Hi,
What is the proper value for L1_CACHE_SHIFT on sparc64?
The cache itself is two chunks of 16 bytes each, makes up 32 bytes. On
i386, L1_CACHE_BYTES == (1 << L1_CACHE_SHIFT), so if this applies here,
too, we have a L1_CACHE_SHIFT of... 5?
Regards,
Thunder
--
(Use http://www.ebb.org/ungeek if you can't decode)
------BEGIN GEEK CODE BLOCK------
Version: 3.12
GCS/E/G/S/AT d- s++:-- a? C++$ ULAVHI++++$ P++$ L++++(+++++)$ E W-$
N--- o? K? w-- O- M V$ PS+ PE- Y- PGP+ t+ 5+ X+ R- !tv b++ DI? !D G
e++++ h* r--- y-
------END GEEK CODE BLOCK------
From: Thunder from the hill <[email protected]>
Date: Fri, 12 Jul 2002 15:58:28 -0600 (MDT)
What is the proper value for L1_CACHE_SHIFT on sparc64?
The cache itself is two chunks of 16 bytes each, makes up 32 bytes. On
i386, L1_CACHE_BYTES == (1 << L1_CACHE_SHIFT), so if this applies here,
too, we have a L1_CACHE_SHIFT of... 5?
Right. But who needs L1_CACHE_SHIFT? Nothing generic should
reference it. Did something get added to 2.5.x that needs it now?
I wouldn't have noticed yet as I've been away for nearly half a month
on vaction until a day or two ago.
Hi,
On Fri, 12 Jul 2002, David S. Miller wrote:
> Right. But who needs L1_CACHE_SHIFT?
fs/dquot.c uses it.
> Nothing generic should reference it. Did something get added to 2.5.x
> that needs it now?
Must have been 2.5, but not since 2.5.24.
> I wouldn't have noticed yet as I've been away for nearly half a month
> on vaction until a day or two ago.
Welcome back!
Regards,
Thunder
--
(Use http://www.ebb.org/ungeek if you can't decode)
------BEGIN GEEK CODE BLOCK------
Version: 3.12
GCS/E/G/S/AT d- s++:-- a? C++$ ULAVHI++++$ P++$ L++++(+++++)$ E W-$
N--- o? K? w-- O- M V$ PS+ PE- Y- PGP+ t+ 5+ X+ R- !tv b++ DI? !D G
e++++ h* r--- y-
------END GEEK CODE BLOCK------