Merge branch 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 platform changes from Ingo Molnar:
 "The main changes in this cycle were:

   - SGI UV updates (Andrew Banman)

   - Intel MID updates (Andy Shevchenko)

   - Initial Mellanox systems platform (Vadim Pasternak)"

* 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/platform/mellanox: Fix return value check in mlxplat_init()
  x86/platform/mellanox: Introduce support for Mellanox systems platform
  x86/platform/uv/BAU: Add UV4-specific functions
  x86/platform/uv/BAU: Fix payload queue setup on UV4 hardware
  x86/platform/uv/BAU: Disable software timeout on UV4 hardware
  x86/platform/uv/BAU: Populate ->uvhub_version with UV4 version information
  x86/platform/uv/BAU: Use generic function pointers
  x86/platform/uv/BAU: Add generic function pointers
  x86/platform/uv/BAU: Convert uv_physnodeaddr() use to uv_gpa_to_offset()
  x86/platform/uv/BAU: Clean up pq_init()
  x86/platform/uv/BAU: Clean up and update printks
  x86/platform/uv/BAU: Clean up vertical alignment
  x86/platform/intel-mid: Keep SRAM powered on at boot
  x86/platform/intel-mid: Add Intel Penwell to ID table
  x86/cpu: Rename Merrifield2 to Moorefield
  x86/platform/intel-mid: Implement power off sequence
  x86/platform/intel-mid: Enable SD card detection on Merrifield
  x86/platform/intel-mid: Enable WiFi on Intel Edison
  x86/platform/intel-mid: Run PWRMU command immediately

15 files changed, 655 insertions, 85 deletions

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index ac5944fa6da2..9b2d50a73a11 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -550,6 +550,18 @@ config X86_INTEL_QUARK
 	  Say Y here if you have a Quark based system such as the Arduino
 	  compatible Intel Galileo.
 
+config MLX_PLATFORM
+	tristate "Mellanox Technologies platform support"
+	depends on X86_64
+	depends on X86_EXTENDED_PLATFORM
+	---help---
+	  This option enables system support for the Mellanox Technologies
+	  platform.
+
+	  Say Y here if you are building a kernel for Mellanox system.
+
+	  Otherwise, say N.
+
 config X86_INTEL_LPSS
 	bool "Intel Low Power Subsystem Support"
 	depends on X86 && ACPI
diff --git a/arch/x86/include/asm/intel-family.h b/arch/x86/include/asm/intel-family.h
index 627719475457..9ae5ab80a497 100644
--- a/arch/x86/include/asm/intel-family.h
+++ b/arch/x86/include/asm/intel-family.h
@@ -56,8 +56,8 @@
 #define INTEL_FAM6_ATOM_SILVERMONT1	0x37 /* BayTrail/BYT / Valleyview */
 #define INTEL_FAM6_ATOM_SILVERMONT2	0x4D /* Avaton/Rangely */
 #define INTEL_FAM6_ATOM_AIRMONT		0x4C /* CherryTrail / Braswell */
-#define INTEL_FAM6_ATOM_MERRIFIELD1	0x4A /* Tangier */
-#define INTEL_FAM6_ATOM_MERRIFIELD2	0x5A /* Annidale */
+#define INTEL_FAM6_ATOM_MERRIFIELD	0x4A /* Tangier */
+#define INTEL_FAM6_ATOM_MOOREFIELD	0x5A /* Annidale */
 #define INTEL_FAM6_ATOM_GOLDMONT	0x5C
 #define INTEL_FAM6_ATOM_DENVERTON	0x5F /* Goldmont Microserver */
 
diff --git a/arch/x86/include/asm/intel-mid.h b/arch/x86/include/asm/intel-mid.h
index 9d6b097aa73d..5b6753d1f7f4 100644
--- a/arch/x86/include/asm/intel-mid.h
+++ b/arch/x86/include/asm/intel-mid.h
@@ -18,6 +18,8 @@
 extern int intel_mid_pci_init(void);
 extern int intel_mid_pci_set_power_state(struct pci_dev *pdev, pci_power_t state);
 
+extern void intel_mid_pwr_power_off(void);
+
 #define INTEL_MID_PWR_LSS_OFFSET	4
 #define INTEL_MID_PWR_LSS_TYPE		(1 << 7)
 
diff --git a/arch/x86/include/asm/intel_scu_ipc.h b/arch/x86/include/asm/intel_scu_ipc.h
index 925b605eb5c6..4fb1d0abef95 100644
--- a/arch/x86/include/asm/intel_scu_ipc.h
+++ b/arch/x86/include/asm/intel_scu_ipc.h
@@ -3,6 +3,8 @@
 
 #include <linux/notifier.h>
 
+#define IPCMSG_COLD_OFF		0x80	/* Only for Tangier */
+
 #define IPCMSG_WARM_RESET	0xF0
 #define IPCMSG_COLD_RESET	0xF1
 #define IPCMSG_SOFT_RESET	0xF2
diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h
index cc44d926c17e..57ab86d94d64 100644
--- a/arch/x86/include/asm/uv/uv_bau.h
+++ b/arch/x86/include/asm/uv/uv_bau.h
@@ -49,14 +49,12 @@
 #define UV_NET_ENDPOINT_INTD		(is_uv1_hub() ?			\
 			UV1_NET_ENDPOINT_INTD : UV2_NET_ENDPOINT_INTD)
 #define UV_DESC_PSHIFT			49
-#define UV_PAYLOADQ_PNODE_SHIFT		49
+#define UV_PAYLOADQ_GNODE_SHIFT		49
 #define UV_PTC_BASENAME			"sgi_uv/ptc_statistics"
 #define UV_BAU_BASENAME			"sgi_uv/bau_tunables"
 #define UV_BAU_TUNABLES_DIR		"sgi_uv"
 #define UV_BAU_TUNABLES_FILE		"bau_tunables"
 #define WHITESPACE			" \t\n"
-#define uv_mmask			((1UL << uv_hub_info->m_val) - 1)
-#define uv_physnodeaddr(x)		((__pa((unsigned long)(x)) & uv_mmask))
 #define cpubit_isset(cpu, bau_local_cpumask) \
 	test_bit((cpu), (bau_local_cpumask).bits)
 
@@ -387,6 +385,17 @@ struct uv2_3_bau_msg_header {
 	/* bits 127:120 */
 };
 
