From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Jon Nettleton Date: Fri, 10 Dec 2021 11:24:01 +0100 Subject: [PATCH 01/13] PCI: layerscape: Add LX2160a MCFG quirks for ECAM errata The PCIe controller in Layerscape LX2160a SoC is not 100% ECAM-compliant. For both V1 and V2 of the SOC which have different PCIe implementations the devices behind the bus can be enumerated via ECAM, however the root port is only accessible via the CCSR address space. By default the firmware only exposes the devices so that most PCIe devices will work out of the box on most distributions, however some users may want to also have the root port exposed as well, especially if working with SR-IOV. This quirk will work with the default firmware as a normal ecam setup, but if the firmware exposes the root port as bus 0 (the default) then this quirk will also allow access to a more traditional PCIe layout. Signed-off-by: Jon Nettleton --- drivers/acpi/pci_mcfg.c | 10 +++ drivers/pci/controller/Makefile | 1 + drivers/pci/controller/pcie-layerscape-ecam.c | 89 +++++++++++++++++++ include/linux/pci-ecam.h | 1 + 4 files changed, 101 insertions(+) create mode 100644 drivers/pci/controller/pcie-layerscape-ecam.c diff --git a/drivers/acpi/pci_mcfg.c b/drivers/acpi/pci_mcfg.c index 860014b89b8e..c80b94af1601 100644 --- a/drivers/acpi/pci_mcfg.c +++ b/drivers/acpi/pci_mcfg.c @@ -55,6 +55,16 @@ static struct mcfg_fixup mcfg_quirks[] = { AL_ECAM("GRAVITON", 0, 6, &al_pcie_ops), AL_ECAM("GRAVITON", 0, 7, &al_pcie_ops), +#define NXP_ECAM(seg) \ + { "NXP ", "LX2160 ", 1, seg, MCFG_BUS_ANY, &ls_pcie_ecam_ops } + + NXP_ECAM(0), + NXP_ECAM(1), + NXP_ECAM(2), + NXP_ECAM(3), + NXP_ECAM(4), + NXP_ECAM(5), + #define QCOM_ECAM32(seg) \ { "QCOM ", "QDF2432 ", 1, seg, MCFG_BUS_ANY, &pci_32b_ops } diff --git a/drivers/pci/controller/Makefile b/drivers/pci/controller/Makefile index 37c8663de7fe..46b28acfff55 100644 --- a/drivers/pci/controller/Makefile +++ b/drivers/pci/controller/Makefile @@ -57,6 +57,7 @@ obj-y += mobiveil/ ifdef CONFIG_ACPI ifdef CONFIG_PCI_QUIRKS +obj-$(CONFIG_ARM64) += pcie-layerscape-ecam.o obj-$(CONFIG_ARM64) += pci-thunder-ecam.o obj-$(CONFIG_ARM64) += pci-thunder-pem.o obj-$(CONFIG_ARM64) += pci-xgene.o diff --git a/drivers/pci/controller/pcie-layerscape-ecam.c b/drivers/pci/controller/pcie-layerscape-ecam.c new file mode 100644 index 000000000000..c580bae170e0 --- /dev/null +++ b/drivers/pci/controller/pcie-layerscape-ecam.c @@ -0,0 +1,89 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * PCIe ecam driver for NXP's Layerscape SOCs, adopted from + * Amazon's Graviton driver. + * + * Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved. + * Copyright 2021 SolidRun Ltd. All Rights Reserved. + * + * Author: Jonathan Chocron + * Author: Jon Nettleton + */ + +#include +#include +#include +#include "../pci.h" + +#if defined(CONFIG_ACPI) && defined(CONFIG_PCI_QUIRKS) + +struct ls_pcie_ecam { + void __iomem *ccsr_base; +}; + +static void __iomem *ls_pcie_ecam_map_bus(struct pci_bus *bus, unsigned int devfn, + int where) +{ + struct pci_config_window *cfg = bus->sysdata; + struct ls_pcie_ecam *pcie = cfg->priv; + void __iomem *ccsr_base = pcie->ccsr_base; + + if (bus->number == 0) { + /* + * + * No devices/functions on the root bus num, so we do this here. + */ + if (PCI_SLOT(devfn) > 0) + return NULL; + else + return ccsr_base + where; + } + + return pci_ecam_map_bus(bus, devfn, where); +} + +static int ls_pcie_ecam_init(struct pci_config_window *cfg) +{ + struct device *dev = cfg->parent; + struct acpi_device *adev = to_acpi_device(dev); + struct acpi_pci_root *root = acpi_driver_data(adev); + struct ls_pcie_ecam *ls_pcie; + struct resource *res; + int ret; + + ls_pcie = devm_kzalloc(dev, sizeof(*ls_pcie), GFP_KERNEL); + if (!ls_pcie) + return -ENOMEM; + + res = devm_kzalloc(dev, sizeof(*res), GFP_KERNEL); + if (!res) + return -ENOMEM; + + ret = acpi_get_rc_resources(dev, "NXP0016", root->segment, res); + if (ret) { + dev_err(dev, "can't get rc csr base address for SEG %d\n", + root->segment); + return ret; + } + + dev_dbg(dev, "Root port ccsr res: %pR\n", res); + + ls_pcie->ccsr_base = devm_pci_remap_cfg_resource(dev, res); + if (IS_ERR(ls_pcie->ccsr_base)) + return PTR_ERR(ls_pcie->ccsr_base); + + cfg->priv = ls_pcie; + + return 0; +} + +const struct pci_ecam_ops ls_pcie_ecam_ops = { + .init = ls_pcie_ecam_init, + .pci_ops = { + .map_bus = ls_pcie_ecam_map_bus, + .read = pci_generic_config_read, + .write = pci_generic_config_write, + } +}; + +#endif /* defined(CONFIG_ACPI) && defined(CONFIG_PCI_QUIRKS) */ diff --git a/include/linux/pci-ecam.h b/include/linux/pci-ecam.h index 6b1301e2498e..e1d70dec5e89 100644 --- a/include/linux/pci-ecam.h +++ b/include/linux/pci-ecam.h @@ -88,6 +88,7 @@ extern const struct pci_ecam_ops xgene_v2_pcie_ecam_ops; /* APM X-Gene PCIe v2.x extern const struct pci_ecam_ops al_pcie_ops; /* Amazon Annapurna Labs PCIe */ extern const struct pci_ecam_ops tegra194_pcie_ops; /* Tegra194 PCIe */ extern const struct pci_ecam_ops loongson_pci_ecam_ops; /* Loongson PCIe */ +extern const struct pci_ecam_ops ls_pcie_ecam_ops; /* NXP Layerscape LX2160a PCIe */ #endif #if IS_ENABLED(CONFIG_PCI_HOST_COMMON) -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 24 Dec 2019 14:46:48 +0000 Subject: [PATCH 02/13] bus: fsl-mc: fix dprc object reading race When modifying the objects attached to a DPRC, we may end up reading the list of objects from the firmware while another thread is changing changing the list. Since we read the objects via: - Read the number of DPRC objects - Iterate over this number of objects retrieving their details and objects can be added in the middle of the list, this causes the last few objects to unexpectedly disappear. The side effect of this is if network interfaces are added after boot, they come and go. This can result in already configured interfaces unexpectedly disappearing. This has been easy to provoke with the restool interface added, and a script which adds network interfaces one after each other; the kernel rescanning runs asynchronously to restool. NXP's approach to fixing this was to introduce a sysfs "attribute" in their vendor tree, /sys/bus/fsl-mc/rescan, which userspace poked at to request the kernel to rescan the DPRC object tree each time the "restool" command completed (whether or not the tool changed anything.) This has the effect of making the kernel's rescan synchronous with a scripted restool, but still fails if we have