Thread (7 messages) flat view 7 messages, 2 authors, 2013-05-28

Re: [PATCH 3/3 v3] net/macb: Try to optimize struct macb layout

From: Nicolas Ferre <hidden>
Date: 2013-05-28 08:08:47
Also in: linux-arm-kernel, lkml

On 15/05/2013 11:37, David Laight :
quoted
From: Havard Skinnemoen <redacted>
quoted
Move the NAPI struct below that since it is used
from the interrupt handler. This field is also marked as
___cacheline_aligned_in_smp.
Is that a good idea if the cache line size is 256 bytes?
Indeed is that ever sane - except for a few special structures
that really need it, and are guaranteed to be allocated
with that much alignment.
I do not know if that was really needed, it was a recommendation by Ben.
(previous email here:
https://patchwork.kernel.org/patch/1887101/
)

Anyway, it seems that this patch needs more work whereas the others of 
the series are not commented.

So maybe, I can resend patches 1, 2 for making this big enhancement go 
forward (as it is in review process for quite a long time...).
David, do you want me to resend or can you retrieve these v3 patches in 
patchwork?

Best regards,
-- 
Nicolas Ferre
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help