+/* Abstracted BAU functions */
+struct bau_operations {
+	unsigned long (*read_l_sw_ack)(void);
+	unsigned long (*read_g_sw_ack)(int pnode);
+	unsigned long (*bau_gpa_to_offset)(unsigned long vaddr);
+	void (*write_l_sw_ack)(unsigned long mmr);
+	void (*write_g_sw_ack)(int pnode, unsigned long mmr);
+	void (*write_payload_first)(int pnode, unsigned long mmr);
+	void (*write_payload_last)(int pnode, unsigned long mmr);
+};
+
 /*
  * The activation descriptor:
  * The format of the message to send, plus all accompanying control
@@ -655,6 +664,16 @@ static inline void write_gmmr_activation(int pnode, unsigned long mmr_image)
 	write_gmmr(pnode, UVH_LB_BAU_SB_ACTIVATION_CONTROL, mmr_image);
 }
 
+static inline void write_mmr_proc_payload_first(int pnode, unsigned long mmr_image)
+{
+	write_gmmr(pnode, UV4H_LB_PROC_INTD_QUEUE_FIRST, mmr_image);
+}
+
+static inline void write_mmr_proc_payload_last(int pnode, unsigned long mmr_image)
+{
+	write_gmmr(pnode, UV4H_LB_PROC_INTD_QUEUE_LAST, mmr_image);
+}
+
 static inline void write_mmr_payload_first(int pnode, unsigned long mmr_image)
 {
 	write_gmmr(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST, mmr_image);
@@ -700,6 +719,26 @@ static inline unsigned long read_gmmr_sw_ack(int pnode)
 	return read_gmmr(pnode, UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE);
 }
 
+static inline void write_mmr_proc_sw_ack(unsigned long mr)
+{
+	uv_write_local_mmr(UV4H_LB_PROC_INTD_SOFT_ACK_CLEAR, mr);
+}
+
+static inline void write_gmmr_proc_sw_ack(int pnode, unsigned long mr)
+{
+	write_gmmr(pnode, UV4H_LB_PROC_INTD_SOFT_ACK_CLEAR, mr);
+}
+
+static inline unsigned long read_mmr_proc_sw_ack(void)
+{
+	return read_lmmr(UV4H_LB_PROC_INTD_SOFT_ACK_PENDING);
+}
+
+static inline unsigned long read_gmmr_proc_sw_ack(int pnode)
+{
+	return read_gmmr(pnode, UV4H_LB_PROC_INTD_SOFT_ACK_PENDING);
+}
+
 static inline void write_mmr_data_config(int pnode, unsigned long mr)
 {
 	uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, mr);
diff --git a/arch/x86/platform/Makefile b/arch/x86/platform/Makefile
index 184842ef332e..3c3c19ea94df 100644
--- a/arch/x86/platform/Makefile
+++ b/arch/x86/platform/Makefile
@@ -8,6 +8,7 @@ obj-y	+= iris/
 obj-y	+= intel/
 obj-y	+= intel-mid/
 obj-y	+= intel-quark/
+obj-y	+= mellanox/
 obj-y	+= olpc/
 obj-y	+= scx200/
 obj-y	+= sfi/
diff --git a/arch/x86/platform/atom/punit_atom_debug.c b/arch/x86/platform/atom/punit_atom_debug.c
index 8ff7b9355416..d49d3be81953 100644
--- a/arch/x86/platform/atom/punit_atom_debug.c
+++ b/arch/x86/platform/atom/punit_atom_debug.c
@@ -155,7 +155,7 @@ static void punit_dbgfs_unregister(void)
 
