Re: [PATCH v4] powerpc: 64K page support for kexec

From: Haren Myneni <haren@linux.vnet.ibm.com>
To: Luke Browning <lukebr@linux.vnet.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>, Milton Miller <miltonm@bga.com>,
	linuxppc-dev@ozlabs.org, Paul Mackerras <paulus@samba.org>,
	Olof Johansson <olof@lixom.net>,
	cbe-oss-dev@ozlabs.org
Subject: Re: [PATCH v4] powerpc: 64K page support for kexec
Date: Fri, 27 Apr 2007 11:23:27 -0700	[thread overview]
Message-ID: <46323F9F.2050205@linux.vnet.ibm.com> (raw)
In-Reply-To: <1177695045.24866.135.camel@luke-laptop>

Luke Browning wrote:
> On Fri, 2007-04-27 at 11:59 -0500, Milton Miller wrote:
>
>   
>> I see I've achived my nefarious goal of making these bugs never happen.
>>
>> (panic is a function, so you are checking that the staticly linked
>> non-weak function is available.   If you want to check on when
>> if its a panic kdump or not, you need to decode the flag and pass
>> it to kexec_sequence, pass it back here, and update all platforms
>> for the new parameter).
>>
>> But I still say the check while clearing the table is too late,
>> it should be a debug scan before clearing any mappings.
>>     
>
> How about the following as an alternative. 
>
>   BUG_ON((hpte_v & 0x4000000000000000UL) && (crashing_cpus == -1));
>   BUG_ON((size == MMU_PAGE_16G) && (crashing_cpus == -1));
>   BUG_ON((size == MMU_PAGE_64K_AP) && (crashing_cpus == -1));
>   
should be crashing_cpu -  contains the panic cpu ID

> I don't have time to work on a multi-platform solution.
>
> Regards,
> Luke
>
>
>
>
>
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@ozlabs.org
> https://ozlabs.org/mailman/listinfo/linuxppc-dev
>