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 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id C0EE4C77B7E for ; Sun, 28 May 2023 19:23:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230020AbjE1TX6 (ORCPT ); Sun, 28 May 2023 15:23:58 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41216 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230011AbjE1TXv (ORCPT ); Sun, 28 May 2023 15:23:51 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [139.178.84.217]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 60A91C7 for ; Sun, 28 May 2023 12:23:50 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id F309C61BA6 for ; Sun, 28 May 2023 19:23:49 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 1CFCAC433EF; Sun, 28 May 2023 19:23:48 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1685301829; bh=6QvBvkH///NvBvLTrTuGtXcNfrdX1zFcZbHf+0xjjgY=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=gBmPIyKgb0+/i6bcAAWYz8kN07Di7pok0dsslsvWkt8a09OeHuCWyfo+ajTwrUQgM qdPE9+ODkZMZ3M6SCHSMpynn92hPPBPIP0Yp+TVnTTfabU/9+bLzq6kgpwh4ulJzMD hyWDgnlLEfpwZlXxV72PvofNhA8gYTfCQMh4Baz4= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Tomas Krcka , Will Deacon , Sasha Levin Subject: [PATCH 5.4 053/161] iommu/arm-smmu-v3: Acknowledge pri/event queue overflow if any Date: Sun, 28 May 2023 20:09:37 +0100 Message-Id: <20230528190838.883784640@linuxfoundation.org> X-Mailer: git-send-email 2.40.1 In-Reply-To: <20230528190837.051205996@linuxfoundation.org> References: <20230528190837.051205996@linuxfoundation.org> User-Agent: quilt/0.67 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: stable@vger.kernel.org From: Tomas Krcka [ Upstream commit 67ea0b7ce41844eae7c10bb04dfe66a23318c224 ] When an overflow occurs in the PRI queue, the SMMU toggles the overflow flag in the PROD register. To exit the overflow condition, the PRI thread is supposed to acknowledge it by toggling this flag in the CONS register. Unacknowledged overflow causes the queue to stop adding anything new. Currently, the priq thread always writes the CONS register back to the SMMU after clearing the queue. The writeback is not necessary if the OVFLG in the PROD register has not been changed, no overflow has occured. This commit checks the difference of the overflow flag between CONS and PROD register. If it's different, toggles the OVACKFLG flag in the CONS register and write it to the SMMU. The situation is similar for the event queue. The acknowledge register is also toggled after clearing the event queue but never propagated to the hardware. This would only be done the next time when executing evtq thread. Unacknowledged event queue overflow doesn't affect the event queue, because the SMMU still adds elements to that queue when the overflow condition is active. But it feel nicer to keep SMMU in sync when possible, so use the same way here as well. Signed-off-by: Tomas Krcka Link: https://lore.kernel.org/r/20230329123420.34641-1-tomas.krcka@gmail.com Signed-off-by: Will Deacon Signed-off-by: Sasha Levin --- drivers/iommu/arm-smmu-v3.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 4f64c3a9ee88d..b3c5d7b4547a4 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -760,6 +760,18 @@ static void queue_inc_cons(struct arm_smmu_ll_queue *q) q->cons = Q_OVF(q->cons) | Q_WRP(q, cons) | Q_IDX(q, cons); } +static void queue_sync_cons_ovf(struct arm_smmu_queue *q) +{ + struct arm_smmu_ll_queue *llq = &q->llq; + + if (likely(Q_OVF(llq->prod) == Q_OVF(llq->cons))) + return; + + llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) | + Q_IDX(llq, llq->cons); + queue_sync_cons_out(q); +} + static int queue_sync_prod_in(struct arm_smmu_queue *q) { int ret = 0; @@ -1720,8 +1732,7 @@ static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev) } while (!queue_empty(llq)); /* Sync our overflow flag, as we believe we're up to speed */ - llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) | - Q_IDX(llq, llq->cons); + queue_sync_cons_ovf(q); return IRQ_HANDLED; } @@ -1779,9 +1790,7 @@ static irqreturn_t arm_smmu_priq_thread(int irq, void *dev) } while (!queue_empty(llq)); /* Sync our overflow flag, as we believe we're up to speed */ - llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) | - Q_IDX(llq, llq->cons); - queue_sync_cons_out(q); + queue_sync_cons_ovf(q); return IRQ_HANDLED; } -- 2.39.2