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 9F306C35FF3 for ; Mon, 17 Mar 2025 16:16:35 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 430FE10E189; Mon, 17 Mar 2025 16:16:35 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="mpd3JamW"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.7]) by gabe.freedesktop.org (Postfix) with ESMTPS id 438FE10E189 for ; Mon, 17 Mar 2025 16:16:33 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1742228193; x=1773764193; h=message-id:subject:from:to:cc:date:in-reply-to: references:content-transfer-encoding:mime-version; bh=J86iIWvgMOd4pBvP9IS535KTuQ0xOj4UBBGzbyXvvH4=; b=mpd3JamWpuT8rkqMLpBjthW8OQkmhmxBhIZvlzIVyFwl6oE4rTxiqyeK KPPY38Yoi/OclOCx03x0Nx9j8bxsZG13fEhcH6w0g1O4OJojAn/u92mjh X+nUYNKH0nLyEird0ccMf70gLMCM7EE0RCZ9KkR+n4KuD9m/5Y45lbgHb uCQa91V0rPPlW9r/6Zdl6graLAq5a07lpNA28BuCYxyQOUNY/QlF7SA2v 11oPtnUlscMIW4FpTtuo15/QBkFUmxCFJodCX01/bIOQbIvJZ6uWQDcFV NaGwmYqdOIDx6Rx6AecytySe/SdMflOn12ucTs0eMuutjctO//0HbFm4C A==; X-CSE-ConnectionGUID: i7A3xIkXT7yTwCM4Ww3Nzw== X-CSE-MsgGUID: 7VJEmHCDS8edgZDZ9U2U+A== X-IronPort-AV: E=McAfee;i="6700,10204,11376"; a="68689648" X-IronPort-AV: E=Sophos;i="6.14,254,1736841600"; d="scan'208";a="68689648" Received: from fmviesa009.fm.intel.com ([10.60.135.149]) by fmvoesa101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 17 Mar 2025 09:16:33 -0700 X-CSE-ConnectionGUID: bJoiP5AKRNu8ddGRnaFuCA== X-CSE-MsgGUID: GrdueWCgQhS4pFHetIZeLw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.14,254,1736841600"; d="scan'208";a="122715554" Received: from hrotuna-mobl2.ger.corp.intel.com (HELO [10.245.246.97]) ([10.245.246.97]) by fmviesa009-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 17 Mar 2025 09:16:31 -0700 Message-ID: <1d313655ac3c7e0444ccd354166fa83d2dc6244a.camel@linux.intel.com> Subject: Re: [PATCH 3/5] drm/xe/bo: Add a bo remove callback From: Thomas =?ISO-8859-1?Q?Hellstr=F6m?= To: Matthew Auld , intel-xe@lists.freedesktop.org Cc: himal.prasad.ghimiray@intel.com, Matthew Brost Date: Mon, 17 Mar 2025 17:16:29 +0100 In-Reply-To: <46faf433-a7d7-480e-be87-4989c5162ea1@intel.com> References: <20250317104132.8200-1-thomas.hellstrom@linux.intel.com> <20250317104132.8200-4-thomas.hellstrom@linux.intel.com> <46faf433-a7d7-480e-be87-4989c5162ea1@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 3.54.3 (3.54.3-1.fc41) 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 Mon, 2025-03-17 at 15:58 +0000, Matthew Auld wrote: > On 17/03/2025 10:41, Thomas Hellstr=C3=B6m wrote: > > On device unbind, migrate exported bos, including pagemap bos to > > system. This allows importers to take proper action without > > disruption. In particular, SVM clients on remote devices may > > continue as if nothing happened, and can chose a different > > placement. > >=20 > > The evict_flags() placement is chosen in such a way that bos that > > aren't exported are purged. > >=20 > > Signed-off-by: Thomas Hellstr=C3=B6m >=20 > > --- > > =C2=A0 drivers/gpu/drm/xe/xe_bo.c=C2=A0=C2=A0=C2=A0=C2=A0 | 96 > > +++++++++++++++++++++++++++++++++- > > =C2=A0 drivers/gpu/drm/xe/xe_bo.h=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0 