Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Andi Shyti <andi.shyti@linux.intel.com>
To: Nirmoy Das <nirmoy.das@linux.intel.com>
Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com>,
	Andi Shyti <andi.shyti@linux.intel.com>,
	intel-gfx <intel-gfx@lists.freedesktop.org>,
	dri-devel <dri-devel@lists.freedesktop.org>,
	Andi Shyti <andi.shyti@kernel.org>,
	Andrzej Hajda <andrzej.hajda@intel.com>,
	Chris Wilson <chris.p.wilson@linux.intel.com>,
	Michal Mrozek <michal.mrozek@intel.com>,
	Nirmoy Das <nirmoy.das@intel.com>,
	stable@vger.kernel.org
Subject: Re: [PATCH] drm/i915/gt: Report full vm address range
Date: Fri, 15 Mar 2024 18:08:51 +0100	[thread overview]
Message-ID: <ZfSAo791UDRnBSwc@ashyti-mobl2.lan> (raw)
In-Reply-To: <35df0767-384f-49f2-806a-f83765ca7c4c@linux.intel.com>

Hi Nirmoy,

> > In Mesa we've been relying on I915_CONTEXT_PARAM_GTT_SIZE so as long as
> > that is adjusted by the kernel
> 
> What do you mean by adjusted by, should it be a aligned size?
> 
> I915_CONTEXT_PARAM_GTT_SIZE ioctl is returning vm->total which is
> adjusted(reduced by a page).
> 
> This patch might cause silent error as it is not removing WABB which is
> using the reserved page to add dummy blt and if userspace is using that
> 
> page then it will be overwritten.

yes, I think this could happen, but there is no solution,
unfortunately. We need to fail at some point.

On the other hand, I think mesa is miscalculating the vm size. In
userspace the total size is derived by the bit size
(maxNBitValue()).

By doing so, I guess there will always be cases of
miscalculation.

There are two solutions here:

 1. we track two sizes, one the true available size and one the
    total size. But this looks like a dirty hack to me.
 2. UMD fixes the size calculation by taking for granted what the
    driver provides and we don't have anything to do in KMD.

Lionel, Michal, thoughts?

Andi

  reply	other threads:[~2024-03-15 17:09 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-13 19:39 [PATCH] drm/i915/gt: Report full vm address range Andi Shyti
2024-03-14  2:40 ` ✗ Fi.CI.BAT: failure for " Patchwork
2024-03-14  5:21 ` [PATCH] " Mrozek, Michal
2024-03-14 14:04 ` Lionel Landwerlin
2024-03-14 16:05   ` Nirmoy Das
2024-03-15 17:08     ` Andi Shyti [this message]
2024-03-18  5:21       ` Mrozek, Michal
2024-03-20 18:30         ` Andi Shyti
2024-03-21  4:29           ` Mrozek, Michal
2024-03-14 20:13 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gt: Report full vm address range (rev2) Patchwork
2024-03-14 20:33 ` ✗ Fi.CI.BAT: failure " Patchwork

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=ZfSAo791UDRnBSwc@ashyti-mobl2.lan \
    --to=andi.shyti@linux.intel.com \
    --cc=andi.shyti@kernel.org \
    --cc=andrzej.hajda@intel.com \
    --cc=chris.p.wilson@linux.intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=lionel.g.landwerlin@intel.com \
    --cc=michal.mrozek@intel.com \
    --cc=nirmoy.das@intel.com \
    --cc=nirmoy.das@linux.intel.com \
    --cc=stable@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox