Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
	Michal Wajdeczko <michal.wajdeczko@intel.com>,
	intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 1/2] drm/i915/guc: Limit number of scratch registers used for H2G
Date: Tue, 16 Oct 2018 09:26:54 -0700	[thread overview]
Message-ID: <40463a81-a0a1-2cac-4426-40de56416a35@intel.com> (raw)
In-Reply-To: <153969028853.22931.17968946219589774177@skylake-alporthouse-com>



On 10/16/2018 4:44 AM, Chris Wilson wrote:
> Quoting Michal Wajdeczko (2018-10-16 10:42:06)
>> We wrongly assumed that GuC is only using last scratch register
>> for G2H messages, but in fact it is also using register [14] to
>> report sleep state status. Remove that register from our H2G
>> send registers pool.
> I'll leave it up to you and Daniele to work out if it is being
> exclusively used for this purpose or if it is just part of the response
> for the suspend/resume message.
> -Chris

AFAICS in the GuC FW, the same register is reused for other purposes in 
other actions (that we don't use/need), but always as a GuC to host 
response.
No message from host to guc uses more than 8 registers and the GuC FW 
itself uses an 8-element array to store the H2G message, so we could 
reduce our send array to just 8 regs. The GEM_BUG_ON() in guc_send_reg() 
would avoid troubles if a command requiring more than 8 register was 
added later, but I'd say that's extremely unlikely since the old SKL FW 
is not getting any major updates and the new one uses CT buffers so we 
don't care about the registers.

Daniele
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

      reply	other threads:[~2018-10-16 16:26 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-16  9:42 [PATCH 1/2] drm/i915/guc: Limit number of scratch registers used for H2G Michal Wajdeczko
2018-10-16  9:42 ` [PATCH 2/2] HAX enable GuC for CI Michal Wajdeczko
2018-10-19  7:05   ` [LKP] 262bd453bf: drm:i915_gem_init_hw[i915]] kernel test robot
2018-10-16 10:08 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/2] drm/i915/guc: Limit number of scratch registers used for H2G Patchwork
2018-10-16 10:30 ` ✗ Fi.CI.BAT: failure " Patchwork
2018-10-16 11:44 ` [PATCH 1/2] " Chris Wilson
2018-10-16 16:26   ` Daniele Ceraolo Spurio [this message]

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=40463a81-a0a1-2cac-4426-40de56416a35@intel.com \
    --to=daniele.ceraolospurio@intel.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=michal.wajdeczko@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