Resguardo inicial

linux-7.2.3/arch/x86/platform/Makefile

@@ -0,0 +1,14 @@
+# SPDX-License-Identifier: GPL-2.0
+# Platform specific code goes here
+obj-y	+= atom/
+obj-y	+= ce4100/
+obj-y	+= efi/
+obj-y	+= geode/
+obj-y	+= iris/
+obj-y	+= intel/
+obj-y	+= intel-mid/
+obj-y	+= intel-quark/
+obj-y	+= olpc/
+obj-y	+= scx200/
+obj-y	+= ts5500/
+obj-y	+= uv/

linux-7.2.3/arch/x86/platform/atom/Makefile

@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_PUNIT_ATOM_DEBUG)	+= punit_atom_debug.o

linux-7.2.3/arch/x86/platform/atom/punit_atom_debug.c

@@ -0,0 +1,207 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Intel SOC Punit device state debug driver
+ * Punit controls power management for North Complex devices (Graphics
+ * blocks, Image Signal Processing, video processing, display, DSP etc.)
+ *
+ * Copyright (c) 2015, Intel Corporation.
+ */
+
+#define pr_fmt(fmt) "punit_atom: " fmt
+
+#include <linux/acpi.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/io.h>
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include <asm/iosf_mbi.h>
+
+/* Subsystem config/status Video processor */
+#define VED_SS_PM0		0x32
+/* Subsystem config/status ISP (Image Signal Processor) */
+#define ISP_SS_PM0		0x39
+/* Subsystem config/status Input/output controller */
+#define MIO_SS_PM		0x3B
+/* Shift bits for getting status for video, isp and i/o */
+#define SSS_SHIFT		24
+
+/* Power gate status reg */
+#define PWRGT_STATUS		0x61
+/* Shift bits for getting status for graphics rendering */
+#define RENDER_POS		0
+/* Shift bits for getting status for media control */
+#define MEDIA_POS		2
+/* Shift bits for getting status for Valley View/Baytrail display */
+#define VLV_DISPLAY_POS		6
+
+/* Subsystem config/status display for Cherry Trail SOC */
+#define CHT_DSP_SSS		0x36
+/* Shift bits for getting status for display */
+#define CHT_DSP_SSS_POS		16
+
+struct punit_device {
+	char *name;
+	int reg;
+	int sss_pos;
+};
+
+static const struct punit_device punit_device_tng[] = {
+	{ "DISPLAY",	CHT_DSP_SSS,	SSS_SHIFT },
+	{ "VED",	VED_SS_PM0,	SSS_SHIFT },
+	{ "ISP",	ISP_SS_PM0,	SSS_SHIFT },
+	{ "MIO",	MIO_SS_PM,	SSS_SHIFT },
+	{ NULL }
+};
+
+static const struct punit_device punit_device_byt[] = {
+	{ "GFX RENDER",	PWRGT_STATUS,	RENDER_POS },
+	{ "GFX MEDIA",	PWRGT_STATUS,	MEDIA_POS },
+	{ "DISPLAY",	PWRGT_STATUS,	VLV_DISPLAY_POS },
+	{ "VED",	VED_SS_PM0,	SSS_SHIFT },
+	{ "ISP",	ISP_SS_PM0,	SSS_SHIFT },
+	{ "MIO",	MIO_SS_PM,	SSS_SHIFT },
+	{ NULL }
+};
+
+static const struct punit_device punit_device_cht[] = {
+	{ "GFX RENDER",	PWRGT_STATUS,	RENDER_POS },
+	{ "GFX MEDIA",	PWRGT_STATUS,	MEDIA_POS },
+	{ "DISPLAY",	CHT_DSP_SSS,	CHT_DSP_SSS_POS },
+	{ "VED",	VED_SS_PM0,	SSS_SHIFT },
+	{ "ISP",	ISP_SS_PM0,	SSS_SHIFT },
+	{ "MIO",	MIO_SS_PM,	SSS_SHIFT },
+	{ NULL }
+};
+
+static const char * const dstates[] = {"D0", "D0i1", "D0i2", "D0i3"};
+
+static int punit_dev_state_show(struct seq_file *seq_file, void *unused)
+{
+	u32 punit_pwr_status;
+	struct punit_device *punit_devp = seq_file->private;
+	int index;
+	int status;
+
+	seq_puts(seq_file, "\n\nPUNIT NORTH COMPLEX DEVICES :\n");
+	while (punit_devp->name) {
+		status = iosf_mbi_read(BT_MBI_UNIT_PMC, MBI_REG_READ,
+				       punit_devp->reg, &punit_pwr_status);
+		if (status) {
+			seq_printf(seq_file, "%9s : Read Failed\n",
+				   punit_devp->name);
+		} else  {
+			index = (punit_pwr_status >> punit_devp->sss_pos) & 3;
+			seq_printf(seq_file, "%9s : %s\n", punit_devp->name,
+				   dstates[index]);
+		}
+		punit_devp++;
+	}
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(punit_dev_state);
+
+static struct dentry *punit_dbg_file;
+
+static void punit_dbgfs_register(struct punit_device *punit_device)
+{
+	punit_dbg_file = debugfs_create_dir("punit_atom", NULL);
+
+	debugfs_create_file("dev_power_state", 0444, punit_dbg_file,
+			    punit_device, &punit_dev_state_fops);
+}
+
+static void punit_dbgfs_unregister(void)
+{
+	debugfs_remove_recursive(punit_dbg_file);
+}
+
+#if defined(CONFIG_ACPI) && defined(CONFIG_SUSPEND)
+static const struct punit_device *punit_dev;
+
+static void punit_s2idle_check(void)
+{
+	const struct punit_device *punit_devp;
+	u32 punit_pwr_status, dstate;
+	int status;
+
+	for (punit_devp = punit_dev; punit_devp->name; punit_devp++) {
+		/* Skip MIO, it is on till the very last moment */
+		if (punit_devp->reg == MIO_SS_PM)
+			continue;
+
+		status = iosf_mbi_read(BT_MBI_UNIT_PMC, MBI_REG_READ,
+				       punit_devp->reg, &punit_pwr_status);
+		if (status) {
+			pr_err("%s read failed\n", punit_devp->name);
+		} else  {
+			dstate = (punit_pwr_status >> punit_devp->sss_pos) & 3;
+			if (!dstate)
+				pr_err("%s is in D0 prior to s2idle\n", punit_devp->name);
+		}
+	}
+}
+
+static struct acpi_s2idle_dev_ops punit_s2idle_ops = {
+	.check = punit_s2idle_check,
+};
+
+static void punit_s2idle_check_register(struct punit_device *punit_device)
+{
+	punit_dev = punit_device;
+	acpi_register_lps0_dev(&punit_s2idle_ops);
+}
+
+static void punit_s2idle_check_unregister(void)
+{
+	acpi_unregister_lps0_dev(&punit_s2idle_ops);
+}
+#else
+static void punit_s2idle_check_register(struct punit_device *punit_device) {}
+static void punit_s2idle_check_unregister(void) {}
+#endif
+
+#define X86_MATCH(vfm, data)					 \
+	X86_MATCH_VFM_FEATURE(vfm, X86_FEATURE_MWAIT, data)
+
+static const struct x86_cpu_id intel_punit_cpu_ids[] = {
+	X86_MATCH(INTEL_ATOM_SILVERMONT,	&punit_device_byt),
+	X86_MATCH(INTEL_ATOM_SILVERMONT_MID,	&punit_device_tng),
+	X86_MATCH(INTEL_ATOM_AIRMONT,		&punit_device_cht),
+	{}
+};
+MODULE_DEVICE_TABLE(x86cpu, intel_punit_cpu_ids);
+
+static int __init punit_atom_debug_init(void)
+{
+	struct punit_device *punit_device;
+	const struct x86_cpu_id *id;
+
+	id = x86_match_cpu(intel_punit_cpu_ids);
+	if (!id)
+		return -ENODEV;
+
+	punit_device = (struct punit_device *)id->driver_data;
+	punit_dbgfs_register(punit_device);
+	punit_s2idle_check_register(punit_device);
+
+	return 0;
+}
+
+static void __exit punit_atom_debug_exit(void)
+{
+	punit_s2idle_check_unregister();
+	punit_dbgfs_unregister();
+}
+
+module_init(punit_atom_debug_init);
+module_exit(punit_atom_debug_exit);
+
+MODULE_AUTHOR("Kumar P, Mahesh <mahesh.kumar.p@intel.com>");
+MODULE_AUTHOR("Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>");
+MODULE_DESCRIPTION("Driver for Punit devices states debugging");
+MODULE_LICENSE("GPL v2");

linux-7.2.3/arch/x86/platform/ce4100/Makefile

@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_X86_INTEL_CE)	+= ce4100.o

linux-7.2.3/arch/x86/platform/ce4100/ce4100.c

@@ -0,0 +1,60 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Intel CE4100  platform specific setup code
+ *
+ * (C) Copyright 2010 Intel Corporation
+ */
+#include <linux/init.h>
+#include <linux/reboot.h>
+
+#include <asm/ce4100.h>
+#include <asm/prom.h>
+#include <asm/setup.h>
+#include <asm/io.h>
+
+/*
+ * The CE4100 platform has an internal 8051 Microcontroller which is
+ * responsible for signaling to the external Power Management Unit the
+ * intention to reset, reboot or power off the system. This 8051 device has
+ * its command register mapped at I/O port 0xcf9 and the value 0x4 is used
+ * to power off the system.
+ */
+static void ce4100_power_off(void)
+{
+	outb(0x4, 0xcf9);
+}
+
+static void __init sdv_arch_setup(void)
+{
+	sdv_serial_fixup();
+}
+
+static void sdv_pci_init(void)
+{
+	x86_of_pci_init();
+}
+
+/*
+ * CE4100 specific x86_init function overrides and early setup
+ * calls.
+ */
+void __init x86_ce4100_early_setup(void)
+{
+	x86_init.oem.arch_setup			= sdv_arch_setup;
+	x86_init.resources.probe_roms		= x86_init_noop;
+	x86_init.mpparse.find_mptable		= x86_init_noop;
+	x86_init.mpparse.early_parse_smp_cfg	= x86_init_noop;
+	x86_init.pci.init			= ce4100_pci_init;
+	x86_init.pci.init_irq			= sdv_pci_init;
+
+	/*
+	 * By default, the reboot method is ACPI which is supported by the
+	 * CE4100 bootloader CEFDK using FADT.ResetReg Address and ResetValue
+	 * the bootloader will however issue a system power off instead of
+	 * reboot. By using BOOT_KBD we ensure proper system reboot as
+	 * expected.
+	 */
+	reboot_type = BOOT_KBD;
+
+	pm_power_off = ce4100_power_off;
+}

linux-7.2.3/arch/x86/platform/ce4100/falconfalls.dts

@@ -0,0 +1,430 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * CE4100 on Falcon Falls
+ *
+ * (c) Copyright 2010 Intel Corporation
+ */
+/dts-v1/;
+/ {
+	model = "intel,falconfalls";
+	compatible = "intel,falconfalls";
+	#address-cells = <1>;
+	#size-cells = <1>;
+
+	cpus {
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		cpu@0 {
+			device_type = "cpu";
+			compatible = "intel,ce4100";
+			reg = <0>;
+			lapic = <&lapic0>;
+		};
+	};
+
+	soc@0 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "intel,ce4100-cp";
+		ranges;
+
+		ioapic1: interrupt-controller@fec00000 {
+			#interrupt-cells = <2>;
+			compatible = "intel,ce4100-ioapic";
+			interrupt-controller;
+			reg = <0xfec00000 0x1000>;
+		};
+
+		timer@fed00000 {
+			compatible = "intel,ce4100-hpet";
+			reg = <0xfed00000 0x200>;
+		};
+
+		lapic0: interrupt-controller@fee00000 {
+			compatible = "intel,ce4100-lapic";
+			reg = <0xfee00000 0x1000>;
+		};
+
+		pci@3fc {
+			#address-cells = <3>;
+			#size-cells = <2>;
+			compatible = "intel,ce4100-pci", "pci";
+			device_type = "pci";
+			bus-range = <0 0>;
+			ranges = <0x2000000 0 0xbffff000 0xbffff000 0 0x1000
+				  0x2000000 0 0xdffe0000 0xdffe0000 0 0x1000
+				  0x0000000 0 0x0	 0x0	    0 0x100>;
+
+			/* Secondary IO-APIC */
+			ioapic2: interrupt-controller@0,1 {
+				#interrupt-cells = <2>;
+				compatible = "intel,ce4100-ioapic";
+				interrupt-controller;
+				reg = <0x100 0x0 0x0 0x0 0x0>;
+				assigned-addresses = <0x02000000 0x0 0xbffff000 0x0 0x1000>;
+			};
+
+			pci@1,0 {
+				#address-cells = <3>;
+				#size-cells = <2>;
+				compatible = "intel,ce4100-pci", "pci";
+				device_type = "pci";
+				bus-range = <1 1>;
+				reg = <0x0800 0x0 0x0 0x0 0x0>;
+				ranges = <0x2000000 0 0xdffe0000 0x2000000 0 0xdffe0000 0 0x1000>;
+
+				interrupt-parent = <&ioapic2>;
+
+				display@2,0 {
+					compatible = "pci8086,2e5b.2",
+						   "pci8086,2e5b",
+						   "pciclass038000",
+						   "pciclass0380";
+
+					reg = <0x11000 0x0 0x0 0x0 0x0>;
+					interrupts = <0 1>;
+				};
+
+				multimedia@3,0 {
+					compatible = "pci8086,2e5c.2",
+						   "pci8086,2e5c",
+						   "pciclass048000",
+						   "pciclass0480";
+
+					reg = <0x11800 0x0 0x0 0x0 0x0>;
+					interrupts = <2 1>;
+				};
+
+				multimedia@4,0 {
+					compatible = "pci8086,2e5d.2",
+						   "pci8086,2e5d",
+						   "pciclass048000",
+						   "pciclass0480";
+
+					reg = <0x12000 0x0 0x0 0x0 0x0>;
+					interrupts = <4 1>;
+				};
+
+				multimedia@4,1 {
+					compatible = "pci8086,2e5e.2",
+						   "pci8086,2e5e",
+						   "pciclass048000",
+						   "pciclass0480";
+
+					reg = <0x12100 0x0 0x0 0x0 0x0>;
+					interrupts = <5 1>;
+				};
+
+				sound@6,0 {
+					compatible = "pci8086,2e5f.2",
+						   "pci8086,2e5f",
+						   "pciclass040100",
+						   "pciclass0401";
+
+					reg = <0x13000 0x0 0x0 0x0 0x0>;
+					interrupts = <6 1>;
+				};
+
+				sound@6,1 {
+					compatible = "pci8086,2e5f.2",
+						   "pci8086,2e5f",
+						   "pciclass040100",
+						   "pciclass0401";
+
+					reg = <0x13100 0x0 0x0 0x0 0x0>;
+					interrupts = <7 1>;
+				};
+
+				sound@6,2 {
+					compatible = "pci8086,2e60.2",
+						   "pci8086,2e60",
+						   "pciclass040100",
+						   "pciclass0401";
+
+					reg = <0x13200 0x0 0x0 0x0 0x0>;
+					interrupts = <8 1>;
+				};
+
+				display@8,0 {
+					compatible = "pci8086,2e61.2",
+						   "pci8086,2e61",
+						   "pciclass038000",
+						   "pciclass0380";
+
+					reg = <0x14000 0x0 0x0 0x0 0x0>;
+					interrupts = <9 1>;
+				};
+
+				display@8,1 {
+					compatible = "pci8086,2e62.2",
+						   "pci8086,2e62",
+						   "pciclass038000",
+						   "pciclass0380";
+
+					reg = <0x14100 0x0 0x0 0x0 0x0>;
+					interrupts = <10 1>;
+				};
+
+				multimedia@8,2 {
+					compatible = "pci8086,2e63.2",
+						   "pci8086,2e63",
+						   "pciclass048000",
+						   "pciclass0480";
+
+					reg = <0x14200 0x0 0x0 0x0 0x0>;
+					interrupts = <11 1>;
+				};
+
+				entertainment-encryption@9,0 {
+					compatible = "pci8086,2e64.2",
+						   "pci8086,2e64",
+						   "pciclass101000",
+						   "pciclass1010";
+
+					reg = <0x14800 0x0 0x0 0x0 0x0>;
+					interrupts = <12 1>;
+				};
+
+				localbus@a,0 {
+					compatible = "pci8086,2e65.2",
+						   "pci8086,2e65",
+						   "pciclassff0000",
+						   "pciclassff00";
+
+					reg = <0x15000 0x0 0x0 0x0 0x0>;
+				};
+
+				serial@b,0 {
+					compatible = "pci8086,2e66.2",
+						   "pci8086,2e66",
+						   "pciclass070003",
+						   "pciclass0700";
+
+					reg = <0x15800 0x0 0x0 0x0 0x0>;
+					interrupts = <14 1>;
+				};
+
+				pcigpio: gpio@b,1 {
+					#gpio-cells = <2>;
+					#interrupt-cells = <2>;
+					compatible = "pci8086,2e67.2",
+						   "pci8086,2e67",
+						   "pciclassff0000",
+						   "pciclassff00";
+
+					reg = <0x15900 0x0 0x0 0x0 0x0>;
+					interrupts = <15 1>;
+					interrupt-controller;
+					gpio-controller;
+					intel,muxctl = <0>;
+				};
+
+				i2c-controller@b,2 {
+					#address-cells = <2>;
+					#size-cells = <1>;
+					compatible = "pci8086,2e68.2",
+						   "pci8086,2e68",
+						   "pciclass,ff0000",
+						   "pciclass,ff00";
+
+					reg = <0x15a00 0x0 0x0 0x0 0x0>;
+					interrupts = <16 1>;
+					ranges = <0 0	0x02000000 0 0xdffe0500	0x100
+						  1 0	0x02000000 0 0xdffe0600	0x100
+						  2 0	0x02000000 0 0xdffe0700	0x100>;
+
+					i2c@0 {
+						#address-cells = <1>;
+						#size-cells = <0>;
+						compatible = "intel,ce4100-i2c-controller";
+						reg = <0 0 0x100>;
+					};
+
+					i2c@1 {
+						#address-cells = <1>;
+						#size-cells = <0>;
+						compatible = "intel,ce4100-i2c-controller";
+						reg = <1 0 0x100>;
+
+						gpio@26 {
+							#gpio-cells = <2>;
+							compatible = "nxp,pcf8575";
+							reg = <0x26>;
+							gpio-controller;
+						};
+					};
+
+					i2c@2 {
+						#address-cells = <1>;
+						#size-cells = <0>;
+						compatible = "intel,ce4100-i2c-controller";
+						reg = <2 0 0x100>;
+
+						gpio@26 {
+							#gpio-cells = <2>;
+							compatible = "nxp,pcf8575";
+							reg = <0x26>;
+							gpio-controller;
+						};
+					};
+				};
+
+				smard-card@b,3 {
+					compatible = "pci8086,2e69.2",
+						   "pci8086,2e69",
+						   "pciclass070500",
+						   "pciclass0705";
+
+					reg = <0x15b00 0x0 0x0 0x0 0x0>;
+					interrupts = <15 1>;
+				};
+
+				spi-controller@b,4 {
+					#address-cells = <1>;
+					#size-cells = <0>;
+					compatible =
+						"pci8086,2e6a.2",
+						"pci8086,2e6a",
+						"pciclass,ff0000",
+						"pciclass,ff00";
+
+					reg = <0x15c00 0x0 0x0 0x0 0x0>;
+					interrupts = <15 1>;
+
+					dac@0 {
+						compatible = "ti,pcm1755";
+						reg = <0>;
+						spi-max-frequency = <115200>;
+					};
+
+					dac@1 {
+						compatible = "ti,pcm1609a";
+						reg = <1>;
+						spi-max-frequency = <115200>;
+					};
+
+					eeprom@2 {
+						compatible = "atmel,at93c46";
+						reg = <2>;
+						spi-max-frequency = <115200>;
+					};
+				};
+
+				multimedia@b,7 {
+					compatible = "pci8086,2e6d.2",
+						   "pci8086,2e6d",
+						   "pciclassff0000",
+						   "pciclassff00";
+
+					reg = <0x15f00 0x0 0x0 0x0 0x0>;
+				};
+
+				ethernet@c,0 {
+					compatible = "pci8086,2e6e.2",
+						   "pci8086,2e6e",
+						   "pciclass020000",
+						   "pciclass0200";
+
+					reg = <0x16000 0x0 0x0 0x0 0x0>;
+					interrupts = <21 1>;
+				};
+
+				clock@c,1 {
+					compatible = "pci8086,2e6f.2",
+						   "pci8086,2e6f",
+						   "pciclassff0000",
+						   "pciclassff00";
+
+					reg = <0x16100 0x0 0x0 0x0 0x0>;
+					interrupts = <3 1>;
+				};
+
+				usb@d,0 {
+					compatible = "pci8086,2e70.2",
+						   "pci8086,2e70",
+						   "pciclass0c0320",
+						   "pciclass0c03";
+
+					reg = <0x16800 0x0 0x0 0x0 0x0>;
+					interrupts = <22 1>;
+				};
+
+				usb@d,1 {
+					compatible = "pci8086,2e70.2",
+						   "pci8086,2e70",
+						   "pciclass0c0320",
+						   "pciclass0c03";
+
+					reg = <0x16900 0x0 0x0 0x0 0x0>;
+					interrupts = <22 1>;
+				};
+
+				sata@e,0 {
+					compatible = "pci8086,2e71.0",
+						   "pci8086,2e71",
+						   "pciclass010601",
+						   "pciclass0106";
+
+					reg = <0x17000 0x0 0x0 0x0 0x0>;
+					interrupts = <23 1>;
+				};
+
+				flash@f,0 {
+					compatible = "pci8086,701.1",
+						   "pci8086,701",
+						   "pciclass050100",
+						   "pciclass0501";
+
+					reg = <0x17800 0x0 0x0 0x0 0x0>;
+					interrupts = <13 1>;
+				};
+
+				entertainment-encryption@10,0 {
+					compatible = "pci8086,702.1",
+						   "pci8086,702",
+						   "pciclass101000",
+						   "pciclass1010";
+
+					reg = <0x18000 0x0 0x0 0x0 0x0>;
+				};
+
+				co-processor@11,0 {
+					compatible = "pci8086,703.1",
+						   "pci8086,703",
+						   "pciclass0b4000",
+						   "pciclass0b40";
+
+					reg = <0x18800 0x0 0x0 0x0 0x0>;
+					interrupts = <1 1>;
+				};
+
+				multimedia@12,0 {
+					compatible = "pci8086,704.0",
+						   "pci8086,704",
+						   "pciclass048000",
+						   "pciclass0480";
+
+					reg = <0x19000 0x0 0x0 0x0 0x0>;
+				};
+			};
+
+			isa@1f,0 {
+				#address-cells = <2>;
+				#size-cells = <1>;
+				compatible = "isa";
+				reg = <0xf800 0x0 0x0 0x0 0x0>;
+				ranges = <1 0 0 0 0 0x100>;
+
+				rtc@70 {
+					compatible = "intel,ce4100-rtc", "motorola,mc146818";
+					interrupts = <8 3>;
+					interrupt-parent = <&ioapic1>;
+					ctrl-reg = <2>;
+					freq-reg = <0x26>;
+					reg = <1 0x70 2>;
+				};
+			};
+		};
+	};
+};