multiple restools running concurrently. This patch takes a different approach: - Read the number of DPRC objects - Iterate over this number of objects retrieving their details - Re-read the number of DPRC objects - If the number of DPRC objects has changed while reading, repeat. This solves the issue where network interfaces unexpectedly disappear while adding others via ls-addni, because they've fallen off the end of the object list. This does *not* solve the issue that if an object is deleted while another is added while we are reading the objects - that requires firmware modification, or a more elaborate solution on the Linux side (e.g., CRCing the object details and reading all objects at least twice to check the CRC is stable.) However, without firmware modification, this is probably the best way to ensure that we read all the objects. Signed-off-by: Russell King --- drivers/bus/fsl-mc/dprc-driver.c | 32 +++++++++++++++++++++++++++++--- 1 file changed, 29 insertions(+), 3 deletions(-) diff --git a/drivers/bus/fsl-mc/dprc-driver.c b/drivers/bus/fsl-mc/dprc-driver.c index 5e70f9775a0e..a3c8d4869132 100644 --- a/drivers/bus/fsl-mc/dprc-driver.c +++ b/drivers/bus/fsl-mc/dprc-driver.c @@ -240,11 +240,11 @@ static void dprc_add_new_devices(struct fsl_mc_device *mc_bus_dev, int dprc_scan_objects(struct fsl_mc_device *mc_bus_dev, bool alloc_interrupts) { - int num_child_objects; + int num_child_objects, num_child_objects2; int dprc_get_obj_failures; int error; - unsigned int irq_count = mc_bus_dev->obj_desc.irq_count; - struct fsl_mc_obj_desc *child_obj_desc_array = NULL; + unsigned int irq_count; + struct fsl_mc_obj_desc *child_obj_desc_array; struct fsl_mc_bus *mc_bus = to_fsl_mc_bus(mc_bus_dev); error = dprc_get_obj_count(mc_bus_dev->mc_io, @@ -257,6 +257,9 @@ int dprc_scan_objects(struct fsl_mc_device *mc_bus_dev, return error; } +retry: + irq_count = mc_bus_dev->obj_desc.irq_count; + child_obj_desc_array = NULL; if (num_child_objects != 0) { int i; @@ -315,6 +318,29 @@ int dprc_scan_objects(struct fsl_mc_device *mc_bus_dev, } } + error = dprc_get_obj_count(mc_bus_dev->mc_io, + 0, + mc_bus_dev->mc_handle, + &num_child_objects2); + if (error < 0) { + if (child_obj_desc_array) + devm_kfree(&mc_bus_dev->dev, child_obj_desc_array); + dev_err(&mc_bus_dev->dev, "dprc_get_obj_count() failed: %d\n", + error); + return error; + } + + if (num_child_objects != num_child_objects2) { + /* + * Something changed while reading the number of objects. + * Retry reading the child object list. + */ + if (child_obj_desc_array) + devm_kfree(&mc_bus_dev->dev, child_obj_desc_array); + num_child_objects = num_child_objects2; + goto retry; + } + /* * Allocate IRQ's before binding the scanned devices with their * respective drivers. -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 24 Jan 2020 17:59:49 +0000 Subject: [PATCH 03/13] iommu: silence iommu group prints On the LX2160A, there are lots (about 160) of IOMMU messages produced during boot; this is excessive. Reduce the severity of these messages to debug level. Signed-off-by: Russell King --- drivers/iommu/iommu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index bfb2f163c691..d88ae5d10716 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -996,7 +996,7 @@ int iommu_group_add_device(struct iommu_group *group, struct device *dev) trace_add_device_to_group(group->id, dev); - dev_info(dev, "Adding to iommu group %d\n", group->id); + dev_dbg(dev, "Adding to iommu group %d\n", group->id); return 0; @@ -1033,7 +1033,7 @@ void iommu_group_remove_device(struct device *dev) if (!group) return; - dev_info(dev, "Removing from iommu group %d\n", group->id); + dev_dbg(dev, "Removing from iommu group %d\n", group->id); mutex_lock(&group->mutex); list_for_each_entry(tmp_device, &group->devices, list) { -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Jon Nettleton Date: Fri, 2 Jul 2021 07:28:21 -0400 Subject: [PATCH 04/13] ACPI: APD: Allow apd device to override fixed_clk_rate Currently by default the apd drivers are always using the fixed_clk_rate assigned in the matched acpi_device_desc. This causes an issue on the LX2160a platform because the NXP0001 settings do not match the platform and instead the I2C bus is only running at 24000kHZ rather than the expect 100000. Instead of patching the source with more static numbers that may or may not change instead add a check for the device property "fixed-clock-rate" that can be added to the ACPI tables to instruct the driver what rate to use. I have chosen fixed-clock-rate because clock-frequency is already used by I2C devices in acpi and device-tree to specify by bus speed, and fixed-clock-rate matches the fixed_clk_rate used by the apd_device_desc. If this device property is not set then the default static values are used so this should cause no regressions. Signed-off-by: Jon Nettleton --- drivers/acpi/acpi_apd.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/acpi_apd.c b/drivers/acpi/acpi_apd.c index 3bbe2276cac7..80ff4495e5b1 100644 --- a/drivers/acpi/acpi_apd.c +++ b/drivers/acpi/acpi_apd.c @@ -46,12 +46,21 @@ struct apd_private_data { static int acpi_apd_setup(struct apd_private_data *pdata) { const struct apd_device_desc *dev_desc = pdata->dev_desc; + struct acpi_device *adev = pdata->adev; + const union acpi_object *obj; + unsigned int fixed_clk_rate; struct clk *clk; - if (dev_desc->fixed_clk_rate) { + if (!acpi_dev_get_property(adev, "uefi-clock-frequency", ACPI_TYPE_INTEGER, &obj)) { + fixed_clk_rate = obj->integer.value; + } else if (dev_desc->fixed_clk_rate) { + fixed_clk_rate = dev_desc->fixed_clk_rate; + } + + if (fixed_clk_rate) { clk = clk_register_fixed_rate(&pdata->adev->dev, dev_name(&pdata->adev->dev), - NULL, 0, dev_desc->fixed_clk_rate); + NULL, 0, fixed_clk_rate); clk_register_clkdev(clk, NULL, dev_name(&pdata->adev->dev)); pdata->clk = clk; } -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Meenakshi Aggarwal Date: Wed, 27 May 2020 21:35:11 +0530 Subject: [PATCH 05/13] mmc: sdhci-of-esdhc: Add ACPI support This patch is to add acpi support in esdhc controller driver Signed-off-by: Meenakshi Aggarwal --- drivers/mmc/host/sdhci-of-esdhc.c | 62 +++++++++++++++++++------------ 1 file changed, 39 insertions(+), 23 deletions(-) diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c index e0266638381d..602ec77936ae 100644 --- a/drivers/mmc/host/sdhci-of-esdhc.c +++ b/drivers/mmc/host/sdhci-of-esdhc.c @@ -10,6 +10,7 @@ * Anton Vorontsov */ +#include #include #include #include @@ -73,6 +74,14 @@ static const struct of_device_id sdhci_esdhc_of_match[] = { }; MODULE_DEVICE_TABLE(of, sdhci_esdhc_of_match); +#ifdef CONFIG_ACPI +static const struct acpi_device_id sdhci_esdhc_ids[] = { + {"NXP0003" }, + { } +}; +MODULE_DEVICE_TABLE(acpi, sdhci_esdhc_ids); +#endif + struct sdhci_esdhc { u8 vendor_ver; u8 spec_ver; @@ -1371,29 +1380,35 @@ static void esdhc_init(struct platform_device *pdev, struct sdhci_host *host) esdhc->clk_fixup = match->data; np = pdev->dev.of_node; - if (of_device_is_compatible(np, "fsl,p2020-esdhc")) { - esdhc->quirk_delay_before_data_reset = true; - esdhc->quirk_trans_complete_erratum = true; - } + /* in case of device tree, get clock from framework */ + if (np) { + if (of_device_is_compatible(np, "fsl,p2020-esdhc")) { + esdhc->quirk_delay_before_data_reset = true; + esdhc->quirk_trans_complete_erratum = true; + } - clk = of_clk_get(np, 0); - if (!IS_ERR(clk)) { - /* - * esdhc->peripheral_clock would be assigned with a value - * which is eSDHC base clock when use periperal clock. - * For some platforms, the clock value got by common clk - * API is peripheral clock while the eSDHC base clock is - * 1/2 peripheral clock. - */ - if (of_device_is_compatible(np, "fsl,ls1046a-esdhc") || - of_device_is_compatible(np, "fsl,ls1028a-esdhc") || - of_device_is_compatible(np, "fsl,ls1088a-esdhc")) - esdhc->peripheral_clock = clk_get_rate(clk) / 2; - else - esdhc->peripheral_clock = clk_get_rate(clk); - - clk_put(clk); - } + clk = of_clk_get(np, 0); + if (!IS_ERR(clk)) { + /* + * esdhc->peripheral_clock would be assigned with a value + * which is eSDHC base clock when use periperal clock. + * For some platforms, the clock value got by common clk + * API is peripheral clock while the eSDHC base clock is + * 1/2 peripheral clock. + */ + if (of_device_is_compatible(np, "fsl,ls1046a-esdhc") || + of_device_is_compatible(np, "fsl,ls1028a-esdhc") || + of_device_is_compatible(np, "fsl,ls1088a-esdhc")) + esdhc->peripheral_clock = clk_get_rate(clk) / 2; + else + esdhc->peripheral_clock = clk_get_rate(clk); + + clk_put(clk); + } + } else { + device_property_read_u32(&pdev->dev, "clock-frequency", + &esdhc->peripheral_clock); + } esdhc_clock_enable(host, false); val = sdhci_readl(host, ESDHC_DMA_SYSCTL); @@ -1426,7 +1441,7 @@ static int sdhci_esdhc_probe(struct platform_device *pdev) np = pdev->dev.of_node; - if (of_property_read_bool(np, "little-endian")) + if (device_property_read_bool(&pdev->dev, "little-endian")) host = sdhci_pltfm_init(pdev, &sdhci_esdhc_le_pdata, sizeof(struct sdhci_esdhc)); else @@ -1513,6 +1528,7 @@ static struct platform_driver sdhci_esdhc_driver = { .name = "sdhci-esdhc", .probe_type = PROBE_PREFER_ASYNCHRONOUS, .of_match_table = sdhci_esdhc_of_match, + .acpi_match_table = sdhci_esdhc_ids, .pm = &esdhc_of_dev_pm_ops, }, .probe = sdhci_esdhc_probe, -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Meharbaan Date: Tue, 28 Jul 2020 17:41:31 +0530 Subject: [PATCH 06/13] drivers/mmc/host/sdhci-of-esdhc : Fix DMA coherent check in ACPI mode. DMA-coherent check to set ESDHC_DMA_SNOOP mask was bypassed when booted in ACPI mode. Now it also checks the acpi device and its parents for _CCA property in the device, and sets the flag accordingly. Signed-off-by: Meharbaan --- drivers/base/property.c | 38 +++++++++++++++++++++++++++++++ drivers/mmc/host/sdhci-of-esdhc.c | 7 ++++-- include/linux/property.h | 2 ++ 3 files changed, 45 insertions(+), 2 deletions(-) diff --git a/drivers/base/property.c b/drivers/base/property.c index 7f338cb4fb7b..68bf74957c55 100644 --- a/drivers/base/property.c +++ b/drivers/base/property.c @@ -16,6 +16,7 @@ #include #include #include +#include struct fwnode_handle *dev_fwnode(const struct device *dev) { @@ -879,6 +880,43 @@ enum dev_dma_attr device_get_dma_attr(struct device *dev) } EXPORT_SYMBOL_GPL(device_get_dma_attr); +/** + * device_match_fw_node - Check if the device is the parent node. + * @dev: Pointer to the device. + * @parent_fwnode Pointer to the parent's firmware node. + * + * The function returns true if the device has no parent. + * + */ +static int device_match_fw_node(struct device *dev, const void *parent_fwnode) +{ + return dev->fwnode == parent_fwnode; +} + +/** + * dev_dma_is_coherent - Check if the device or any of its parents has + * dma support enabled. + * @dev: Pointer to the device. + * + * The function gets the device pointer and check for device_dma_supported() + * on the device pointer passed and then recursively on its parent nodes. + */ + +bool dev_dma_is_coherent(struct device *dev) +{ + struct fwnode_handle *parent_fwnode; + + while (dev) { + if (device_dma_supported(dev)) + return true; + parent_fwnode = fwnode_get_next_parent(dev->fwnode); + dev = bus_find_device(&platform_bus_type, NULL, parent_fwnode, + device_match_fw_node); + } + return false; +} +EXPORT_SYMBOL_GPL(dev_dma_is_coherent); + /** * fwnode_get_phy_mode - Get phy mode for given firmware node * @fwnode: Pointer to the given node diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c index 602ec77936ae..308353675dd6 100644 --- a/drivers/mmc/host/sdhci-of-esdhc.c +++ b/drivers/mmc/host/sdhci-of-esdhc.c @@ -545,8 +545,11 @@ static int esdhc_of_enable_dma(struct sdhci_host *host) } value = sdhci_readl(host, ESDHC_DMA_SYSCTL); - - if (of_dma_is_coherent(dev->of_node)) + /* + * of_dma_is_coherent() returns false in case of acpi hence + * dev_dma_is_coherent() is used along with it. + */ + if (of_dma_is_coherent(dev->of_node) || dev_dma_is_coherent(dev)) value |= ESDHC_DMA_SNOOP; else value &= ~ESDHC_DMA_SNOOP; diff --git a/include/linux/property.h b/include/linux/property.h index 117cc200c656..e0c7799519df 100644 --- a/include/linux/property.h +++ b/include/linux/property.h @@ -387,6 +387,8 @@ bool device_dma_supported(struct device *dev); enum dev_dma_attr device_get_dma_attr(struct device *dev); +bool dev_dma_is_coherent(struct device *dev); + const void *device_get_match_data(const struct device *dev); int device_get_phy_mode(struct device *dev); -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Jon Nettleton Date: Thu, 18 Aug 2022 09:00:51 +0200 Subject: [PATCH 07/13] aarch64: Optimize memcpy for Cortex-A72 This variant of memcpy not only works around some hardware issues but also outperforms the current memcpy used by the kernel and glibc on Cortex-A72 hardware. Signed-off-by: Jon Nettleton --- arch/arm64/lib/memcpy.S | 49 ++++++++++++++++++----------------------- 1 file changed, 