SWIOTLB for ppc/mpc86xx
Subsystems (2)
cover SWIOTLB for ppc/mpc86xx
This patch series allows the use of swiotlb on mpc86xx, with 85xx support soon to follow. The most notable change here is the addition of addr_needs_map to dma_ops. This is used to tell if a device can reach an address without bounce buffering. I believe that patches will be coming out soon to make dma_ops non-arch-specific, and I expect to see something similar there as well.
I've also changed sg_dma_len to be the same on both 32 and 64-bit. We now look at sg->dma_length in both cases (see patch log).
The dma_window* elements in the pci_controller struct are hoisted out of an #ifdef CONFIG_PPC64 so we can use them to store off information about how much memory is mapped via the inbound pci windows.
A dma-swiotlb.c is created in arch/powerpc/kernel to contain all the platform-specific iotlb code.
Finally, hooks are provided to allow enabling of this feature on 86xx via Kconfig, and a 36-bit device tree for 8641hpcn is provided.
Ye Olde Diffstat:
arch/powerpc/Kconfig | 2 +- arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts | 597 ++++++++++++++++++++++++++++ arch/powerpc/include/asm/dma-mapping.h | 11 + arch/powerpc/include/asm/pci-bridge.h | 6 +- arch/powerpc/include/asm/scatterlist.h | 6 +- arch/powerpc/include/asm/swiotlb.h | 24 ++ arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/dma-swiotlb.c | 161 ++++++++ arch/powerpc/kernel/dma.c | 2 +- arch/powerpc/kernel/setup_32.c | 4 + arch/powerpc/sysdev/fsl_pci.c | 4 + 11 files changed, 809 insertions(+), 9 deletions(-)
Cheers,
Becky
cover enable swiotlb on ppc/86xx
GIT: Please enter your email below.
GIT: Lines beginning in "GIT: " will be removed.
GIT: Consider including an overall diffstat or table of contents
GIT: for the patch you are writing.
cover Allow swiotlb use on ppc/mpc86xx
GIT: Please enter your email below.
GIT: Lines beginning in "GIT: " will be removed.
GIT: Consider including an overall diffstat or table of contents
GIT: for the patch you are writing.
PATCH 1/5 powerpc: Use sg->dma_length in sg_dma_len() macro on 32-bit
Currently, the 32-bit code uses sg->length instead of sg->dma_lentgh to report sg_dma_len. However, since the default dma code for 32-bit (the dma_direct case) sets dma_length and length to the same thing, we should be able to use dma_length there as well. This gets rid of some 32-vs-64-bit ifdefs, and is needed by the swiotlb code which actually distinguishes between dma_length and length.
--- arch/powerpc/include/asm/scatterlist.h | 6 +----- 1 files changed, 1 insertions(+), 5 deletions(-)
arch/powerpc/include/asm/scatterlist.h +1 -5
@@ -21,7 +21,7 @@ struct scatterlist {unsigned int offset;unsigned int length; -/* For TCE support */ +/* For TCE or SWIOTLB support */ dma_addr_t dma_address;u32 dma_length; };
@@ -34,11 +34,7 @@ struct scatterlist {* is 0. */ #define sg_dma_address (sg) ((sg)->dma_address) -#ifdef __powerpc64__#define sg_dma_len (sg) ((sg)->dma_length) -#else -#define sg_dma_len (sg) ((sg)->length) -#endif #ifdef __powerpc64__#define ISA_DMA_THRESHOLD (~0UL)
Signature · 1 line
1.6.0.6
PATCH 2/5 powerpc: Add 36-bit device tree for mpc8641hpcn
The new dts places most of the devices in physical address space above 32-bits, which allows us to have more than 4GB of RAM present.
--- arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts | 597 ++++++++++++++++++++++++++++ 1 files changed, 597 insertions(+), 0 deletions(-) create mode 100644 arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts new file +597 -0
@@ -0,0 +1,597 @@ +/* + * MPC8641 HPCN Device Tree Source + * + * Copyright 2006 Freescale Semiconductor Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +/dts-v1/; + +/ { + model = "MPC8641HPCN"; + compatible = "fsl,mpc8641hpcn"; + #address-cells = <2>; + #size-cells = <2>; + + aliases { + ethernet0 = &enet0; + ethernet1 = &enet1; + ethernet2 = &enet2; + ethernet3 = &enet3; + serial0 = &serial0; + serial1 = &serial1; + pci0 = &pci0; + pci1 = &pci1; + }; + + cpus { + #address-cells = <1>; + #size-cells = <0>; + + PowerPC,8641@0 { + device_type = "cpu"; + reg = <0>; + d-cache-line-size = <32>; // 32 bytes + i-cache-line-size = <32>; // 32 bytes + d-cache-size = <32768>; // L1, 32K + i-cache-size = <32768>; // L1, 32K + timebase-frequency = <0>; // 33 MHz, from uboot + bus-frequency = <0>; // From uboot + clock-frequency = <0>; // From uboot + }; + PowerPC,8641@1 { + device_type = "cpu"; + reg = <1>; + d-cache-line-size = <32>; // 32 bytes + i-cache-line-size = <32>; // 32 bytes + d-cache-size = <32768>; // L1, 32K + i-cache-size = <32768>; // L1, 32K + timebase-frequency = <0>; // 33 MHz, from uboot + bus-frequency = <0>; // From uboot + clock-frequency = <0>; // From uboot + }; + }; + + memory { + device_type = "memory"; + reg = <0x0 0x00000000 0x0 0x40000000>; // 1G at 0x0 + }; + + localbus@fffe05000 { + #address-cells = <2>; + #size-cells = <1>; + compatible = "fsl,mpc8641-localbus", "simple-bus"; + reg = <0x0f 0xffe05000 0x0 0x1000>; + interrupts = <19 2>; + interrupt-parent = <&mpic>; + + ranges = <0 0 0xf 0xef800000 0x00800000 + 2 0 0xf 0xffdf8000 0x00008000 + 3 0 0xf 0xffdf0000 0x00008000>; + + flash@0,0 { + compatible = "cfi-flash"; + reg = <0 0 0x00800000>; + bank-width = <2>; + device-width = <2>; + #address-cells = <1>; + #size-cells = <1>; + partition@0 { + label = "kernel"; + reg = <0x00000000 0x00300000>; + }; + partition@300000 { + label = "firmware b"; + reg = <0x00300000 0x00100000>; + read-only; + }; + partition@400000 { + label = "fs"; + reg = <0x00400000 0x00300000>; + }; + partition@700000 { + label = "firmware a"; + reg = <0x00700000 0x00100000>; + read-only; + }; + }; + }; + + soc8641@fffe00000 { + #address-cells = <1>; + #size-cells = <1>; + device_type = "soc"; + compatible = "simple-bus"; + ranges = <0x00000000 0x0f 0xffe00000 0x00100000>; + reg = <0x0f 0xffe00000 0x0 0x00001000>; // CCSRBAR + bus-frequency = <0>; + + i2c@3000 { + #address-cells = <1>; + #size-cells = <0>; + cell-index = <0>; + compatible = "fsl-i2c"; + reg = <0x3000 0x100>; + interrupts = <43 2>; + interrupt-parent = <&mpic>; + dfsrr; + }; + + i2c@3100 { + #address-cells = <1>; + #size-cells = <0>; + cell-index = <1>; + compatible = "fsl-i2c"; + reg = <0x3100 0x100>; + interrupts = <43 2>; + interrupt-parent = <&mpic>; + dfsrr; + }; + + dma@21300 { + #address-cells = <1>; + #size-cells = <1>; + compatible = "fsl,mpc8641-dma", "fsl,eloplus-dma"; + reg = <0x21300 0x4>; + ranges = <0x0 0x21100 0x200>; + cell-index = <0>; + dma-channel@0 { + compatible = "fsl,mpc8641-dma-channel", + "fsl,eloplus-dma-channel"; + reg = <0x0 0x80>; + cell-index = <0>; + interrupt-parent = <&mpic>; + interrupts = <20 2>; + }; + dma-channel@80 { + compatible = "fsl,mpc8641-dma-channel", + "fsl,eloplus-dma-channel"; + reg = <0x80 0x80>; + cell-index = <1>; + interrupt-parent = <&mpic>; + interrupts = <21 2>; + }; + dma-channel@100 { + compatible = "fsl,mpc8641-dma-channel", + "fsl,eloplus-dma-channel"; + reg = <0x100 0x80>; + cell-index = <2>; + interrupt-parent = <&mpic>; + interrupts = <22 2>; + }; + dma-channel@180 { + compatible = "fsl,mpc8641-dma-channel", + "fsl,eloplus-dma-channel"; + reg = <0x180 0x80>; + cell-index = <3>; + interrupt-parent = <&mpic>; + interrupts = <23 2>; + }; + }; + + enet0: ethernet@24000 { + #address-cells = <1>; + #size-cells = <1>; + cell-index = <0>; + device_type = "network"; + model = "TSEC"; + compatible = "gianfar"; + reg = <0x24000 0x1000>; + ranges = <0x0 0x24000 0x1000>; + local-mac-address = [ 00 00 00 00 00 00 ]; + interrupts = <29 2 30 2 34 2>; + interrupt-parent = <&mpic>; + tbi-handle = <&tbi0>; + phy-handle = <&phy0>; + phy-connection-type = "rgmii-id"; + + mdio@520 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "fsl,gianfar-mdio"; + reg = <0x520 0x20>; + + phy0: ethernet-phy@0 { + interrupt-parent = <&mpic>; + interrupts = <10 1>; + reg = <0>; + device_type = "ethernet-phy"; + }; + phy1: ethernet-phy@1 { + interrupt-parent = <&mpic>; + interrupts = <10 1>; + reg = <1>; + device_type = "ethernet-phy"; + }; + phy2: ethernet-phy@2 { + interrupt-parent = <&mpic>; + interrupts = <10 1>; + reg = <2>; + device_type = "ethernet-phy"; + }; + phy3: ethernet-phy@3 { + interrupt-parent = <&mpic>; + interrupts = <10 1>; + reg = <3>; + device_type = "ethernet-phy"; + }; + tbi0: tbi-phy@11 { + reg = <0x11>; + device_type = "tbi-phy"; + }; + }; + }; + + enet1: ethernet@25000 { + #address-cells = <1>; + #size-cells = <1>; + cell-index = <1>; + device_type = "network"; + model = "TSEC"; + compatible = "gianfar"; + reg = <0x25000 0x1000>; + ranges = <0x0 0x25000 0x1000>; + local-mac-address = [ 00 00 00 00 00 00 ]; + interrupts = <35 2 36 2 40 2>; + interrupt-parent = <&mpic>; + tbi-handle = <&tbi1>; + phy-handle = <&phy1>; + phy-connection-type = "rgmii-id"; + + mdio@520 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "fsl,gianfar-tbi"; + reg = <0x520 0x20>; + + tbi1: tbi-phy@11 { + reg = <0x11>; + device_type = "tbi-phy"; + }; + }; + }; + + enet2: ethernet@26000 { + #address-cells = <1>; + #size-cells = <1>; + cell-index = <2>; + device_type = "network"; + model = "TSEC"; + compatible = "gianfar"; + reg = <0x26000 0x1000>; + ranges = <0x0 0x26000 0x1000>; + local-mac-address = [ 00 00 00 00 00 00 ]; + interrupts = <31 2 32 2 33 2>; + interrupt-parent = <&mpic>; + tbi-handle = <&tbi2>; + phy-handle = <&phy2>; + phy-connection-type = "rgmii-id"; + + mdio@520 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "fsl,gianfar-tbi"; + reg = <0x520 0x20>; + + tbi2: tbi-phy@11 { + reg = <0x11>; + device_type = "tbi-phy"; + }; + }; + }; + + enet3: ethernet@27000 { + #address-cells = <1>; + #size-cells = <1>; + cell-index = <3>; + device_type = "network"; + model = "TSEC"; + compatible = "gianfar"; + reg = <0x27000 0x1000>; + ranges = <0x0 0x27000 0x1000>; + local-mac-address = [ 00 00 00 00 00 00 ]; + interrupts = <37 2 38 2 39 2>; + interrupt-parent = <&mpic>; + tbi-handle = <&tbi3>; + phy-handle = <&phy3>; + phy-connection-type = "rgmii-id"; + + mdio@520 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "fsl,gianfar-tbi"; + reg = <0x520 0x20>; + + tbi3: tbi-phy@11 { + reg = <0x11>; + device_type = "tbi-phy"; + }; + }; + }; + + serial0: serial@4500 { + cell-index = <0>; + device_type = "serial"; + compatible = "ns16550"; + reg = <0x4500 0x100>; + clock-frequency = <0>; + interrupts = <42 2>; + interrupt-parent = <&mpic>; + }; + + serial1: serial@4600 { + cell-index = <1>; + device_type = "serial"; + compatible = "ns16550"; + reg = <0x4600 0x100>; + clock-frequency = <0>; + interrupts = <28 2>; + interrupt-parent = <&mpic>; + }; + + mpic: pic@40000 { + interrupt-controller; + #address-cells = <0>; + #interrupt-cells = <2>; + reg = <0x40000 0x40000>; + compatible = "chrp,open-pic"; + device_type = "open-pic"; + }; + + global-utilities@e0000 { + compatible = "fsl,mpc8641-guts"; + reg = <0xe0000 0x1000>; + fsl,has-rstcr; + }; + }; + + pci0: pcie@fffe08000 { + cell-index = <0>; + compatible = "fsl,mpc8641-pcie"; + device_type = "pci"; + #interrupt-cells = <1>; + #size-cells = <2>; + #address-cells = <3>; + reg = <0x0f 0xffe08000 0x0 0x1000>; + bus-range = <0x0 0xff>; + ranges = <0x02000000 0x0 0xc0000000 0x0c 0x00000000 0x0 0x20000000 + 0x01000000 0x0 0x00000000 0x0f 0xffc00000 0x0 0x00010000>; + clock-frequency = <33333333>; + interrupt-parent = <&mpic>; + interrupts = <24 2>; + interrupt-map-mask = <0xff00 0 0 7>; + interrupt-map = < + /* IDSEL 0x11 func 0 - PCI slot 1 */ + 0x8800 0 0 1 &mpic 2 1 + 0x8800 0 0 2 &mpic 3 1 + 0x8800 0 0 3 &mpic 4 1 + 0x8800 0 0 4 &mpic 1 1 + + /* IDSEL 0x11 func 1 - PCI slot 1 */ + 0x8900 0 0 1 &mpic 2 1 + 0x8900 0 0 2 &mpic 3 1 + 0x8900 0 0 3 &mpic 4 1 + 0x8900 0 0 4 &mpic 1 1 + + /* IDSEL 0x11 func 2 - PCI slot 1 */ + 0x8a00 0 0 1 &mpic 2 1 + 0x8a00 0 0 2 &mpic 3 1 + 0x8a00 0 0 3 &mpic 4 1 + 0x8a00 0 0 4 &mpic 1 1 + + /* IDSEL 0x11 func 3 - PCI slot 1 */ + 0x8b00 0 0 1 &mpic 2 1 + 0x8b00 0 0 2 &mpic 3 1 + 0x8b00 0 0 3 &mpic 4 1 + 0x8b00 0 0 4 &mpic 1 1 + + /* IDSEL 0x11 func 4 - PCI slot 1 */ + 0x8c00 0 0 1 &mpic 2 1 + 0x8c00 0 0 2 &mpic 3 1 + 0x8c00 0 0 3 &mpic 4 1 + 0x8c00 0 0 4 &mpic 1 1 + + /* IDSEL 0x11 func 5 - PCI slot 1 */ + 0x8d00 0 0 1 &mpic 2 1 + 0x8d00 0 0 2 &mpic 3 1 + 0x8d00 0 0 3 &mpic 4 1 + 0x8d00 0 0 4 &mpic 1 1 + + /* IDSEL 0x11 func 6 - PCI slot 1 */ + 0x8e00 0 0 1 &mpic 2 1 + 0x8e00 0 0 2 &mpic 3 1 + 0x8e00 0 0 3 &mpic 4 1 + 0x8e00 0 0 4 &mpic 1 1 + + /* IDSEL 0x11 func 7 - PCI slot 1 */ + 0x8f00 0 0 1 &mpic 2 1 + 0x8f00 0 0 2 &mpic 3 1 + 0x8f00 0 0 3 &mpic 4 1 + 0x8f00 0 0 4 &mpic 1 1 + + /* IDSEL 0x12 func 0 - PCI slot 2 */ + 0x9000 0 0 1 &mpic 3 1 + 0x9000 0 0 2 &mpic 4 1 + 0x9000 0 0 3 &mpic 1 1 + 0x9000 0 0 4 &mpic 2 1 + + /* IDSEL 0x12 func 1 - PCI slot 2 */ + 0x9100 0 0 1 &mpic 3 1 + 0x9100 0 0 2 &mpic 4 1 + 0x9100 0 0 3 &mpic 1 1 + 0x9100 0 0 4 &mpic 2 1 + + /* IDSEL 0x12 func 2 - PCI slot 2 */ + 0x9200 0 0 1 &mpic 3 1 + 0x9200 0 0 2 &mpic 4 1 + 0x9200 0 0 3 &mpic 1 1 + 0x9200 0 0 4 &mpic 2 1 + + /* IDSEL 0x12 func 3 - PCI slot 2 */ + 0x9300 0 0 1 &mpic 3 1 + 0x9300 0 0 2 &mpic 4 1 + 0x9300 0 0 3 &mpic 1 1 + 0x9300 0 0 4 &mpic 2 1 + + /* IDSEL 0x12 func 4 - PCI slot 2 */ + 0x9400 0 0 1 &mpic 3 1 + 0x9400 0 0 2 &mpic 4 1 + 0x9400 0 0 3 &mpic 1 1 + 0x9400 0 0 4 &mpic 2 1 + + /* IDSEL 0x12 func 5 - PCI slot 2 */ + 0x9500 0 0 1 &mpic 3 1 + 0x9500 0 0 2 &mpic 4 1 + 0x9500 0 0 3 &mpic 1 1 + 0x9500 0 0 4 &mpic 2 1 + + /* IDSEL 0x12 func 6 - PCI slot 2 */ + 0x9600 0 0 1 &mpic 3 1 + 0x9600 0 0 2 &mpic 4 1 + 0x9600 0 0 3 &mpic 1 1 + 0x9600 0 0 4 &mpic 2 1 + + /* IDSEL 0x12 func 7 - PCI slot 2 */ + 0x9700 0 0 1 &mpic 3 1 + 0x9700 0 0 2 &mpic 4 1 + 0x9700 0 0 3 &mpic 1 1 + 0x9700 0 0 4 &mpic 2 1 + + // IDSEL 0x1c USB + 0xe000 0 0 1 &i8259 12 2 + 0xe100 0 0 2 &i8259 9 2 + 0xe200 0 0 3 &i8259 10 2 + 0xe300 0 0 4 &i8259 11 2 + + // IDSEL 0x1d Audio + 0xe800 0 0 1 &i8259 6 2 + + // IDSEL 0x1e Legacy + 0xf000 0 0 1 &i8259 7 2 + 0xf100 0 0 1 &i8259 7 2 + + // IDSEL 0x1f IDE/SATA + 0xf800 0 0 1 &i8259 14 2 + 0xf900 0 0 1 &i8259 5 2 + >; + + pcie@0 { + reg = <0 0 0 0 0>; + #size-cells = <2>; + #address-cells = <3>; + device_type = "pci"; + ranges = <0x02000000 0x0 0xc0000000 + 0x02000000 0x0 0xc0000000 + 0x0 0x20000000 + + 0x01000000 0x0 0x00000000 + 0x01000000 0x0 0x00000000 + 0x0 0x00010000>; + uli1575@0 { + reg = <0 0 0 0 0>; + #size-cells = <2>; + #address-cells = <3>; + ranges = <0x02000000 0x0 0xc0000000 + 0x02000000 0x0 0xc0000000 + 0x0 0x20000000 + 0x01000000 0x0 0x00000000 + 0x01000000 0x0 0x00000000 + 0x0 0x00010000>; + isa@1e { + device_type = "isa"; + #interrupt-cells = <2>; + #size-cells = <1>; + #address-cells = <2>; + reg = <0xf000 0 0 0 0>; + ranges = <1 0 0x01000000 0 0 + 0x00001000>; + interrupt-parent = <&i8259>; + + i8259: interrupt-controller@20 { + reg = <1 0x20 2 + 1 0xa0 2 + 1 0x4d0 2>; + interrupt-controller; + device_type = "interrupt-controller"; + #address-cells = <0>; + #interrupt-cells = <2>; + compatible = "chrp,iic"; + interrupts = <9 2>; + interrupt-parent = <&mpic>; + }; + + i8042@60 { + #size-cells = <0>; + #address-cells = <1>; + reg = <1 0x60 1 1 0x64 1>; + interrupts = <1 3 12 3>; + interrupt-parent = + <&i8259>; + + keyboard@0 { + reg = <0>; + compatible = "pnpPNP,303"; + }; + + mouse@1 { + reg = <1>; + compatible = "pnpPNP,f03"; + }; + }; + + rtc@70 { + compatible = + "pnpPNP,b00"; + reg = <1 0x70 2>; + }; + + gpio@400 { + reg = <1 0x400 0x80>; + }; + }; + }; + }; + + }; + + pci1: pcie@fffe09000 { + cell-index = <1>; + compatible = "fsl,mpc8641-pcie"; + device_type = "pci"; + #interrupt-cells = <1>; + #size-cells = <2>; + #address-cells = <3>; + reg = <0x0f 0xffe09000 0x0 0x1000>; + bus-range = <0x0 0xff>; + ranges = <0x02000000 0x0 0xc0000000 0x0c 0x20000000 0x0 0x20000000 + 0x01000000 0x0 0x00000000 0x0f 0xffc10000 0x0 0x00010000>; + clock-frequency = <33333333>; + interrupt-parent = <&mpic>; + interrupts = <25 2>; + interrupt-map-mask = <0xf800 0 0 7>; + interrupt-map = < + /* IDSEL 0x0 */ + 0x0000 0 0 1 &mpic 4 1 + 0x0000 0 0 2 &mpic 5 1 + 0x0000 0 0 3 &mpic 6 1 + 0x0000 0 0 4 &mpic 7 1 + >; + pcie@0 { + reg = <0 0 0 0 0>; + #size-cells = <2>; + #address-cells = <3>; + device_type = "pci"; + ranges = <0x02000000 0x0 0xc0000000 + 0x02000000 0x0 0xc0000000 + 0x0 0x20000000 + + 0x01000000 0x0 0x00000000 + 0x01000000 0x0 0x00000000 + 0x0 0x00010000>; + }; + }; +};
Signature · 1 line
1.6.0.6
PATCH 3/5 powerpc: make dma_window_* in pci_controller struct avail on 32b
Also, convert them to resource_size_t (which is unsigned long on 64-bit, so it's not a change there).
We will be using these on fsl 32b to indicate the start and size address of memory that the pci controller can actually reach - this is needed to determine if an address requires bounce buffering. For now, initialize them to a standard value; in the near future, the value will be calculated based on how the inbound windows are programmed.
--- arch/powerpc/include/asm/pci-bridge.h | 6 ++++-- arch/powerpc/sysdev/fsl_pci.c | 4 ++++ 2 files changed, 8 insertions(+), 2 deletions(-)
arch/powerpc/include/asm/pci-bridge.h +4 -2
@@ -140,10 +140,12 @@ struct pci_controller {struct resource io_resource;struct resource mem_resources[3 ];int global_number;/* PCI domain number */ + +resource_size_t dma_window_base_cur; +resource_size_t dma_window_size; +#ifdef CONFIG_PPC64 unsigned long buid; -unsigned long dma_window_base_cur; -unsigned long dma_window_size;void *private_data;#endif /* CONFIG_PPC64 */
arch/powerpc/sysdev/fsl_pci.c +4 -0
@@ -152,6 +152,10 @@ static void __init setup_pci_atmu(struct pci_controller *hose,out_be32 (&pci->piw[2 ].piwbar,0x00000000 );out_be32 (&pci->piw[2 ].piwar,PIWAR_2G ); +/* Save the base address and size covered by inbound window mappings */ + hose->dma_window_base_cur =0x00000000 ; + hose->dma_window_size =0x80000000 ; +iounmap (pci); }
Signature · 1 line
1.6.0.6
PATCH 4/5 powerpc: Add support for swiotlb on 32-bit
This patch includes the basic infrastructure to use swiotlb bounce buffering on 32-bit powerpc. It is not yet enabled on any platforms.
--- arch/powerpc/Kconfig | 2 +- arch/powerpc/include/asm/dma-mapping.h | 11 ++ arch/powerpc/include/asm/swiotlb.h | 24 +++++ arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/dma-swiotlb.c | 161 ++++++++++++++++++++++++++++++++ arch/powerpc/kernel/dma.c | 2 +- arch/powerpc/kernel/setup_32.c | 4 + 7 files changed, 203 insertions(+), 2 deletions(-) create mode 100644 arch/powerpc/include/asm/swiotlb.h create mode 100644 arch/powerpc/kernel/dma-swiotlb.c
arch/powerpc/Kconfig +1 -1
@@ -294,7 +294,7 @@ config IOMMU_HELPER config PPC_NEED_DMA_SYNC_OPS def_bool y - depends on NOT_COHERENT_CACHE + depends on (NOT_COHERENT_CACHE || SWIOTLB) config HOTPLUG_CPU bool "Support for enabling/disabling CPUs"
arch/powerpc/include/asm/dma-mapping.h +11 -0
@@ -15,9 +15,18 @@#include <linux/scatterlist.h> #include <linux/dma-attrs.h> #include <asm/io.h> +#include <asm/swiotlb.h> #define DMA_ERROR_CODE (~(dma_addr_t)0x0) +/* Some dma direct funcs must be visible for use in other dma_ops */ +extern void *dma_direct_alloc_coherent (struct device *dev,size_t size, +dma_addr_t *dma_handle,gfp_t flag); +extern void dma_direct_free_coherent (struct device *dev,size_t size, +void *vaddr,dma_addr_t dma_handle); + +extern unsigned long get_dma_direct_offset (struct device *dev); +#ifdef CONFIG_NOT_COHERENT_CACHE /* * DMA-consistent mapping functions for PowerPCs that don't support
@@ -76,6 +85,8 @@ struct dma_mapping_ops { dma_addr_t dma_address, size_t size, enum dma_data_direction direction, struct dma_attrs *attrs); +int (*addr_needs_map)(struct device *dev,dma_addr_t addr, +size_t size);#ifdef CONFIG_PPC_NEED_DMA_SYNC_OPS void (*sync_single_range_for_cpu)(struct device *hwdev, dma_addr_t dma_handle, unsigned long offset,
arch/powerpc/include/asm/swiotlb.h new file +24 -0
@@ -0,0 +1,24 @@ +/* +* Copyright (C) 2009 Becky Bruce, Freescale Semiconductor +* +* This program is free software; you can redistribute it and/or modify it +* under the terms of the GNU General Public License as published by the +* Free Software Foundation; either version 2 of the License, or (at your +* option) any later version. +* +*/ + +#ifndef __ASM_SWIOTLB_H +#define __ASM_SWIOTLB_H + +#include <linux/swiotlb.h> + +extern struct dma_mapping_ops swiotlb_dma_ops; +extern struct dma_mapping_ops swiotlb_pci_dma_ops; + +int swiotlb_arch_address_needs_mapping (struct device *,dma_addr_t , +size_t size); + +static inline void dma_mark_clean (void *addr,size_t size) {} + +#endif /* __ASM_SWIOTLB_H */
arch/powerpc/kernel/Makefile +1 -0
@@ -82,6 +82,7 @@ obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_PPC_UDBG_16550) += legacy_serial.o udbg_16550.o obj-$(CONFIG_STACKTRACE) += stacktrace.o +obj-$(CONFIG_SWIOTLB) += dma-swiotlb.o pci64-$(CONFIG_PPC64) += pci_dn.o isa-bridge.o obj-$(CONFIG_PCI) += pci_$(CONFIG_WORD_SIZE).o $(pci64-y) \
arch/powerpc/kernel/dma-swiotlb.c new file +161 -0
161 lines not on this page · raw message
arch/powerpc/kernel/dma.c +1 -1
@@ -19,7 +19,7 @@* default the offset is PCI_DRAM_OFFSET. */ -static unsigned long get_dma_direct_offset (struct device *dev) +unsigned long get_dma_direct_offset (struct device *dev) {if (dev)return (unsigned long )dev->archdata.dma_data;
arch/powerpc/kernel/setup_32.c +4 -0
@@ -332,6 +332,10 @@ void __init setup_arch(char **cmdline_p) ppc_md.setup_arch ();if ( ppc_md.progress ) ppc_md.progress ("arch: exit" ,0x3eab ); +/* Allow iotlb to do it's setup */ +#ifdef CONFIG_SWIOTLB +swiotlb_init (); +#endif paging_init ();/* Initialize the MMU context management stuff */
Signature · 1 line
1.6.0.6
PATCH 5/5 powerpc: Add 86xx support for SWIOTLB
Minor code to allow enabling swiotlb on mpc86xx, including Kconfig addition for SWIOTLB.
--- arch/powerpc/Kconfig | 10 ++++++++++ arch/powerpc/kernel/dma-swiotlb.c | 2 ++ arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 3 +++ 3 files changed, 15 insertions(+), 0 deletions(-)
arch/powerpc/Kconfig +10 -0
@@ -292,6 +292,16 @@ config IOMMU_VMERGE config IOMMU_HELPER def_bool PPC64 +config SWIOTLB + bool "SWIOTLB support" + depends on PPC_86xx + select IOMMU_HELPER + ---help--- + Support for IO bounce buffering for systems without an IOMMU. + This allows us to DMA to the full physical address space on + platforms where the size of a physical address is larger + than the bus address. + config PPC_NEED_DMA_SYNC_OPS def_bool y depends on (NOT_COHERENT_CACHE || SWIOTLB)
arch/powerpc/kernel/dma-swiotlb.c +2 -0
@@ -159,3 +159,5 @@ static int __init setup_bus_notifier(void)return 0 ; } + +machine_arch_initcall (mpc86xx_hpcn, setup_bus_notifier);
arch/powerpc/platforms/86xx/mpc86xx_hpcn.c +3 -0
@@ -88,6 +88,9 @@ mpc86xx_hpcn_setup_arch(void) ppc_md.pci_exclude_device = mpc86xx_exclude_device; +#ifdef CONFIG_SWIOTLB +set_pci_dma_ops (&swiotlb_pci_dma_ops); +#endif #endif printk ("MPC86xx HPCN board from Freescale Semiconductor\n" );
Signature · 1 line
1.6.0.6
Re: [PATCH 5/5] powerpc: Add 86xx support for SWIOTLB
Quoted 18 lines · Becky Bruce
Minor code to allow enabling swiotlb on mpc86xx, including
Kconfigaddition for SWIOTLB.Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/Kconfig | 10 ++++++++++ arch/powerpc/kernel/dma-swiotlb.c | 2 ++ arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 3 +++ 3 files changed, 15 insertions(+), 0 deletions(-)arch/powerpc/Kconfig +0
-0@@ -292,6 +292,16 @@ config IOMMU_VMERGEconfig
IOMMU_HELPER
def_boolPPC64
arch/powerpc/Kconfig config IOMMU_VMERGE · line 304
Quoted 10 lines
arch/powerpc/Kconfig +10
-0+config SWIOTLB + bool "SWIOTLB support" + depends on PPC_86xx + select IOMMU_HELPER + ---help--- + Support for IO bounce buffering for systems without an IOMMU. + This allows us to DMA to the full physical address space on + platforms where the size of a physical address is larger + than the bus address. +
As stated on previous patch, move the bulk of this into 4/5.
Quoted 9 lines
config
PPC_NEED_DMA_SYNC_OPSdef_bool y depends on (NOT_COHERENT_CACHE || SWIOTLB)arch/powerpc/kernel/dma-swiotlb.c +0
-0
arch/powerpc/kernel/dma-swiotlb.c static int __init setup_bus_notifier(void) · line 163
Quoted 5 lines
arch/powerpc/kernel/dma-swiotlb.c +2
-0@@ -159,3 +159,5 @@ static int __init setup_bus_notifier(void)return 0 ; } + +machine_arch_initcall (mpc86xx_hpcn, setup_bus_notifier);
Hmm, not sure what we chatted about here, but I don't want to have to add every board into this file to register the bus notifiers.
Quoted 22 lines
arch/powerpc/platforms/86xx/mpc86xx_hpcn.c +3
-0@@ -88,6 +88,9 @@ mpc86xx_hpcn_setup_arch(void) ppc_md.pci_exclude_device = mpc86xx_exclude_device; +#ifdef CONFIG_SWIOTLB +set_pci_dma_ops (&swiotlb_pci_dma_ops); +#endif #endif printk ("MPC86xx HPCN board from Freescale Semiconductor\n" );Signature · 6 lines
1.6.0.6 _______________________________________________ Linuxppc-dev mailing list [email protected] https://ozlabs.org/mailman/listinfo/linuxppc-dev
Re: [PATCH 5/5] powerpc: Add 86xx support for SWIOTLB
Quoted 21 lines · Kumar Gala
Quoted 18 lines · Becky Bruce
Minor code to allow enabling swiotlb on mpc86xx, including
Kconfigaddition for SWIOTLB.Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/Kconfig | 10 ++++++++++ arch/powerpc/kernel/dma-swiotlb.c | 2 ++ arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 3 +++ 3 files changed, 15 insertions(+), 0 deletions(-)arch/powerpc/Kconfig +0
-0@@ -292,6 +292,16 @@ config IOMMU_VMERGEconfig
IOMMU_HELPER
def_boolPPC64
Quoted 12 lines
arch/powerpc/Kconfig config IOMMU_VMERGE · line 304
Quoted 10 lines
arch/powerpc/Kconfig +10
-0+config SWIOTLB + bool "SWIOTLB support" + depends on PPC_86xx + select IOMMU_HELPER + ---help--- + Support for IO bounce buffering for systems without an IOMMU. + This allows us to DMA to the full physical address space on + platforms where the size of a physical address is larger + than the bus address. +As stated on previous patch, move the bulk of this into 4/5.
Yep.
Quoted 12 lines
Quoted 12 lines
config
PPC_NEED_DMA_SYNC_OPSdef_bool y depends on (NOT_COHERENT_CACHE || SWIOTLB)arch/powerpc/kernel/dma-swiotlb.c +0
-0@@ -159,3 +159,5 @@ static int __init setup_bus_notifier(void)
Quoted 6 lines
arch/powerpc/kernel/dma-swiotlb.c static int __init setup_bus_notifier(void) · line 163
Quoted 4 lines
arch/powerpc/kernel/dma-swiotlb.c +2
-0return 0 ; } + +machine_arch_initcall (mpc86xx_hpcn, setup_bus_notifier);Hmm, not sure what we chatted about here, but I don't want to have to add every board into this file to register the bus notifiers.
We talked about this, and this was what we decided on - I don't really like the idea, either, but there's a lot of precedent for it. I'd like to do this differently, but I"m not sure what the solution is - we'd need to look into that more (or perhaps someone here will have some sage advice).
Cheers,
B
Re: [PATCH 5/5] powerpc: Add 86xx support for SWIOTLB
Quoted 21 lines · Becky Bruce
Quoted 12 lines · Kumar Gala
Quoted 11 lines · Becky Bruce
config
PPC_NEED_DMA_SYNC_OPSdef_bool y depends on (NOT_COHERENT_CACHE || SWIOTLB)arch/powerpc/ renamed from arch/powerpc/kernel/dma-swiotlb.c +0
-0
kernel/dma-swiotlb.c
index 29a68e6..3065d03 100644arch/powerpc/kernel/dma-swiotlb.c +0
-0@@ -159,3 +159,5 @@ static int __init setup_bus_notifier(void)Quoted 7 lines
arch/powerpc/kernel/dma-swiotlb.c static int __init setup_bus_notifier(void) · line 163
Quoted 4 lines
arch/powerpc/kernel/dma-swiotlb.c +2
-0return 0 ; } + +machine_arch_initcall (mpc86xx_hpcn, setup_bus_notifier);Hmm, not sure what we chatted about here, but I don't want to have to add every board into this file to register the bus notifiers.
Quoted 5 lines
We talked about this, and this was what we decided on - I don't really like the idea, either, but there's a lot of precedent for it. I'd like to do this differently, but I"m not sure what the solution is - we'd need to look into that more (or perhaps someone here will have some sage advice).
Give it a better name, export it, and call it from the board setup file?
Actually what depends on the board anyway? It's just the dma window config in the pci_controller isn't it? So maybe you can always call the notifier, and if the dma mask isn't 36 bits, and the controller has the window configured then you use swiotlb?
cheers
Re: [PATCH 4/5] powerpc: Add support for swiotlb on 32-bit
In reply to Becky BruceQuoted 26 lines · Becky Bruce
This patch includes the basic infrastructure to use swiotlb bounce buffering on 32-bit powerpc. It is not yet enabled on any platforms.
Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/Kconfig | 2 +- arch/powerpc/include/asm/dma-mapping.h | 11 ++ arch/powerpc/include/asm/swiotlb.h | 24 +++++ arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/dma-swiotlb.c | 161 +++++++++++++++++++++++ +++++++++ arch/powerpc/kernel/dma.c | 2 +- arch/powerpc/kernel/setup_32.c | 4 + 7 files changed, 203 insertions(+), 2 deletions(-)create mode 100644
arch/powerpc/include/asm/swiotlb.hcreate mode 100644arch/powerpc/kernel/dma-swiotlb.carch/powerpc/Kconfig +0
-0@@ -294,7 +294,7 @@ config IOMMU_HELPERconfig
PPC_NEED_DMA_SYNC_OPS
def_booly
arch/powerpc/Kconfig config IOMMU_HELPER · line 297
Quoted 2 lines
arch/powerpc/Kconfig +1
-1- depends on NOT_COHERENT_CACHE + depends on (NOT_COHERENT_CACHE || SWIOTLB)
This isn't right, since you haven't introduced the SWIOTLB Kconfig.
Why don't we just put the SWIOTLB Kconfig option in here and default it no (and remove the dep on 86xx).
Quoted 24 lines
config
HOTPLUG_CPU
bool "Support for enabling/disabling CPUs"arch/powerpc/kernel/dma-swiotlb.c new file +6
-0@@ -0,0 +1,161 @@ +/* +* Contains routines needed to support swiotlb for ppc. +* +* Copyright (C) 2009 Becky Bruce, Freescale Semiconductor +* +* This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ */
[snip]
Quoted 4 lines
arch/powerpc/kernel/dma-swiotlb.c new file +3
-0+ +static int ppc_swiotlb_bus_notify (struct notifier_block *nb, +unsigned long action,void *data)
arch/powerpc/kernel/dma-swiotlb.c line 139
Quoted 10 lines
arch/powerpc/kernel/dma-swiotlb.c new file +10
-0+{ +struct device *dev = data; + +/* We are only intereted in device addition */ +if (action !=BUS_NOTIFY_ADD_DEVICE ) +return 0 ; + +if (dma_get_mask (dev) <DMA_BIT_MASK (36 )) +set_dma_ops (dev, &swiotlb_dma_ops); +
this isn't right. Isn't possible for a PCI dev to have a DMA_BIT_MASK(64) but us still not be able to dma to it? Also, I dont like the assumption of 36-bit physical here.
Quoted 23 lines
arch/powerpc/kernel/dma-swiotlb.c new file +22
-0+return NOTIFY_DONE ; +} + +static struct notifier_block ppc_swiotlb_plat_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +}; + +static struct notifier_block ppc_swiotlb_of_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +}; + +static int __init setup_bus_notifier (void ) +{ +bus_register_notifier (&platform_bus_type, + &ppc_swiotlb_plat_bus_notifier); +bus_register_notifier (&of_platform_bus_type, + &ppc_swiotlb_of_bus_notifier); + +return 0 ; +}
Re: [PATCH 4/5] powerpc: Add support for swiotlb on 32-bit
Quoted 35 lines · Kumar Gala
Quoted 26 lines · Becky Bruce
This patch includes the basic infrastructure to use swiotlb bounce buffering on 32-bit powerpc. It is not yet enabled on any platforms.
Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/Kconfig | 2 +- arch/powerpc/include/asm/dma-mapping.h | 11 ++ arch/powerpc/include/asm/swiotlb.h | 24 +++++ arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/dma-swiotlb.c | 161 ++++++++++++++++++++++ ++++++++++ arch/powerpc/kernel/dma.c | 2 +- arch/powerpc/kernel/setup_32.c | 4 + 7 files changed, 203 insertions(+), 2 deletions(-)create mode 100644
arch/powerpc/include/asm/swiotlb.hcreate mode 100644arch/powerpc/kernel/dma-swiotlb.carch/powerpc/Kconfig +0
-0@@ -294,7 +294,7 @@ config IOMMU_HELPERconfig
PPC_NEED_DMA_SYNC_OPS
def_boolyarch/powerpc/Kconfig config IOMMU_HELPER · line 297
Quoted 2 lines
arch/powerpc/Kconfig +1
-1- depends on NOT_COHERENT_CACHE + depends on (NOT_COHERENT_CACHE || SWIOTLB)This isn't right, since you haven't introduced the SWIOTLB
Kconfig.
Quoted 1 line
Why don't we just put the SWIOTLB
Kconfigoption in here and default it no (and remove the dep on 86xx).
Sure. I'll probably add a comment or something though so people don't think they can just enable it on anything and expect it to work. Too many people seem to read the config files and decide things are possible that actually aren't :)
Quoted 30 lines
Quoted 26 lines
config
HOTPLUG_CPU
bool "Support for enabling/disabling CPUs"arch/powerpc/kernel/dma-swiotlb.c new file +6
-0@@ -0,0 +1,161 @@ +/* +* Contains routines needed to support swiotlb for ppc. +* +* Copyright (C) 2009 Becky Bruce, Freescale Semiconductor +* +* This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ */[snip]
Quoted 1 line
Quoted 15 lines
arch/powerpc/kernel/dma-swiotlb.c line 139
Quoted 13 lines
arch/powerpc/kernel/dma-swiotlb.c new file +13
-0+ +static int ppc_swiotlb_bus_notify (struct notifier_block *nb, +unsigned long action,void *data) +{ +struct device *dev = data; + +/* We are only intereted in device addition */ +if (action !=BUS_NOTIFY_ADD_DEVICE ) +return 0 ; + +if (dma_get_mask (dev) <DMA_BIT_MASK (36 )) +set_dma_ops (dev, &swiotlb_dma_ops); +this isn't right. Isn't possible for a PCI dev to have a
DMA_BIT_MASK(64)but us still not be able to dma to it? Also, I dont like the assumption of 36-bit physical here.
You're right - this is just handling the basic case and for now I'm assuming that we don't bounce 64-bit devices (or any device that can handle 36 bits). We'll need to talk in more detail about a solution to that problem, because knowing if a 64b dev *might* need to bounce (which is all this is really saying) requires more information. We also don't really have infrastructure to deal with holes in the visible dev memory, and I don't know if we want to handle that as well. I don't want to set the dma_ops to swiotlb unless we have to because there's a slight perf hit in running all the checks to see if an address needs to be bounced.
Thanks,
B
Re: [PATCH 4/5] powerpc: Add support for swiotlb on 32-bit
Quoted 95 lines · On Mon, 20 Apr 2009 13:03:20 -0500 Becky Bruce <[email protected]>…
Quoted 35 lines · Kumar Gala
Quoted 26 lines · Becky Bruce
This patch includes the basic infrastructure to use swiotlb bounce buffering on 32-bit powerpc. It is not yet enabled on any platforms.
Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/Kconfig | 2 +- arch/powerpc/include/asm/dma-mapping.h | 11 ++ arch/powerpc/include/asm/swiotlb.h | 24 +++++ arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/dma-swiotlb.c | 161 ++++++++++++++++++++++ ++++++++++ arch/powerpc/kernel/dma.c | 2 +- arch/powerpc/kernel/setup_32.c | 4 + 7 files changed, 203 insertions(+), 2 deletions(-)create mode 100644
arch/powerpc/include/asm/swiotlb.hcreate mode 100644arch/powerpc/kernel/dma-swiotlb.carch/powerpc/Kconfig +0
-0@@ -294,7 +294,7 @@ config IOMMU_HELPERconfig
PPC_NEED_DMA_SYNC_OPS
def_boolyarch/powerpc/Kconfig config IOMMU_HELPER · line 297
Quoted 2 lines
arch/powerpc/Kconfig +1
-1- depends on NOT_COHERENT_CACHE + depends on (NOT_COHERENT_CACHE || SWIOTLB)This isn't right, since you haven't introduced the SWIOTLB
Kconfig.Quoted 2 lines
Why don't we just put the SWIOTLB
Kconfigoption in here and default it no (and remove the dep on 86xx).Sure. I'll probably add a comment or something though so people don't think they can just enable it on anything and expect it to work. Too many people seem to read the config files and decide things are possible that actually aren't :)
Quoted 40 lines
Quoted 27 lines
config
HOTPLUG_CPU
bool "Support for enabling/disabling CPUs"arch/powerpc/ renamed from arch/powerpc/kernel/dma-swiotlb.c +0
-0
kernel/dma-swiotlb.c
new file mode 100644
index 0000000..29a68e6arch/powerpc/kernel/dma-swiotlb.c new file +6
-0@@ -0,0 +1,161 @@ +/* +* Contains routines needed to support swiotlb for ppc. +* +* Copyright (C) 2009 Becky Bruce, Freescale Semiconductor +* +* This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ */[snip]
Quoted 10 lines
arch/powerpc/kernel/dma-swiotlb.c new file +9
-0+ +static int ppc_swiotlb_bus_notify (struct notifier_block *nb, +unsigned long action,void *data) +{ +struct device *dev = data; + +/* We are only intereted in device addition */ +if (action !=BUS_NOTIFY_ADD_DEVICE ) +return 0 ;Quoted 8 lines
arch/powerpc/kernel/dma-swiotlb.c line 139
Quoted 4 lines
arch/powerpc/kernel/dma-swiotlb.c new file +4
-0+ +if (dma_get_mask (dev) <DMA_BIT_MASK (36 )) +set_dma_ops (dev, &swiotlb_dma_ops); +this isn't right. Isn't possible for a PCI dev to have a
DMA_BIT_MASK(64)but us still not be able to dma to it? Also, I dont like the assumption of 36-bit physical here.
Quoted 10 lines
You're right - this is just handling the basic case and for now I'm assuming that we don't bounce 64-bit devices (or any device that can
handle 36 bits). We'll need to talk in more detail about a solution
to that problem, because knowing if a 64b dev *might* need to bounce
(which is all this is really saying) requires more information. We
also don't really have infrastructure to deal with holes in the visible dev memory, and I don't know if we want to handle that as well. I don't want to set thedma_opsto swiotlb unless we have to because there's a slight perf hit in running all the checks to see if an address needs to be bounced.
I think that you always just set the dma_ops to swiotlb. It's the cleanest solution. Do you really see the performance drop due to the checking?
Re: [PATCH 4/5] powerpc: Add support for swiotlb on 32-bit
In reply to Becky BruceQuoted 24 lines · Becky Bruce
arch/powerpc/kernel/dma-swiotlb.c new file +24
-0+static int ppc_swiotlb_bus_notify (struct notifier_block *nb, +unsigned long action,void *data) +{ +struct device *dev = data; + +/* We are only intereted in device addition */ +if (action !=BUS_NOTIFY_ADD_DEVICE ) +return 0 ; + +if (dma_get_mask (dev) <DMA_BIT_MASK (36 )) +set_dma_ops (dev, &swiotlb_dma_ops); + +return NOTIFY_DONE ; +} + +static struct notifier_block ppc_swiotlb_plat_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +}; + +static struct notifier_block ppc_swiotlb_of_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +};
arch/powerpc/kernel/dma-swiotlb.c line 161
Quoted 10 lines
arch/powerpc/kernel/dma-swiotlb.c new file +10
-0+ +static int __init setup_bus_notifier (void ) +{ +bus_register_notifier (&platform_bus_type, + &ppc_swiotlb_plat_bus_notifier); +bus_register_notifier (&of_platform_bus_type, + &ppc_swiotlb_of_bus_notifier); + +return 0 ; +}
I think we should move all this into the platform code for now. I don't like having to duplicate it but that gives us the proper flexibility for now.