linux-7.2.3/arch/x86/platform/efi/Makefile

@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+KASAN_SANITIZE := n
+GCOV_PROFILE := n
+
+obj-$(CONFIG_EFI) 		+= memmap.o quirks.o efi.o efi_$(BITS).o \
+				   efi_stub_$(BITS).o
+obj-$(CONFIG_EFI_MIXED)		+= efi_thunk_$(BITS).o
+obj-$(CONFIG_EFI_RUNTIME_MAP)	+= runtime-map.o

linux-7.2.3/arch/x86/platform/efi/efi.c

@@ -0,0 +1,926 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Common EFI (Extensible Firmware Interface) support functions
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 1999 VA Linux Systems
+ * Copyright (C) 1999 Walt Drummond <drummond@valinux.com>
+ * Copyright (C) 1999-2002 Hewlett-Packard Co.
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
+ *	Stephane Eranian <eranian@hpl.hp.com>
+ * Copyright (C) 2005-2008 Intel Co.
+ *	Fenghua Yu <fenghua.yu@intel.com>
+ *	Bibo Mao <bibo.mao@intel.com>
+ *	Chandramouli Narayanan <mouli@linux.intel.com>
+ *	Huang Ying <ying.huang@intel.com>
+ * Copyright (C) 2013 SuSE Labs
+ *	Borislav Petkov <bp@suse.de> - runtime services VA mapping
+ *
+ * Copied from efi_32.c to eliminate the duplicated code between EFI
+ * 32/64 support code. --ying 2007-10-26
+ *
+ * All EFI Runtime Services are not implemented yet as EFI only
+ * supports physical mode addressing on SoftSDV. This is to be fixed
+ * in a future version.  --drummond 1999-07-20
+ *
+ * Implemented EFI runtime services and virtual mode calls.  --davidm
+ *
+ * Goutham Rao: <goutham.rao@intel.com>
+ *	Skip non-WB memory and ignore empty memory ranges.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/efi.h>
+#include <linux/efi-bgrt.h>
+#include <linux/export.h>
+#include <linux/memblock.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include <linux/time.h>
+#include <linux/io.h>
+#include <linux/reboot.h>
+#include <linux/bcd.h>
+
+#include <asm/setup.h>
+#include <asm/efi.h>
+#include <asm/e820/api.h>
+#include <asm/time.h>
+#include <asm/tlbflush.h>
+#include <asm/x86_init.h>
+#include <asm/uv/uv.h>
+
+static unsigned long efi_systab_phys __initdata;
+static unsigned long efi_runtime, efi_nr_tables;
+
+unsigned long efi_fw_vendor, efi_config_table;
+
+static const efi_config_table_type_t arch_tables[] __initconst = {
+#ifdef CONFIG_X86_UV
+	{UV_SYSTEM_TABLE_GUID,		&uv_systab_phys,	"UVsystab"	},
+#endif
+	{},
+};
+
+static const unsigned long * const efi_tables[] = {
+	&efi.acpi,
+	&efi.acpi20,
+	&efi.smbios,
+	&efi.smbios3,
+#ifdef CONFIG_X86_UV
+	&uv_systab_phys,
+#endif
+	&efi_fw_vendor,
+	&efi_runtime,
+	&efi_config_table,
+	&efi.esrt,
+	&efi_mem_attr_table,
+#ifdef CONFIG_EFI_RCI2_TABLE
+	&rci2_table_phys,
+#endif
+	&efi.tpm_log,
+	&efi.tpm_final_log,
+	&efi_rng_seed,
+#ifdef CONFIG_LOAD_UEFI_KEYS
+	&efi.mokvar_table,
+#endif
+#ifdef CONFIG_EFI_COCO_SECRET
+	&efi.coco_secret,
+#endif
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	&efi.unaccepted,
+#endif
+};
+
+u64 efi_setup;		/* efi setup_data physical address */
+
+static int add_efi_memmap __initdata;
+static int __init setup_add_efi_memmap(char *arg)
+{
+	add_efi_memmap = 1;
+	return 0;
+}
+early_param("add_efi_memmap", setup_add_efi_memmap);
+
+/*
+ * Tell the kernel about the EFI memory map.  This might include
+ * more than the max 128 entries that can fit in the passed in e820
+ * legacy (zeropage) memory map, but the kernel's e820 table can hold
+ * E820_MAX_ENTRIES.
+ */
+
+static void __init do_add_efi_memmap(void)
+{
+	efi_memory_desc_t *md;
+
+	if (!efi_enabled(EFI_MEMMAP))
+		return;
+
+	for_each_efi_memory_desc(md) {
+		unsigned long long start = md->phys_addr;
+		unsigned long long size = md->num_pages << EFI_PAGE_SHIFT;
+		int e820_type;
+
+		switch (md->type) {
+		case EFI_LOADER_CODE:
+		case EFI_LOADER_DATA:
+		case EFI_BOOT_SERVICES_CODE:
+		case EFI_BOOT_SERVICES_DATA:
+		case EFI_CONVENTIONAL_MEMORY:
+			if (efi_soft_reserve_enabled()
+			    && (md->attribute & EFI_MEMORY_SP))
+				e820_type = E820_TYPE_SOFT_RESERVED;
+			else if (md->attribute & EFI_MEMORY_WB)
+				e820_type = E820_TYPE_RAM;
+			else
+				e820_type = E820_TYPE_RESERVED;
+			break;
+		case EFI_ACPI_RECLAIM_MEMORY:
+			e820_type = E820_TYPE_ACPI;
+			break;
+		case EFI_ACPI_MEMORY_NVS:
+			e820_type = E820_TYPE_NVS;
+			break;
+		case EFI_UNUSABLE_MEMORY:
+			e820_type = E820_TYPE_UNUSABLE;
+			break;
+		case EFI_PERSISTENT_MEMORY:
+			e820_type = E820_TYPE_PMEM;
+			break;
+		default:
+			/*
+			 * EFI_RESERVED_TYPE EFI_RUNTIME_SERVICES_CODE
+			 * EFI_RUNTIME_SERVICES_DATA EFI_MEMORY_MAPPED_IO
+			 * EFI_MEMORY_MAPPED_IO_PORT_SPACE EFI_PAL_CODE
+			 */
+			e820_type = E820_TYPE_RESERVED;
+			break;
+		}
+
+		e820__range_add(start, size, e820_type);
+	}
+	e820__update_table(e820_table);
+}
+
+/*
+ * Given add_efi_memmap defaults to 0 and there is no alternative
+ * e820 mechanism for soft-reserved memory, import the full EFI memory
+ * map if soft reservations are present and enabled. Otherwise, the
+ * mechanism to disable the kernel's consideration of EFI_MEMORY_SP is
+ * the efi=nosoftreserve option.
+ */
+static bool do_efi_soft_reserve(void)
+{
+	efi_memory_desc_t *md;
+
+	if (!efi_enabled(EFI_MEMMAP))
+		return false;
+
+	if (!efi_soft_reserve_enabled())
+		return false;
+
+	for_each_efi_memory_desc(md)
+		if (md->type == EFI_CONVENTIONAL_MEMORY &&
+		    (md->attribute & EFI_MEMORY_SP))
+			return true;
+	return false;
+}
+
+int __init efi_memblock_x86_reserve_range(void)
+{
+	struct efi_info *e = &boot_params.efi_info;
+	struct efi_memory_map_data data;
+	phys_addr_t pmap;
+	int rv;
+
+	if (efi_enabled(EFI_PARAVIRT))
+		return 0;
+
+	/* Can't handle firmware tables above 4GB on i386 */
+	if (IS_ENABLED(CONFIG_X86_32) && e->efi_memmap_hi > 0) {
+		pr_err("Memory map is above 4GB, disabling EFI.\n");
+		return -EINVAL;
+	}
+	pmap = (phys_addr_t)(e->efi_memmap | ((u64)e->efi_memmap_hi << 32));
+
+	data.phys_map		= pmap;
+	data.size 		= e->efi_memmap_size;
+	data.desc_size		= e->efi_memdesc_size;
+	data.desc_version	= e->efi_memdesc_version;
+
+	if (!efi_enabled(EFI_PARAVIRT)) {
+		rv = efi_memmap_init_early(&data);
+		if (rv)
+			return rv;
+	}
+
+	if (add_efi_memmap || do_efi_soft_reserve())
+		do_add_efi_memmap();
+
+	WARN(efi.memmap.desc_version != 1,
+	     "Unexpected EFI_MEMORY_DESCRIPTOR version %ld",
+	     efi.memmap.desc_version);
+
+	memblock_reserve(pmap, efi.memmap.nr_map * efi.memmap.desc_size);
+	set_bit(EFI_PRESERVE_BS_REGIONS, &efi.flags);
+
+	return 0;
+}
+
+#define OVERFLOW_ADDR_SHIFT	(64 - EFI_PAGE_SHIFT)
+#define OVERFLOW_ADDR_MASK	(U64_MAX << OVERFLOW_ADDR_SHIFT)
+#define U64_HIGH_BIT		(~(U64_MAX >> 1))
+
+static bool __init efi_memmap_entry_valid(const efi_memory_desc_t *md, int i)
+{
+	u64 end = (md->num_pages << EFI_PAGE_SHIFT) + md->phys_addr - 1;
+	u64 end_hi = 0;
+	char buf[64];
+
+	if (md->num_pages == 0) {
+		end = 0;
+	} else if (md->num_pages > EFI_PAGES_MAX ||
+		   EFI_PAGES_MAX - md->num_pages <
+		   (md->phys_addr >> EFI_PAGE_SHIFT)) {
+		end_hi = (md->num_pages & OVERFLOW_ADDR_MASK)
+			>> OVERFLOW_ADDR_SHIFT;
+
+		if ((md->phys_addr & U64_HIGH_BIT) && !(end & U64_HIGH_BIT))
+			end_hi += 1;
+	} else {
+		return true;
+	}
+
+	pr_warn_once(FW_BUG "Invalid EFI memory map entries:\n");
+
+	if (end_hi) {
+		pr_warn("mem%02u: %s range=[0x%016llx-0x%llx%016llx] (invalid)\n",
+			i, efi_md_typeattr_format(buf, sizeof(buf), md),
+			md->phys_addr, end_hi, end);
+	} else {
+		pr_warn("mem%02u: %s range=[0x%016llx-0x%016llx] (invalid)\n",
+			i, efi_md_typeattr_format(buf, sizeof(buf), md),
+			md->phys_addr, end);
+	}
+	return false;
+}
+
+static void __init efi_clean_memmap(void)
+{
+	efi_memory_desc_t *out = efi.memmap.map;
+	const efi_memory_desc_t *in = out;
+	const efi_memory_desc_t *end = efi.memmap.map_end;
+	int i, n_removal;
+
+	for (i = n_removal = 0; in < end; i++) {
+		if (efi_memmap_entry_valid(in, i)) {
+			if (out != in)
+				memcpy(out, in, efi.memmap.desc_size);
+			out = (void *)out + efi.memmap.desc_size;
+		} else {
+			n_removal++;
+		}
+		in = (void *)in + efi.memmap.desc_size;
+	}
+
+	if (n_removal > 0) {
+		struct efi_memory_map_data data = {
+			.phys_map	= efi.memmap.phys_map,
+			.desc_version	= efi.memmap.desc_version,
+			.desc_size	= efi.memmap.desc_size,
+			.size		= efi.memmap.desc_size * (efi.memmap.nr_map - n_removal),
+			.flags		= 0,
+		};
+
+		pr_warn("Removing %d invalid memory map entries.\n", n_removal);
+		efi_memmap_install(&data);
+	}
+}
+
+/*
+ * Firmware can use EfiMemoryMappedIO to request that MMIO regions be
+ * mapped by the OS so they can be accessed by EFI runtime services, but
+ * should have no other significance to the OS (UEFI r2.10, sec 7.2).
+ * However, most bootloaders and EFI stubs convert EfiMemoryMappedIO
+ * regions to E820_TYPE_RESERVED entries, which prevent Linux from
+ * allocating space from them (see remove_e820_regions()).
+ *
+ * Some platforms use EfiMemoryMappedIO entries for PCI MMCONFIG space and
+ * PCI host bridge windows, which means Linux can't allocate BAR space for
+ * hot-added devices.
+ *
+ * Remove large EfiMemoryMappedIO regions from the E820 map to avoid this
+ * problem.
+ *
+ * Retain small EfiMemoryMappedIO regions because on some platforms, these
+ * describe non-window space that's included in host bridge _CRS.  If we
+ * assign that space to PCI devices, they don't work.
+ */
+static void __init efi_remove_e820_mmio(void)
+{
+	efi_memory_desc_t *md;
+	u64 size, start, end;
+	int i = 0;
+
+	for_each_efi_memory_desc(md) {
+		if (md->type == EFI_MEMORY_MAPPED_IO) {
+			size = md->num_pages << EFI_PAGE_SHIFT;
+			start = md->phys_addr;
+			end = start + size - 1;
+			if (size >= 256*1024) {
+				pr_info("Remove mem%02u: MMIO range=[0x%08llx-0x%08llx] (%lluMB) from e820 map\n",
+					i, start, end, size >> 20);
+				e820__range_remove(start, size, E820_TYPE_RESERVED);
+			} else {
+				pr_info("Not removing mem%02u: MMIO range=[0x%08llx-0x%08llx] (%lluKB) from e820 map\n",
+					i, start, end, size >> 10);
+			}
+		}
+		i++;
+	}
+}
+
+void __init efi_print_memmap(void)
+{
+	efi_memory_desc_t *md;
+	int i = 0;
+
+	for_each_efi_memory_desc(md) {
+		char buf[64];
+
+		pr_info("mem%02u: %s range=[0x%016llx-0x%016llx] (%lluMB)\n",
+			i++, efi_md_typeattr_format(buf, sizeof(buf), md),
+			md->phys_addr,
+			md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) - 1,
+			(md->num_pages >> (20 - EFI_PAGE_SHIFT)));
+	}
+}
+
+static int __init efi_systab_init(unsigned long phys)
+{
+	int size = efi_enabled(EFI_64BIT) ? sizeof(efi_system_table_64_t)
+					  : sizeof(efi_system_table_32_t);
+	const efi_table_hdr_t *hdr;
+	bool over4g = false;
+	void *p;
+	int ret;
+
+	hdr = p = early_memremap_ro(phys, size);
+	if (p == NULL) {
+		pr_err("Couldn't map the system table!\n");
+		return -ENOMEM;
+	}
+
+	ret = efi_systab_check_header(hdr);
+	if (ret) {
+		early_memunmap(p, size);
+		return ret;
+	}
+
+	if (efi_enabled(EFI_64BIT)) {
+		const efi_system_table_64_t *systab64 = p;
+
+		efi_runtime	= systab64->runtime;
+		over4g		= systab64->runtime > U32_MAX;
+
+		if (efi_setup) {
+			struct efi_setup_data *data;
+
+			data = early_memremap_ro(efi_setup, sizeof(*data));
+			if (!data) {
+				early_memunmap(p, size);
+				return -ENOMEM;
+			}
+
+			efi_fw_vendor		= (unsigned long)data->fw_vendor;
+			efi_config_table	= (unsigned long)data->tables;
+
+			over4g |= data->fw_vendor	> U32_MAX ||
+				  data->tables		> U32_MAX;
+
+			early_memunmap(data, sizeof(*data));
+		} else {
+			efi_fw_vendor		= systab64->fw_vendor;
+			efi_config_table	= systab64->tables;
+
+			over4g |= systab64->fw_vendor	> U32_MAX ||
+				  systab64->tables	> U32_MAX;
+		}
+		efi_nr_tables = systab64->nr_tables;
+	} else {
+		const efi_system_table_32_t *systab32 = p;
+
+		efi_fw_vendor		= systab32->fw_vendor;
+		efi_runtime		= systab32->runtime;
+		efi_config_table	= systab32->tables;
+		efi_nr_tables		= systab32->nr_tables;
+	}
+
+	efi.runtime_version = hdr->revision;
+
+	efi_systab_report_header(hdr, efi_fw_vendor);
+	early_memunmap(p, size);
+
+	if (IS_ENABLED(CONFIG_X86_32) && over4g) {
+		pr_err("EFI data located above 4GB, disabling EFI.\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int __init efi_config_init(const efi_config_table_type_t *arch_tables)
+{
+	void *config_tables;
+	int sz, ret;
+
+	if (efi_nr_tables == 0)
+		return 0;
+
+	if (efi_enabled(EFI_64BIT))
+		sz = sizeof(efi_config_table_64_t);
+	else
+		sz = sizeof(efi_config_table_32_t);
+
+	/*
+	 * Let's see what config tables the firmware passed to us.
+	 */
+	config_tables = early_memremap(efi_config_table, efi_nr_tables * sz);
+	if (config_tables == NULL) {
+		pr_err("Could not map Configuration table!\n");
+		return -ENOMEM;
+	}
+
+	ret = efi_config_parse_tables(config_tables, efi_nr_tables,
+				      arch_tables);
+
+	early_memunmap(config_tables, efi_nr_tables * sz);
+	return ret;
+}
+
+void __init efi_init(void)
+{
+	if (IS_ENABLED(CONFIG_X86_32) &&
+	    (boot_params.efi_info.efi_systab_hi ||
+	     boot_params.efi_info.efi_memmap_hi)) {
+		pr_info("Table located above 4GB, disabling EFI.\n");
+		return;
+	}
+
+	efi_systab_phys = boot_params.efi_info.efi_systab |
+			  ((__u64)boot_params.efi_info.efi_systab_hi << 32);
+
+	if (efi_systab_init(efi_systab_phys))
+		return;
+
+	if (efi_reuse_config(efi_config_table, efi_nr_tables))
+		return;
+
+	if (efi_config_init(arch_tables))
+		return;
+
+	/*
+	 * Note: We currently don't support runtime services on an EFI
+	 * that doesn't match the kernel 32/64-bit mode.
+	 */
+
+	if (!efi_runtime_supported())
+		pr_err("No EFI runtime due to 32/64-bit mismatch with kernel\n");
+
+	if (!efi_runtime_supported() || efi_runtime_disabled()) {
+		efi_memmap_unmap();
+		return;
+	}
+
+	set_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+	efi_clean_memmap();
+
+	efi_remove_e820_mmio();
+
+	if (efi_enabled(EFI_DBG))
+		efi_print_memmap();
+}
+
+/* Merge contiguous regions of the same type and attribute */
+static void __init efi_merge_regions(void)
+{
+	efi_memory_desc_t *md, *prev_md = NULL;
+
+	for_each_efi_memory_desc(md) {
+		u64 prev_size;
+
+		if (!prev_md) {
+			prev_md = md;
+			continue;
+		}
+
+		if (prev_md->type != md->type ||
+		    prev_md->attribute != md->attribute) {
+			prev_md = md;
+			continue;
+		}
+
+		prev_size = prev_md->num_pages << EFI_PAGE_SHIFT;
+
+		if (md->phys_addr == (prev_md->phys_addr + prev_size)) {
+			prev_md->num_pages += md->num_pages;
+			md->type = EFI_RESERVED_TYPE;
+			md->attribute = 0;
+			continue;
+		}
+		prev_md = md;
+	}
+}
+
+static void *realloc_pages(void *old_memmap, int old_shift)
+{
+	void *ret;
+
+	ret = (void *)__get_free_pages(GFP_KERNEL, old_shift + 1);
+	if (!ret)
+		goto out;
+
+	/*
+	 * A first-time allocation doesn't have anything to copy.
+	 */
+	if (!old_memmap)
+		return ret;
+
+	memcpy(ret, old_memmap, PAGE_SIZE << old_shift);
+
+out:
+	free_pages((unsigned long)old_memmap, old_shift);
+	return ret;
+}
+
+/*
+ * Iterate the EFI memory map in reverse order because the regions
+ * will be mapped top-down. The end result is the same as if we had
+ * mapped things forward, but doesn't require us to change the
+ * existing implementation of efi_map_region().
+ */
+static inline void *efi_map_next_entry_reverse(void *entry)
+{
+	/* Initial call */
+	if (!entry)
+		return efi.memmap.map_end - efi.memmap.desc_size;
+
+	entry -= efi.memmap.desc_size;
+	if (entry < efi.memmap.map)
+		return NULL;
+
+	return entry;
+}
+
+/*
+ * efi_map_next_entry - Return the next EFI memory map descriptor
+ * @entry: Previous EFI memory map descriptor
+ *
+ * This is a helper function to iterate over the EFI memory map, which
+ * we do in different orders depending on the current configuration.
+ *
+ * To begin traversing the memory map @entry must be %NULL.
+ *
+ * Returns %NULL when we reach the end of the memory map.
+ */
+static void *efi_map_next_entry(void *entry)
+{
+	if (efi_enabled(EFI_64BIT)) {
+		/*
+		 * Starting in UEFI v2.5 the EFI_PROPERTIES_TABLE
+		 * config table feature requires us to map all entries
+		 * in the same order as they appear in the EFI memory
+		 * map. That is to say, entry N must have a lower
+		 * virtual address than entry N+1. This is because the
+		 * firmware toolchain leaves relative references in
+		 * the code/data sections, which are split and become
+		 * separate EFI memory regions. Mapping things
+		 * out-of-order leads to the firmware accessing
+		 * unmapped addresses.