 static const struct x86_cpu_id intel_punit_cpu_ids[] = {
 	ICPU(INTEL_FAM6_ATOM_SILVERMONT1, punit_device_byt),
-	ICPU(INTEL_FAM6_ATOM_MERRIFIELD1, punit_device_tng),
+	ICPU(INTEL_FAM6_ATOM_MERRIFIELD,  punit_device_tng),
 	ICPU(INTEL_FAM6_ATOM_AIRMONT,	  punit_device_cht),
 	{}
 };
diff --git a/arch/x86/platform/intel-mid/device_libs/Makefile b/arch/x86/platform/intel-mid/device_libs/Makefile
index fc135bf70511..429d08be7848 100644
--- a/arch/x86/platform/intel-mid/device_libs/Makefile
+++ b/arch/x86/platform/intel-mid/device_libs/Makefile
@@ -1,5 +1,9 @@
 # Family-Level Interface Shim (FLIS)
 obj-$(subst m,y,$(CONFIG_PINCTRL_MERRIFIELD)) += platform_mrfld_pinctrl.o
+# SDHCI Devices
+obj-$(subst m,y,$(CONFIG_MMC_SDHCI_PCI)) += platform_mrfld_sd.o
+# WiFi
+obj-$(subst m,y,$(CONFIG_BRCMFMAC_SDIO)) += platform_bcm43xx.o
 # IPC Devices
 obj-y += platform_ipc.o
 obj-$(subst m,y,$(CONFIG_MFD_INTEL_MSIC)) += platform_msic.o
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_bcm43xx.c b/arch/x86/platform/intel-mid/device_libs/platform_bcm43xx.c
new file mode 100644
index 000000000000..4392c15ed9e0
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_bcm43xx.c
@@ -0,0 +1,95 @@
+/*
+ * platform_bcm43xx.c: bcm43xx platform data initilization file
+ *
+ * (C) Copyright 2016 Intel Corporation
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * 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; version 2
+ * of the License.
+ */
+
+#include <linux/gpio.h>
+#include <linux/platform_device.h>
+#include <linux/regulator/machine.h>
+#include <linux/regulator/fixed.h>
+#include <linux/sfi.h>
+
+#include <asm/intel-mid.h>
+
+#define WLAN_SFI_GPIO_IRQ_NAME		"WLAN-interrupt"
+#define WLAN_SFI_GPIO_ENABLE_NAME	"WLAN-enable"
+
+#define WLAN_DEV_NAME			"0000:00:01.3"
+
+static struct regulator_consumer_supply bcm43xx_vmmc_supply = {
+	.dev_name		= WLAN_DEV_NAME,
+	.supply			= "vmmc",
+};
+
+static struct regulator_init_data bcm43xx_vmmc_data = {
+	.constraints = {
+		.valid_ops_mask		= REGULATOR_CHANGE_STATUS,
+	},
+	.num_consumer_supplies	= 1,
+	.consumer_supplies	= &bcm43xx_vmmc_supply,
+};
+
+static struct fixed_voltage_config bcm43xx_vmmc = {
+	.supply_name		= "bcm43xx-vmmc-regulator",
+	/*
+	 * Announce 2.0V here to be compatible with SDIO specification. The
+	 * real voltage and signaling are still 1.8V.
+	 */
+	.microvolts		= 2000000,		/* 1.8V */
+	.gpio			= -EINVAL,
+	.startup_delay		= 250 * 1000,		/* 250ms */
+	.enable_high		= 1,			/* active high */
+	.enabled_at_boot	= 0,			/* disabled at boot */
+	.init_data		= &bcm43xx_vmmc_data,
+};
+
+static struct platform_device bcm43xx_vmmc_regulator = {
+	.name		= "reg-fixed-voltage",
+	.id		= PLATFORM_DEVID_AUTO,
+	.dev = {
+		.platform_data	= &bcm43xx_vmmc,
+	},
+};
+
+static int __init bcm43xx_regulator_register(void)
+{
+	int ret;
+
+	bcm43xx_vmmc.gpio = get_gpio_by_name(WLAN_SFI_GPIO_ENABLE_NAME);
+	ret = platform_device_register(&bcm43xx_vmmc_regulator);
+	if (ret) {
+		pr_err("%s: vmmc regulator register failed\n", __func__);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void __init *bcm43xx_platform_data(void *info)
+{
+	int ret;
+
+	ret = bcm43xx_regulator_register();
+	if (ret)
+		return NULL;
+
+	pr_info("Using generic wifi platform data\n");
+
+	/* For now it's empty */
+	return NULL;
+}
+
+static const struct devs_id bcm43xx_clk_vmmc_dev_id __initconst = {
+	.name			= "bcm43xx_clk_vmmc",
+	.type			= SFI_DEV_TYPE_SD,
+	.get_platform_data	= &bcm43xx_platform_data,
+};
+
+sfi_device(bcm43xx_clk_vmmc_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mrfld_sd.c b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_sd.c
new file mode 100644
index 000000000000..00c4a034ad93
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_sd.c
@@ -0,0 +1,47 @@
+/*
+ * SDHCI platform data initilisation file
+ *
+ * (C) Copyright 2016 Intel Corporation
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * 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; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/pci.h>
+
+#include <linux/mmc/sdhci-pci-data.h>
+
+#include <asm/intel-mid.h>
+
+#define INTEL_MRFLD_SD			2
+#define INTEL_MRFLD_SD_CD_GPIO		77
+
+static struct sdhci_pci_data mrfld_sdhci_pci_data = {
+	.rst_n_gpio	= -EINVAL,
+	.cd_gpio	= INTEL_MRFLD_SD_CD_GPIO,
+};
+
+static struct sdhci_pci_data *
+mrfld_sdhci_pci_get_data(struct pci_dev *pdev, int slotno)
+{
+	unsigned int func = PCI_FUNC(pdev->devfn);
+
+	if (func == INTEL_MRFLD_SD)
+		return &mrfld_sdhci_pci_data;
+
+	return NULL;
+}
+
+static int __init mrfld_sd_init(void)
+{
+	if (intel_mid_identify_cpu() != INTEL_MID_CPU_CHIP_TANGIER)
+		return -ENODEV;
+
+	sdhci_pci_get_data = mrfld_sdhci_pci_get_data;
+	return 0;
+}
+arch_initcall(mrfld_sd_init);
diff --git a/arch/x86/platform/intel-mid/intel-mid.c b/arch/x86/platform/intel-mid/intel-mid.c
index ce119d2ba0d0..7850128f0026 100644
--- a/arch/x86/platform/intel-mid/intel-mid.c
+++ b/arch/x86/platform/intel-mid/intel-mid.c
@@ -70,6 +70,11 @@ EXPORT_SYMBOL_GPL(__intel_mid_cpu_chip);
 
 static void intel_mid_power_off(void)
 {
+	/* Shut down South Complex via PWRMU */
+	intel_mid_pwr_power_off();
+
+	/* Only for Tangier, the rest will ignore this command */
+	intel_scu_ipc_simple_command(IPCMSG_COLD_OFF, 1);
 };
 
 static void intel_mid_reboot(void)
diff --git a/arch/x86/platform/intel-mid/pwr.c b/arch/x86/platform/intel-mid/pwr.c
index c901a3423772..5d3b45ad1c03 100644
--- a/arch/x86/platform/intel-mid/pwr.c
+++ b/arch/x86/platform/intel-mid/pwr.c
@@ -44,7 +44,19 @@
 /* Bits in PM_CMD */
 #define PM_CMD_CMD(x)		((x) << 0)
 #define PM_CMD_IOC		(1 << 8)
-#define PM_CMD_D3cold		(1 << 21)
+#define PM_CMD_CM_NOP		(0 << 9)
+#define PM_CMD_CM_IMMEDIATE	(1 << 9)
+#define PM_CMD_CM_DELAY		(2 << 9)
+#define PM_CMD_CM_TRIGGER	(3 << 9)
+
+/* System states */
+#define PM_CMD_SYS_STATE_S5	(5 << 16)
+
+/* Trigger variants */
+#define PM_CMD_CFG_TRIGGER_NC	(3 << 19)
+
+/* Message to wait for TRIGGER_NC case */
+#define TRIGGER_NC_MSG_2	(2 << 22)
 
 /* List of commands */
 #define CMD_SET_CFG		0x01
@@ -137,7 +149,7 @@ static int mid_pwr_wait(struct mid_pwr *pwr)
 
 static int mid_pwr_wait_for_cmd(struct mid_pwr *pwr, u8 cmd)
 {
-	writel(PM_CMD_CMD(cmd), pwr->regs + PM_CMD);
+	writel(PM_CMD_CMD(cmd) | PM_CMD_CM_IMMEDIATE, pwr->regs + PM_CMD);
 	return mid_pwr_wait(pwr);
 }
 
@@ -260,6 +272,20 @@ int intel_mid_pci_set_power_state(struct pci_dev *pdev, pci_power_t state)
 }
 EXPORT_SYMBOL_GPL(intel_mid_pci_set_power_state);
 
+void intel_mid_pwr_power_off(void)
+{
+	struct mid_pwr *pwr = midpwr;
+	u32 cmd = PM_CMD_SYS_STATE_S5 |
+		  PM_CMD_CMD(CMD_SET_CFG) |
+		  PM_CMD_CM_TRIGGER |
+		  PM_CMD_CFG_TRIGGER_NC |
+		  TRIGGER_NC_MSG_2;
+
+	/* Send command to SCU */
+	writel(cmd, pwr->regs + PM_CMD);
+	mid_pwr_wait(pwr);
+}
+
 int intel_mid_pwr_get_lss_id(struct pci_dev *pdev)
 {
 	int vndr;
@@ -354,7 +380,7 @@ static int mid_pwr_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	return 0;
 }
 
