[PATCH v2 07/11] powerpc/8xx: macro for handling CPU15 errata

leroy christophe christophe.leroy at c-s.fr
Tue Jan 20 22:32:39 AEDT 2015


Le 20/01/2015 12:09, David Laight a écrit :
>  From Christophe Leroy
>> Having a macro will help keep clear code.
> It might remove an #if but it doesn't really help.
> All it means is that anyone reading the code has to hunt for
> the definition before proceeding.
>
> Some comment about what (and why) the extra code is needed
> might help.
The main reason is because of patch 09/11 where we have to duplicate 
this code. I prefer to just duplicate one line rather than duplicate the 
whole code (especially because in v1 of the PATCHset, it was duplicated 
twice):

-    DO_8xx_CPU15(r11, r10)
[...]
#ifdef CONFIG_MODULES
[...]
+    DO_8xx_CPU15(r10, r11)
[...]
+#else
+    mfspr    r10, SPRN_SRR0    /* Get effective address of fault */
+    DO_8xx_CPU15(r11, r10)

Is this approach wrong ?


>
> ...
>> +
>> +#ifdef CONFIG_8xx_CPU15
>> +#define DO_8xx_CPU15(tmp, addr)	\
>> +	addi	tmp, addr, PAGE_SIZE;	\
>> +	tlbie	tmp;			\
>> +	addi	tmp, addr, PAGE_SIZE;	\
> You've even transcribed this incorrectly.
Oops
>
> Clearly not tested :-)
Indeed it's been tested, but tests can only show that the code is not 
worth than before.
This code is there to fix a chip errata which (almost?) never happens.
In my production version, I have not activated this errata, and he have 
never seen the problem on any of the more than 200 boards that have run 
for at least 4 years.

Christophe
>
> 	David
>
>> +	tlbie	tmp
>> +#else
>> +#define DO_8xx_CPU15(tmp, addr)
>> +#endif
>> +
>>   InstructionTLBMiss:
>>   #ifdef CONFIG_8xx_CPU6
>>   	mtspr	SPRN_DAR, r3
>> @@ -304,12 +315,7 @@ InstructionTLBMiss:
>>   	EXCEPTION_PROLOG_0
>>   	mtspr	SPRN_SPRG_SCRATCH2, r10
>>   	mfspr	r10, SPRN_SRR0	/* Get effective address of fault */
>> -#ifdef CONFIG_8xx_CPU15
>> -	addi	r11, r10, PAGE_SIZE
>> -	tlbie	r11
>> -	addi	r11, r10, -PAGE_SIZE
>> -	tlbie	r11
>> -#endif
>> +	DO_8xx_CPU15(r11, r10)
>>
>>   	/* If we are faulting a kernel address, we have to use the
>>   	 * kernel page tables.
>> --
>> 2.1.0
>>
>> _______________________________________________
>> Linuxppc-dev mailing list
>> Linuxppc-dev at lists.ozlabs.org
>> https://lists.ozlabs.org/listinfo/linuxppc-dev



More information about the Linuxppc-dev mailing list