Re: [PATCH 00/11] Add compression support to pstore

[Aruna Balakrishnaiah <aruna@linux.vnet.ibm.com>]

On Thursday 08 August 2013 09:38 AM, Aruna Balakrishnaiah wrote:
> Hi Tony,
>
> On Thursday 08 August 2013 03:52 AM, Tony Luck wrote:
>> On Tue, Aug 6, 2013 at 10:35 PM, Tony Luck <tony.luck@gmail.com> wrote:
>>> ERST is at the whim of the BIOS writer (the ACPI standard doesn't provide any
>>> suggestions on record sizes).  My systems support ~6K record size.
>> Off by a little - 7896 bytes on my current machine.
>>
>>> efivars has, IIRC, a 1k limit coded in the Linux back end.
>> My memory was correct for this one.
>>
>> Adding a little tracing to pstore_getrecords() I see this:
>>
>> pstore: inflated 3880 bytes compressed to 17459 bytes
>> pstore: inflated 2567 bytes compressed to 17531 bytes
>> pstore: inflated 4018 bytes compressed to 17488 bytes
>>
>> Which isn't at all what I expected.  The ERST backend
>> advertised a bufsize of 7896, and I have the default
>> kmsg_bytes of 10240.  So on my forced panic the code
>> decided to create a three part pstore dump.  The sum of
>> the pieces is close to, but a little over the target of 10K.
>> But I don't understand why the compressed sizes are so
>> much smaller that the ERST backend block size.
>
> The sizes of compressed text depends on the nature of uncompressed
> data that is captured from kmsg_dump, considering the worst
> case of plain text based on experiments 45% was thecompression achieved.
> So we chose a buffer of size psinfo->bufsize * 100/45.
> If the uncompressed data captured was more of plain text nature then it
> would take up size close to ERST backend block size. Thats the reason
> you see compressed data of 2.5k to 4.0k. 2.5k would have more
> repeated occurrences than 4.0k.
>
> The sum of 3 pstore records should not have exceeded kmsg_bytes.
> Is it after adding total_len in the fix patch? Will take a look at it.

The sum of first two records is less than kmsg_bytes, so it captures the 3rd record.
Only after 3rd record is captured and written total is evaluated against kmsg_bytes
when itexceeds the limit it stops capturing the next one.

This shall happen even without compression right? If total is checked before
write this can be avoided.

- Aruna

>
>> The uncompressed sizes appear to be close to constant.
>> The compression ratios vary from 14% to 23%
>>
>> Why do we get three small parts instead of two bigger
>> ones close the the 7896 ERST bufsize?
>
> Same explanation as given above.
>
>>
>> -Tony
>>
>
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/linuxppc-dev
>