[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230421112355.GAZEJyS+w84odQ9s2i@fat_crate.local>
Date: Fri, 21 Apr 2023 13:23:55 +0200
From: Borislav Petkov <bp@...en8.de>
To: Juergen Gross <jgross@...e.com>
Cc: linux-kernel@...r.kernel.org, x86@...nel.org,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
Dave Hansen <dave.hansen@...ux.intel.com>,
"H. Peter Anvin" <hpa@...or.com>,
Michael Kelley <mikelley@...rosoft.com>
Subject: Re: [PATCH v5 11/15] x86/mtrr: construct a memory map with cache
modes
On Thu, Apr 20, 2023 at 05:10:03PM +0200, Juergen Gross wrote:
> I think this will need another final loop over the MTRRs to check against the
> constructed map if a MTRR is completely useless.
Yeah, I slept on it: so I think there should be a patch ontop which does
add debug output - disabled by default and controllable by adding
"mtrr=debug" on the cmdline or so - which dumps the cache map operations
(add/remove) and the final table.
The reason being, when this cache_map thing hits upstream, we would need
a way to debug any potential issues which people might report so asking
them to do a "mtrr=debug" boot would be a good help.
And pls make the prints pr_info() and not pr_debug() because people
should not have to recompile in order to enable that.
> Another question: in case we detect such a hidden MTRR, should it be disabled
> in order to have more MTRRs available for run-time adding?
Let's not do anything for now and address this if really needed.
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
Powered by blists - more mailing lists