-static int mid_set_initial_state(struct mid_pwr *pwr)
+static int mid_set_initial_state(struct mid_pwr *pwr, const u32 *states)
 {
 	unsigned int i, j;
 	int ret;
@@ -379,10 +405,10 @@ static int mid_set_initial_state(struct mid_pwr *pwr)
 	 * NOTE: The actual device mapping is provided by a platform at run
 	 * time using vendor capability of PCI configuration space.
 	 */
-	mid_pwr_set_state(pwr, 0, 0xffffffff);
-	mid_pwr_set_state(pwr, 1, 0xffffffff);
-	mid_pwr_set_state(pwr, 2, 0xffffffff);
-	mid_pwr_set_state(pwr, 3, 0xffffffff);
+	mid_pwr_set_state(pwr, 0, states[0]);
+	mid_pwr_set_state(pwr, 1, states[1]);
+	mid_pwr_set_state(pwr, 2, states[2]);
+	mid_pwr_set_state(pwr, 3, states[3]);
 
 	/* Send command to SCU */
 	ret = mid_pwr_wait_for_cmd(pwr, CMD_SET_CFG);
@@ -397,13 +423,41 @@ static int mid_set_initial_state(struct mid_pwr *pwr)
 	return 0;
 }
 
-static const struct mid_pwr_device_info mid_info = {
-	.set_initial_state = mid_set_initial_state,
+static int pnw_set_initial_state(struct mid_pwr *pwr)
+{
+	/* On Penwell SRAM must stay powered on */
+	const u32 states[] = {
+		0xf00fffff,		/* PM_SSC(0) */
+		0xffffffff,		/* PM_SSC(1) */
+		0xffffffff,		/* PM_SSC(2) */
+		0xffffffff,		/* PM_SSC(3) */
+	};
+	return mid_set_initial_state(pwr, states);
+}
+
+static int tng_set_initial_state(struct mid_pwr *pwr)
+{
+	const u32 states[] = {
+		0xffffffff,		/* PM_SSC(0) */
+		0xffffffff,		/* PM_SSC(1) */
+		0xffffffff,		/* PM_SSC(2) */
+		0xffffffff,		/* PM_SSC(3) */
+	};
+	return mid_set_initial_state(pwr, states);
+}
+
+static const struct mid_pwr_device_info pnw_info = {
+	.set_initial_state = pnw_set_initial_state,
+};
+
+static const struct mid_pwr_device_info tng_info = {
+	.set_initial_state = tng_set_initial_state,
 };
 
+/* This table should be in sync with the one in drivers/pci/pci-mid.c */
 static const struct pci_device_id mid_pwr_pci_ids[] = {
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_PENWELL), (kernel_ulong_t)&mid_info },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_TANGIER), (kernel_ulong_t)&mid_info },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_PENWELL), (kernel_ulong_t)&pnw_info },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_TANGIER), (kernel_ulong_t)&tng_info },
 	{}
 };
 
