lkml.org 
[lkml]   [2010]   [Aug]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
DateTue, 10 Aug 2010 10:25:49 +0530
FromNitin Gupta <>
SubjectRe: [PATCH 05/10] Reduce per table entry overhead by 4 bytes
On 08/10/2010 12:29 AM, Pekka Enberg wrote:
> On Mon, Aug 9, 2010 at 8:26 PM, Nitin Gupta <ngupta@vflare.org> wrote:
>> Each zram device maintains an array (table) that maps
>> index within the device to the location of corresponding
>> compressed chunk. Currently we store 'struct page' pointer,
>> offset with page and various flags separately which takes
>> 12 bytes per table entry. Now all these are encoded in a
>> single 'phys_add_t' value which results in savings of 4 bytes
>> per entry (except on PAE systems).
>>
>> Unfortunately, cleanups related to some variable renames
>> were mixed in this patch. So, please bear some additional
>> noise.
>
> The noise makes this patch pretty difficult to review properly. Care
> to spilt the patch into two pieces?
>

Ok, I will split them as separate patches.

Thanks for all the reviews and Acks.
Nitin



\
 
 \ /
  Last update: 2010-08-10 06:57    [from the cache]
©2003-2010