- k
Re: [PATCH 4/5] powerpc: Add support for swiotlb on 32-bit
Quoted 31 lines · Kumar Gala
Quoted 28 lines · Becky Bruce
arch/powerpc/kernel/dma-swiotlb.c new file +28
-0+static int ppc_swiotlb_bus_notify (struct notifier_block *nb, +unsigned long action,void *data) +{ +struct device *dev = data; + +/* We are only intereted in device addition */ +if (action !=BUS_NOTIFY_ADD_DEVICE ) +return 0 ; + +if (dma_get_mask (dev) <DMA_BIT_MASK (36 )) +set_dma_ops (dev, &swiotlb_dma_ops); + +return NOTIFY_DONE ; +} + +static struct notifier_block ppc_swiotlb_plat_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +}; + +static struct notifier_block ppc_swiotlb_of_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +}; + +static int __init setup_bus_notifier (void ) +{ + bus_register_notifier(&platform_bus_type,
Quoted 8 lines
arch/powerpc/kernel/dma-swiotlb.c line 161
Quoted 6 lines
arch/powerpc/kernel/dma-swiotlb.c new file +6
-0+ &ppc_swiotlb_plat_bus_notifier); +bus_register_notifier (&of_platform_bus_type, + &ppc_swiotlb_of_bus_notifier); + +return 0 ; +}I think we should move all this into the platform code for now. I don't like having to duplicate it but that gives us the proper flexibility for now.
Ugh, gross. I'd like to think about this some more.
-B
Re: [PATCH 4/5] powerpc: Add support for swiotlb on 32-bit
Quoted 43 lines · Becky Bruce
Quoted 31 lines · Kumar Gala
Quoted 28 lines · Becky Bruce
arch/powerpc/kernel/dma-swiotlb.c new file +28
-0+static int ppc_swiotlb_bus_notify (struct notifier_block *nb, +unsigned long action,void *data) +{ +struct device *dev = data; + +/* We are only intereted in device addition */ +if (action !=BUS_NOTIFY_ADD_DEVICE ) +return 0 ; + +if (dma_get_mask (dev) <DMA_BIT_MASK (36 )) +set_dma_ops (dev, &swiotlb_dma_ops); + +return NOTIFY_DONE ; +} + +static struct notifier_block ppc_swiotlb_plat_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +}; + +static struct notifier_block ppc_swiotlb_of_bus_notifier = { + .notifier_call = ppc_swiotlb_bus_notify, + .priority =0 , +}; + +static int __init setup_bus_notifier (void ) +{ + bus_register_notifier(&platform_bus_type,Quoted 8 lines
arch/powerpc/kernel/dma-swiotlb.c line 161
Quoted 6 lines
arch/powerpc/kernel/dma-swiotlb.c new file +6
-0+ &ppc_swiotlb_plat_bus_notifier); +bus_register_notifier (&of_platform_bus_type, + &ppc_swiotlb_of_bus_notifier); + +return 0 ; +}I think we should move all this into the platform code for now. I don't like having to duplicate it but that gives us the proper flexibility for now.
Quoted 1 line
Ugh, gross. I'd like to think about this some more.
I'm suggesting we do it one for FSL in fsl_soc.c, the 4xx guys can do it once, etc. Since the behavior desired is going to be a bit unique to SoCs/chipsets.
- k
Re: [PATCH 4/5] powerpc: Add support for swiotlb on 32-bit
Quoted 1 line · Kumar Gala
I'm suggesting we do it one for FSL in
fsl_soc.c, the 4xx guys can do it once, etc. Since the behavior desired is going to be a bit unique to SoCs/chipsets.
Perhaps we should have a dma_mask in platform/of_platform device structs? The driver knows best how many bits it can shove into a DMA address register, and it would let us avoid hardcoding 36 bits into this code.
What other platform-specific behavior do you have in mind? Could we supply a default implementation that platforms can override if they need something weird, rather than duplicating it per soc family?
-Scott
Re: [PATCH 3/5] powerpc: make dma_window_* in pci_controller struct avail on 32b
In reply to Becky BruceQuoted 10 lines · Becky Bruce
Also, convert them to
resource_size_t(which is unsigned long on 64-bit, so it's not a change there).We will be using these on fsl 32b to indicate the start and size address of memory that the pci controller can actually reach - this is needed to determine if an address requires bounce buffering. For now, initialize them to a standard value; in the near future, the value will be calculated based on how the inbound windows are programmed.
Quoted 5 lines
Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/include/asm/pci-bridge.h | 6 ++++-- arch/powerpc/sysdev/fsl_pci.c | 4 ++++ 2 files changed, 8 insertions(+), 2 deletions(-)
Ben, your ack on this patch would be appreciated as I've got some other FSL PCI patches that I want to fold this into my tree for.
- k
Re: [PATCH 2/5] powerpc: Add 36-bit device tree for mpc8641hpcn
In reply to Becky BruceQuoted 25 lines · Becky Bruce
The new dts places most of the devices in physical address space above 32-bits, which allows us to have more than 4GB of RAM present.
Signed-off-by Becky Bruce <[email protected]>--- arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts | 597 ++++++++++++++++++++++++++++ 1 files changed, 597 insertions(+), 0 deletions(-) create mode 100644 arch/powerpc/boot/dts/mpc8641_hpcn_36b.dtsarch/powerpc/boot/dts/mpc8641_hpcn_36b.dts new file +10
-0@@ -0,0 +1,597 @@ +/* + * MPC8641 HPCN Device Tree Source + * + * Copyright 2006 Freescale Semiconductor Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */
[snip]
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts line 108
Quoted 5 lines
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts new file +5
-0+ soc8641@fffe00000 { + #address-cells = <1>; + #size-cells = <1>; + device_type = "soc"; + compatible = "simple-bus";
Uh, you definitely need something more specific in the compatible property before "simple-bus".
Signature · 4 lines
David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson
Re: [PATCH 2/5] powerpc: Add 36-bit device tree for mpc8641hpcn
Quoted 32 lines · David Gibson
Quoted 29 lines · Becky Bruce
The new dts places most of the devices in physical address space above 32-bits, which allows us to have more than 4GB of RAM present.
Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts| 597 ++++++++++++++++++
++++++++++1 files changed, 597 insertions(+), 0 deletions(-)create mode 100644
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dtsarch/powerpc/boot/dts/mpc8641_hpcn_36b.dts new file +6
-0@@ -0,0 +1,597 @@ +/* + * MPC8641 HPCN Device Tree Source + * + * Copyright 2006 Freescale Semiconductor Inc. + * + * This program is free software; you can redistribute it and/ormodify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */[snip]
Quoted 8 lines
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts line 108
Quoted 5 lines
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts new file +5
-0+ soc8641@fffe00000 { + #address-cells = <1>; + #size-cells = <1>; + device_type = "soc"; + compatible = "simple-bus";Uh, you definitely need something more specific in the compatible property before "simple-bus".
This is a copy of the existing mpc8641hpcn dts file, with just physical address changes, so if there's a problem here it definitely exists in the current 8641hpcn dts, and possibly other dts files as well. I think the correct solution is for me to go look at that .dts (and any others that may be similar), and put out a followup to fix them all.
Thanks,
Becky
Re: [PATCH 2/5] powerpc: Add 36-bit device tree for mpc8641hpcn
Quoted 45 lines · Becky Bruce
Quoted 33 lines · David Gibson
Quoted 30 lines · Becky Bruce
The new dts places most of the devices in physical address space above 32-bits, which allows us to have more than 4GB of RAM present.
Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts| 597 ++++++++++++++++++
++++++++++1 files changed, 597 insertions(+), 0 deletions(-)create mode 100644
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dtsarch/ renamed from arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts +0
-0
powerpc/boot/dts/mpc8641_hpcn_36b.dts
new file mode 100644
index 0000000..baa3dbaarch/powerpc/boot/dts/mpc8641_hpcn_36b.dts new file +6
-0@@ -0,0 +1,597 @@ +/* + * MPC8641 HPCN Device Tree Source + * + * Copyright 2006 Freescale Semiconductor Inc. + * + * This program is free software; you can redistribute it and/ormodify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */[snip]
Quoted 8 lines
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts line 108
Quoted 5 lines
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts new file +5
-0+ soc8641@fffe00000 { + #address-cells = <1>; + #size-cells = <1>; + device_type = "soc"; + compatible = "simple-bus";Uh, you definitely need something more specific in the compatible property before "simple-bus".