21 insertions(+), 28 deletions(-) diff --git a/arch/arm64/lib/memcpy.S b/arch/arm64/lib/memcpy.S index 4ab48d49c451..ec867fa90eb3 100644 --- a/arch/arm64/lib/memcpy.S +++ b/arch/arm64/lib/memcpy.S @@ -36,12 +36,8 @@ #define D_h x13 #define E_l x14 #define E_h x15 -#define F_l x16 -#define F_h x17 -#define G_l count -#define G_h dst -#define H_l src -#define H_h srcend +#define F_l count +#define F_h dst #define tmp1 x14 /* This implementation handles overlaps and supports both memcpy and memmove @@ -60,10 +56,10 @@ SYM_FUNC_START(__pi_memcpy) add srcend, src, count add dstend, dstin, count - cmp count, 128 + cmp count, 96 b.hi L(copy_long) cmp count, 32 - b.hi L(copy32_128) + b.hi L(copy32_96) /* Small copies: 0..32 bytes. */ cmp count, 16 @@ -107,14 +103,14 @@ L(copy0): ret .p2align 4 - /* Medium copies: 33..128 bytes. */ -L(copy32_128): + /* Medium copies: 33..96 bytes. */ +L(copy32_96): ldp A_l, A_h, [src] ldp B_l, B_h, [src, 16] ldp C_l, C_h, [srcend, -32] ldp D_l, D_h, [srcend, -16] cmp count, 64 - b.hi L(copy128) + b.hi L(copy96) stp A_l, A_h, [dstin] stp B_l, B_h, [dstin, 16] stp C_l, C_h, [dstend, -32] @@ -122,17 +118,10 @@ L(copy32_128): ret .p2align 4 - /* Copy 65..128 bytes. */ -L(copy128): + /* Copy 65..96 bytes. */ +L(copy96): ldp E_l, E_h, [src, 32] ldp F_l, F_h, [src, 48] - cmp count, 96 - b.ls L(copy96) - ldp G_l, G_h, [srcend, -64] - ldp H_l, H_h, [srcend, -48] - stp G_l, G_h, [dstend, -64] - stp H_l, H_h, [dstend, -48] -L(copy96): stp A_l, A_h, [dstin] stp B_l, B_h, [dstin, 16] stp E_l, E_h, [dstin, 32] @@ -161,7 +150,8 @@ L(copy_long): stp D_l, D_h, [dstin] ldp B_l, B_h, [src, 32] ldp C_l, C_h, [src, 48] - ldp D_l, D_h, [src, 64]! + ldp D_l, D_h, [src, 64] + add src, src, 64 subs count, count, 128 + 16 /* Test and readjust count. */ b.ls L(copy64_from_end) @@ -173,7 +163,8 @@ L(loop64): stp C_l, C_h, [dst, 48] ldp C_l, C_h, [src, 48] stp D_l, D_h, [dst, 64]! - ldp D_l, D_h, [src, 64]! + ldp D_l, D_h, [src, 64] + add src, src, 64 subs count, count, 64 b.hi L(loop64) @@ -206,7 +197,8 @@ L(copy_long_backwards): stp D_l, D_h, [dstend, -16] ldp B_l, B_h, [srcend, -32] ldp C_l, C_h, [srcend, -48] - ldp D_l, D_h, [srcend, -64]! + ldp D_l, D_h, [srcend, -64] + sub srcend, srcend, 64 sub dstend, dstend, tmp1 subs count, count, 128 b.ls L(copy64_from_start) @@ -219,13 +211,14 @@ L(loop64_backwards): stp C_l, C_h, [dstend, -48] ldp C_l, C_h, [srcend, -48] stp D_l, D_h, [dstend, -64]! - ldp D_l, D_h, [srcend, -64]! + ldp D_l, D_h, [srcend, -64] + sub srcend, srcend, 64 subs count, count, 64 b.hi L(loop64_backwards) /* Write the last iteration and copy 64 bytes from the start. */ L(copy64_from_start): - ldp G_l, G_h, [src, 48] + ldp F_l, F_h, [src, 48] stp A_l, A_h, [dstend, -16] ldp A_l, A_h, [src, 32] stp B_l, B_h, [dstend, -32] @@ -233,10 +226,10 @@ L(copy64_from_start): stp C_l, C_h, [dstend, -48] ldp C_l, C_h, [src] stp D_l, D_h, [dstend, -64] - stp G_l, G_h, [dstin, 48] - stp A_l, A_h, [dstin, 32] - stp B_l, B_h, [dstin, 16] stp C_l, C_h, [dstin] + stp B_l, B_h, [dstin, 16] + stp A_l, A_h, [dstin, 32] + stp F_l, F_h, [dstin, 48] ret SYM_FUNC_END(__pi_memcpy) -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Jon Nettleton Date: Mon, 5 Sep 2022 13:30:44 +0200 Subject: [PATCH 08/13] arm64: optimize context switch for Cortex-A72 The Cortex-A72 has a performance errata regarding writing to the register file. Therefore we change all the store and load addressing to offsets and manually increment the src and dst with an add. Signed-off-by: Jon Nettleton --- arch/arm64/kernel/entry.S | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S index e28137d64b76..23757249bed0 100644 --- a/arch/arm64/kernel/entry.S +++ b/arch/arm64/kernel/entry.S @@ -826,20 +826,22 @@ SYM_FUNC_START(cpu_switch_to) mov x10, #THREAD_CPU_CONTEXT add x8, x0, x10 mov x9, sp - stp x19, x20, [x8], #16 // store callee-saved registers - stp x21, x22, [x8], #16 - stp x23, x24, [x8], #16 - stp x25, x26, [x8], #16 - stp x27, x28, [x8], #16 - stp x29, x9, [x8], #16 + stp x19, x20, [x8] // store callee-saved registers + stp x21, x22, [x8, #16] + stp x23, x24, [x8, #32] + stp x25, x26, [x8, #48] + stp x27, x28, [x8, #64] + stp x29, x9, [x8, #80] + add x8, x8, #96 str lr, [x8] add x8, x1, x10 - ldp x19, x20, [x8], #16 // restore callee-saved registers - ldp x21, x22, [x8], #16 - ldp x23, x24, [x8], #16 - ldp x25, x26, [x8], #16 - ldp x27, x28, [x8], #16 - ldp x29, x9, [x8], #16 + ldp x19, x20, [x8] // restore callee-saved registers + ldp x21, x22, [x8, #16] + ldp x23, x24, [x8, #32] + ldp x25, x26, [x8, #48] + ldp x27, x28, [x8, #64] + ldp x29, x9, [x8, #80] + add x8, x8, #96 ldr lr, [x8] mov sp, x9 msr sp_el0, x1 -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Wed, 28 Sep 2022 12:58:51 +0100 Subject: [PATCH 09/13] arm64: Update copy_from_user() Replace the old mangled-beyond-hope copy_from_user() routine with a shiny new one based on our newest memcpy() implementation. When plumbed into the standalone memcpy benchmark from the Arm Optimized Routines library, this routine outperforms the old one by up to ~1.7x for small copies, with comparable gains across a range of microarchitectures, levelling off once sizes get up to ~2KB and general load/store throughput starts to dominate. Some of this is paid for by pushing more complexity into the fixup handlers, but much of that could be recovered again with cleverer exception records that can give more information about the original access to the handler itself. For now though, the label sleds are at least entertaining. Signed-off-by: Robin Murphy --- arch/arm64/lib/copy_from_user.S | 266 ++++++++++++++++++++++++++------ 1 file changed, 219 insertions(+), 47 deletions(-) diff --git a/arch/arm64/lib/copy_from_user.S b/arch/arm64/lib/copy_from_user.S index 34e317907524..a4b9bd73a5a8 100644 --- a/arch/arm64/lib/copy_from_user.S +++ b/arch/arm64/lib/copy_from_user.S @@ -1,73 +1,245 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (C) 2012 ARM Ltd. + * Copyright (c) 2012-2022, Arm Limited. */ #include #include #include -#include + +/* Assumptions: + * + * ARMv8-a, AArch64, unaligned accesses. + * + */ + +#define L(label) .L ## label + +#define dstin x0 +#define src x1 +#define count x2 +#define dst x3 +#define srcend x4 +#define dstend x5 +#define A_l x6 +#define A_lw w6 +#define A_h x7 +#define B_l x8 +#define B_lw w8 +#define B_h x9 +#define C_l x10 +#define C_lw w10 +#define C_h x11 +#define D_l x12 +#define D_h x13 +#define E_l x14 +#define