From: Ingo Molnar <mingo@elte.hu>
To: Andi Kleen <ak@suse.de>
Cc: linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>,
Venki Pallipadi <venkatesh.pallipadi@intel.com>,
suresh.b.siddha@intel.com, Arjan van de Ven <arjan@infradead.org>,
Dave Jones <davej@redhat.com>
Subject: Re: CPA patchset
Date: Thu, 10 Jan 2008 11:57:26 +0100 [thread overview]
Message-ID: <20080110105726.GD28209@elte.hu> (raw)
In-Reply-To: <20080110100712.GO25945@bingen.suse.de>
* Andi Kleen <ak@suse.de> wrote:
> > > > WBINVD isnt particular fast (takes a few msecs), but why is
> > > > that a problem? Drivers dont do high-frequency ioremap-ing.
> > > > It's typically only done at driver/device startup and that's
> > > > it.
> > >
> > > Actually graphics drivers can do higher frequency allocation of WC
> > > memory (with PAT) support.
> >
> > but that's not too smart: why dont they use WB plus cflush instead?
>
> Because they need to access it WC for performance.
I think you have it fundamentally backwards: the best for performance is
WB + cflush. What would WC offer for performance that cflush cannot do?
also, it's irrelevant to change_page_attr() call frequency. Just map in
everything from the card and use it. In graphics, if you remap anything
on the fly and it's not a slowpath you've lost the performance game even
before you began it.
Ingo
next prev parent reply other threads:[~2008-01-10 10:57 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-03 15:24 [PATCH CPA] [1/28] Shrink __PAGE_KERNEL/__PAGE_KERNEL_EXEC on non PAE kernels Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [2/28] CPA: Do a simple self test at boot Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [3/28] Add pte accessors for the global bit Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [4/28] Add pte_clrhuge on i386 Andi Kleen
2008-01-05 6:48 ` Jeremy Fitzhardinge
2008-01-03 15:24 ` [PATCH CPA] [5/28] Add pte_pgprot " Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [6/28] CPA: Undo white space changes Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [7/28] Extract page table dumping code from i386 fault handler into dump_pagetable() Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [8/28] CPA: Return the page table level in lookup_address() Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [9/28] CPA: Add simple self test at boot Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [10/28] CPA: Change kernel_map_pages to not use c_p_a() Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [11/28] CPA: Change 32bit back to init_mm semaphore locking Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [12/28] CPA: CLFLUSH support in change_page_attr() Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [13/28] CPA: Use macros to modify the PG_arch_1 page flags in change_page_attr Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [14/28] CPA: Use page granuality TLB flushing " Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [15/28] CPA: Don't flush the caches when the CPU supports self-snoop Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [16/28] CPA: Use wbinvd() macro instead of inline assembly in 64bit c_p_a() Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [17/28] CPA: Reorder TLB / cache flushes to follow Intel recommendation Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [18/28] CPA: Make change_page_attr() more robust against use of PAT bits Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [19/28] CPA: Limit cache flushing to pages that really change caching Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [20/28] CPA: Fix inaccurate comments in 64bit change_page_attr() Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [21/28] CPA: Dump pagetable when inconsistency is detected Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [22/28] CPA: Only queue actually unused page table pages for freeing Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [23/28] CPA: Remove unnecessary masking of address Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [24/28] CPA: Only unmap kernel init pages in text mapping when CONFIG_DEBUG_RODATA is set Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [25/28] CPA: Always do full TLB flush when splitting large pages Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [26/28] CPA: Fix reference counting when changing already changed pages Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [27/28] CPA: Change comments of external interfaces to kerneldoc format Andi Kleen
2008-01-03 15:24 ` [PATCH CPA] [28/28] Make kernel_text test match boot mapping initialization Andi Kleen
2008-01-10 9:31 ` CPA patchset Ingo Molnar
2008-01-10 9:44 ` Dave Airlie
2008-01-10 9:55 ` Andi Kleen
2008-01-10 10:20 ` Dave Airlie
2008-01-10 10:50 ` Andi Kleen
2008-01-10 10:02 ` Ingo Molnar
2008-01-10 9:53 ` Andi Kleen
2008-01-10 10:04 ` Ingo Molnar
2008-01-10 10:07 ` Andi Kleen
2008-01-10 10:57 ` Ingo Molnar [this message]
2008-01-10 11:12 ` Andi Kleen
2008-01-11 7:19 ` Ingo Molnar
2008-01-11 7:33 ` Ingo Molnar
2008-01-11 11:28 ` Andi Kleen
2008-01-11 11:26 ` Andi Kleen
2008-01-11 17:02 ` dean gaudet
2008-01-11 17:18 ` Arjan van de Ven
2008-01-11 17:56 ` dean gaudet
2008-01-10 10:43 ` Ingo Molnar
2008-01-10 11:07 ` Andi Kleen
2008-01-10 12:22 ` Ingo Molnar
2008-01-10 12:39 ` Andi Kleen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20080110105726.GD28209@elte.hu \
--to=mingo@elte.hu \
--cc=ak@suse.de \
--cc=arjan@infradead.org \
--cc=davej@redhat.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=suresh.b.siddha@intel.com \
--cc=tglx@linutronix.de \
--cc=venkatesh.pallipadi@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox