Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 563A2CD98CE for ; Thu, 11 Jun 2026 14:09:52 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 0399310EF50; Thu, 11 Jun 2026 14:09:52 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="hud9iiyz"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.14]) by gabe.freedesktop.org (Postfix) with ESMTPS id 48BBD10EF50 for ; Thu, 11 Jun 2026 14:09:50 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1781186990; x=1812722990; h=message-id:subject:from:to:cc:date:in-reply-to: references:content-transfer-encoding:mime-version; bh=6IsuKw0bBOrtcyCSs93WMhL19+37CQE3kaLggWc0u3Y=; b=hud9iiyzU7dXJ7S74c/xfhBDEjjjVMX9DUEj5wJs7kOxaF429tcQ4Xud PNTo1qiuKLfO3bcp/zDRtYwJzrFqiLdId6uy11sn6SyApoG7CNWDTbqpg w4UyGVF2CBrlvXL3oq3gs2lGo9+hVVsVkhwpMDMp7QhTOi+y31MVSRZD6 kSmbCFjzHeE3lrsPgZAwSeesMMGiOusF0TbM8xZarzFKLkyGnXVeOitz4 uj7tg3+rVGOTJhjONle70w+PLICcBP8XGfUndQPfnupN6zEDIWVKjcv1b eDXekjakrLc36Os0pmhu03GTUQP3MSmKsOQEm+wm/1+/G8p+6uepLf5w6 A==; X-CSE-ConnectionGUID: htsXKNWASTiHnbtDpvTVNg== X-CSE-MsgGUID: PN4yYfofRwmR54iVMTP4ww== X-IronPort-AV: E=McAfee;i="6800,10657,11813"; a="82045355" X-IronPort-AV: E=Sophos;i="6.24,199,1774335600"; d="scan'208";a="82045355" Received: from fmviesa005.fm.intel.com ([10.60.135.145]) by fmvoesa108.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jun 2026 07:09:50 -0700 X-CSE-ConnectionGUID: LBhT/5HqStu0nO8ad/Beog== X-CSE-MsgGUID: mn/++YxQTV++xM8IpKhgLw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.24,199,1774335600"; d="scan'208";a="251581604" Received: from amilburn-desk.amilburn-desk (HELO [10.245.244.169]) ([10.245.244.169]) by fmviesa005-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Jun 2026 07:09:49 -0700 Message-ID: <43c31bca8f6f8fded74c5610aa20059958e74394.camel@linux.intel.com> Subject: Re: [RFC PATCH 1/2] drm/gpusvm: Add a DMA-mapping accounting callback From: Thomas =?ISO-8859-1?Q?Hellstr=F6m?= To: Matthew Brost Cc: intel-xe@lists.freedesktop.org Date: Thu, 11 Jun 2026 16:09:46 +0200 In-Reply-To: References: <20260610133451.8930-1-thomas.hellstrom@linux.intel.com> Organization: Intel Sweden AB, Registration Number: 556189-6027 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable User-Agent: Evolution 3.58.3 (3.58.3-1.fc43) MIME-Version: 1.0 X-BeenThere: intel-xe@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel Xe graphics driver List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-xe-bounces@lists.freedesktop.org Sender: "Intel-xe" On Thu, 2026-06-11 at 06:53 -0700, Matthew Brost wrote: > On Thu, Jun 11, 2026 at 09:42:01AM +0200, Thomas Hellstr=C3=B6m wrote: > > On Wed, 2026-06-10 at 08:37 -0700, Matthew Brost wrote: > > > On Wed, Jun 10, 2026 at 03:34:50PM +0200, Thomas Hellstr=C3=B6m wrote= : > > > > Drivers that use the GPU SVM core to DMA-map system memory for > > > > GPU > > > > access may want to keep track of how many pages they have > > > > mapped, > > > > for > > > > example to expose statistics or to detect leaks. Doing this > > > > accounting > > > > in the driver around drm_gpusvm_get_pages() / > > > > drm_gpusvm_unmap_pages() > > > > is error prone: the number, order and kind of the dma_addr[] > > > > entries can > > > > change between map and unmap (migration, partial unmaps, the > > > > iova > > > > vs > > > > non-iova paths), which makes symmetric accounting hard to get > > > > right. > > > >=20 > > > > Add an optional @dma_map_account