E_h x15 +#define F_l x16 +#define F_h x17 +#define tmp1 x14 /* - * Copy from user space to a kernel buffer (alignment handled by the hardware) + * Derived from memcpy with various adjustments: * - * Parameters: - * x0 - to - * x1 - from - * x2 - n - * Returns: - * x0 - bytes not copied + * - memmove parts are removed since user and kernel pointers won't overlap. + * - The main loop is scaled down to 48 bytes per iteration since the increase + * in load ops changes the balance; little cores barely notice the difference, + * so big cores can benefit from keeping the loop relatively short. + * - Similarly, preferring source rather than destination alignment works out + * better on average. + * - The 33-128 byte cases are reworked to better balance the stores with the + * doubled-up load ops, and keep a more consistent access pattern. + * - The 0-3 byte sequence is replaced with the one borrowed from clear_user, + * since LDTRB lacks a register-offset addressing mode. */ - .macro ldrb1 reg, ptr, val - user_ldst 9998f, ldtrb, \reg, \ptr, \val - .endm +#define U_pre(x...) USER(L(fixup_pre), x) +#define U_dst(x...) USER(L(fixup_dst), x) +#define U_S1(x...) USER(L(fixup_s1), x) +#define U_M16(x...) USER(L(fixup_m16), x) +#define U_M32(x...) USER(L(fixup_m32), x) +#define U_M64(x...) USER(L(fixup_m64), x) +#define U_L32(x...) USER(L(fixup_l32), x) +#define U_L48(x...) USER(L(fixup_l48), x) +#define U_L64(x...) USER(L(fixup_l64), x) - .macro strb1 reg, ptr, val - strb \reg, [\ptr], \val - .endm +SYM_FUNC_START(__arch_copy_from_user) + add srcend, src, count + add dstend, dstin, count + cmp count, 128 + b.hi L(copy_long) + cmp count, 32 + b.hi L(copy32_128) - .macro ldrh1 reg, ptr, val - user_ldst 9997f, ldtrh, \reg, \ptr, \val - .endm + /* Small copies: 0..32 bytes. */ + cmp count, 16 + b.lo L(copy16) +U_pre( ldtr A_l, [src]) +U_pre( ldtr A_h, [src, 8]) +U_pre( ldtr D_l, [srcend, -16]) +U_pre( ldtr D_h, [srcend, -8]) + stp A_l, A_h, [dstin] + stp D_l, D_h, [dstend, -16] + mov x0, #0 + ret - .macro strh1 reg, ptr, val - strh \reg, [\ptr], \val - .endm + /* Copy 8-15 bytes. */ +L(copy16): + tbz count, 3, L(copy8) +U_pre( ldtr A_l, [src]) +U_pre( ldtr A_h, [srcend, -8]) + str A_l, [dstin] + str A_h, [dstend, -8] + mov x0, #0 + ret - .macro ldr1 reg, ptr, val - user_ldst 9997f, ldtr, \reg, \ptr, \val - .endm + .p2align 3 + /* Copy 4-7 bytes. */ +L(copy8): + tbz count, 2, L(copy4) +U_pre( ldtr A_lw, [src]) +U_pre( ldtr B_lw, [srcend, -4]) + str A_lw, [dstin] + str B_lw, [dstend, -4] + mov x0, #0 + ret - .macro str1 reg, ptr, val - str \reg, [\ptr], \val - .endm + /* Copy 0..3 bytes. */ +L(copy4): + tbz count, #1, L(copy1) +U_pre( ldtrh A_lw, [src]) + strh A_lw, [dstin] +L(copy1): + tbz count, #0, L(copy0) +U_S1( ldtrb A_lw, [srcend, -1]) + strb A_lw, [dstend, -1] +L(copy0): + mov x0, #0 + ret - .macro ldp1 reg1, reg2, ptr, val - user_ldp 9997f, \reg1, \reg2, \ptr, \val - .endm + .p2align 4 + /* Medium copies: 33..128 bytes. */ +L(copy32_128): +U_pre( ldtr A_l, [src]) +U_pre( ldtr A_h, [src, 8]) +U_pre( ldtr B_l, [src, 16]) +U_pre( ldtr B_h, [src, 24]) + stp A_l, A_h, [dstin] + stp B_l, B_h, [dstin, 16] +U_M32( ldtr C_l, [srcend, -32]) +U_M32( ldtr C_h, [srcend, -24]) +U_M32( ldtr D_l, [srcend, -16]) +U_M32( ldtr D_h, [srcend, -8]) + cmp count, 64 + b.ls L(copy64) +U_M32( ldtr E_l, [src, 32]) +U_M32( ldtr E_h, [src, 40]) +U_M32( ldtr F_l, [src, 48]) +U_M32( ldtr F_h, [src, 56]) + stp E_l, E_h, [dstin, 32] + stp F_l, F_h, [dstin, 48] +U_M64( ldtr A_l, [srcend, -64]) +U_M64( ldtr A_h, [srcend, -56]) +U_M64( ldtr B_l, [srcend, -48]) +U_M64( ldtr B_h, [srcend, -40]) + stp A_l, A_h, [dstend, -64] + stp B_l, B_h, [dstend, -48] +L(copy64): + stp C_l, C_h, [dstend, -32] + stp D_l, D_h, [dstend, -16] + mov x0, #0 + ret - .macro stp1 reg1, reg2, ptr, val - stp \reg1, \reg2, [\ptr], \val - .endm + .p2align 4 + /* Copy more than 128 bytes. */ +L(copy_long): + /* Copy 16 bytes and then align src to 16-byte alignment. */ -end .req x5 -srcin .req x15 -SYM_FUNC_START(__arch_copy_from_user) - add end, x0, x2 - mov srcin, x1 -#include "copy_template.S" - mov x0, #0 // Nothing to copy +U_pre( ldtr D_l, [src]) +U_pre( ldtr D_h, [src, 8]) + and tmp1, src, 15 + bic src, src, 15 + sub dst, dstin, tmp1 + add count, count, tmp1 /* Count is now 16 too large. */ +U_pre( ldtr A_l, [src, 16]) +U_pre( ldtr A_h, [src, 24]) + stp D_l, D_h, [dstin] +U_M16( ldtr B_l, [src, 32]) +U_M16( ldtr B_h, [src, 40]) +U_M16( ldtr C_l, [src, 48]) +U_M16( ldtr C_h, [src, 56]) + add src, src, #48 + subs count, count, 96 + 16 /* Test and readjust count. */ + b.ls L(copy48_from_end) + +L(loop48): + stp A_l, A_h, [dst, 16] +U_L32( ldtr A_l, [src, 16]) +U_L32( ldtr A_h, [src, 24]) + stp B_l, B_h, [dst, 32] +U_L48( ldtr B_l, [src, 32]) +U_L48( ldtr B_h, [src, 40]) + stp C_l, C_h, [dst, 48]! +U_dst( ldtr C_l, [src, 48]) +U_dst( ldtr C_h, [src, 56]) + add src, src, #48 + subs count, count, 48 + b.hi L(loop48) + + /* Write the last iteration and copy 48 bytes from the end. */ +L(copy48_from_end): + stp A_l, A_h, [dst, 16] +U_L32( ldtr A_l, [srcend, -48]) +U_L32( ldtr A_h, [srcend, -40]) + stp B_l, B_h, [dst, 32] +U_L48( ldtr B_l, [srcend, -32]) +U_L48( ldtr B_h, [srcend, -24]) + stp C_l, C_h, [dst, 48] +U_L64( ldtr C_l, [srcend, -16]) +U_L64( ldtr C_h, [srcend, -8]) + stp A_l, A_h, [dstend, -48] + stp B_l, B_h, [dstend, -32] + stp C_l, C_h, [dstend, -16] + mov x0, #0 + ret + + /* Fixups... */ + + /* + * Fault before anything has been written, but progress may have + * been possible; realign dst and retry a single byte to confirm. + */ +L(fixup_pre): + mov dst, dstin +U_dst( ldtrb A_lw, [src]) + strb A_lw, [dst], #1 +L(fixup_dst): + sub x0, dstend, dst ret - // Exception fixups -9997: cmp dst, dstin - b.ne 9998f - // Before being absolutely sure we couldn't copy anything, try harder -USER(9998f, ldtrb tmp1w, [srcin]) - strb tmp1w, [dst], #1 -9998: sub x0, end, dst // bytes not copied + /* Small: Fault with 1 byte remaining, regardless of count */ +L(fixup_s1): + mov x0, #1 ret + + /* Medium: Faults after n bytes beyond dstin have been written */ +L(fixup_m64): + add dstin, dstin, #32 +L(fixup_m32): + add dstin, dstin, #16 +L(fixup_m16): + add dst, dstin, #16 + b L(fixup_dst) + + /* Large: Faults after n bytes beyond dst have been written */ +L(fixup_l64): + add dst, dst, #16 +L(fixup_l48): + add dst, dst, #16 +L(fixup_l32): + add dst, dst, #32 + b L(fixup_dst) + SYM_FUNC_END(__arch_copy_from_user) EXPORT_SYMBOL(__arch_copy_from_user) -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Wed, 