From mboxrd@z Thu Jan 1 00:00:00 1970 From: Murali Karicheri Subject: Re: [PATCH v3 2/4] of: move of_dma_configure() to device,c to help re-use Date: Thu, 8 Jan 2015 18:44:31 -0500 Message-ID: <54AF165F.7070909@ti.com> References: <1420656594-8908-1-git-send-email-m-karicheri2@ti.com> <3861625.CJ6m15sZMh@wuerfel> <54AED9EC.1020703@ti.com> <2083932.v1GzgAfexK@wuerfel> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <2083932.v1GzgAfexK@wuerfel> Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Arnd Bergmann Cc: Rob Herring , Joerg Roedel , Grant Likely , Rob Herring , Linux IOMMU , "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , Bjorn Helgaas , "linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org" , Will Deacon , Russell King - ARM Linux List-Id: devicetree@vger.kernel.org On 01/08/2015 05:24 PM, Arnd Bergmann wrote: > On Thursday 08 January 2015 14:26:36 Murali Karicheri wrote: >> On 01/08/2015 03:40 AM, Arnd Bergmann wrote: >>> On Wednesday 07 January 2015 17:37:56 Rob Herring wrote: >>>> On Wed, Jan 7, 2015 at 12:49 PM, Murali Karicheri wrote: >>>> >>>>> + ret = of_dma_get_range(np,&dma_addr,&paddr,&size); >>>>> + if (ret< 0) { >>>>> + dma_addr = offset = 0; >>>>> + size = dev->coherent_dma_mask + 1; >>>> >>>> If coherent_dma_mask is DMA_BIT_MASK(64), then you will overflow and >>>> have a size of 0. There may also be a problem when the mask is only >>>> 32-bit type. >>> >>> The mask is always a 64-bit type, it's not optional. But you are right, >>> the 64-bit mask case is broken, so I guess we have to fix it differently >>> by always passing the smaller value into arch_setup_dma_ops and >>> adapting that function instead. >> Arnd, >> >> What is the smaller value you are referring to in the below code? >> between *dev->dma_mask and size from DT? But overflow can still happen >> when size is to be calculated in arch_setup_dma_ops() for Non DT case or >> when DT size is configured to be equivalent of DMA_BIT_MASK(64) + 1. Can >> we discuss the code change you have in mind when you get a chance? > > I meant changing every function that the size values gets passed into > to take a mask like 0xffffffff instead of a size like 0x100000000, so > we can represent a 64-bit capable bus correctly. The function here refers to arch_setup_dma_ops() and anything that is called by this API, right? > > This means we also need to adapt the value returned from of_dma_get_range. > A minor complication here is that the DT properties sometimes already > contain the mask value, in particular when we want to represent a > full mapping like > The grep of dma-ranges for arch/arm shows the size used is mask + 1 as ./boot/dts/keystone.dtsi: dma-ranges = <0x80000000 0x8 0x00000000 0x80000000>; ./boot/dts/keystone.dtsi: dma-ranges; ./boot/dts/keystone.dtsi: dma-ranges; ./boot/dts/r8a7790.dtsi: dma-ranges = <0x42000000 0 0x40000000 0 0x40000000 0 0x80000000 ./boot/dts/integratorap.dts: dma-ranges = <0x80000000 0x0 0x80000000>; ./boot/dts/r8a7791.dtsi: dma-ranges = <0x42000000 0 0x40000000 0 0x40000000 0 0x80000000 ./boot/dts/.k2hk-evm.dtb.dts.tmp: dma-ranges = <0x80000000 0x8 0x00000000 0x80000000>; ./boot/dts/.k2hk-evm.dtb.dts.tmp: dma-ranges; ./boot/dts/.k2l-evm.dtb.dts.tmp: dma-ranges = <0x80000000 0x8 0x00000000 0x80000000>; ./boot/dts/.k2l-evm.dtb.dts.tmp: dma-ranges; ./boot/dts/.k2e-evm.dtb.dts.tmp: dma-ranges = <0x80000000 0x8 0x00000000 0x80000000>; ./boot/dts/.k2e-evm.dtb.dts.tmp: dma-ranges; ./boot/dts/.k2e-evm.dtb.dts.tmp: dma-ranges; ./boot/dts/k2e.dtsi: dma-ranges; ./boot/dts/k2e.dtsi: dma-ranges; So for ARM 32 the below case doesn't seem to apply. > bus { > #address-cells =<1>; > #size-cells =<1>; > dma-ranges =<0 0 0xffffffff>; /* all 4 GB, DMA_BIT_MASK(32) */ > }; > > as opposed to > A search under arch/arm64 and arch/powerpc showed another format for dma-ranges for PCI. Now I am bit nervous on how of_pci_dma_configure() behave for these platforms as PCI nodes on these platforms may have dma-ranges in a different format than that in platform bus nodes. These DT property has additional cell for resource identification (example 0x42000000 as the first cell). Also one of them, amd-seattle-soc.dtsi uses the format of size (0x7fffffffff) similar to what you refered above. ./boot/dts/apm/apm-storm.dtsi: dma-ranges = <0x42000000 0x80 0x00000000 0x80 0x00000000 0x00 0x80000000 ./boot/dts/apm/apm-storm.dtsi: dma-ranges = <0x42000000 0x80 0x00000000 0x80 0x00000000 0x00 0x80000000 ./boot/dts/apm/apm-storm.dtsi: dma-ranges = <0x42000000 0x80 0x00000000 0x80 0x00000000 0x00 0x80000000 ./boot/dts/apm/apm-storm.dtsi: dma-ranges = <0x42000000 0x80 0x00000000 0x80 0x00000000 0x00 0x80000000 ./boot/dts/apm/apm-storm.dtsi: dma-ranges = <0x42000000 0x80 0x00000000 0x80 0x00000000 0x00 0x80000000 ./boot/dts/amd/amd-seattle-soc.dtsi: dma-ranges = <0x80 0x0 0x80 0x0 0x7f 0xffffffff>; ./boot/dts/amd/amd-seattle-soc.dtsi: dma-ranges = <0x43000000 0x80 0x0 0x80 0x0 0x7f 0xffffffff>; Or I got this wrong. Any comment? For ARM 32, DT size value can be extracted and size-1 can be passed to arch_setup_dma_ops() provided it handles this value properly. > bus { > #address-cells =<1>; > #size-cells =<1>; > dma-ranges =<0 0 0x80000000>; /* only lower 2GB, DMA_BIT_MASK(31) */ > }; > > or > > bus { > #address-cells =<2>; > #size-cells =<2>; > dma-ranges =<0 0 0x0000000100000000>; /* 4GB of 64-bit address space */ > }; > > > Arnd -- Murali Karicheri Linux Kernel, Texas Instruments -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html