2 + > > =C2=A0 drivers/gpu/drm/xe/xe_device.c |=C2=A0 2 + > > =C2=A0 3 files changed, 99 insertions(+), 1 deletion(-) > >=20 > > diff --git a/drivers/gpu/drm/xe/xe_bo.c > > b/drivers/gpu/drm/xe/xe_bo.c > > index 64f9c936eea0..c7c206041632 100644 > > --- a/drivers/gpu/drm/xe/xe_bo.c > > +++ b/drivers/gpu/drm/xe/xe_bo.c > > @@ -55,6 +55,8 @@ static struct ttm_placement sys_placement =3D { > > =C2=A0=C2=A0 .placement =3D &sys_placement_flags, > > =C2=A0 }; > > =C2=A0=20 > > +static struct ttm_placement purge_placement; > > + > > =C2=A0 static const struct ttm_place tt_placement_flags[] =3D { > > =C2=A0=C2=A0 { > > =C2=A0=C2=A0 .fpfn =3D 0, > > @@ -281,6 +283,8 @@ int xe_bo_placement_for_flags(struct xe_device > > *xe, struct xe_bo *bo, > > =C2=A0 static void xe_evict_flags(struct ttm_buffer_object *tbo, > > =C2=A0=C2=A0 =C2=A0=C2=A0 struct ttm_placement *placement) > > =C2=A0 { > > + struct xe_device *xe =3D container_of(tbo->bdev, > > typeof(*xe), ttm); > > + bool device_unplugged =3D drm_dev_is_unplugged(&xe->drm); > > =C2=A0=C2=A0 struct xe_bo *bo; > > =C2=A0=20 > > =C2=A0=C2=A0 if (!xe_bo_is_xe_bo(tbo)) { > > @@ -290,7 +294,7 @@ static void xe_evict_flags(struct > > ttm_buffer_object *tbo, > > =C2=A0=C2=A0 return; > > =C2=A0=C2=A0 } > > =C2=A0=20 > > - *placement =3D sys_placement; > > + *placement =3D device_unplugged ? purge_placement : > > sys_placement; > > =C2=A0=C2=A0 return; > > =C2=A0=C2=A0 } > > =C2=A0=20 > > @@ -300,6 +304,11 @@ static void xe_evict_flags(struct > > ttm_buffer_object *tbo, > > =C2=A0=C2=A0 return; > > =C2=A0=C2=A0 } > > =C2=A0=20 > > + if (device_unplugged && !tbo->base.dma_buf) { > > + *placement =3D purge_placement; > > + return; > > + } >=20 > > + > > =C2=A0=C2=A0 /* > > =C2=A0=C2=A0 * For xe, sg bos that are evicted to system just triggers > > a > > =C2=A0=C2=A0 * rebind of the sg list upon subsequent validation to > > XE_PL_TT. > > @@ -657,11 +666,20 @@ static int xe_bo_move_dmabuf(struct > > ttm_buffer_object *ttm_bo, > > =C2=A0=C2=A0 struct xe_ttm_tt *xe_tt =3D container_of(ttm_bo->ttm, stru= ct > > xe_ttm_tt, > > =C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 ttm); > > =C2=A0=C2=A0 struct xe_device *xe =3D ttm_to_xe_device(ttm_bo->bdev); > > + bool device_unplugged =3D drm_dev_is_unplugged(&xe->drm); > > =C2=A0=C2=A0 struct sg_table *sg; > > =C2=A0=20 > > =C2=A0=C2=A0 xe_assert(xe, attach); > > =C2=A0=C2=A0 xe_assert(xe, ttm_bo->ttm); > > =C2=A0=20 > > + if (device_unplugged && new_res->mem_type =3D=3D XE_PL_SYSTEM > > && > > + =C2=A0=C2=A0=C2=A0 ttm_bo->sg) { > > + dma_resv_wait_timeout(ttm_bo->base.resv, > > DMA_RESV_USAGE_BOOKKEEP, > > + =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 false, > > MAX_SCHEDULE_TIMEOUT); > > + dma_buf_unmap_attachment(attach, ttm_bo->sg, > > DMA_BIDIRECTIONAL); > > + ttm_bo->sg =3D NULL; > > + } >=20 > > + > > =C2=A0=C2=A0 if (new_res->mem_type =3D=3D XE_PL_SYSTEM) > > =C2=A0=C2=A0 goto out; > > =C2=A0=20 > > @@ -2945,6 +2963,82 @@ void > > xe_bo_runtime_pm_release_mmap_offset(struct xe_bo *bo) > > =C2=A0=C2=A0 list_del_init(&bo->vram_userfault_link); > > =C2=A0 } > > =C2=A0=20 > > +static void xe_bo_dma_unmap_pinned(struct