Re: [pre-2.6.23 REGRESSION] 2.6.23-rc3-git1 crash/stuck on VIA CN700 system

[Dave Jones <davej@redhat.com>]

On Wed, Sep 05, 2007 at 10:24:07PM +0300, Stefan Becker wrote:

 > #if 0
 > 	if (cpu_has_clflush)
 > 		asm("clflush (%0) " :: "r" (addr) : "memory");
 > #endif
 > }
 > 
 > This kernel boots up OK. Looking at the preprocessed C code the 
 > following code in alternative_instructions() is compiled in:
 > 
 > #ifdef CONFIG_SMP
 > 	if (smp_alt_once) {
 > 		if (1 == num_possible_cpus()) {
 > 			printk(KERN_INFO "SMP alternatives: switching to UP code\n");
 > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ This I still see at bootup
 > 
 > 			set_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
 > 			set_bit(X86_FEATURE_UP, cpu_data[0].x86_capability);
 > 			alternatives_smp_unlock(__smp_locks, __smp_locks_end,
 > 						_text, _etext);
 > 
 > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ this function uses text_poke()
 > ---> BOOM
 > 		}
 > 		free_init_pages("SMP alternatives",
 > 				(unsigned long)__smp_locks,
 > 				(unsigned long)__smp_locks_end);
 > 
 > So what can we do about the clflush on this CPU?

Could be that even though it advertises clflush support there are
errata on some revs of the CPU.  Can you paste your /proc/cpuinfo,
and I'll check with VIA to find out if they're aware of any
known problems and if so, find out the range of steppings we
need to not use clflush on.

	Dave

-- 
http://www.codemonkey.org.uk