+		 *
+		 * Since we need to map things this way whether or not
+		 * the kernel actually makes use of
+		 * EFI_PROPERTIES_TABLE, let's just switch to this
+		 * scheme by default for 64-bit.
+		 */
+		return efi_map_next_entry_reverse(entry);
+	}
+
+	/* Initial call */
+	if (!entry)
+		return efi.memmap.map;
+
+	entry += efi.memmap.desc_size;
+	if (entry >= efi.memmap.map_end)
+		return NULL;
+
+	return entry;
+}
+
+static bool should_map_region(efi_memory_desc_t *md)
+{
+	/*
+	 * Runtime regions always require runtime mappings (obviously).
+	 */
+	if (md->attribute & EFI_MEMORY_RUNTIME)
+		return true;
+
+	/*
+	 * 32-bit EFI doesn't suffer from the bug that requires us to
+	 * reserve boot services regions, and mixed mode support
+	 * doesn't exist for 32-bit kernels.
+	 */
+	if (IS_ENABLED(CONFIG_X86_32))
+		return false;
+
+	/*
+	 * EFI specific purpose memory may be reserved by default
+	 * depending on kernel config and boot options.
+	 */
+	if (md->type == EFI_CONVENTIONAL_MEMORY &&
+	    efi_soft_reserve_enabled() &&
+	    (md->attribute & EFI_MEMORY_SP))
+		return false;
+
+	/*
+	 * Map all of RAM so that we can access arguments in the 1:1
+	 * mapping when making EFI runtime calls.
+	 */
+	if (efi_is_mixed()) {
+		if (md->type == EFI_CONVENTIONAL_MEMORY ||
+		    md->type == EFI_LOADER_DATA ||
+		    md->type == EFI_LOADER_CODE)
+			return true;
+	}
+
+	/*
+	 * Map boot services regions as a workaround for buggy
+	 * firmware that accesses them even when they shouldn't.
+	 *
+	 * See efi_{reserve,free}_boot_services().
+	 */
+	if (md->type == EFI_BOOT_SERVICES_CODE ||
+	    md->type == EFI_BOOT_SERVICES_DATA)
+		return true;
+
+	return false;
+}
+
+/*
+ * Map the efi memory ranges of the runtime services and update new_mmap with
+ * virtual addresses.
+ */
+static void * __init efi_map_regions(int *count, int *pg_shift)
+{
+	void *p, *new_memmap = NULL;
+	unsigned long left = 0;
+	unsigned long desc_size;
+	efi_memory_desc_t *md;
+
+	desc_size = efi.memmap.desc_size;
+
+	p = NULL;
+	while ((p = efi_map_next_entry(p))) {
+		md = p;
+
+		if (!should_map_region(md))
+			continue;
+
+		efi_map_region(md);
+
+		if (left < desc_size) {
+			new_memmap = realloc_pages(new_memmap, *pg_shift);
+			if (!new_memmap)
+				return NULL;
+
+			left += PAGE_SIZE << *pg_shift;
+			(*pg_shift)++;
+		}
+
+		memcpy(new_memmap + (*count * desc_size), md, desc_size);
+
+		left -= desc_size;
+		(*count)++;
+	}
+
+	return new_memmap;
+}
+
+static void __init kexec_enter_virtual_mode(void)
+{
+#ifdef CONFIG_KEXEC_CORE
+	efi_memory_desc_t *md;
+	unsigned int num_pages;
+
+	/*
+	 * We don't do virtual mode, since we don't do runtime services, on
+	 * non-native EFI.
+	 */
+	if (efi_is_mixed()) {
+		efi_memmap_unmap();
+		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+		return;
+	}
+
+	if (efi_alloc_page_tables()) {
+		pr_err("Failed to allocate EFI page tables\n");
+		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+		return;
+	}
+
+	/*
+	* Map efi regions which were passed via setup_data. The virt_addr is a
+	* fixed addr which was used in first kernel of a kexec boot.
+	*/
+	for_each_efi_memory_desc(md)
+		efi_map_region_fixed(md); /* FIXME: add error handling */
+
+	/*
+	 * Unregister the early EFI memmap from efi_init() and install
+	 * the new EFI memory map.
+	 */
+	efi_memmap_unmap();
+
+	if (efi_memmap_init_late(efi.memmap.phys_map,
+				 efi.memmap.desc_size * efi.memmap.nr_map)) {
+		pr_err("Failed to remap late EFI memory map\n");
+		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+		return;
+	}
+
+	num_pages = ALIGN(efi.memmap.nr_map * efi.memmap.desc_size, PAGE_SIZE);
+	num_pages >>= PAGE_SHIFT;
+
+	if (efi_setup_page_tables(efi.memmap.phys_map, num_pages)) {
+		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+		return;
+	}
+
+	efi_sync_low_kernel_mappings();
+	efi_native_runtime_setup();
+	efi_runtime_update_mappings();
+#endif
+}
+
+/*
+ * This function will switch the EFI runtime services to virtual mode.
+ * Essentially, we look through the EFI memmap and map every region that
+ * has the runtime attribute bit set in its memory descriptor into the
+ * efi_pgd page table.
+ *
+ * The new method does a pagetable switch in a preemption-safe manner
+ * so that we're in a different address space when calling a runtime
+ * function. For function arguments passing we do copy the PUDs of the
+ * kernel page table into efi_pgd prior to each call.
+ *
+ * Specially for kexec boot, efi runtime maps in previous kernel should
+ * be passed in via setup_data. In that case runtime ranges will be mapped
+ * to the same virtual addresses as the first kernel, see
+ * kexec_enter_virtual_mode().
+ */
+static void __init __efi_enter_virtual_mode(void)
+{
+	int count = 0, pg_shift = 0;
+	void *new_memmap = NULL;
+	efi_status_t status;
+	unsigned long pa;
+
+	if (efi_alloc_page_tables()) {
+		pr_err("Failed to allocate EFI page tables\n");
+		goto err;
+	}
+
+	efi_merge_regions();
+	new_memmap = efi_map_regions(&count, &pg_shift);
+	if (!new_memmap) {
+		pr_err("Error reallocating memory, EFI runtime non-functional!\n");
+		goto err;
+	}
+
+	pa = __pa(new_memmap);
+
+	/*
+	 * Unregister the early EFI memmap from efi_init() and install
+	 * the new EFI memory map that we are about to pass to the
+	 * firmware via SetVirtualAddressMap().
+	 */
+	efi_memmap_unmap();
+
+	if (efi_memmap_init_late(pa, efi.memmap.desc_size * count)) {
+		pr_err("Failed to remap late EFI memory map\n");
+		goto err;
+	}
+
+	if (efi_enabled(EFI_DBG)) {
+		pr_info("EFI runtime memory map:\n");
+		efi_print_memmap();
+	}
+
+	if (efi_setup_page_tables(pa, 1 << pg_shift))
+		goto err;
+
+	efi_sync_low_kernel_mappings();
+
+	status = efi_set_virtual_address_map(efi.memmap.desc_size * count,
+					     efi.memmap.desc_size,
+					     efi.memmap.desc_version,
+					     (efi_memory_desc_t *)pa,
+					     efi_systab_phys);
+	if (status != EFI_SUCCESS) {
+		pr_err("Unable to switch EFI into virtual mode (status=%lx)!\n",
+		       status);
+		goto err;
+	}
+
+	efi_check_for_embedded_firmwares();
+	efi_unmap_boot_services();
+
+	if (!efi_is_mixed())
+		efi_native_runtime_setup();
+	else
+		efi_thunk_runtime_setup();
+
+	/*
+	 * Apply more restrictive page table mapping attributes now that
+	 * SVAM() has been called and the firmware has performed all
+	 * necessary relocation fixups for the new virtual addresses.
+	 */
+	efi_runtime_update_mappings();
+
+	/* clean DUMMY object */
+	efi_delete_dummy_variable();
+	return;
+
+err:
+	clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+}
+
+void __init efi_enter_virtual_mode(void)
+{
+	if (efi_enabled(EFI_PARAVIRT))
+		return;
+
+	efi.runtime = (efi_runtime_services_t *)efi_runtime;
+
+	if (efi_setup)
+		kexec_enter_virtual_mode();
+	else
+		__efi_enter_virtual_mode();
+
+	efi_dump_pagetable();
+}
+
+bool efi_is_table_address(unsigned long phys_addr)
+{
+	unsigned int i;
+
+	if (phys_addr == EFI_INVALID_TABLE_ADDR)
+		return false;
+
+	for (i = 0; i < ARRAY_SIZE(efi_tables); i++)
+		if (*(efi_tables[i]) == phys_addr)
+			return true;
+
+	return false;
+}
+
+#define EFI_FIELD(var) efi_ ## var
+
+#define EFI_ATTR_SHOW(name) \
+static ssize_t name##_show(struct kobject *kobj, \
+				struct kobj_attribute *attr, char *buf) \
+{ \
+	return sprintf(buf, "0x%lx\n", EFI_FIELD(name)); \
+}
+
+EFI_ATTR_SHOW(fw_vendor);
+EFI_ATTR_SHOW(runtime);
+EFI_ATTR_SHOW(config_table);
+
+struct kobj_attribute efi_attr_fw_vendor = __ATTR_RO(fw_vendor);
+struct kobj_attribute efi_attr_runtime = __ATTR_RO(runtime);
+struct kobj_attribute efi_attr_config_table = __ATTR_RO(config_table);
+
+umode_t efi_attr_is_visible(struct kobject *kobj, struct attribute *attr, int n)
+{
+	if (attr == &efi_attr_fw_vendor.attr) {
+		if (efi_enabled(EFI_PARAVIRT) ||
+				efi_fw_vendor == EFI_INVALID_TABLE_ADDR)
+			return 0;
+	} else if (attr == &efi_attr_runtime.attr) {
+		if (efi_runtime == EFI_INVALID_TABLE_ADDR)
+			return 0;
+	} else if (attr == &efi_attr_config_table.attr) {
+		if (efi_config_table == EFI_INVALID_TABLE_ADDR)
+			return 0;
+	}
+	return attr->mode;
+}
+
+enum efi_secureboot_mode __x86_efi_boot_mode(void)
+{
+	return boot_params.secure_boot;
+}

linux-7.2.3/arch/x86/platform/efi/efi_32.c

@@ -0,0 +1,154 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Extensible Firmware Interface
+ *
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 1999 VA Linux Systems
+ * Copyright (C) 1999 Walt Drummond <drummond@valinux.com>
+ * Copyright (C) 1999-2002 Hewlett-Packard Co.
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
+ *	Stephane Eranian <eranian@hpl.hp.com>
+ *
+ * All EFI Runtime Services are not implemented yet as EFI only
+ * supports physical mode addressing on SoftSDV. This is to be fixed
+ * in a future version.  --drummond 1999-07-20
+ *
+ * Implemented EFI runtime services and virtual mode calls.  --davidm
+ *
+ * Goutham Rao: <goutham.rao@intel.com>
+ *	Skip non-WB memory and ignore empty memory ranges.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/ioport.h>
+#include <linux/efi.h>
+#include <linux/pgtable.h>
+
+#include <asm/io.h>
+#include <asm/desc.h>
+#include <asm/page.h>
+#include <asm/set_memory.h>
+#include <asm/tlbflush.h>
+#include <asm/efi.h>
+
+void __init efi_map_region(efi_memory_desc_t *md)
+{
+	u64 start_pfn, end_pfn, end;
+	unsigned long size;
+	void *va;
+
+	start_pfn	= PFN_DOWN(md->phys_addr);
+	size		= md->num_pages << PAGE_SHIFT;
+	end		= md->phys_addr + size;
+	end_pfn 	= PFN_UP(end);
+
+	if (pfn_range_is_mapped(start_pfn, end_pfn)) {
+		va = __va(md->phys_addr);
+
+		if (!(md->attribute & EFI_MEMORY_WB))
+			set_memory_uc((unsigned long)va, md->num_pages);
+	} else {
+		va = ioremap_cache(md->phys_addr, size);
+	}
+
+	md->virt_addr = (unsigned long)va;
+	if (!va)
+		pr_err("ioremap of 0x%llX failed!\n", md->phys_addr);
+}
+
+/*
+ * To make EFI call EFI runtime service in physical addressing mode we need
+ * prolog/epilog before/after the invocation to claim the EFI runtime service
+ * handler exclusively and to duplicate a memory mapping in low memory space,
+ * say 0 - 3G.
+ */
+
+int __init efi_alloc_page_tables(void)
+{
+	return 0;
+}
+
+void efi_sync_low_kernel_mappings(void) {}
+
+void __init efi_dump_pagetable(void)
+{
+#ifdef CONFIG_EFI_PGT_DUMP
+	ptdump_walk_pgd_level(NULL, &init_mm);
+#endif
+}
+
+int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
+{
+	return 0;
+}
+
+void __init efi_map_region_fixed(efi_memory_desc_t *md) {}
+void __init parse_efi_setup(u64 phys_addr, u32 data_len) {}
+
+efi_status_t efi_call_svam(efi_runtime_services_t * const *,
+			   u32, u32, u32, void *, u32);
+
+efi_status_t __init efi_set_virtual_address_map(unsigned long memory_map_size,
+						unsigned long descriptor_size,
+						u32 descriptor_version,
+						efi_memory_desc_t *virtual_map,
+						unsigned long systab_phys)
+{
+	const efi_system_table_t *systab = (efi_system_table_t *)systab_phys;
+	struct desc_ptr gdt_descr;
+	efi_status_t status;
+	unsigned long flags;
+	pgd_t *save_pgd;
+
+	/* Current pgd is swapper_pg_dir, we'll restore it later: */
+	save_pgd = swapper_pg_dir;
+	load_cr3(initial_page_table);
+	__flush_tlb_all();
+
+	gdt_descr.address = get_cpu_gdt_paddr(0);
+	gdt_descr.size = GDT_SIZE - 1;
+	load_gdt(&gdt_descr);
+
+	/* Disable interrupts around EFI calls: */
+	local_irq_save(flags);
+	status = efi_call_svam(&systab->runtime,
+			       memory_map_size, descriptor_size,
+			       descriptor_version, virtual_map,
+			       __pa(&efi.runtime));
+	local_irq_restore(flags);
+
+	load_fixmap_gdt(0);
+	load_cr3(save_pgd);
+	__flush_tlb_all();
+
+	return status;
+}
+
+void __init efi_runtime_update_mappings(void)
+{
+	if (__supported_pte_mask & _PAGE_NX) {
+		efi_memory_desc_t *md;
+
+		/* Make EFI runtime service code area executable */
+		for_each_efi_memory_desc(md) {
+			if (md->type != EFI_RUNTIME_SERVICES_CODE)
+				continue;
+
+			set_memory_x(md->virt_addr, md->num_pages);
+		}
+	}
+}
+
+void arch_efi_call_virt_setup(void)
+{
+	efi_fpu_begin();
+	firmware_restrict_branch_speculation_start();
+}
+
+void arch_efi_call_virt_teardown(void)
+{
+	firmware_restrict_branch_speculation_end();
+	efi_fpu_end();
+}