callback to struct > > > > drm_gpusvm_ops > > > > and > > > > invoke it once per &drm_pagemap_addr entry at the exact points > > > > where the > > > > entry is DMA-mapped (sign =3D=3D +1) in drm_gpusvm_get_pages() and > > > > unmapped > > > > (sign =3D=3D -1) in __drm_gpusvm_unmap_pages(). Since the map error > > > > path and > > > > every unmap/free path funnel through __drm_gpusvm_unmap_pages() > > > > for > > > > the > > > > entries that were actually mapped, the accounting is symmetric > > > > by > > > > construction. The callback receives the full &drm_pagemap_addr > > > > so > > > > the > > > > driver can use the order and the proto field to distinguish > > > > system > > > > memory mappings from device interconnect mappings. > > > >=20 > > > > The callback must also be usable by the core drm_gpusvm_pages- > > > > only > > > > API > > > > (mm =3D=3D NULL), as used e.g. for userptr-style mappings. Relax > > > > drm_gpusvm_init() to allow such users to pass a restricted @ops > > > > that > > > > only implements the page-level hooks; the full-SVM hooks > > > > (@invalidate) > > > > and the chunk configuration must still be unset in that mode. > > > >=20 > > > > Assisted-by: GitHub_Copilot:claude-opus-4.8 > > > > Signed-off-by: Thomas Hellstr=C3=B6m > > > > > > > > --- > > > > =C2=A0drivers/gpu/drm/drm_gpusvm.c | 17 +++++++++++++++-- > > >=20 > > > I believe this makes sense and is useful for debugging/profiling > > > =E2=80=94 > > > the > > > same applies to the next patch. > > >=20 > > > I do have a question, though: drm_pagemap also performs DMA > > > mapping > > > for > > > migrations. Should we include those in the accounting as well, or > > > was > > > there a reason this was intentionally omitted? > >=20 > > SVM migration was initially intentionally omitted, because the > > primary > > intention here is to debug UMD size-aligning's influence on the > > number > > of 2M pages we get. > >=20 > > But for completeness before we merge this we might want to add the > > transient migration entries as well. > >=20 >=20 > This makes sense. However, one benefit of including transient DMA > mappings in the stats is that if we leak a DMA mapping, a warning > will > appear when the driver is unloaded. >=20 > Either option works in my opinion, but if transient DMA mappings are > omitted, please mention this in the commit message letter. >=20 > So for this version /w an updated commit message: > Revieved-by: Matthew Brost =20 > > We might also want to add similar stats for the PPGTT PTE sizes. > >=20 >=20 > Yes, that is probably a good idea but PPGTT is per-user/VM, so unsure > if > global stats make sense there. IIRC at one point we had a patch to > dump > VM's via debugfs, we could look at reviving that. >=20 > Matt > =C2=A0 > > Thanks, > > Thomas > >=20 > >=20 > >=20 > > >=20 > > > Matt > > >=20 > > > > =C2=A0include/drm/drm_gpusvm.h=C2=A0=C2=A0=C2=A0=C2=A0 | 19 +++++++= ++++++++++++ > > > > =C2=A02 files changed, 34 insertions(+), 2 deletions(-) > > > >=20 > > > > diff --git a/drivers/gpu/drm/drm_gpusvm.c > > > > b/drivers/gpu/drm/drm_gpusvm.c > > > > index 958cb605aedd..c18f15c77e94 100644 > > > > --- a/drivers/gpu/drm/drm_gpusvm.c > > > > +++ b/drivers/gpu/drm/drm_gpusvm.c > > > > @@ -393,8 +393,15 @@ int drm_gpusvm_init(struct drm_gpusvm > > > > *gpusvm, > > > > =C2=A0 return -EINVAL; > > > > =C2=A0 mmgrab(mm); > > > > =C2=A0 } else { > > > > - /* No full SVM