From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-1.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_PASS,URIBL_BLOCKED autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id E35AFC433F4 for ; Fri, 24 Aug 2018 19:26:14 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id A956F208EB for ; Fri, 24 Aug 2018 19:26:14 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org A956F208EB Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=redhat.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727741AbeHXXB5 (ORCPT ); Fri, 24 Aug 2018 19:01:57 -0400 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:59936 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1727507AbeHXXBx (ORCPT ); Fri, 24 Aug 2018 19:01:53 -0400 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 0B0564023475; Fri, 24 Aug 2018 19:25:56 +0000 (UTC) Received: from localhost.localdomain.com (ovpn-122-125.rdu2.redhat.com [10.10.122.125]) by smtp.corp.redhat.com (Postfix) with ESMTP id 8AF432026D6D; Fri, 24 Aug 2018 19:25:55 +0000 (UTC) From: jglisse@redhat.com To: linux-mm@kvack.org Cc: Andrew Morton , linux-kernel@vger.kernel.org, =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= , Michal Hocko , Ralph Campbell , John Hubbard Subject: [PATCH 7/7] mm/hmm: proper support for blockable mmu_notifier Date: Fri, 24 Aug 2018 15:25:49 -0400 Message-Id: <20180824192549.30844-8-jglisse@redhat.com> In-Reply-To: <20180824192549.30844-1-jglisse@redhat.com> References: <20180824192549.30844-1-jglisse@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Scanned-By: MIMEDefang 2.78 on 10.11.54.4 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.6]); Fri, 24 Aug 2018 19:25:56 +0000 (UTC) X-Greylist: inspected by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.6]); Fri, 24 Aug 2018 19:25:56 +0000 (UTC) for IP:'10.11.54.4' DOMAIN:'int-mx04.intmail.prod.int.rdu2.redhat.com' HELO:'smtp.corp.redhat.com' FROM:'jglisse@redhat.com' RCPT:'' Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Jérôme Glisse When mmu_notifier calls invalidate_range_start callback with blockable set to false we should not sleep. Properly propagate this to HMM users. Signed-off-by: Jérôme Glisse Cc: Michal Hocko Cc: Ralph Campbell Cc: John Hubbard Cc: Andrew Morton --- include/linux/hmm.h | 12 +++++++++--- mm/hmm.c | 39 ++++++++++++++++++++++++++++----------- 2 files changed, 37 insertions(+), 14 deletions(-) diff --git a/include/linux/hmm.h b/include/linux/hmm.h index 064924bce75c..c783916f8732 100644 --- a/include/linux/hmm.h +++ b/include/linux/hmm.h @@ -287,11 +287,13 @@ enum hmm_update_event { * @start: virtual start address of the range to update * @end: virtual end address of the range to update * @event: event triggering the update (what is happening) + * @blockable: can the callback block/sleep ? */ struct hmm_update { unsigned long start; unsigned long end; enum hmm_update_event event; + bool blockable; }; /* @@ -314,6 +316,8 @@ struct hmm_mirror_ops { * * @mirror: pointer to struct hmm_mirror * @update: update informations (see struct hmm_update) + * Returns: -EAGAIN if update.blockable false and callback need to + * block, 0 otherwise. * * This callback ultimately originates from mmu_notifiers when the CPU * page table is updated. The device driver must update its page table @@ -322,10 +326,12 @@ struct hmm_mirror_ops { * * The device driver must not return from this callback until the device * page tables are completely updated (TLBs flushed, etc); this is a - * synchronous call. + * synchronous call. If driver need to sleep and update->blockable is + * false then you need to abort (do not do anything that would sleep or + * block) and return -EAGAIN. */ - void (*sync_cpu_device_pagetables)(struct hmm_mirror *mirror, - const struct hmm_update *update); + int (*sync_cpu_device_pagetables)(struct hmm_mirror *mirror, + const struct hmm_update *update); }; /* diff --git a/mm/hmm.c b/mm/hmm.c index 6fe31e2bfa1e..1d8fcaa0606f 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -123,12 +123,18 @@ void hmm_mm_destroy(struct mm_struct *mm) kfree(mm->hmm); } -static void hmm_invalidate_range(struct hmm *hmm, bool device, - const struct hmm_update *update) +static int hmm_invalidate_range(struct hmm *hmm, bool device, + const struct hmm_update *update) { struct hmm_mirror *mirror; struct hmm_range *range; + /* + * It is fine to wait on lock here even if update->blockable is false + * as the hmm->lock is only held for short period of time (when adding + * or walking the ranges list). We could also convert the range list + * into a lru list and avoid the spinlock all together. + */ spin_lock(&hmm->lock); list_for_each_entry(range, &hmm->ranges, list) { unsigned long addr, idx, npages; @@ -145,12 +151,26 @@ static void hmm_invalidate_range(struct hmm *hmm, bool device, spin_unlock(&hmm->lock); if (!device) - return; + return 0; + /* + * It is fine to wait on mirrors_sem here even if update->blockable is + * false as this semaphore is only taken in write mode for short period + * when adding a new mirror to the list. + */ down_read(&hmm->mirrors_sem); - list_for_each_entry(mirror, &hmm->mirrors, list) - mirror->ops->sync_cpu_device_pagetables(mirror, update); + list_for_each_entry(mirror, &hmm->mirrors, list) { + int ret; + + ret = mirror->ops->sync_cpu_device_pagetables(mirror, update); + if (!update->blockable && ret == -EAGAIN) { + up_read(&hmm->mirrors_sem); + return -EAGAIN; + } + } up_read(&hmm->mirrors_sem); + + return 0; } static void hmm_release(struct mmu_notifier *mn, struct mm_struct *mm) @@ -188,17 +208,13 @@ static int hmm_invalidate_range_start(struct mmu_notifier *mn, struct hmm_update update; struct hmm *hmm = mm->hmm; - if (!blockable) - return -EAGAIN; - VM_BUG_ON(!hmm); update.start = start; update.end = end; update.event = HMM_UPDATE_INVALIDATE; - hmm_invalidate_range(hmm, true, &update); - - return 0; + update.blockable = blockable; + return hmm_invalidate_range(hmm, true, &update); } static void hmm_invalidate_range_end(struct mmu_notifier *mn, @@ -214,6 +230,7 @@ static void hmm_invalidate_range_end(struct mmu_notifier *mn, update.start = start; update.end = end; update.event = HMM_UPDATE_INVALIDATE; + update.blockable = true; hmm_invalidate_range(hmm, false, &update); } -- 2.17.1