On Mon, Feb 15, 2021 at 09:03:54PM -0800, Vineet Gupta wrote:
> On 2/10/21 3:28 PM, Kees Cook wrote:
> > As started by commit 05a5f51ca566 ("Documentation: Replace lkml.org
> > links with lore"), replace lkml.org links with lore to better use a
> > single source that's more likely to stay available
On Wed, Dec 15, 2021 at 12:41 AM David Laight wrote:
>
> From: Yury Norov
> > Sent: 14 December 2021 19:43
> ...
> >
> > I think that for long bitmaps the most time consuming operation is moving
> > data to L1, and for short bitmaps the difference between approaches is
> > barely measurable.
> >
>
On 12/14/21 3:42 PM, Vladimir Isaev wrote:
> On Dec 14, 2021 3:18 PM, Barbaros Tokaoglu wrote:
>> For 16 kB data cache with 4 ways and 32 byte cache lines we have tag LSB
>> [(128 locations for each way: 7 bits) + byte offset (32 bytes: 5 bits)]
>> lower than page offset (13 bits for 8 kB pages).
For 16 kB data cache with 4 ways and 32 byte cache lines we have tag LSB
[(128 locations for each way: 7 bits) + byte offset (32 bytes: 5 bits)]
lower than page offset (13 bits for 8 kB pages).
So it is possible for PTAG to change even for single page loop. Hence
update ARC_REG_*C_PTAG register on
From: Yury Norov
> Sent: 14 December 2021 19:43
...
>
> I think that for long bitmaps the most time consuming operation is moving
> data to L1, and for short bitmaps the difference between approaches is
> barely measurable.
>
> But hweght_long on each iteration can't be more effective than the cu