xe_device *xe) >=20 > Should we prefix this with xe_device_ or something? > xe_device_unmap_pinned_bo() ? Hmm, yes, while it doesn't take a bo as argument, I think it should reside in the bo subsystem. Let me check whether we should split this up and have the loops elsewhere. >=20 > > +{ > > + struct list_head still_in_list; > > + > > + INIT_LIST_HEAD(&still_in_list); >=20 > LIST_HEAD(still_in_list); Right. Will fix. >=20 >=20 > > + spin_lock(&xe->pinned.lock); > > + for (;;) { > > + struct xe_bo *bo =3D list_first_entry_or_null(&xe- > > >pinned.kernel_bo_present, > > + =C2=A0=C2=A0=C2=A0 > > typeof(*bo), pinned_link); > > + struct ttm_buffer_object *ttm_bo; > > + struct ttm_tt *tt; > > + struct xe_ttm_tt *xe_tt; > > + > > + if (!bo) > > + break; > > + > > + list_move_tail(&bo->pinned_link, &still_in_list); > > + xe_bo_get(bo); > > + spin_unlock(&xe->pinned.lock); > > + > > + xe_bo_lock(bo, false); > > + ttm_bo =3D &bo->ttm; > > + tt =3D ttm_bo->ttm; > > + if (tt) { > > + xe_ttm_bo_delete_mem_notify(ttm_bo); >=20 > Do we actually need this? It's for imported pinned dma-bufs, I don't think we have any of these ATM, but might in the future I guess. OFC I can structure that a bit nicer. >=20 > > + xe_tt =3D container_of(tt, typeof(*xe_tt), > > ttm); > > + if (xe_tt->sg) { > > + dma_unmap_sgtable(xe_tt->xe- > > >drm.dev, xe_tt->sg, > > + =C2=A0 > > DMA_BIDIRECTIONAL, 0); > > + sg_free_table(xe_tt->sg); > > + xe_tt->sg =3D NULL; > > + } >=20 > Would it make sense to also call ttm_tt_unpopulate() here, so we nuke > the pages? I think we arrived at not nuking the pages for exported dma-bufs. For the other use-cases I guess it's hard to tell: Is there a chance that whatever needed the buffer pinned can issue a stray write? I still think the best approach would to ensure that those subsystems used devm_ managed resources rather than drmm_-managed resources. Meanwhile, I can perhaps add a comment about that? >=20 > > + } > > + > > + xe_bo_unlock(bo); > > + xe_bo_put(bo); > > + spin_lock(&xe->pinned.lock); > > + } > > + list_splice_tail(&still_in_list, &xe- > > >pinned.kernel_bo_present); > > + spin_unlock(&xe->pinned.lock); > > +} >=20 > We could potentially move this type of thing into xe_bo_evict.c? Could perhaps be the best choice, actually. >=20 > > + > > +/** > > + * xe_bo_remove() - Handle bos when the pci_device is about to be > > removed > > + * @xe: The xe device. > > + * > > + * On pci_device removal we need to drop all dma mappings and move > > + * the data of exported bos out to system. This includes SVM bos > > and > > + * exported dma-buf bos. This is done by evicting all bos, but > > + * the evict placement in xe_evict_flags() is chosen such that all > > + * bos except those mentioned are purged, and thus their memory > > + * is released. > > + * > > + * For pinned bos, we're unmapping dma. > > + */ > > +void xe_bo_remove(struct xe_device *xe) >=20 > xe_device_remove_bo() ? Will look into it. >=20 > > +{ > > + unsigned int mem_type; > > + > > + /* > > + * Move pagemap bos and exported dma-buf to system. >=20 > ..and purge everything else." Will add. Thanks for reviewing! /Thomas