diff --git a/arch/x86/platform/mellanox/Makefile b/arch/x86/platform/mellanox/Makefile
new file mode 100644
index 000000000000..f43c93188a1d
--- /dev/null
+++ b/arch/x86/platform/mellanox/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_MLX_PLATFORM)	+= mlx-platform.o
diff --git a/arch/x86/platform/mellanox/mlx-platform.c b/arch/x86/platform/mellanox/mlx-platform.c
new file mode 100644
index 000000000000..7dcfcca97399
--- /dev/null
+++ b/arch/x86/platform/mellanox/mlx-platform.c
@@ -0,0 +1,266 @@
+/*
+ * arch/x86/platform/mellanox/mlx-platform.c
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Vadim Pasternak <vadimp@mellanox.com>
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ *    contributors may be used to endorse or promote products derived from
+ *    this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/device.h>
+#include <linux/dmi.h>
+#include <linux/i2c.h>
+#include <linux/i2c-mux.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/platform_data/i2c-mux-reg.h>
+
+#define MLX_PLAT_DEVICE_NAME		"mlxplat"
+
+/* LPC bus IO offsets */
+#define MLXPLAT_CPLD_LPC_I2C_BASE_ADRR		0x2000
+#define MLXPLAT_CPLD_LPC_REG_BASE_ADRR		0x2500
+#define MLXPLAT_CPLD_LPC_IO_RANGE		0x100
+#define MLXPLAT_CPLD_LPC_I2C_CH1_OFF		0xdb
+#define MLXPLAT_CPLD_LPC_I2C_CH2_OFF		0xda
+#define MLXPLAT_CPLD_LPC_PIO_OFFSET		0x10000UL
+#define MLXPLAT_CPLD_LPC_REG1	((MLXPLAT_CPLD_LPC_REG_BASE_ADRR + \
+				  MLXPLAT_CPLD_LPC_I2C_CH1_OFF) | \
+				  MLXPLAT_CPLD_LPC_PIO_OFFSET)
+#define MLXPLAT_CPLD_LPC_REG2	((MLXPLAT_CPLD_LPC_REG_BASE_ADRR + \
+				  MLXPLAT_CPLD_LPC_I2C_CH2_OFF) | \
+				  MLXPLAT_CPLD_LPC_PIO_OFFSET)
+
+/* Start channel numbers */
+#define MLXPLAT_CPLD_CH1			2
+#define MLXPLAT_CPLD_CH2			10
+
+/* Number of LPC attached MUX platform devices */
+#define MLXPLAT_CPLD_LPC_MUX_DEVS		2
+
+/* mlxplat_priv - platform private data
+ * @pdev_i2c - i2c controller platform device
+ * @pdev_mux - array of mux platform devices
+ */
+struct mlxplat_priv {
+	struct platform_device *pdev_i2c;
+	struct platform_device *pdev_mux[MLXPLAT_CPLD_LPC_MUX_DEVS];
+};
+
+/* Regions for LPC I2C controller and LPC base register space */
+static const struct resource mlxplat_lpc_resources[] = {
+	[0] = DEFINE_RES_NAMED(MLXPLAT_CPLD_LPC_I2C_BASE_ADRR,
+			       MLXPLAT_CPLD_LPC_IO_RANGE,
+			       "mlxplat_cpld_lpc_i2c_ctrl", IORESOURCE_IO),
+	[1] = DEFINE_RES_NAMED(MLXPLAT_CPLD_LPC_REG_BASE_ADRR,
+			       MLXPLAT_CPLD_LPC_IO_RANGE,
+			       "mlxplat_cpld_lpc_regs",
+			       IORESOURCE_IO),
+};
+
+/* Platform default channels */
+static const int mlxplat_default_channels[][8] = {
+	{
+		MLXPLAT_CPLD_CH1, MLXPLAT_CPLD_CH1 + 1, MLXPLAT_CPLD_CH1 + 2,
+		MLXPLAT_CPLD_CH1 + 3, MLXPLAT_CPLD_CH1 + 4, MLXPLAT_CPLD_CH1 +
+		5, MLXPLAT_CPLD_CH1 + 6, MLXPLAT_CPLD_CH1 + 7
+	},
+	{
+		MLXPLAT_CPLD_CH2, MLXPLAT_CPLD_CH2 + 1, MLXPLAT_CPLD_CH2 + 2,
+		MLXPLAT_CPLD_CH2 + 3, MLXPLAT_CPLD_CH2 + 4, MLXPLAT_CPLD_CH2 +
+		5, MLXPLAT_CPLD_CH2 + 6, MLXPLAT_CPLD_CH2 + 7
+	},
+};
+
+/* Platform channels for MSN21xx system family */
+static const int mlxplat_msn21xx_channels[] = { 1, 2, 3, 4, 5, 6, 7, 8 };
+
+/* Platform mux data */
+static struct i2c_mux_reg_platform_data mlxplat_mux_data[] = {
+	{
+		.parent = 1,
+		.base_nr = MLXPLAT_CPLD_CH1,
+		.write_only = 1,
+		.reg = (void __iomem *)MLXPLAT_CPLD_LPC_REG1,
+		.reg_size = 1,
+		.idle_in_use = 1,
+	},
+	{
+		.parent = 1,
+		.base_nr = MLXPLAT_CPLD_CH2,
+		.write_only = 1,
+		.reg = (void __iomem *)MLXPLAT_CPLD_LPC_REG2,
+		.reg_size = 1,
+		.idle_in_use = 1,
+	},
+
+};
+
+static struct platform_device *mlxplat_dev;
+
+static int __init mlxplat_dmi_default_matched(const struct dmi_system_id *dmi)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(mlxplat_mux_data); i++) {
+		mlxplat_mux_data[i].values = mlxplat_default_channels[i];
+		mlxplat_mux_data[i].n_values =
+				ARRAY_SIZE(mlxplat_default_channels[i]);
+	}
+
+	return 1;
+};
+
+static int __init mlxplat_dmi_msn21xx_matched(const struct dmi_system_id *dmi)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(mlxplat_mux_data); i++) {
+		mlxplat_mux_data[i].values = mlxplat_msn21xx_channels;
+		mlxplat_mux_data[i].n_values =
+				ARRAY_SIZE(mlxplat_msn21xx_channels);
+	}
+
+	return 1;
+};
+
+static struct dmi_system_id mlxplat_dmi_table[] __initdata = {
+	{
+		.callback = mlxplat_dmi_default_matched,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Mellanox Technologies"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MSN24"),
+		},
+	},
+	{
+		.callback = mlxplat_dmi_default_matched,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Mellanox Technologies"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MSN27"),
+		},
+	},
+	{
+		.callback = mlxplat_dmi_default_matched,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Mellanox Technologies"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MSB"),
+		},
+	},
+	{
+		.callback = mlxplat_dmi_default_matched,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Mellanox Technologies"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MSX"),
+		},
+	},
+	{
+		.callback = mlxplat_dmi_msn21xx_matched,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Mellanox Technologies"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MSN21"),
+		},
+	},
+	{ }
+};
+
+static int __init mlxplat_init(void)
+{
+	struct mlxplat_priv *priv;
+	int i, err;
+
+	if (!dmi_check_system(mlxplat_dmi_table))
+		return -ENODEV;
+
+	mlxplat_dev = platform_device_register_simple(MLX_PLAT_DEVICE_NAME, -1,
+					mlxplat_lpc_resources,
+					ARRAY_SIZE(mlxplat_lpc_resources));
+
+	if (IS_ERR(mlxplat_dev))
+		return PTR_ERR(mlxplat_dev);
+
+	priv = devm_kzalloc(&mlxplat_dev->dev, sizeof(struct mlxplat_priv),
+			    GFP_KERNEL);
+	if (!priv) {
+		err = -ENOMEM;
+		goto fail_alloc;
+	}
+	platform_set_drvdata(mlxplat_dev, priv);
+
+	priv->pdev_i2c = platform_device_register_simple("i2c_mlxcpld", -1,
+							 NULL, 0);
+	if (IS_ERR(priv->pdev_i2c)) {
+		err = PTR_ERR(priv->pdev_i2c);
+		goto fail_alloc;
+	};
+
+	for (i = 0; i < ARRAY_SIZE(mlxplat_mux_data); i++) {
+		priv->pdev_mux[i] = platform_device_register_resndata(
+						&mlxplat_dev->dev,
+						"i2c-mux-reg", i, NULL,
+						0, &mlxplat_mux_data[i],
+						sizeof(mlxplat_mux_data[i]));
+		if (IS_ERR(priv->pdev_mux[i])) {
+			err = PTR_ERR(priv->pdev_mux[i]);
+			goto fail_platform_mux_register;
+		}
+	}
+
+	return 0;
+
+fail_platform_mux_register:
+	for (i--; i > 0 ; i--)
+		platform_device_unregister(priv->pdev_mux[i]);
+	platform_device_unregister(priv->pdev_i2c);
+fail_alloc:
+	platform_device_unregister(mlxplat_dev);
+
+	return err;
+}
+module_init(mlxplat_init);
+
+static void __exit mlxplat_exit(void)
+{
+	struct mlxplat_priv *priv = platform_get_drvdata(mlxplat_dev);
+	int i;
+
+	for (i = ARRAY_SIZE(mlxplat_mux_data) - 1; i >= 0 ; i--)
+		platform_device_unregister(priv->pdev_mux[i]);
+
+	platform_device_unregister(priv->pdev_i2c);
+	platform_device_unregister(mlxplat_dev);
+}
+module_exit(mlxplat_exit);
+
+MODULE_AUTHOR("Vadim Pasternak (vadimp@mellanox.com)");
+MODULE_DESCRIPTION("Mellanox platform driver");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_ALIAS("dmi:*:*Mellanox*:MSN24*:");
+MODULE_ALIAS("dmi:*:*Mellanox*:MSN27*:");
+MODULE_ALIAS("dmi:*:*Mellanox*:MSB*:");
+MODULE_ALIAS("dmi:*:*Mellanox*:MSX*:");
+MODULE_ALIAS("dmi:*:*Mellanox*:MSN21*:");
diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c
index 276e1b7bba3c..9e42842e924a 100644
--- a/arch/x86/platform/uv/tlb_uv.c
+++ b/arch/x86/platform/uv/tlb_uv.c
@@ -24,6 +24,29 @@
 #include <asm/irq_vectors.h>
 #include <asm/timer.h>
 
