From nobody Sun Apr 28 01:10:12 2024 Delivered-To: importer@patchew.org Received-SPF: none (zoho.com: 198.145.21.10 is neither permitted nor denied by domain of lists.01.org) client-ip=198.145.21.10; envelope-from=edk2-devel-bounces@lists.01.org; helo=ml01.01.org; Authentication-Results: mx.zohomail.com; dkim=fail; spf=none (zoho.com: 198.145.21.10 is neither permitted nor denied by domain of lists.01.org) smtp.mailfrom=edk2-devel-bounces@lists.01.org Return-Path: Received: from ml01.01.org (ml01.01.org [198.145.21.10]) by mx.zohomail.com with SMTPS id 1502972759374395.82336450822436; Thu, 17 Aug 2017 05:25:59 -0700 (PDT) Received: from [127.0.0.1] (localhost [IPv6:::1]) by ml01.01.org (Postfix) with ESMTP id D1BA421CFA5F2; Thu, 17 Aug 2017 05:23:29 -0700 (PDT) Received: from mail-wr0-x231.google.com (mail-wr0-x231.google.com [IPv6:2a00:1450:400c:c0c::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ml01.01.org (Postfix) with ESMTPS id B9F902095B9D7 for ; Thu, 17 Aug 2017 05:23:27 -0700 (PDT) Received: by mail-wr0-x231.google.com with SMTP id b65so42201404wrd.0 for ; Thu, 17 Aug 2017 05:25:54 -0700 (PDT) Received: from localhost.localdomain ([154.146.161.128]) by smtp.gmail.com with ESMTPSA id 5sm3959480wre.5.2017.08.17.05.25.50 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 17 Aug 2017 05:25:52 -0700 (PDT) X-Original-To: edk2-devel@lists.01.org DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id; bh=/pBDIXWYVvygxMbyf3DRkEZYy5ynHqm8zhoOCL8g8Oc=; b=AxX/EwlHYHxKVwd2TMjP2MjdeVuvZejvUxjqPrQpUpShxo7oaqaMMdDbkWzDbXmc05 I24BTFku35VYJvaGCJV7xrfB1/FmgWTum2YZ+czL/kvj5E42kJc5l7lGy1Wxvey6VSbG ni7Dt92zK6XgJQZrHvfsOaYYm5mvLZLq0/Nig= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id; bh=/pBDIXWYVvygxMbyf3DRkEZYy5ynHqm8zhoOCL8g8Oc=; b=itILBM8t1xkp/dLMkXSdTZAN7j6V+wzsqQc3oxucdsJTH8sC9PLyuFoNzhyz0V9EjZ nX0/05LCi0K4gRGXsVdWrBYK0lz8aviGqHP6oXCJZaRzZdNZns7Bg3SQYvKhuCSp+Thl 2ldpy0Gq8Nf6M2GUdZVHoisd6WVeDLW2sgZQ0yioZ0+GsWWbJNOk7t2niGajvYBByNbY WU9RkbhZOxyIy+DoV3VfZeqy9za0nureKtCR1TcgyRDOMQSE7X/zdsPVw2aefq/usj93 0srTkN03iH5CPHMzP3x2pl981rl8VXOZcFsrTjNmq2hhOUVpzw9oEaY6RIJywXpU+wzf zShQ== X-Gm-Message-State: AHYfb5jr10nXwDpRftz+vw47mQj+hzKy6J5oRjvjbwdi32Sb35P06qrA GgYTrP6xL+dfshOc1JzYzA== X-Received: by 10.28.94.19 with SMTP id s19mr1106250wmb.15.1502972752933; Thu, 17 Aug 2017 05:25:52 -0700 (PDT) From: Ard Biesheuvel To: edk2-devel@lists.01.org, leif.lindholm@linaro.org, lersek@redhat.com Date: Thu, 17 Aug 2017 13:25:46 +0100 Message-Id: <20170817122546.17683-1-ard.biesheuvel@linaro.org> X-Mailer: git-send-email 2.11.0 Subject: [edk2] [PATCH] ArmPkg/ArmDmaLib: use double buffering only for bus master write X-BeenThere: edk2-devel@lists.01.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: EDK II Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Ard Biesheuvel MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Errors-To: edk2-devel-bounces@lists.01.org Sender: "edk2-devel" X-ZohoMail-DKIM: fail (Header signature does not verify) X-ZohoMail: RDKM_2 RSF_4 Z_629925259 SPT_0 Content-Type: text/plain; charset="utf-8" The ArmPkg implementation of DmaLib uses double buffering to ensure that any attempt to perform non-coherent DMA on unaligned buffers cannot corrupt adjacent unrelated data which happens to share cachelines with the data we are exchanging with the device. Such corruption can only occur on bus master read, in which case we have to invalidate the caches to ensure the CPU will see the data written to memory by the device. In the bus master write case, we can simply clean and invalidate at the same time, which may purge unrelated adjacent data from the caches, but will not corrupt its contents. Also, this double buffer does not necessarily have to be allocated from uncached memory: by the same reasoning, we can perform cache invalidation on an ordinary pool allocation as long as we take the same alignment constraints into account. So update our code accordingly: remove double buffering from the bus master read path, and switch to a pool allocation for the double buffer. Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Ard Biesheuvel Reviewed-by: Leif Lindholm --- ArmPkg/Library/ArmDmaLib/ArmDmaLib.c | 47 ++++++++++++-------- 1 file changed, 28 insertions(+), 19 deletions(-) diff --git a/ArmPkg/Library/ArmDmaLib/ArmDmaLib.c b/ArmPkg/Library/ArmDmaLi= b/ArmDmaLib.c index f4ee9e4c5ea2..61d70614bff0 100644 --- a/ArmPkg/Library/ArmDmaLib/ArmDmaLib.c +++ b/ArmPkg/Library/ArmDmaLib/ArmDmaLib.c @@ -80,6 +80,7 @@ DmaMap ( MAP_INFO_INSTANCE *Map; VOID *Buffer; EFI_GCD_MEMORY_SPACE_DESCRIPTOR GcdDescriptor; + UINTN AllocSize; =20 if (HostAddress =3D=3D NULL || NumberOfBytes =3D=3D NULL || DeviceAddres= s =3D=3D NULL || Mapping =3D=3D NULL ) { return EFI_INVALID_PARAMETER; @@ -104,8 +105,9 @@ DmaMap ( return EFI_OUT_OF_RESOURCES; } =20 - if ((((UINTN)HostAddress & (mCpu->DmaBufferAlignment - 1)) !=3D 0) || - ((*NumberOfBytes & (mCpu->DmaBufferAlignment - 1)) !=3D 0)) { + if (Operation !=3D MapOperationBusMasterRead && + ((((UINTN)HostAddress & (mCpu->DmaBufferAlignment - 1)) !=3D 0) || + ((*NumberOfBytes & (mCpu->DmaBufferAlignment - 1)) !=3D 0))) { =20 // Get the cacheability of the region Status =3D gDS->GetMemorySpaceDescriptor ((UINTN)HostAddress, &GcdDesc= riptor); @@ -129,21 +131,24 @@ DmaMap ( } =20 // - // If the buffer does not fill entire cache lines we must double buf= fer into - // uncached memory. Device (PCI) address becomes uncached page. + // If the buffer does not fill entire cache lines we must double buf= fer + // into a suitably aligned allocation that allows us to invalidate t= he + // cache without running the risk of corrupting adjacent unrelated d= ata. + // Note that pool allocations are guaranteed to be 8 byte aligned, so + // we only have to add (alignment - 8) worth of padding. // - Map->DoubleBuffer =3D TRUE; - Status =3D DmaAllocateBuffer (EfiBootServicesData, EFI_SIZE_TO_PAGES= (*NumberOfBytes), &Buffer); - if (EFI_ERROR (Status)) { + Map->DoubleBuffer =3D TRUE; + AllocSize =3D ALIGN_VALUE (*NumberOfBytes, mCpu->DmaBufferAlignment)= + + (mCpu->DmaBufferAlignment - 8); + Map->BufferAddress =3D AllocatePool (AllocSize); + if (Map->BufferAddress =3D=3D NULL) { + Status =3D EFI_OUT_OF_RESOURCES; goto FreeMapInfo; } =20 - if (Operation =3D=3D MapOperationBusMasterRead) { - CopyMem (Buffer, HostAddress, *NumberOfBytes); - } - + Buffer =3D ALIGN_POINTER (Map->BufferAddress, mCpu->DmaBufferAlignme= nt); *DeviceAddress =3D HostToDeviceAddress (ConvertToPhysicalAddress (Bu= ffer)); - Map->BufferAddress =3D Buffer; + } else { Map->DoubleBuffer =3D FALSE; } @@ -207,6 +212,7 @@ DmaUnmap ( { MAP_INFO_INSTANCE *Map; EFI_STATUS Status; + VOID *Buffer; =20 if (Mapping =3D=3D NULL) { ASSERT (FALSE); @@ -217,17 +223,20 @@ DmaUnmap ( =20 Status =3D EFI_SUCCESS; if (Map->DoubleBuffer) { - ASSERT (Map->Operation !=3D MapOperationBusMasterCommonBuffer); + ASSERT (Map->Operation =3D=3D MapOperationBusMasterWrite); =20 - if (Map->Operation =3D=3D MapOperationBusMasterCommonBuffer) { + if (Map->Operation !=3D MapOperationBusMasterWrite) { Status =3D EFI_INVALID_PARAMETER; - } else if (Map->Operation =3D=3D MapOperationBusMasterWrite) { - CopyMem ((VOID *)(UINTN)Map->HostAddress, Map->BufferAddress, - Map->NumberOfBytes); - } + } else { + Buffer =3D ALIGN_POINTER (Map->BufferAddress, mCpu->DmaBufferAlignme= nt); + + mCpu->FlushDataCache (mCpu, (UINTN)Buffer, Map->NumberOfBytes, + EfiCpuFlushTypeInvalidate); =20 - DmaFreeBuffer (EFI_SIZE_TO_PAGES (Map->NumberOfBytes), Map->BufferAddr= ess); + CopyMem ((VOID *)(UINTN)Map->HostAddress, Buffer, Map->NumberOfBytes= ); =20 + FreePool (Map->BufferAddress); + } } else { if (Map->Operation =3D=3D MapOperationBusMasterWrite) { // --=20 2.11.0 _______________________________________________ edk2-devel mailing list edk2-devel@lists.01.org https://lists.01.org/mailman/listinfo/edk2-devel