Re: [PATCH 2/5] x86: PAT: fix ambiguous paranoia check in pat_init()

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Andreas Herrmann <andreas.herrmann3@...>
Cc: Thomas Gleixner <tglx@...>, Ingo Molnar <mingo@...>, H. Peter Anvin <hpa@...>, <linux-kernel@...>, Venkatesh Pallipadi <venkatesh.pallipadi@...>, Suresh B Siddha <suresh.b.siddha@...>
Date: Tuesday, June 10, 2008 - 6:55 pm

On 10-06-08 16:05, Andreas Herrmann wrote:


No, it's correct. This is to be read as "The PAT feature is enabled in 
the kernel but the CPU claims to not have PAT". The message is not a 
leftover from the old flag-clearing setup or anything; it was introduced 
in the patch that disabled that.


Which is okay-ish, or at least by design it seems. The paranoia check in 
pat.c will catch this case.

The current setup is okay really. The only thing it still wants is a 
commandline whitelist switch but that needs a somewhat different setup 
as validate_pat_support() is too early for __setup() or early_param().

Rene.
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[PATCH 2/5] x86: PAT: fix ambiguous paranoia check in pat_in..., Andreas Herrmann, (Tue Jun 10, 10:05 am)
Re: [PATCH 2/5] x86: PAT: fix ambiguous paranoia check in pa..., Rene Herman, (Tue Jun 10, 6:55 pm)
Re: [PATCH 2/5] x86: PAT: fix ambiguous paranoia check in pa..., Andreas Herrmann, (Wed Jun 11, 12:12 pm)
[PATCH] x86: enable PAT on (almost) all CPUs that advertise it, Andreas Herrmann, (Wed Jun 11, 11:05 am)