+static struct bau_operations ops;
+
+static struct bau_operations uv123_bau_ops = {
+	.bau_gpa_to_offset       = uv_gpa_to_offset,
+	.read_l_sw_ack           = read_mmr_sw_ack,
+	.read_g_sw_ack           = read_gmmr_sw_ack,
+	.write_l_sw_ack          = write_mmr_sw_ack,
+	.write_g_sw_ack          = write_gmmr_sw_ack,
+	.write_payload_first     = write_mmr_payload_first,
+	.write_payload_last      = write_mmr_payload_last,
+};
+
+static struct bau_operations uv4_bau_ops = {
+	.bau_gpa_to_offset       = uv_gpa_to_soc_phys_ram,
+	.read_l_sw_ack           = read_mmr_proc_sw_ack,
+	.read_g_sw_ack           = read_gmmr_proc_sw_ack,
+	.write_l_sw_ack          = write_mmr_proc_sw_ack,
+	.write_g_sw_ack          = write_gmmr_proc_sw_ack,
+	.write_payload_first     = write_mmr_proc_payload_first,
+	.write_payload_last      = write_mmr_proc_payload_last,
+};
+
+
 /* timeouts in nanoseconds (indexed by UVH_AGING_PRESCALE_SEL urgency7 30:28) */
 static int timeout_base_ns[] = {
 		20,
@@ -55,16 +78,16 @@ static int congested_reps	= CONGESTED_REPS;
 static int disabled_period	= DISABLED_PERIOD;
 
 static struct tunables tunables[] = {
-	{&max_concurr, MAX_BAU_CONCURRENT}, /* must be [0] */
-	{&plugged_delay, PLUGGED_DELAY},
-	{&plugsb4reset, PLUGSB4RESET},
-	{&timeoutsb4reset, TIMEOUTSB4RESET},
-	{&ipi_reset_limit, IPI_RESET_LIMIT},
-	{&complete_threshold, COMPLETE_THRESHOLD},
-	{&congested_respns_us, CONGESTED_RESPONSE_US},
-	{&congested_reps, CONGESTED_REPS},
-	{&disabled_period, DISABLED_PERIOD},
-	{&giveup_limit, GIVEUP_LIMIT}
+	{&max_concurr,           MAX_BAU_CONCURRENT}, /* must be [0] */
+	{&plugged_delay,         PLUGGED_DELAY},
+	{&plugsb4reset,          PLUGSB4RESET},
+	{&timeoutsb4reset,       TIMEOUTSB4RESET},
+	{&ipi_reset_limit,       IPI_RESET_LIMIT},
+	{&complete_threshold,    COMPLETE_THRESHOLD},
+	{&congested_respns_us,   CONGESTED_RESPONSE_US},
+	{&congested_reps,        CONGESTED_REPS},
+	{&disabled_period,       DISABLED_PERIOD},
+	{&giveup_limit,          GIVEUP_LIMIT}
 };
 
 static struct dentry *tunables_dir;
@@ -216,7 +239,7 @@ static void reply_to_message(struct msg_desc *mdp, struct bau_control *bcp,
 	msg = mdp->msg;
 	if (!msg->canceled && do_acknowledge) {
 		dw = (msg->swack_vec << UV_SW_ACK_NPENDING) | msg->swack_vec;
-		write_mmr_sw_ack(dw);
+		ops.write_l_sw_ack(dw);
 	}
 	msg->replied_to = 1;
 	msg->swack_vec = 0;
@@ -252,7 +275,7 @@ static void bau_process_retry_msg(struct msg_desc *mdp,
 			msg->swack_vec) == 0) &&
 		    (msg2->sending_cpu == msg->sending_cpu) &&
 		    (msg2->msg_type != MSG_NOOP)) {
-			mmr = read_mmr_sw_ack();
+			mmr = ops.read_l_sw_ack();
 			msg_res = msg2->swack_vec;
 			/*
 			 * This is a message retry; clear the resources held
@@ -270,7 +293,7 @@ static void bau_process_retry_msg(struct msg_desc *mdp,
 				stat->d_canceled++;
 				cancel_count++;
 				mr = (msg_res << UV_SW_ACK_NPENDING) | msg_res;
-				write_mmr_sw_ack(mr);
+				ops.write_l_sw_ack(mr);
 			}
 		}
 	}
@@ -403,12 +426,12 @@ static void do_reset(void *ptr)
 			/*
 			 * only reset the resource if it is still pending
 			 */
-			mmr = read_mmr_sw_ack();
+			mmr = ops.read_l_sw_ack();
 			msg_res = msg->swack_vec;
 			mr = (msg_res << UV_SW_ACK_NPENDING) | msg_res;
 			if (mmr & msg_res) {
 				stat->d_rcanceled++;
-				write_mmr_sw_ack(mr);
+				ops.write_l_sw_ack(mr);
 			}
 		}
 	}
@@ -1198,7 +1221,7 @@ void process_uv2_message(struct msg_desc *mdp, struct bau_control *bcp)
 	struct bau_pq_entry *msg = mdp->msg;
 	struct bau_pq_entry *other_msg;
 
-	mmr_image = read_mmr_sw_ack();
+	mmr_image = ops.read_l_sw_ack();
 	swack_vec = msg->swack_vec;
 
 	if ((swack_vec & mmr_image) == 0) {
@@ -1427,7 +1450,7 @@ static int ptc_seq_show(struct seq_file *file, void *data)
 		/* destination side statistics */
 		seq_printf(file,
 			"%lx %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld\n",
-			   read_gmmr_sw_ack(uv_cpu_to_pnode(cpu)),
+			   ops.read_g_sw_ack(uv_cpu_to_pnode(cpu)),
 			   stat->d_requestee, cycles_2_us(stat->d_time),
 			   stat->d_alltlb, stat->d_onetlb, stat->d_multmsg,
 			   stat->d_nomsg, stat->d_retries, stat->d_canceled,
@@ -1493,16 +1516,16 @@ static ssize_t ptc_proc_write(struct file *file, const char __user *user,
 	}
 
 	if (kstrtol(optstr, 10, &input_arg) < 0) {
-		printk(KERN_DEBUG "%s is invalid\n", optstr);
+		pr_debug("%s is invalid\n", optstr);
 		return -EINVAL;
 	}
 
 	if (input_arg == 0) {
 		elements = ARRAY_SIZE(stat_description);
-		printk(KERN_DEBUG "# cpu:      cpu number\n");
-		printk(KERN_DEBUG "Sender statistics:\n");
+		pr_debug("# cpu:      cpu number\n");
+		pr_debug("Sender statistics:\n");
 		for (i = 0; i < elements; i++)
-			printk(KERN_DEBUG "%s\n", stat_description[i]);
+			pr_debug("%s\n", stat_description[i]);
 	} else if (input_arg == -1) {
 		for_each_present_cpu(cpu) {
 			stat = &per_cpu(ptcstats, cpu);
@@ -1550,7 +1573,7 @@ static int parse_tunables_write(struct bau_control *bcp, char *instr,
 			break;
 	}
 	if (cnt != e) {
-		printk(KERN_INFO "bau tunable error: should be %d values\n", e);
+		pr_info("bau tunable error: should be %d values\n", e);
 		return -EINVAL;
 	}
 