Quoted 5 lines
This is a copy of the existing mpc8641hpcn dts file, with just physical address changes, so if there's a problem here it definitely exists in the current 8641hpcn dts, and possibly other dts files as well. I think the correct solution is for me to go look at that .dts (and any others that may be similar), and put out a followup to fix them all.
Ok, fair enough.
Signature · 4 lines
David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson
Kumar Gala
AppliedRe: [PATCH 2/5] powerpc: Add 36-bit device tree for mpc8641hpcn
In reply to Becky BruceQuoted 9 lines · Becky Bruce
The new dts places most of the devices in physical address space above 32-bits, which allows us to have more than 4GB of RAM present.
Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts| 597 +++++++++++++++++++
+++++++++1 files changed, 597 insertions(+), 0 deletions(-)create mode 100644
arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts
applied to next. We can deal with any simple-bus issues in a follow up patch.
- k
Re: [PATCH 1/5] powerpc: Use sg->dma_length in sg_dma_len() macro on 32-bit
In reply to Becky BruceQuoted 17 lines · Becky Bruce
Currently, the 32-bit code uses
sg->lengthinstead ofsg->dma_lentghto reportsg_dma_len. However, since the default dma code for 32-bit (thedma_directcase) setsdma_lengthand length to the same thing, we should be able to usedma_lengththere as well. This gets rid of some 32-vs-64-bit ifdefs, and is needed by the swiotlb code which actually distinguishes betweendma_lengthand length.Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/include/asm/scatterlist.h| 6 +-----1 files changed, 1 insertions(+), 5 deletions(-)arch/powerpc/include/asm/scatterlist.h +0
-0
arch/powerpc/include/asm/scatterlist.h struct scatterlist { · line 22
Quoted 2 lines
arch/powerpc/include/asm/scatterlist.h +0
-0@@ -21,7 +21,7 @@ struct scatterlist {unsigned int offset;unsigned int length;
can we get rid of length?
Quoted 21 lines
arch/powerpc/include/asm/scatterlist.h +1
-5-/* For TCE support */ +/* For TCE or SWIOTLB support */ dma_addr_t dma_address;u32 dma_length; };@@ -34,11 +34,7 @@ struct scatterlist {* is 0. */ #define sg_dma_address (sg) ((sg)->dma_address) -#ifdef __powerpc64__#define sg_dma_len (sg) ((sg)->dma_length) -#else -#define sg_dma_len (sg) ((sg)->length) -#endif #ifdef __powerpc64__#define ISA_DMA_THRESHOLD (~0UL)Signature · 1 line
1.6.0.6
- k
Re: [PATCH 1/5] powerpc: Use sg->dma_length in sg_dma_len() macro on 32-bit
Quoted 15 lines · On Mon, 20 Apr 2009 15:06:16 -0500 Kumar Gala <[email protected]> wrote:
Quoted 12 lines · Becky Bruce
Currently, the 32-bit code uses
sg->lengthinstead ofsg->dma_lentghto reportsg_dma_len. However, since the default dma code for 32-bit (thedma_directcase) setsdma_lengthand length to the same thing, we should be able to usedma_lengththere as well. This gets rid of some 32-vs-64-bit ifdefs, and is needed by the swiotlb code which actually distinguishes betweendma_lengthand length.Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/include/asm/scatterlist.h| 6 +-----1 files changed, 1 insertions(+), 5 deletions(-)
Quoted 10 lines
arch/powerpc/include/asm/scatterlist.h struct scatterlist { · line 22
Quoted 8 lines
arch/powerpc/ renamed from arch/powerpc/include/asm/scatterlist.h +0
-0
include/asm/scatterlist.h
index fcf7d55..912bf59 100644arch/powerpc/include/asm/scatterlist.h +0
-0@@ -21,7 +21,7 @@ struct scatterlist {unsigned int offset;unsigned int length;can we get rid of length?
You can't.
Quoted 25 lines
Quoted 21 lines
arch/powerpc/include/asm/scatterlist.h +1
-5-/* For TCE support */ +/* For TCE or SWIOTLB support */ dma_addr_t dma_address;u32 dma_length; };@@ -34,11 +34,7 @@ struct scatterlist {* is 0. */ #define sg_dma_address (sg) ((sg)->dma_address) -#ifdef __powerpc64__#define sg_dma_len (sg) ((sg)->dma_length) -#else -#define sg_dma_len (sg) ((sg)->length) -#endif #ifdef __powerpc64__#define ISA_DMA_THRESHOLD (~0UL)Signature · 1 line
1.6.0.6- k
Re: [PATCH 1/5] powerpc: Use sg->dma_length in sg_dma_len() macro on 32-bit
In reply to Kumar GalaQuoted 15 lines · Kumar Gala
Quoted 12 lines · Becky Bruce
Currently, the 32-bit code uses
sg->lengthinstead ofsg->dma_lentghto reportsg_dma_len. However, since the default dma code for 32-bit (thedma_directcase) setsdma_lengthand length to the same thing, we should be able to usedma_lengththere as well. This gets rid of some 32-vs-64-bit ifdefs, and is needed by the swiotlb code which actually distinguishes betweendma_lengthand length.Signed-off-by Becky Bruce <[email protected]>---
arch/powerpc/include/asm/scatterlist.h| 6 +-----1 files changed, 1 insertions(+), 5 deletions(-)
Quoted 9 lines
arch/powerpc/include/asm/scatterlist.h struct scatterlist { · line 22
Quoted 7 lines
arch/powerpc/include/asm/scatterlist.h +0
-0@@ -21,7 +21,7 @@ struct scatterlist {unsigned int offset;unsigned int length;can we get rid of length?
No - they're both used by the iotlb code and are conceptually different. "dma_length" can get set to less than "length" to indicate that something went wrong with a dma request - the iotlb code sets it to 0 if we can't allocate a bounce buffer, for example. It probably has other uses as well - this is just the one I'm familiar with. In most cases they are equal.
Cheers,
B
Re: [PATCH 0/5] Allow swiotlb use on ppc/mpc86xx
In reply to Becky BruceFAIL.
Here's the text that should have been there :)
-B
This patch series allows the use of swiotlb on mpc86xx, with 85xx upport soon to follow. The most notable change here is the addition of addr_needs_map to dma_ops. This is used to tell if a device can reach an address without bounce buffering. I believe that patches will be coming out soon to make dma_ops non-arch-specific, and I expect to see something similar there as well.
I've also changed sg_dma_len to be the same on both 32 and 64-bit. We now look at sg->dma_length in both cases (see patch log).
The dma_window* elements in the pci_controller struct are hoisted out of an #ifdef CONFIG_PPC64 so we can use them to store off information about how much memory is mapped via the inbound pci windows.
A dma-swiotlb.c is created in arch/powerpc/kernel to contain all the platform-specific iotlb code.
Finally, hooks are provided to allow enabling of this feature on 86xx via Kconfig, and a 36-bit device tree for 8641hpcn is provided.
Ye Olde Diffstat:
arch/powerpc/Kconfig | 2 +- arch/powerpc/boot/dts/mpc8641_hpcn_36b.dts | 597 +++++++++++++++++++ +++++++++ arch/powerpc/include/asm/dma-mapping.h | 11 + arch/powerpc/include/asm/pci-bridge.h | 6 +- arch/powerpc/include/asm/scatterlist.h | 6 +- arch/powerpc/include/asm/swiotlb.h | 24 ++ arch/powerpc/kernel/Makefile | 1 + arch/powerpc/kernel/dma-swiotlb.c | 161 ++++++++ arch/powerpc/kernel/dma.c | 2 +- arch/powerpc/kernel/setup_32.c | 4 + arch/powerpc/sysdev/fsl_pci.c | 4 + 11 files changed, 809 insertions(+), 9 deletions(-)
Cheers,
Becky