Re: [PATCH] x86, UV: Fix NMI handler for UV platforms

[Cyrill Gorcunov <gorcunov@gmail.com>]

On 03/21/2011 07:14 PM, Ingo Molnar wrote:
> 
> * Jack Steiner <steiner@sgi.com> wrote:
> 
>> This fixes a problem seen on UV systems handling NMIs from the node controller.
>> The original code used the DIE notifier as the hook to get to the UV NMI
>> handler. This does not work if performance counters are active - the hw_perf
>> code consumes the NMI and the UV handler is not called.
> 
> Sigh:
> 
>> --- linux.orig/arch/x86/kernel/traps.c	2011-03-21 09:05:43.000000000 -0500
>> +++ linux/arch/x86/kernel/traps.c	2011-03-21 09:13:01.306555675 -0500
>> @@ -57,6 +57,7 @@
>>  #include <asm/mce.h>
>>  
>>  #include <asm/mach_traps.h>
>> +#include <asm/uv/uv.h>
>>  
>>  #ifdef CONFIG_X86_64
>>  #include <asm/x86_init.h>
>> @@ -397,13 +398,16 @@ unknown_nmi_error(unsigned char reason,
>>  static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
>>  {
>>  	unsigned char reason = 0;
>> +	int handled;
>>  
>>  	/*
>>  	 * CPU-specific NMI must be processed before non-CPU-specific
>>  	 * NMI, otherwise we may lose it, because the CPU-specific
>>  	 * NMI can not be detected/processed on other CPUs.
>>  	 */
>> -	if (notify_die(DIE_NMI, "nmi", regs, 0, 2, SIGINT) == NOTIFY_STOP)
>> +	handled = uv_handle_nmi(regs, reason);
>> +	if (notify_die(DIE_NMI, "nmi", regs, 0, 2, SIGINT) == NOTIFY_STOP ||
>> +	    		handled)
>>  		return;
> 
> Such code is extremely ugly. Please *reduce* the number of is_uv_system() type 
> of hacks in core x86 code, not increase it!
> 
> Any reason why a higher priority for the UV NMI handler cannot solve the 'perf 
> eats the NMI' problem?
> 
> Thanks,
> 
> 	Ingo

Yeah, Don has made priority system for NMI notifiers. Need to look in.

-- 
    Cyrill