@@ -1567,7 +1590,7 @@ static int parse_tunables_write(struct bau_control *bcp, char *instr,
 				continue;
 			}
 			if (val < 1 || val > bcp->cpus_in_uvhub) {
-				printk(KERN_DEBUG
+				pr_debug(
 				"Error: BAU max concurrent %d is invalid\n",
 				val);
 				return -EINVAL;
@@ -1615,17 +1638,17 @@ static ssize_t tunables_write(struct file *file, const char __user *user,
 
 	for_each_present_cpu(cpu) {
 		bcp = &per_cpu(bau_control, cpu);
-		bcp->max_concurr =		max_concurr;
-		bcp->max_concurr_const =	max_concurr;
-		bcp->plugged_delay =		plugged_delay;
-		bcp->plugsb4reset =		plugsb4reset;
-		bcp->timeoutsb4reset =		timeoutsb4reset;
-		bcp->ipi_reset_limit =		ipi_reset_limit;
-		bcp->complete_threshold =	complete_threshold;
-		bcp->cong_response_us =		congested_respns_us;
-		bcp->cong_reps =		congested_reps;
-		bcp->disabled_period =		sec_2_cycles(disabled_period);
-		bcp->giveup_limit =		giveup_limit;
+		bcp->max_concurr         = max_concurr;
+		bcp->max_concurr_const   = max_concurr;
+		bcp->plugged_delay       = plugged_delay;
+		bcp->plugsb4reset        = plugsb4reset;
+		bcp->timeoutsb4reset     = timeoutsb4reset;
+		bcp->ipi_reset_limit     = ipi_reset_limit;
+		bcp->complete_threshold  = complete_threshold;
+		bcp->cong_response_us    = congested_respns_us;
+		bcp->cong_reps           = congested_reps;
+		bcp->disabled_period     = sec_2_cycles(disabled_period);
+		bcp->giveup_limit        = giveup_limit;
 	}
 	return count;
 }
@@ -1672,21 +1695,21 @@ static int __init uv_ptc_init(void)
 	proc_uv_ptc = proc_create(UV_PTC_BASENAME, 0444, NULL,
 				  &proc_uv_ptc_operations);
 	if (!proc_uv_ptc) {
-		printk(KERN_ERR "unable to create %s proc entry\n",
+		pr_err("unable to create %s proc entry\n",
 		       UV_PTC_BASENAME);
 		return -EINVAL;
 	}
 
 	tunables_dir = debugfs_create_dir(UV_BAU_TUNABLES_DIR, NULL);
 	if (!tunables_dir) {
-		printk(KERN_ERR "unable to create debugfs directory %s\n",
+		pr_err("unable to create debugfs directory %s\n",
 		       UV_BAU_TUNABLES_DIR);
 		return -EINVAL;
 	}
 	tunables_file = debugfs_create_file(UV_BAU_TUNABLES_FILE, 0600,
 					tunables_dir, NULL, &tunables_fops);
 	if (!tunables_file) {
-		printk(KERN_ERR "unable to create debugfs file %s\n",
+		pr_err("unable to create debugfs file %s\n",
 		       UV_BAU_TUNABLES_FILE);
 		return -EINVAL;
 	}
@@ -1721,7 +1744,7 @@ static void activation_descriptor_init(int node, int pnode, int base_pnode)
 
 	gpa = uv_gpa(bau_desc);
 	n = uv_gpa_to_gnode(gpa);
-	m = uv_gpa_to_offset(gpa);
+	m = ops.bau_gpa_to_offset(gpa);
 	if (is_uv1_hub())
 		uv1 = 1;
 
@@ -1736,7 +1759,7 @@ static void activation_descriptor_init(int node, int pnode, int base_pnode)
 		memset(bd2, 0, sizeof(struct bau_desc));
 		if (uv1) {
 			uv1_hdr = &bd2->header.uv1_hdr;
-			uv1_hdr->swack_flag =	1;
+			uv1_hdr->swack_flag = 1;
 			/*
 			 * The base_dest_nasid set in the message header
 			 * is the nasid of the first uvhub in the partition.
@@ -1745,10 +1768,10 @@ static void activation_descriptor_init(int node, int pnode, int base_pnode)
 			 * if nasid striding is being used.
 			 */
 			uv1_hdr->base_dest_nasid =
-						UV_PNODE_TO_NASID(base_pnode);
-			uv1_hdr->dest_subnodeid =	UV_LB_SUBNODEID;
-			uv1_hdr->command =		UV_NET_ENDPOINT_INTD;
-			uv1_hdr->int_both =		1;
+			                          UV_PNODE_TO_NASID(base_pnode);
+			uv1_hdr->dest_subnodeid  = UV_LB_SUBNODEID;
+			uv1_hdr->command         = UV_NET_ENDPOINT_INTD;
+			uv1_hdr->int_both        = 1;
 			/*
 			 * all others need to be set to zero:
 			 *   fairness chaining multilevel count replied_to
@@ -1759,11 +1782,11 @@ static void activation_descriptor_init(int node, int pnode, int base_pnode)
 			 * uses native mode for selective broadcasts.
 			 */
 			uv2_3_hdr = &bd2->header.uv2_3_hdr;
-			uv2_3_hdr->swack_flag =	1;
+			uv2_3_hdr->swack_flag      = 1;
 			uv2_3_hdr->base_dest_nasid =
-						UV_PNODE_TO_NASID(base_pnode);
-			uv2_3_hdr->dest_subnodeid =	UV_LB_SUBNODEID;
-			uv2_3_hdr->command =		UV_NET_ENDPOINT_INTD;
+			                          UV_PNODE_TO_NASID(base_pnode);
+			uv2_3_hdr->dest_subnodeid  = UV_LB_SUBNODEID;
+			uv2_3_hdr->command         = UV_NET_ENDPOINT_INTD;
 		}
 	}
 	for_each_present_cpu(cpu) {
@@ -1786,10 +1809,7 @@ static void pq_init(int node, int pnode)
 	size_t plsize;
 	char *cp;
 	void *vp;
-	unsigned long pn;
-	unsigned long first;
-	unsigned long pn_first;
-	unsigned long last;
+	unsigned long gnode, first, last, tail;
 	struct bau_pq_entry *pqp;
 	struct bau_control *bcp;
 