linux-7.2.3/arch/x86/platform/efi/efi_64.c

@@ -0,0 +1,875 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * x86_64 specific EFI support functions
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 2005-2008 Intel Co.
+ *	Fenghua Yu <fenghua.yu@intel.com>
+ *	Bibo Mao <bibo.mao@intel.com>
+ *	Chandramouli Narayanan <mouli@linux.intel.com>
+ *	Huang Ying <ying.huang@intel.com>
+ *
+ * Code to convert EFI to E820 map has been implemented in elilo bootloader
+ * based on a EFI patch by Edgar Hucek. Based on the E820 map, the page table
+ * is setup appropriately for EFI runtime code.
+ * - mouli 06/14/2007.
+ *
+ */
+
+#define pr_fmt(fmt) "efi: " fmt
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+#include <linux/spinlock.h>
+#include <linux/memblock.h>
+#include <linux/ioport.h>
+#include <linux/mc146818rtc.h>
+#include <linux/efi.h>
+#include <linux/export.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
+#include <linux/reboot.h>
+#include <linux/slab.h>
+#include <linux/ucs2_string.h>
+#include <linux/cc_platform.h>
+#include <linux/sched/task.h>
+
+#include <asm/setup.h>
+#include <asm/page.h>
+#include <asm/e820/api.h>
+#include <asm/tlbflush.h>
+#include <asm/proto.h>
+#include <asm/efi.h>
+#include <asm/cacheflush.h>
+#include <asm/fixmap.h>
+#include <asm/realmode.h>
+#include <asm/time.h>
+#include <asm/pgalloc.h>
+#include <asm/sev.h>
+
+/*
+ * We allocate runtime services regions top-down, starting from -4G, i.e.
+ * 0xffff_ffff_0000_0000 and limit EFI VA mapping space to 64G.
+ */
+static u64 efi_va = EFI_VA_START;
+static struct mm_struct *efi_prev_mm;
+static unsigned long efi_cr4_lass;
+
+/*
+ * We need our own copy of the higher levels of the page tables
+ * because we want to avoid inserting EFI region mappings (EFI_VA_END
+ * to EFI_VA_START) into the standard kernel page tables. Everything
+ * else can be shared, see efi_sync_low_kernel_mappings().
+ *
+ * We don't want the pgd on the pgd_list and cannot use pgd_alloc() for the
+ * allocation.
+ */
+int __init efi_alloc_page_tables(void)
+{
+	pgd_t *pgd, *efi_pgd;
+	p4d_t *p4d;
+	pud_t *pud;
+	gfp_t gfp_mask;
+
+	gfp_mask = GFP_KERNEL | __GFP_ZERO;
+	efi_pgd = (pgd_t *)__get_free_pages(gfp_mask, pgd_allocation_order());
+	if (!efi_pgd)
+		goto fail;
+
+	pgd = efi_pgd + pgd_index(EFI_VA_END);
+	p4d = p4d_alloc(&init_mm, pgd, EFI_VA_END);
+	if (!p4d)
+		goto free_pgd;
+
+	pud = pud_alloc(&init_mm, p4d, EFI_VA_END);
+	if (!pud)
+		goto free_p4d;
+
+	efi_mm.pgd = efi_pgd;
+	mm_init_cpumask(&efi_mm);
+	init_new_context(NULL, &efi_mm);
+	set_notrack_mm(&efi_mm);
+
+	return 0;
+
+free_p4d:
+	if (pgtable_l5_enabled())
+		free_page((unsigned long)pgd_page_vaddr(*pgd));
+free_pgd:
+	free_pages((unsigned long)efi_pgd, pgd_allocation_order());
+fail:
+	return -ENOMEM;
+}
+
+/*
+ * Add low kernel mappings for passing arguments to EFI functions.
+ */
+void efi_sync_low_kernel_mappings(void)
+{
+	unsigned num_entries;
+	pgd_t *pgd_k, *pgd_efi;
+	p4d_t *p4d_k, *p4d_efi;
+	pud_t *pud_k, *pud_efi;
+	pgd_t *efi_pgd = efi_mm.pgd;
+
+	pgd_efi = efi_pgd + pgd_index(PAGE_OFFSET);
+	pgd_k = pgd_offset_k(PAGE_OFFSET);
+
+	num_entries = pgd_index(EFI_VA_END) - pgd_index(PAGE_OFFSET);
+	memcpy(pgd_efi, pgd_k, sizeof(pgd_t) * num_entries);
+
+	pgd_efi = efi_pgd + pgd_index(EFI_VA_END);
+	pgd_k = pgd_offset_k(EFI_VA_END);
+	p4d_efi = p4d_offset(pgd_efi, 0);
+	p4d_k = p4d_offset(pgd_k, 0);
+
+	num_entries = p4d_index(EFI_VA_END);
+	memcpy(p4d_efi, p4d_k, sizeof(p4d_t) * num_entries);
+
+	/*
+	 * We share all the PUD entries apart from those that map the
+	 * EFI regions. Copy around them.
+	 */
+	BUILD_BUG_ON((EFI_VA_START & ~PUD_MASK) != 0);
+	BUILD_BUG_ON((EFI_VA_END & ~PUD_MASK) != 0);
+
+	p4d_efi = p4d_offset(pgd_efi, EFI_VA_END);
+	p4d_k = p4d_offset(pgd_k, EFI_VA_END);
+	pud_efi = pud_offset(p4d_efi, 0);
+	pud_k = pud_offset(p4d_k, 0);
+
+	num_entries = pud_index(EFI_VA_END);
+	memcpy(pud_efi, pud_k, sizeof(pud_t) * num_entries);
+
+	pud_efi = pud_offset(p4d_efi, EFI_VA_START);
+	pud_k = pud_offset(p4d_k, EFI_VA_START);
+
+	num_entries = PTRS_PER_PUD - pud_index(EFI_VA_START);
+	memcpy(pud_efi, pud_k, sizeof(pud_t) * num_entries);
+}
+
+/*
+ * Wrapper for slow_virt_to_phys() that handles NULL addresses.
+ */
+static inline phys_addr_t
+virt_to_phys_or_null_size(void *va, unsigned long size)
+{
+	phys_addr_t pa;
+
+	if (!va)
+		return 0;
+
+	if (virt_addr_valid(va))
+		return virt_to_phys(va);
+
+	pa = slow_virt_to_phys(va);
+
+	/* check if the object crosses a page boundary */
+	if (WARN_ON((pa ^ (pa + size - 1)) & PAGE_MASK))
+		return 0;
+
+	return pa;
+}
+
+#define virt_to_phys_or_null(addr)				\
+	virt_to_phys_or_null_size((addr), sizeof(*(addr)))
+
+int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
+{
+	extern const u8 __efi64_thunk_ret_tramp[];
+	unsigned long pfn, text, pf, rodata, tramp;
+	struct page *page;
+	unsigned npages;
+	pgd_t *pgd = efi_mm.pgd;
+
+	/*
+	 * It can happen that the physical address of new_memmap lands in memory
+	 * which is not mapped in the EFI page table. Therefore we need to go
+	 * and ident-map those pages containing the map before calling
+	 * phys_efi_set_virtual_address_map().
+	 */
+	pfn = pa_memmap >> PAGE_SHIFT;
+	pf = _PAGE_NX | _PAGE_RW | _PAGE_ENC;
+	if (kernel_map_pages_in_pgd(pgd, pfn, pa_memmap, num_pages, pf)) {
+		pr_err("Error ident-mapping new memmap (0x%lx)!\n", pa_memmap);
+		return 1;
+	}
+
+	/*
+	 * Certain firmware versions are way too sentimental and still believe
+	 * they are exclusive and unquestionable owners of the first physical page,
+	 * even though they explicitly mark it as EFI_CONVENTIONAL_MEMORY
+	 * (but then write-access it later during SetVirtualAddressMap()).
+	 *
+	 * Create a 1:1 mapping for this page, to avoid triple faults during early
+	 * boot with such firmware. We are free to hand this page to the BIOS,
+	 * as trim_bios_range() will reserve the first page and isolate it away
+	 * from memory allocators anyway.
+	 */
+	if (kernel_map_pages_in_pgd(pgd, 0x0, 0x0, 1, pf)) {
+		pr_err("Failed to create 1:1 mapping for the first page!\n");
+		return 1;
+	}
+
+	/*
+	 * When SEV-ES is active, the GHCB as set by the kernel will be used
+	 * by firmware. Create a 1:1 unencrypted mapping for each GHCB.
+	 */
+	if (sev_es_efi_map_ghcbs_cas(pgd)) {
+		pr_err("Failed to create 1:1 mapping for the GHCBs and CAs!\n");
+		return 1;
+	}
+
+	/*
+	 * When making calls to the firmware everything needs to be 1:1
+	 * mapped and addressable with 32-bit pointers. Map the kernel
+	 * text and allocate a new stack because we can't rely on the
+	 * stack pointer being < 4GB.
+	 */
+	if (!efi_is_mixed())
+		return 0;
+
+	page = alloc_page(GFP_KERNEL|__GFP_DMA32);
+	if (!page) {
+		pr_err("Unable to allocate EFI runtime stack < 4GB\n");
+		return 1;
+	}
+
+	efi_mixed_mode_stack_pa = page_to_phys(page + 1); /* stack grows down */
+
+	npages = (_etext - _text) >> PAGE_SHIFT;
+	text = __pa(_text);
+
+	if (kernel_unmap_pages_in_pgd(pgd, text, npages)) {
+		pr_err("Failed to unmap kernel text 1:1 mapping\n");
+		return 1;
+	}
+
+	npages = (__end_rodata - __start_rodata) >> PAGE_SHIFT;
