Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: Matt Roper <matthew.d.roper@intel.com>,
	Lucas De Marchi <lucas.demarchi@intel.com>
Subject: Re: [Intel-xe] [PATCH] drm/xe/rtp: Fix displayed whitelisted register range
Date: Mon, 13 Nov 2023 13:15:47 -0800	[thread overview]
Message-ID: <87v8a5fj7g.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <877cmqhbg2.wl-ashutosh.dixit@intel.com>

On Thu, 09 Nov 2023 13:19:09 -0800, Dixit, Ashutosh wrote:
>
> On Thu, 09 Nov 2023 09:07:54 -0800, Ashutosh Dixit wrote:
> >
> > Fix displayed whitelisted register range. For example, with:
> >
> >	XE_RTP_ACTIONS(WHITELIST(0xdafc, RING_FORCE_TO_NONPRIV_RANGE_4))
> >
> > Before we would show:
> >	xe REG[0xdaf0-0xdaff]: allow read access
> >
> > With this patch we show:
> >	xe REG[0xdafc-0xdb0b]: allow read access
>
> Please ignore this patch for now. I am still looking into this.

Please ignore this patch. The previous code is correct according to Bspec
45925 / Bspec 60340.

The confusion was there because there seems to be some change in register
whitelisting in Xe2. Prior to Xe2 it appears the alignment mentioned in
bits 1:0 (of the above Bspec links) was not enforced. E.g. the statement:

	XE_RTP_ACTIONS(WHITELIST(0xdafc, RING_FORCE_TO_NONPRIV_RANGE_4))

whitelisted registers 0xdafc, 0xdb00, 0xdb04, 0xdb08 with a single nonpriv
slot.

This seems to have changed in Xe2, where the above statement whitelists
registers 0xdaf0, 0xdaf4, 0xdaf8, 0xdafc instead, following Bspec.

Therefore in Xe2 we need two slots to whitelist these registers as follows:

	XE_RTP_ACTIONS(WHITELIST(0xdafc, RING_FORCE_TO_NONPRIV_RANGE_1))
	XE_RTP_ACTIONS(WHITELIST(0xdb00, RING_FORCE_TO_NONPRIV_RANGE_4))

So the patch assumes the behavior which is present in generations prior to
Xe2 (but has changed Xe2 onwards, at least as far as I can tell).

In any case, please ignore this patch, let's keep the previous code which
follows Bspec from Xe2 onwards.

Thanks.
--
Ashutosh

      reply	other threads:[~2023-11-13 21:15 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-09 17:07 [Intel-xe] [PATCH] drm/xe/rtp: Fix displayed whitelisted register range Ashutosh Dixit
2023-11-09 18:05 ` [Intel-xe] ✓ CI.Patch_applied: success for " Patchwork
2023-11-09 18:06 ` [Intel-xe] ✓ CI.checkpatch: " Patchwork
2023-11-09 18:07 ` [Intel-xe] ✓ CI.KUnit: " Patchwork
2023-11-09 18:16 ` [Intel-xe] ✓ CI.Build: " Patchwork
2023-11-09 18:16 ` [Intel-xe] ✓ CI.Hooks: " Patchwork
2023-11-09 18:18 ` [Intel-xe] ✓ CI.checksparse: " Patchwork
2023-11-09 21:19 ` [Intel-xe] [PATCH] " Dixit, Ashutosh
2023-11-13 21:15   ` Dixit, Ashutosh [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=87v8a5fj7g.wl-ashutosh.dixit@intel.com \
    --to=ashutosh.dixit@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=lucas.demarchi@intel.com \
    --cc=matthew.d.roper@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