@@ -1810,17 +1830,25 @@ static void pq_init(int node, int pnode)
 		bcp->bau_msg_head	= pqp;
 		bcp->queue_last		= pqp + (DEST_Q_SIZE - 1);
 	}
+
+	first = ops.bau_gpa_to_offset(uv_gpa(pqp));
+	last = ops.bau_gpa_to_offset(uv_gpa(pqp + (DEST_Q_SIZE - 1)));
+
 	/*
-	 * need the gnode of where the memory was really allocated
+	 * Pre UV4, the gnode is required to locate the payload queue
+	 * and the payload queue tail must be maintained by the kernel.
 	 */
-	pn = uv_gpa_to_gnode(uv_gpa(pqp));
-	first = uv_physnodeaddr(pqp);
-	pn_first = ((unsigned long)pn << UV_PAYLOADQ_PNODE_SHIFT) | first;
-	last = uv_physnodeaddr(pqp + (DEST_Q_SIZE - 1));
-	write_mmr_payload_first(pnode, pn_first);
-	write_mmr_payload_tail(pnode, first);
-	write_mmr_payload_last(pnode, last);
-	write_gmmr_sw_ack(pnode, 0xffffUL);
+	bcp = &per_cpu(bau_control, smp_processor_id());
+	if (bcp->uvhub_version <= 3) {
+		tail = first;
+		gnode = uv_gpa_to_gnode(uv_gpa(pqp));
+		first = (gnode << UV_PAYLOADQ_GNODE_SHIFT) | tail;
+		write_mmr_payload_tail(pnode, tail);
+	}
+
+	ops.write_payload_first(pnode, first);
+	ops.write_payload_last(pnode, last);
+	ops.write_g_sw_ack(pnode, 0xffffUL);
 
 	/* in effect, all msg_type's are set to MSG_NOOP */
 	memset(pqp, 0, sizeof(struct bau_pq_entry) * DEST_Q_SIZE);
@@ -1910,8 +1938,8 @@ static void __init init_per_cpu_tunables(void)
 		bcp->complete_threshold		= complete_threshold;
 		bcp->cong_response_us		= congested_respns_us;
 		bcp->cong_reps			= congested_reps;
-		bcp->disabled_period =		sec_2_cycles(disabled_period);
-		bcp->giveup_limit =		giveup_limit;
+		bcp->disabled_period		= sec_2_cycles(disabled_period);
+		bcp->giveup_limit		= giveup_limit;
 		spin_lock_init(&bcp->queue_lock);
 		spin_lock_init(&bcp->uvhub_lock);
 		spin_lock_init(&bcp->disable_lock);
@@ -1940,7 +1968,7 @@ static int __init get_cpu_topology(int base_pnode,
 
 		pnode = uv_cpu_hub_info(cpu)->pnode;
 		if ((pnode - base_pnode) >= UV_DISTRIBUTION_SIZE) {
-			printk(KERN_EMERG
+			pr_emerg(
 				"cpu %d pnode %d-%d beyond %d; BAU disabled\n",
 				cpu, pnode, base_pnode, UV_DISTRIBUTION_SIZE);
 			return 1;
@@ -1965,7 +1993,7 @@ static int __init get_cpu_topology(int base_pnode,
 		sdp->cpu_number[sdp->num_cpus] = cpu;
 		sdp->num_cpus++;
 		if (sdp->num_cpus > MAX_CPUS_PER_SOCKET) {
-			printk(KERN_EMERG "%d cpus per socket invalid\n",
+			pr_emerg("%d cpus per socket invalid\n",
 				sdp->num_cpus);
 			return 1;
 		}
@@ -2031,15 +2059,17 @@ static int scan_sock(struct socket_desc *sdp, struct uvhub_desc *bdp,
 			bcp->uvhub_version = 2;
 		else if (is_uv3_hub())
 			bcp->uvhub_version = 3;
+		else if (is_uv4_hub())
+			bcp->uvhub_version = 4;
 		else {
-			printk(KERN_EMERG "uvhub version not 1, 2 or 3\n");
+			pr_emerg("uvhub version not 1, 2, 3, or 4\n");
 			return 1;
 		}
 		bcp->uvhub_master = *hmasterp;
 		bcp->uvhub_cpu = uv_cpu_blade_processor_id(cpu);
 
 		if (bcp->uvhub_cpu >= MAX_CPUS_PER_UVHUB) {
-			printk(KERN_EMERG "%d cpus per uvhub invalid\n",
+			pr_emerg("%d cpus per uvhub invalid\n",
 				bcp->uvhub_cpu);
 			return 1;
 		}
@@ -2094,7 +2124,8 @@ static int __init init_per_cpu(int nuvhubs, int base_part_pnode)
 	void *vp;
 	struct uvhub_desc *uvhub_descs;
 
-	timeout_us = calculate_destination_timeout();
+	if (is_uv3_hub() || is_uv2_hub() || is_uv1_hub())
+		timeout_us = calculate_destination_timeout();
 
 	vp = kmalloc(nuvhubs * sizeof(struct uvhub_desc), GFP_KERNEL);
 	uvhub_descs = (struct uvhub_desc *)vp;
@@ -2134,6 +2165,15 @@ static int __init uv_bau_init(void)
 	if (!is_uv_system())
 		return 0;
 
+	if (is_uv4_hub())
+		ops = uv4_bau_ops;
+	else if (is_uv3_hub())
+		ops = uv123_bau_ops;
+	else if (is_uv2_hub())
+		ops = uv123_bau_ops;
+	else if (is_uv1_hub())
+		ops = uv123_bau_ops;
+
 	for_each_possible_cpu(cur_cpu) {
 		mask = &per_cpu(uv_flush_tlb_mask, cur_cpu);
 		zalloc_cpumask_var_node(mask, GFP_KERNEL, cpu_to_node(cur_cpu));
@@ -2149,7 +2189,9 @@ static int __init uv_bau_init(void)
 			uv_base_pnode = uv_blade_to_pnode(uvhub);
 	}
 
-	enable_timeouts();
+	/* software timeouts are not supported on UV4 */
+	if (is_uv3_hub() || is_uv2_hub() || is_uv1_hub())
+		enable_timeouts();
 
 	if (init_per_cpu(nuvhubs, uv_base_pnode)) {
 		set_bau_off();