Re: [PATCH v3] powerpc/powernv Platform dump interface

[Vasant Hegde <hegdevasant@linux.vnet.ibm.com>]

On 03/05/2014 07:25 AM, Stewart Smith wrote:
> Vasant Hegde <hegdevasant@linux.vnet.ibm.com> writes:
>>> index 0000000..32fe7f5
>>> --- /dev/null
>>> +++ b/Documentation/ABI/stable/sysfs-firmware-opal-dump
>>> @@ -0,0 +1,41 @@
>>> +What:		/sys/firmware/opal/dump
>>> +Date:		Feb 2014
>>> +Contact:	Stewart Smith <stewart@linux.vnet.ibm.com>
>>> +Description:
>>> +		This directory exposes interfaces for interacting with
>>> +		the FSP and platform dumps through OPAL firmware interface.
>>> +
>>> +		This is only for the powerpc/powernv platform.
>>> +
>>> +		initiate_dump:	When '1' is written to it,
>>> +				we will initiate a dump.
>>
>> initiate FSP dump
>
> ack. I can send subsequent patch that fixes up these bits.
>
>>> --- /dev/null
>>> +++ b/arch/powerpc/platforms/powernv/opal-dump.c
>>> @@ -0,0 +1,525 @@
>>> +/*
>>> + * PowerNV OPAL Dump Interface
>>> + *
>>> + * Copyright 2013,2014 IBM Corp.
>>> + *
>>> + * This program is free software; you can redistribute it and/or
>>> + * modify it under the terms of the GNU General Public License
>>> + * as published by the Free Software Foundation; either version
>>> + * 2 of the License, or (at your option) any later version.
>>> + */
>>> +
>>> +#include <linux/kobject.h>
>>> +#include <linux/mm.h>
>>> +#include <linux/slab.h>
>>> +#include <linux/vmalloc.h>
>>> +#include <linux/pagemap.h>
>>> +#include <linux/delay.h>
>>> +
>>> +#include <asm/opal.h>
>>> +
>>> +#define DUMP_TYPE_FSP	0x01
>>
>> Better define other dump type (sysdump etc) here and use it below ?
>
> Likely yes, Is there a definitive list somewhere?

List is depending on what OPAL gives to Linux...For now

#define DUMP_TYPE_FSP	0x01
#define DUMP_TYPE_SYS   0x02
#define DUMP_TYPE_SMA   0x03


>
>>> +static ssize_t init_dump_store(struct dump_obj *dump_obj,
>>> +			       struct dump_attribute *attr,
>>> +			       const char *buf,
>>> +			       size_t count)
>>> +{
>>> +	dump_fips_init(DUMP_TYPE_FSP);
>>> +	pr_info("%s: Initiated FSP dump\n", __func__);
>>
>> This might mislead if OPAL fails to initiate FSP dump.. Better move this to
>> dump_fips_init () ?
>
> good point. I don't think we actually process the response of FSP dump
> init, which IIRC may actually be able to tell us if it failed to
> initiate a dump.
>

Yes..

-Vasant