mode, only core > > > > drm_gpusvm_pages > > > > API. */ > > > > - if (ops || num_chunks || mm_range || > > > > notifier_size) > > > > + /* > > > > + * No full SVM mode, only core > > > > drm_gpusvm_pages > > > > API. A > > > > + * restricted @ops that only implements the > > > > page- > > > > level hooks > > > > + * (e.g. @dma_map_account) is allowed; the > > > > full- > > > > SVM hooks must > > > > + * not be set. > > > > + */ > > > > + if (num_chunks || mm_range || notifier_size) > > > > + return -EINVAL; > > > > + if (ops && ops->invalidate) > > > > =C2=A0 return -EINVAL; > > > > =C2=A0 } > > > > =C2=A0 > > > > @@ -1162,6 +1169,8 @@ static void > > > > __drm_gpusvm_unmap_pages(struct > > > > drm_gpusvm *gpusvm, > > > > =C2=A0 else if (dpagemap && dpagemap->ops- > > > > > device_unmap) > > > > =C2=A0 dpagemap->ops- > > > > > device_unmap(dpagemap, > > > > =C2=A0 =C2=A0=C2=A0=C2=A0 > > > > dev, > > > > addr); > > > > + if (gpusvm->ops && gpusvm->ops- > > > > > dma_map_account) > > > > + gpusvm->ops- > > > > > dma_map_account(gpusvm, addr, -1); > > > > =C2=A0 i +=3D 1 << addr->order; > > > > =C2=A0 } > > > > =C2=A0 > > > > @@ -1584,6 +1593,10 @@ int drm_gpusvm_get_pages(struct > > > > drm_gpusvm > > > > *gpusvm, > > > > =C2=A0 (addr, > > > > DRM_INTERCONNECT_SYSTEM, > > > > order, > > > > =C2=A0 dma_dir); > > > > =C2=A0 } > > > > + if (gpusvm->ops && gpusvm->ops- > > > > >dma_map_account) > > > > + gpusvm->ops->dma_map_account(gpusvm, > > > > + =C2=A0=C2=A0=C2=A0=C2=A0 > > > > &svm_pages- > > > > > dma_addr[j], > > > > + =C2=A0=C2=A0=C2=A0=C2=A0 1); > > > > =C2=A0 i +=3D 1 << order; > > > > =C2=A0 num_dma_mapped =3D i; > > > > =C2=A0 flags.has_dma_mapping =3D true; > > > > diff --git a/include/drm/drm_gpusvm.h > > > > b/include/drm/drm_gpusvm.h > > > > index 8a4d7134a9a7..87ad2dcaa7c4 100644 > > > > --- a/include/drm/drm_gpusvm.h > > > > +++ b/include/drm/drm_gpusvm.h > > > > @@ -75,6 +75,25 @@ struct drm_gpusvm_ops { > > > > =C2=A0 void (*invalidate)(struct drm_gpusvm *gpusvm, > > > > =C2=A0 =C2=A0=C2=A0 struct drm_gpusvm_notifier > > > > *notifier, > > > > =C2=A0 =C2=A0=C2=A0 const struct mmu_notifier_range > > > > *mmu_range); > > > > + > > > > + /** > > > > + * @dma_map_account: Account a DMA-mapping change > > > > (optional) > > > > + * @gpusvm: Pointer to the GPU SVM > > > > + * @addr: The address descriptor of the chunk being > > > > (un)mapped > > > > + * @sign: +1 when @addr has just been DMA-mapped, -1 > > > > when > > > > it is being > > > > + *=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 unmapped > > > > + * > > > > + * Called once per &drm_pagemap_addr entry, when the > > > > entry > > > > is > > > > + * DMA-mapped by drm_gpusvm_get_pages() (@sign =3D=3D +1) > > > > and > > > > when it is > > > > + * unmapped (@sign =3D=3D -1), so the driver can keep > > > > symmetric accounting > > > > + * of the pages it has mapped for GPU access. The > > > > @addr- > > > > > proto field > > > > + * can be used to distinguish system-memory mappings > > > > from > > > > device > > > > + * interconnect mappings. May be provided in both full > > > > SVM > > > > mode and the > > > > + * core drm_gpusvm_pages-only mode. > > > > + */ > > > > + void (*dma_map_account)(struct drm_gpusvm *gpusvm, > > > > + const struct drm_pagemap_addr > > > > *addr, > > > > + int sign); > > > > =C2=A0}; > > > > =C2=A0 > > > > =C2=A0/** > > > > --=20 > > > > 2.54.0 > > > >=20