From: Daniel Vetter <daniel@ffwll.ch>
To: Ben Widawsky <ben@bwidawsk.net>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 06/10] drm/i915: Stop using AGP layer for GEN6+
Date: Tue, 23 Oct 2012 22:03:21 +0200 [thread overview]
Message-ID: <CAKMK7uHf2J-OaWLFW0s=7vqQkWw5s8zaT8LTO9H0+Y4KY1K0nA@mail.gmail.com> (raw)
In-Reply-To: <20121023125724.16b00131@bwidawsk.net>
On Tue, Oct 23, 2012 at 9:57 PM, Ben Widawsky <ben@bwidawsk.net> wrote:
>> On Tue, Oct 23, 2012 at 4:57 PM, Ben Widawsky <ben@bwidawsk.net> wrote:
>> > Actually, after we introduce the FLSH_CNTL patch from Jesse/me later in the
>> > series, I think we just want a POSTING_READ on that register. It is
>> > technically "required" by our desire to some day WC the registers, and
>> > should synchronize everything else for us.
>> >
>> > After a quick read of memory_barriers.txt (again), I think mmiowb is
>> > actually what we might want in addition to the POSTING_READ I'd add.
>>
>> Imo we have special rules for the igd, since clearly not all registers
>> in our 4mb mmio window are equal. So I'd prefer the keep the readback
>> of the last pte write (to ensure those are flushed out) and maybe also
>> add a readback of the gfx_flsh_cntl reg (like I've seen in some
>> internal vlv tree). Just to be paranoid.
>> -Daniel
>
> What's your definition of flush? I think we just need one read to
> satisfy the device I/O flush, and I think the write to the regsiter
> should satisy the TLB flush. That leads me to the conclusion that just
> the POSTING_READ of the FLSH_CNTL register is sufficient.
>
> If you want me to do both, I have no problem with that either, and I'll
> just update the comment to say that we believe it is unnecessary.
>
> I don't really care either way.
I think we need both, otherwise one of the pte writes might not have
arrived at the endpoint when the chip is purging the tlb already,
which might give is a tiny window where we load an invalid entry into
the tlb. Tiny window, but I'm paranoid ;-) And yes, I know that's more
strict than what docs say, but our igd also seems to be less coherent
than what docs claim.
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
next prev parent reply other threads:[~2012-10-23 20:03 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-23 1:34 [PATCH 00/10] Kill AGP dependencies for Gen6+ Ben Widawsky
2012-10-23 1:34 ` [PATCH 01/10] drm/i915: No LLC_MLC for HSW Ben Widawsky
2012-10-23 10:15 ` Mika Kuoppala
2012-10-25 20:47 ` Jesse Barnes
2012-10-26 1:03 ` Ben Widawsky
2012-10-23 1:34 ` [PATCH 02/10] drm/i915: Add dev to ppgtt Ben Widawsky
2012-10-25 20:48 ` Jesse Barnes
2012-10-23 1:34 ` [PATCH 03/10] drm/i915: introduce gtt_pte_t Ben Widawsky
2012-10-25 20:49 ` Jesse Barnes
2012-10-23 1:34 ` [PATCH 04/10] drm/i915: Extract PPGTT pte encoding Ben Widawsky
2012-10-25 20:50 ` Jesse Barnes
2012-10-23 1:34 ` [PATCH 05/10] drm/i915: move more pte encoding to pte encode Ben Widawsky
2012-10-25 20:51 ` Jesse Barnes
2012-10-31 10:52 ` Daniel Vetter
2012-10-23 1:34 ` [PATCH 06/10] drm/i915: Stop using AGP layer for GEN6+ Ben Widawsky
2012-10-23 9:59 ` Chris Wilson
2012-10-23 14:57 ` Ben Widawsky
2012-10-23 14:58 ` Daniel Vetter
2012-10-23 19:57 ` Ben Widawsky
2012-10-23 20:03 ` Daniel Vetter [this message]
2012-10-23 20:27 ` Ben Widawsky
2012-10-25 20:55 ` Jesse Barnes
2012-10-25 20:54 ` Jesse Barnes
[not found] ` <878vaxwk61.fsf@gaia.fi.intel.com>
2012-10-24 17:56 ` Ben Widawsky
2012-10-26 20:54 ` [PATCH 06/10 v2] " Ben Widawsky
2012-10-29 2:08 ` [PATCH 6/9 v3] " Ben Widawsky
2012-10-31 9:57 ` Daniel Vetter
2012-10-31 15:51 ` Ben Widawsky
2012-10-31 10:37 ` [PATCH 06/10 v2] " Chris Wilson
2012-10-23 1:34 ` [PATCH 07/10] drm/i915: Calculate correct stolen size for GEN7+ Ben Widawsky
2012-10-25 21:06 ` Jesse Barnes
2012-10-25 22:15 ` Ben Widawsky
2012-10-25 22:39 ` Jesse Barnes
2012-10-23 1:34 ` [PATCH 08/10] drm/i915: Kill off now unused gen6+ AGP code Ben Widawsky
2012-10-25 21:07 ` Jesse Barnes
2012-10-23 1:34 ` [PATCH 09/10] drm/i915: flush system agent TLBs on SNB Ben Widawsky
2012-10-23 14:17 ` Jesse Barnes
2012-10-23 14:28 ` Daniel Vetter
2012-10-23 14:35 ` Jesse Barnes
2012-10-23 14:42 ` Ben Widawsky
2012-10-25 21:07 ` Jesse Barnes
2012-10-26 20:55 ` [PATCH 09/10 v2] " Ben Widawsky
2012-10-23 1:34 ` [PATCH 10/10] drm/i915: Kill off actually requiring AGP Ben Widawsky
2012-10-23 2:03 ` Ben Widawsky
2012-10-25 21:09 ` Jesse Barnes
2012-10-25 22:15 ` Ben Widawsky
2012-10-25 22:39 ` Jesse Barnes
2012-10-25 22:47 ` Daniel Vetter
2012-10-26 20:52 ` Ben Widawsky
2012-10-26 20:58 ` [PATCH 00/10] Kill AGP dependencies for Gen6+ Ben Widawsky
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='CAKMK7uHf2J-OaWLFW0s=7vqQkWw5s8zaT8LTO9H0+Y4KY1K0nA@mail.gmail.com' \
--to=daniel@ffwll.ch \
--cc=ben@bwidawsk.net \
--cc=intel-gfx@lists.freedesktop.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.