28 Sep 2022 12:58:52 +0100 Subject: [PATCH 10/13] arm64: Update copy_to_user() As with its counterpart, replace copy_to_user() with a new and improved version similarly derived from memcpy(). Different tradeoffs from the base implementation are made relative to copy_from_user() to get the most consistent results across different microarchitectures, but the overall shape of the performance gain ends up about the same. The exception fixups are even more comical this time around, but that's down to now needing to reconstruct the faulting address, and cope with overlapping stores at various points. Again, improvements to the exception mechanism could significantly simplify things in future. Signed-off-by: Robin Murphy --- arch/arm64/lib/copy_to_user.S | 385 +++++++++++++++++++++++++++++----- 1 file changed, 338 insertions(+), 47 deletions(-) diff --git a/arch/arm64/lib/copy_to_user.S b/arch/arm64/lib/copy_to_user.S index 802231772608..b641f00f50d6 100644 --- a/arch/arm64/lib/copy_to_user.S +++ b/arch/arm64/lib/copy_to_user.S @@ -1,73 +1,364 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (C) 2012 ARM Ltd. + * Copyright (c) 2012-2022, Arm Limited. */ #include #include #include -#include + +/* Assumptions: + * + * ARMv8-a, AArch64, unaligned accesses. + * + */ + +#define L(label) .L ## label + +#define dstin x0 +#define src x1 +#define count x2 +#define dst x3 +#define srcend x4 +#define dstend x5 +#define A_l x6 +#define A_lw w6 +#define A_h x7 +#define B_l x8 +#define B_lw w8 +#define B_h x9 +#define C_l x10 +#define C_lw w10 +#define C_h x11 +#define D_l x12 +#define D_h x13 +#define E_l x14 +#define E_h x15 +#define F_l x16 +#define F_h x17 +#define tmp1 x14 /* - * Copy to user space from a kernel buffer (alignment handled by the hardware) + * Derived from memcpy with various adjustments: * - * Parameters: - * x0 - to - * x1 - from - * x2 - n - * Returns: - * x0 - bytes not copied + * - memmove parts are removed since user and kernel pointers won't overlap. + * - Contrary to copy_from_user, although big cores still aren't particularly + * keen on the increased instruction count in the main loop, processing fewer + * than 64 bytes per iteration here hurts little cores more. + * - The medium-size cases are reworked to better balance the loads with the + * doubled-up store ops, avoid potential out-of-sequence faults, and preserve + * the input arguments for the sake of fault handling. + * - The 0-3 byte sequence is replaced with the one borrowed from clear_user, + * since STTRB lacks a register-offset addressing mode. */ - .macro ldrb1 reg, ptr, val - ldrb \reg, [\ptr], \val - .endm - .macro strb1 reg, ptr, val - user_ldst 9998f, sttrb, \reg, \ptr, \val - .endm +#define U_pre(x...) USER(L(fixup_pre), x) +#define U_dst(x...) USER(L(fixup_dst), x) - .macro ldrh1 reg, ptr, val - ldrh \reg, [\ptr], \val - .endm +#define U_S1(x...) USER(L(fixup_s1), x) +#define U_S4(x...) USER(L(fixup_s4), x) +#define U_S8(x...) USER(L(fixup_s8), x) +#define U_ST8(x...) USER(L(fixup_st8), x) +#define U_S16(x...) USER(L(fixup_s16), x) +#define U_M24(x...) USER(L(fixup_m24), x) +#define U_M32(x...) USER(L(fixup_m32), x) +#define U_M40(x...) USER(L(fixup_m40), x) +#define U_M48(x...) USER(L(fixup_m48), x) +#define U_M56(x...) USER(L(fixup_m56), x) +#define U_M64(x...) USER(L(fixup_m64), x) +#define U_MT8(x...) USER(L(fixup_mt8), x) +#define U_MT16(x...) USER(L(fixup_mt16), x) +#define U_MT24(x...) USER(L(fixup_mt24), x) +#define U_MT32(x...) USER(L(fixup_mt32), x) +#define U_MT40(x...) USER(L(fixup_mt40), x) +#define U_MT48(x...) USER(L(fixup_mt48), x) +#define U_MT56(x...) USER(L(fixup_mt56), x) - .macro strh1 reg, ptr, val - user_ldst 9997f, sttrh, \reg, \ptr, \val - .endm +#define U_L16(x...) USER(L(fixup_l16), x) +#define U_L24(x...) USER(L(fixup_l24), x) +#define U_L32(x...) USER(L(fixup_l32), x) +#define U_L40(x...) USER(L(fixup_l40), x) +#define U_L48(x...) USER(L(fixup_l48), x) +#define U_L56(x...) USER(L(fixup_l56), x) +#define U_L64(x...) USER(L(fixup_l64), x) +#define U_L72(x...) USER(L(fixup_l72), x) +#define U_LT8(x...) USER(L(fixup_lt8), x) +#define U_LT16(x...) USER(L(fixup_lt16), x) +#define U_LT24(x...) USER(L(fixup_lt24), x) +#define U_LT32(x...) USER(L(fixup_lt32), x) +#define U_LT40(x...) USER(L(fixup_lt40), x) +#define U_LT48(x...) USER(L(fixup_lt48), x) +#define U_LT56(x...) USER(L(fixup_lt56), x) +#define U_LT64(x...) USER(L(fixup_lt64), x) - .macro ldr1 reg, ptr, val - ldr \reg, [\ptr], \val - .endm +SYM_FUNC_START(__arch_copy_to_user) + add srcend, src, count + add dstend, dstin, count + cmp count, 128 + b.hi L(copy_long) + cmp count, 32 + b.hi L(copy32_128) - .macro str1 reg, ptr, val - user_ldst 9997f, sttr, \reg, \ptr, \val - .endm + /* Small copies: 0..32 bytes. */ + cmp count, 16 + b.lo L(copy16) + ldp A_l, A_h, [src] + ldp D_l, D_h, [srcend, -16] +U_pre( sttr A_l, [dstin]) +U_S8( sttr A_h, [dstin, 8]) +U_S16( sttr D_l, [dstend, -16]) +U_ST8( sttr D_h, [dstend, -8]) + mov x0, #0 + ret - .macro ldp1 reg1, reg2, ptr, val - ldp \reg1, \reg2, [\ptr], \val - .endm + /* Copy 8-15 bytes. */ +L(copy16): + tbz count, 3, L(copy8) + ldr A_l, [src] + ldr A_h, [srcend, -8] +U_pre( sttr A_l, [dstin]) +U_S8( sttr A_h, [dstend, -8]) + mov x0, #0 + ret - .macro stp1 reg1, reg2, ptr, val - user_stp 9997f, \reg1, \reg2, \ptr, \val - .endm + .p2align 3 + /* Copy 4-7 bytes. */ +L(copy8): + tbz count, 2, L(copy4) + ldr A_lw, [src] + ldr B_lw, [srcend, -4] +U_pre( sttr A_lw, [dstin]) +U_S4( sttr B_lw, [dstend, -4]) + mov x0, #0 + ret -end .req x5 -srcin .req x15 -SYM_FUNC_START(__arch_copy_to_user) - add end, x0, x2 - mov srcin, x1 -#include "copy_template.S" + /* Copy 0..3 bytes. */ +L(copy4): + tbz count, #1, L(copy1) + ldrh A_lw, [src] +U_pre( sttrh A_lw, [dstin]) +L(copy1): + tbz count, #0, L(copy0) + ldrb A_lw, [srcend, -1] +U_S1( sttrb A_lw, [dstend, -1]) +L(copy0): mov x0, #0 ret - // Exception fixups -9997: cmp dst, dstin - b.ne 9998f - // Before being absolutely sure we couldn't copy anything, try harder - ldrb tmp1w, [srcin] -USER(9998f, sttrb tmp1w, [dst]) - add dst, dst, #1 -9998: sub x0, end, dst // bytes not copied + .p2align 4 + /* Medium copies: 33..128 bytes. */ +L(copy32_128): + ldp A_l, A_h, [src] + ldp B_l, B_h, [src, 16] +U_pre( sttr A_l, [dstin]) +U_S8( sttr A_h, [dstin, 8]) +U_S16( sttr B_l, [dstin, 16]) +U_M24( sttr B_h, [dstin, 24]) + ldp C_l, C_h, [srcend, -32] + ldp D_l, D_h, [srcend, -16] + cmp