdiff linux-2.6.17-git20-vanilla/arch/i386/kernel/cpu/cpufreq/Kconfig linux-2.6.17-git20/arch/i386/kernel/cpu/cpufreq/Kconfig
@@ -202,7 +202,7 @@ config X86_LONGRUN
 config X86_LONGHAUL
 	tristate "VIA Cyrix III Longhaul"
 	select CPU_FREQ_TABLE
-	depends on BROKEN
+	depends on !SMP && !X86_UP_APIC && ACPI_PROCESSOR
 	help
 	  This adds the CPUFreq driver for VIA Samuel/CyrixIII,
 	  VIA Cyrix Samuel/C3, VIA Cyrix Ezra and VIA Cyrix Ezra-T

------------------------------------------------------------------------
Interaktywne mapy miast na http://map24.interia.pl

Please split this up into easy-to-review single-change-per-diff patchset. This way if it becomes easier to review, and also if there are any objectionable parts, I can drop those whilst still applying the non-contentious parts.

Thanks,

Dave

http://www.codemonkey.org.uk