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 5519AC35FF3 for ; Mon, 17 Mar 2025 17:35:28 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 7352010E2DB; Mon, 17 Mar 2025 17:35:22 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="hPTaBkve"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.18]) by gabe.freedesktop.org (Postfix) with ESMTPS id AD61C10E2DB for ; Mon, 17 Mar 2025 17:35:18 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1742232921; x=1773768921; h=message-id:date:mime-version:subject:to:cc:references: from:in-reply-to:content-transfer-encoding; bh=Q6cXF/ylZa/q/tZcXlF31mrnMbD7xapPZFyt2T15y2U=; b=hPTaBkveZwDpOZPlhyhr0DGEJ9Zx+edi016HchMRWgniGbMx2U4FGe/k QuE8720W8rMPG4tcmIwHBKI81p2ZMz2bw8TkBQ6ZID8TOSUG4q4soDfaq D+VwbMRx3YClJLD0LXXMgpbmVneG3Cvms5tMrZCZFui7YGv2jKMoIwuGp ba/RMsLcVpc6KpeKGHuBYXoLg8xTrmUs11VRIo/0jjSUR69iBNPGW8cpA O1hGSbeueoHcCiFCnTE31C1+3hcl3Dz6Ui7ehG6wFMrXcgvEDl2o0lqU3 USO2awQZBqykwxoxPpou+0Jd1v9FBZ2KR7FaNyFtNrtLdh/JD5e2UzzL0 g==; X-CSE-ConnectionGUID: 92vs/3OHQu6GrhQe1Kwgsw== X-CSE-MsgGUID: JUFJqUSLR4uRfgyqWXAokw== X-IronPort-AV: E=McAfee;i="6700,10204,11376"; a="43515765" X-IronPort-AV: E=Sophos;i="6.14,254,1736841600"; d="scan'208";a="43515765" Received: from orviesa005.jf.intel.com ([10.64.159.145]) by orvoesa110.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 17 Mar 2025 10:35:19 -0700 X-CSE-ConnectionGUID: 4gIg34HjS1u0sH7dqIXeCg== X-CSE-MsgGUID: z6ZDyjk9TpqVWI1nw24MYw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.14,254,1736841600"; d="scan'208";a="127179682" Received: from smoticic-mobl1.ger.corp.intel.com (HELO [10.245.245.8]) ([10.245.245.8]) by orviesa005-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 17 Mar 2025 10:35:17 -0700 Message-ID: <5dd6e196-ba18-45b8-846f-de38c614284e@intel.com> Date: Mon, 17 Mar 2025 17:35:14 +0000 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH 3/5] drm/xe/bo: Add a bo remove callback To: =?UTF-8?Q?Thomas_Hellstr=C3=B6m?= , intel-xe@lists.freedesktop.org Cc: himal.prasad.ghimiray@intel.com, Matthew Brost References: <20250317104132.8200-1-thomas.hellstrom@linux.intel.com> <20250317104132.8200-4-thomas.hellstrom@linux.intel.com> <46faf433-a7d7-480e-be87-4989c5162ea1@intel.com> <1d313655ac3c7e0444ccd354166fa83d2dc6244a.camel@linux.intel.com> <93fe9e47-6cbd-443c-b08f-0c5a62daafa5@intel.com> <3ccce780ff19ffc12e504fbd4a6c719f1f01804b.camel@linux.intel.com> Content-Language: en-GB From: Matthew Auld In-Reply-To: <3ccce780ff19ffc12e504fbd4a6c719f1f01804b.camel@linux.intel.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit 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 17/03/2025 17:07, Thomas Hellström wrote: > On Mon, 2025-03-17 at 16:50 +0000, Matthew Auld wrote: >> On 17/03/2025 16:16, Thomas Hellström wrote: >>> On Mon, 2025-03-17 at 15:58 +0000, Matthew Auld wrote: >>>> On 17/03/2025 10:41, Thomas Hellström 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. >>>>> >>>>> The evict_flags() placement is chosen in such a way that bos >>>>> that >>>>> aren't exported are purged. >>>>> >>>>> Signed-off-by: Thomas Hellström >>>>> >>>> >>>>> --- >>>>>    drivers/gpu/drm/xe/xe_bo.c     | 96 >>>>> +++++++++++++++++++++++++++++++++- >>>>>    drivers/gpu/drm/xe/xe_bo.h     |  2 + >>>>>    drivers/gpu/drm/xe/xe_device.c |  2 + >>>>>    3 files changed, 99 insertions(+), 1 deletion(-) >>>>> >>>>> 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 = { >>>>>     .placement = &sys_placement_flags, >>>>>    }; >>>>> >>>>> +static struct ttm_placement purge_placement; >>>>> + >>>>>    static const struct ttm_place tt_placement_flags[] = { >>>>>     { >>>>>     .fpfn = 0, >>>>> @@ -281,6 +283,8 @@ int xe_bo_placement_for_flags(struct >>>>> xe_device >>>>> *xe, struct xe_bo *bo, >>>>>    static void xe_evict_flags(struct ttm_buffer_object *tbo, >>>>>        struct ttm_placement *placement) >>>>>    { >>>>> + struct xe_device *xe = container_of(tbo->bdev, >>>>> typeof(*xe), ttm); >>>>> + bool device_unplugged = drm_dev_is_unplugged(&xe- >>>>>> drm); >>>>>     struct xe_bo *bo; >>>>> >>>>>     if (!xe_bo_is_xe_bo(tbo)) { >>>>> @@ -290,7 +294,7 @@ static void xe_evict_flags(struct >>>>> ttm_buffer_object *tbo, >>>>>     return; >>>>>     } >>>>> >>>>> - *placement = sys_placement; >>>>> + *placement = device_unplugged ? >>>>> purge_placement : >>>>> sys_placement; >>>>>     return; >>>>>     } >>>>> >>>>> @@ -300,6 +304,11 @@ static void xe_evict_flags(struct >>>>> ttm_buffer_object *tbo, >>>>>     return; >>>>>     } >>>>> >>>>> + if (device_unplugged && !tbo->base.dma_buf) { >>>>> + *placement = purge_placement; >>>>> + return; >>>>> + } >>>> >>>>> + >>>>>     /* >>>>>     * For xe, sg bos that are evicted to system just >>>>> triggers >>>>> a >>>>>     * 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, >>>>>     struct xe_ttm_tt *xe_tt = container_of(ttm_bo->ttm, >>>>> struct >>>>> xe_ttm_tt, >>>>>            ttm); >>>>>     struct xe_device *xe = ttm_to_xe_device(ttm_bo->bdev); >>>>> + bool device_unplugged = drm_dev_is_unplugged(&xe- >>>>>> drm); >>>>>     struct sg_table *sg; >>>>> >>>>>     xe_assert(xe, attach); >>>>>     xe_assert(xe, ttm_bo->ttm); >>>>> >>>>> + if (device_unplugged && new_res->mem_type == >>>>> XE_PL_SYSTEM >>>>> && >>>>> +     ttm_bo->sg) { >>>>> + dma_resv_wait_timeout(ttm_bo->base.resv, >>>>> DMA_RESV_USAGE_BOOKKEEP, >>>>> +       false, >>>>> MAX_SCHEDULE_TIMEOUT); >>>>> + dma_buf_unmap_attachment(attach, ttm_bo->sg, >>>>> DMA_BIDIRECTIONAL); >>>>> + ttm_bo->sg = NULL; >>>>> + } >>>> >>>>> + >>>>>     if (new_res->mem_type == XE_PL_SYSTEM) >>>>>     goto out; >>>>> >>>>> @@ -2945,6 +2963,82 @@ void >>>>> xe_bo_runtime_pm_release_mmap_offset(struct xe_bo *bo) >>>>>     list_del_init(&bo->vram_userfault_link); >>>>>    } >>>>> >>>>> +static void xe_bo_dma_unmap_pinned(struct xe_device *xe) >>>> >>>> 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. >>> >>>> >>>>> +{ >>>>> + struct list_head still_in_list; >>>>> + >>>>> + INIT_LIST_HEAD(&still_in_list); >>>> >>>> LIST_HEAD(still_in_list); >>> >>> Right. Will fix. >>>> >>>> >>>>> + spin_lock(&xe->pinned.lock); >>>>> + for (;;) { >>>>> + struct xe_bo *bo = >>>>> list_first_entry_or_null(&xe- >>>>>> pinned.kernel_bo_present, >>>>> + >>>>> 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 = &bo->ttm; >>>>> + tt = ttm_bo->ttm; >>>>> + if (tt) { >>>>> + xe_ttm_bo_delete_mem_notify(ttm_bo); >>>> >>>> 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. >> >> Ah ok. But here we are looping over kernel_bo_present, do we really >> have >> dma-buf stuff on that? I only see the vram list thing, did you mean >> that? > > That's actually a very good point. I should have paid closer attention > to what the various lists actually hold. So pinned dma-bufs don't sit > on any xe lists as it seems? That's my understanding. Although not sure what is going on with that vram list, which is curretly unused. > > I guess things still work for suspend / resume since dma-bufs are not > supposed to be compressed? Right. > > For this case, though, They need to be added somewhere. Yeah, looks like we are indeed missing that. > > /Thomas > > >> >>> >>>> >>>>> + xe_tt = container_of(tt, >>>>> typeof(*xe_tt), >>>>> ttm); >>>>> + if (xe_tt->sg) { >>>>> + dma_unmap_sgtable(xe_tt->xe- >>>>>> drm.dev, xe_tt->sg, >>>>> + >>>>> DMA_BIDIRECTIONAL, 0); >>>>> + sg_free_table(xe_tt->sg); >>>>> + xe_tt->sg = NULL; >>>>> + } >>>> >>>> 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? >> >> Yeah, for dma-buf leaving the pages alone make sense. I was just >> getting >> thrown off by the kernel_bo_present. I don't really mind either way. >> >>> >>>> >>>>> + } >>>>> + >>>>> + 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); >>>>> +} >>>> >>>> We could potentially move this type of thing into xe_bo_evict.c? >>> >>> Could perhaps be the best choice, actually. >>> >>>> >>>>> + >>>>> +/** >>>>> + * 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) >>>> >>>> xe_device_remove_bo() ? >>> >>> Will look into it. >>> >>>> >>>>> +{ >>>>> + unsigned int mem_type; >>>>> + >>>>> + /* >>>>> + * Move pagemap bos and exported dma-buf to system. >>>> >>>> ..and purge everything else." >>> >>> Will add. >>> >>> Thanks for reviewing! >>> /Thomas >>> >> >