count, 64 + b.ls L(copy64) + ldp E_l, E_h, [src, 32] + ldp F_l, F_h, [src, 48] +U_M32( sttr E_l, [dstin, 32]) +U_M40( sttr E_h, [dstin, 40]) +U_M48( sttr F_l, [dstin, 48]) +U_M56( sttr F_h, [dstin, 56]) + ldp A_l, A_h, [srcend, -64] + ldp B_l, B_h, [srcend, -48] +U_M64( sttr A_l, [dstend, -64]) +U_MT56( sttr A_h, [dstend, -56]) +U_MT48( sttr B_l, [dstend, -48]) +U_MT40( sttr B_h, [dstend, -40]) +L(copy64): +U_MT32( sttr C_l, [dstend, -32]) +U_MT24( sttr C_h, [dstend, -24]) +U_MT16( sttr D_l, [dstend, -16]) +U_MT8( sttr D_h, [dstend, -8]) + mov x0, #0 ret + + .p2align 4 + nop + nop + nop + /* Copy more than 128 bytes. */ +L(copy_long): + /* Copy 16 bytes and then align dst to 16-byte alignment. */ + + ldp D_l, D_h, [src] + and tmp1, dstin, 15 + bic dst, dstin, 15 + sub src, src, tmp1 + add count, count, tmp1 /* Count is now 16 too large. */ + ldp A_l, A_h, [src, 16] +U_pre( sttr D_l, [dstin]) +U_S8( sttr D_h, [dstin, 8]) + ldp B_l, B_h, [src, 32] + ldp C_l, C_h, [src, 48] + ldp D_l, D_h, [src, 64]! + subs count, count, 128 + 16 /* Test and readjust count. */ + b.ls L(copy64_from_end) + +L(loop64): +U_L16( sttr A_l, [dst, 16]) +U_L24( sttr A_h, [dst, 24]) + ldp A_l, A_h, [src, 16] +U_L32( sttr B_l, [dst, 32]) +U_L40( sttr B_h, [dst, 40]) + ldp B_l, B_h, [src, 32] +U_L48( sttr C_l, [dst, 48]) +U_L56( sttr C_h, [dst, 56]) + ldp C_l, C_h, [src, 48] +U_L64( sttr D_l, [dst, 64]) +U_L72( sttr D_h, [dst, 72]) + add dst, dst, #64 + ldp D_l, D_h, [src, 64]! + subs count, count, 64 + b.hi L(loop64) + + /* Write the last iteration and copy 64 bytes from the end. */ +L(copy64_from_end): + ldp E_l, E_h, [srcend, -64] +U_L16( sttr A_l, [dst, 16]) +U_L24( sttr A_h, [dst, 24]) + ldp A_l, A_h, [srcend, -48] +U_L32( sttr B_l, [dst, 32]) +U_L40( sttr B_h, [dst, 40]) + ldp B_l, B_h, [srcend, -32] +U_L48( sttr C_l, [dst, 48]) +U_L56( sttr C_h, [dst, 56]) + ldp C_l, C_h, [srcend, -16] +U_L64( sttr D_l, [dst, 64]) +U_L72( sttr D_h, [dst, 72]) +U_LT64( sttr E_l, [dstend, -64]) +U_LT56( sttr E_h, [dstend, -56]) +U_LT48( sttr A_l, [dstend, -48]) +U_LT40( sttr A_h, [dstend, -40]) +U_LT32( sttr B_l, [dstend, -32]) +U_LT24( sttr B_h, [dstend, -24]) +U_LT16( sttr C_l, [dstend, -16]) +U_LT8( sttr C_h, [dstend, -8]) + mov x0, #0 + ret + + /* Fixups... */ + + /* + * Fault on the first write, but progress may have been possible; + * realign dst and retry a single byte to confirm. + */ +L(fixup_pre): + mov dst, dstin +U_dst( ldtrb A_lw, [src]) + strb A_lw, [dst], #1 +L(fixup_dst): + sub x0, dstend, dst + ret + + /* Small: Fault with 1 byte remaining, regardless of count */ +L(fixup_s1): + mov x0, #1 + ret + + /* Small tail case: Fault 8 bytes before dstend, >=16 bytes written */ +L(fixup_st8): + sub dst, dstend, #8 + add dstin, dstin, #16 +L(fixup_tail): + cmp dst, dstin + csel dst, dst, dstin, hi + b L(fixup_dst) + + /* Small/medium: Faults n bytes past dtsin, that much written */ +L(fixup_m64): + add dstin, dstin, #8 +L(fixup_m56): + add dstin, dstin, #8 +L(fixup_m48): + add dstin, dstin, #8 +L(fixup_m40): + add dstin, dstin, #8 +L(fixup_m32): + add dstin, dstin, #8 +L(fixup_m24): + add dstin, dstin, #8 +L(fixup_s16): + add dstin, dstin, #8 +L(fixup_s8): + add dstin, dstin, #4 +L(fixup_s4): + add dst, dstin, #4 + b L(fixup_dst) + + /* + * Medium tail cases: Faults n bytes before dstend, 64 or 32 bytes + * past dstin written, depending on original count + */ +L(fixup_mt56): + sub count, count, #8 +L(fixup_mt48): + sub count, count, #8 +L(fixup_mt40): + sub count, count, #8 +L(fixup_mt32): + sub count, count, #8 +L(fixup_mt24): + sub count, count, #8 +L(fixup_mt16): + sub count, count, #8 +L(fixup_mt8): + add count, count, #8 + add dst, dstin, count + + sub tmp1, dstend, dstin + cmp tmp1, #64 + add tmp1, dstin, #64 + add dstin, dstin, #32 + csel dstin, dstin, tmp1, ls + b L(fixup_tail) + + /* Large: Faults n bytes past dst, at least 16 bytes past dstin written */ +L(fixup_l72): + add dst, dst, #8 +L(fixup_l64): + add dst, dst, #8 +L(fixup_l56): + add dst, dst, #8 +L(fixup_l48): + add dst, dst, #8 +L(fixup_l40): + add dst, dst, #8 +L(fixup_l32): + add dst, dst, #8 +L(fixup_l24): + add dst, dst, #8 +L(fixup_l16): + add dst, dst, #16 + add dstin, dstin, #16 + b L(fixup_tail) + + /* Large tail: Faults n bytes before dstend, 80 bytes past dst written */ +L(fixup_lt64): + sub count, count, #8 +L(fixup_lt56): + sub count, count, #8 +L(fixup_lt48): + sub count, count, #8 +L(fixup_lt40): + sub count, count, #8 +L(fixup_lt32): + sub count, count, #8 +L(fixup_lt24): + sub count, count, #8 +L(fixup_lt16): + sub count, count, #8 +L(fixup_lt8): + add count, count, #56 /* Count was also off by 64 */ + add dstin, dst, #80 + add dst, dst, count + b L(fixup_tail) + SYM_FUNC_END(__arch_copy_to_user) EXPORT_SYMBOL(__arch_copy_to_user) -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Wed, 28 Sep 2022 12:58:53 +0100 Subject: [PATCH 11/13] arm64: Garbage-collect usercopy leftovers With both usercopy routines replaced, remove the now-unused template and supporting macros. Signed-off-by: Robin Murphy --- arch/arm64/include/asm/asm-uaccess.h | 30 ----- arch/arm64/lib/copy_template.S | 181 --------------------------- 2 files changed, 211 deletions(-) delete mode 100644 arch/arm64/lib/copy_template.S diff --git a/arch/arm64/include/asm/asm-uaccess.h b/arch/arm64/include/asm/asm-uaccess.h index 75b211c98dea..c2dea9b6b9e9 100644 --- a/arch/arm64/include/asm/asm-uaccess.h +++ b/arch/arm64/include/asm/asm-uaccess.h @@ -62,34 +62,4 @@ alternative_else_nop_endif #define USER(l, x...) \ 9999: x; \ _asm_extable_uaccess 9999b, l - -/* - * Generate the assembly for LDTR/STTR with exception table entries. - * This is complicated as there is no post-increment or pair versions of the - * unprivileged instructions, and USER() only works for single instructions. - */ - .macro user_ldp l, reg1, reg2, addr, post_inc -8888: ldtr \reg1, [\addr]; -8889: ldtr \reg2, [\addr, #8]; - add \addr, \addr, \post_inc; - - _asm_extable_uaccess 8888b, \l; - _asm_extable_uaccess 8889b, \l; - .endm - - .macro user_stp l, reg1, reg2, addr, post_inc -8888: sttr \reg1, [\addr]; -8889: sttr \reg2, [\addr, #8]; - add \addr, \addr, \post_inc; - - _asm_extable_uaccess 8888b,\l; - _asm_extable_uaccess 8889b,\l; - .endm - - .macro user_ldst l, inst, reg, addr, post_inc -8888: \inst \reg, [\addr]; - add \addr, \addr, \post_inc; - - _asm_extable_uaccess 8888b, \l; - .endm #endif diff --git a/arch/arm64/lib/copy_template.S b/arch/arm64/lib/copy_template.S deleted file mode 100644 index 488df234c49a..000000000000 --- a/arch/arm64/lib/copy_template.S +++ /dev/null @@ -1,181 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (C) 2013 ARM Ltd. - * Copyright (C) 2013 Linaro. - * - * This code is based on glibc cortex strings work originally authored by Linaro - * be found @ - * - * http://bazaar.launchpad.net/~linaro-toolchain-dev/cortex-strings/trunk/ - * files/head:/src/aarch64/ - */ - - -/* - * Copy a buffer from src to dest (alignment handled by the hardware) - * - * Parameters: - * x0 - dest - * x1 - src - * x2 - n - * Returns: - * x0 - dest - */ -dstin .req x0 -src .req x1 -count .req x2 -tmp1 .req x3 -tmp1w .req w3 -tmp2 .req x4 -tmp2w .req w4 -dst .req x6 - -A_l .req x7 -A_h .req x8 -B_l .req x9 -B_h .req x10 -C_l .req x11 -C_h .req x12 -D_l .req x13 -D_h .req x14 - - mov dst, dstin - cmp count, #16 - /*When memory length is less than 16, the accessed are not aligned.*/ - b.lo .Ltiny15 - - neg tmp2, src - ands tmp2, tmp2, #15/* Bytes to reach alignment. */ - b.eq .LSrcAligned - sub count, count, tmp2 - /* - * Copy the leading memory data from src to dst in an increasing - * address order.By this way,the risk of overwriting the source - * memory data is eliminated when the distance between src and - * dst is less than 16. The memory accesses here are alignment. - */ - tbz tmp2, #0, 1f - ldrb1 tmp1w, src, #1 - strb1 tmp1w, dst, #1 -1: - tbz tmp2, #1, 2f - ldrh1 tmp1w, src, #2 - strh1 tmp1w, dst, #2 -2: - tbz tmp2, #2, 3f - ldr1 tmp1w, src, #4 - str1 tmp1w, dst, #4 -3: - tbz tmp2, #3, .LSrcAligned - ldr1 tmp1, src, #8 - str1 tmp1, dst, #8 - -.LSrcAligned: - cmp count, #64 - b.ge .Lcpy_over64 - /* - * Deal with small copies quickly by dropping straight into the - * exit block. - */ -.Ltail63: - /* - * Copy up to 48 bytes of data. At this point we only need the - * bottom 6 bits of count to be accurate. - */ - ands tmp1, count, #0x30 - b.eq .Ltiny15 - cmp tmp1w, #0x20 - b.eq 1f - b.lt 2f - ldp1 A_l, A_h, src, #16 - stp1 A_l, A_h, dst, #16 -1: - ldp1 A_l, A_h, src, #16 - stp1 A_l, A_h, dst, #16 -2: - ldp1 A_l, A_h, src, #16 - stp1 A_l, A_h, dst, #16 -.Ltiny15: - /* - * Prefer to break one ldp/stp into several load/store to access - * memory in an increasing address order,rather than to load/store 16 - * bytes from (src-16) to (dst-16) and to backward the src to aligned - * address,which way is used in original cortex memcpy. If keeping - * the original memcpy process here, memmove need to satisfy the - * precondition that src address is at least 16 bytes bigger than dst - * address,otherwise some source data will be overwritten when memove - * call memcpy directly. To make memmove simpler and decouple the - * memcpy's dependency on memmove, withdrew the original process. - */ - tbz count, #3, 1f - ldr1 tmp1, src, #8 - str1 tmp1, dst, #8 -1: - tbz count, #2, 2f - ldr1 tmp1w, src, #4 - str1 tmp1w, dst, #4 -2: - tbz count, #1, 3f - ldrh1 tmp1w, src, #2 - strh1 tmp1w, dst, #2 -3: - tbz count, #0, .Lexitfunc - ldrb1 tmp1w, src, #1 - strb1 tmp1w, dst, #1 - - b .Lexitfunc - -.Lcpy_over64: - subs count, count, #128 - b.ge .Lcpy_body_large - /* - * Less than 128 bytes to copy, so handle 64 here and then jump - * to the tail. - */ - ldp1 A_l, A_h, src, #16 - stp1 A_l, A_h, dst, #16 - ldp1 B_l, B_h, src, #16 - ldp1 C_l, C_h, src, #16 - stp1 B_l, B_h, dst, #16 - stp1 C_l, C_h, dst, #16 - ldp1 D_l, D_h, src, #16 - stp1 D_l, D_h, dst, #16 - - tst count, #0x3f - b.ne .Ltail63 - b .Lexitfunc - - /* - * Critical loop. Start at a new cache line boundary. Assuming - * 64 bytes per line this ensures the entire loop is in one line. - */ - .p2align L1_CACHE_SHIFT -.Lcpy_body_large: - /* pre-get 64 bytes data. */ - ldp1 A_l, A_h, src, #16 - ldp1 B_l, B_h, src, #16 - ldp1 C_l, C_h, src, #16 - ldp1 D_l, D_h, src, #16 -1: - /* - * interlace the load of next 64 bytes data block with store of the last - * loaded 64 bytes data. - */ - stp1 A_l, A_h, dst, #16 - ldp1 A_l, A_h, src, #16 - stp1 B_l, B_h, dst, #16 - ldp1 B_l, B_h, src, #16 - stp1 C_l, C_h, dst, #16 - ldp1 C_l, C_h, src, #16 - stp1 D_l, D_h, dst, #16 - ldp1 D_l, D_h, src, #16 - subs count, count, #64 - b.ge 1b - stp1 A_l, A_h, dst, #16 - stp1 B_l, B_h, dst, #16 - stp1 C_l, C_h, dst, #16 - stp1 D_l, D_h, dst, #16 - - tst count, #0x3f - b.ne .Ltail63 -.Lexitfunc: -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Jon Nettleton Date: Tue, 22 Nov 2022 13:31:13 +0100 Subject: [PATCH 12/13] soc: fsl: dpio: Update for better ACPI support Update the lx2160a_soc to match on the soc_id provided by TF-A via the SMCCC_ID request. This allows the DPIO driver to be configured correctly without patches the fsl-guts driver for ACPI support using an in kernel standard mechanism. Signed-off-by: Jon Nettleton --- drivers/soc/fsl/dpio/dpio-driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/soc/fsl/dpio/dpio-driver.c b/drivers/soc/fsl/dpio/dpio-driver.c index 5a2edc48dd79..6c99d20f9d8d 100644 --- a/drivers/soc/fsl/dpio/dpio-driver.c +++ b/drivers/soc/fsl/dpio/dpio-driver.c @@ -50,6 +50,7 @@ static const struct soc_device_attribute ls2088a_soc[] = { static const struct soc_device_attribute lx2160a_soc[] = { {.family = "QorIQ LX2160A"}, + {.soc_id = "jep106:0015:8736"}, { /* sentinel */ } }; -- 2.39.1 From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Jon Nettleton Date: Thu, 5 Jan 2023 20:20:35 +0100 Subject: [PATCH 13/13] bus: fsl-mc-bus: fix dma error message for dpio devices Signed-off-by: Jon Nettleton --- drivers/bus/fsl-mc/fsl-mc-bus.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/bus/fsl-mc/fsl-mc-bus.c b/drivers/bus/fsl-mc/fsl-mc-bus.c index 6143dbf31f31..1a89b9e7ff7c 100644 --- a/drivers/bus/fsl-mc/fsl-mc-bus.c +++ b/drivers/bus/fsl-mc/fsl-mc-bus.c @@ -819,6 +819,9 @@ int fsl_mc_device_add(struct fsl_mc_obj_desc *obj_desc, mc_dev->obj_desc = *obj_desc; mc_dev->mc_io = mc_io; + mc_dev->dma_mask = FSL_MC_DEFAULT_DMA_MASK; + mc_dev->dev.dma_mask = &mc_dev->dma_mask; + mc_dev->dev.coherent_dma_mask = mc_dev->dma_mask; device_initialize(&mc_dev->dev); mc_dev->dev.parent = parent_dev; mc_dev->dev.bus = &fsl_mc_bus_type; @@ -872,9 +875,6 @@ int fsl_mc_device_add(struct fsl_mc_obj_desc *obj_desc, * parent's ICID and interrupt domain. */ mc_dev->icid = parent_mc_dev->icid; - mc_dev->dma_mask = FSL_MC_DEFAULT_DMA_MASK; - mc_dev->dev.dma_mask = &mc_dev->dma_mask; - mc_dev->dev.coherent_dma_mask = mc_dev->dma_mask; dev_set_msi_domain(&mc_dev->dev, dev_get_msi_domain(&parent_mc_dev->dev)); } -- 2.39.1