+	rodata = __pa(__start_rodata);
+	pfn = rodata >> PAGE_SHIFT;
+
+	pf = _PAGE_NX | _PAGE_ENC;
+	if (kernel_map_pages_in_pgd(pgd, pfn, rodata, npages, pf)) {
+		pr_err("Failed to map kernel rodata 1:1\n");
+		return 1;
+	}
+
+	tramp = __pa(__efi64_thunk_ret_tramp);
+	pfn = tramp >> PAGE_SHIFT;
+
+	pf = _PAGE_ENC;
+	if (kernel_map_pages_in_pgd(pgd, pfn, tramp, 1, pf)) {
+		pr_err("Failed to map mixed mode return trampoline\n");
+		return 1;
+	}
+
+	return 0;
+}
+
+static void __init __map_region(efi_memory_desc_t *md, u64 va)
+{
+	unsigned long flags = _PAGE_RW;
+	unsigned long pfn;
+	pgd_t *pgd = efi_mm.pgd;
+
+	/*
+	 * EFI_RUNTIME_SERVICES_CODE regions typically cover PE/COFF
+	 * executable images in memory that consist of both R-X and
+	 * RW- sections, so we cannot apply read-only or non-exec
+	 * permissions just yet. However, modern EFI systems provide
+	 * a memory attributes table that describes those sections
+	 * with the appropriate restricted permissions, which are
+	 * applied in efi_runtime_update_mappings() below. All other
+	 * regions can be mapped non-executable at this point, with
+	 * the exception of boot services code regions, but those will
+	 * be unmapped again entirely in efi_free_boot_services().
+	 */
+	if (md->type != EFI_BOOT_SERVICES_CODE &&
+	    md->type != EFI_RUNTIME_SERVICES_CODE)
+		flags |= _PAGE_NX;
+
+	if (!(md->attribute & EFI_MEMORY_WB))
+		flags |= _PAGE_PCD;
+
+	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT) &&
+	    md->type != EFI_MEMORY_MAPPED_IO)
+		flags |= _PAGE_ENC;
+
+	pfn = md->phys_addr >> PAGE_SHIFT;
+	if (kernel_map_pages_in_pgd(pgd, pfn, va, md->num_pages, flags))
+		pr_warn("Error mapping PA 0x%llx -> VA 0x%llx!\n",
+			   md->phys_addr, va);
+}
+
+void __init efi_map_region(efi_memory_desc_t *md)
+{
+	unsigned long size = md->num_pages << PAGE_SHIFT;
+	u64 pa = md->phys_addr;
+
+	/*
+	 * Make sure the 1:1 mappings are present as a catch-all for b0rked
+	 * firmware which doesn't update all internal pointers after switching
+	 * to virtual mode and would otherwise crap on us.
+	 */
+	__map_region(md, md->phys_addr);
+
+	/*
+	 * Enforce the 1:1 mapping as the default virtual address when
+	 * booting in EFI mixed mode, because even though we may be
+	 * running a 64-bit kernel, the firmware may only be 32-bit.
+	 */
+	if (efi_is_mixed()) {
+		md->virt_addr = md->phys_addr;
+		return;
+	}
+
+	efi_va -= size;
+
+	/* Is PA 2M-aligned? */
+	if (!(pa & (PMD_SIZE - 1))) {
+		efi_va &= PMD_MASK;
+	} else {
+		u64 pa_offset = pa & (PMD_SIZE - 1);
+		u64 prev_va = efi_va;
+
+		/* get us the same offset within this 2M page */
+		efi_va = (efi_va & PMD_MASK) + pa_offset;
+
+		if (efi_va > prev_va)
+			efi_va -= PMD_SIZE;
+	}
+
+	if (efi_va < EFI_VA_END) {
+		pr_warn(FW_WARN "VA address range overflow!\n");
+		return;
+	}
+
+	/* Do the VA map */
+	__map_region(md, efi_va);
+	md->virt_addr = efi_va;
+}
+
+/*
+ * kexec kernel will use efi_map_region_fixed to map efi runtime memory ranges.
+ * md->virt_addr is the original virtual address which had been mapped in kexec
+ * 1st kernel.
+ */
+void __init efi_map_region_fixed(efi_memory_desc_t *md)
+{
+	__map_region(md, md->phys_addr);
+	__map_region(md, md->virt_addr);
+}
+
+void __init parse_efi_setup(u64 phys_addr, u32 data_len)
+{
+	efi_setup = phys_addr + sizeof(struct setup_data);
+}
+
+static int __init efi_update_mappings(efi_memory_desc_t *md, unsigned long pf)
+{
+	unsigned long pfn;
+	pgd_t *pgd = efi_mm.pgd;
+	int err1, err2;
+
+	/* Update the 1:1 mapping */
+	pfn = md->phys_addr >> PAGE_SHIFT;
+	err1 = kernel_map_pages_in_pgd(pgd, pfn, md->phys_addr, md->num_pages, pf);
+	if (err1) {
+		pr_err("Error while updating 1:1 mapping PA 0x%llx -> VA 0x%llx!\n",
+			   md->phys_addr, md->virt_addr);
+	}
+
+	err2 = kernel_map_pages_in_pgd(pgd, pfn, md->virt_addr, md->num_pages, pf);
+	if (err2) {
+		pr_err("Error while updating VA mapping PA 0x%llx -> VA 0x%llx!\n",
+			   md->phys_addr, md->virt_addr);
+	}
+
+	return err1 || err2;
+}
+
+bool efi_disable_ibt_for_runtime __ro_after_init = true;
+
+static int __init efi_update_mem_attr(struct mm_struct *mm, efi_memory_desc_t *md,
+				      bool has_ibt)
+{
+	unsigned long pf = 0;
+
+	efi_disable_ibt_for_runtime |= !has_ibt;
+
+	if (md->attribute & EFI_MEMORY_XP)
+		pf |= _PAGE_NX;
+
+	if (!(md->attribute & EFI_MEMORY_RO))
+		pf |= _PAGE_RW;
+
+	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT))
+		pf |= _PAGE_ENC;
+
+	return efi_update_mappings(md, pf);
+}
+
+void __init efi_runtime_update_mappings(void)
+{
+	if (efi_enabled(EFI_MEM_ATTR)) {
+		efi_disable_ibt_for_runtime = false;
+		efi_memattr_apply_permissions(NULL, efi_update_mem_attr);
+	}
+}
+
+void __init efi_dump_pagetable(void)
+{
+#ifdef CONFIG_EFI_PGT_DUMP
+	ptdump_walk_pgd_level(NULL, &efi_mm);
+#endif
+}
+
+/*
+ * Makes the calling thread switch to/from efi_mm context. Can be used
+ * in a kernel thread and user context. Preemption needs to remain disabled
+ * while the EFI-mm is borrowed. mmgrab()/mmdrop() is not used because the mm
+ * can not change under us.
+ * It should be ensured that there are no concurrent calls to this function.
+ */
+static void efi_enter_mm(void)
+{
+	efi_prev_mm = use_temporary_mm(&efi_mm);
+}
+
+static void efi_leave_mm(void)
+{
+	unuse_temporary_mm(efi_prev_mm);
+}
+
+/*
+ * Toggle LASS to allow EFI to access any 1:1 mapped region in the lower
+ * half.
+ *
+ * Disable LASS only after switching to EFI-mm, as userspace is not
+ * mapped in it. Similar to EFI-mm, these rely on preemption being
+ * disabled and the calls being serialized.
+ */
+
+static void efi_disable_lass(void)
+{
+	if (!cpu_feature_enabled(X86_FEATURE_LASS))
+		return;
+
+	lockdep_assert_preemption_disabled();
+
+	/* Save current CR4.LASS state */
+	efi_cr4_lass = cr4_read_shadow() & X86_CR4_LASS;
+	cr4_clear_bits(efi_cr4_lass);
+}
+
+static void efi_enable_lass(void)
+{
+	if (!cpu_feature_enabled(X86_FEATURE_LASS))
+		return;
+
+	lockdep_assert_preemption_disabled();
+
+	/* Reprogram CR4.LASS only if it was set earlier */
+	cr4_set_bits(efi_cr4_lass);
+}
+
+void arch_efi_call_virt_setup(void)
+{
+	efi_sync_low_kernel_mappings();
+	efi_fpu_begin();
+	firmware_restrict_branch_speculation_start();
+	efi_enter_mm();
+	efi_disable_lass();
+}
+
+void arch_efi_call_virt_teardown(void)
+{
+	efi_enable_lass();
+	efi_leave_mm();
+	firmware_restrict_branch_speculation_end();
+	efi_fpu_end();
+}
+
+static DEFINE_SPINLOCK(efi_runtime_lock);
+
+/*
+ * DS and ES contain user values.  We need to save them.
+ * The 32-bit EFI code needs a valid DS, ES, and SS.  There's no
+ * need to save the old SS: __KERNEL_DS is always acceptable.
+ */
+#define __efi_thunk(func, ...)						\
+({									\
+	unsigned short __ds, __es;					\
+	efi_status_t ____s;						\
+									\
+	savesegment(ds, __ds);						\
+	savesegment(es, __es);						\
+									\
+	loadsegment(ss, __KERNEL_DS);					\
+	loadsegment(ds, __KERNEL_DS);					\
+	loadsegment(es, __KERNEL_DS);					\
+									\
+	____s = efi64_thunk(efi.runtime->mixed_mode.func, __VA_ARGS__);	\
+									\
+	loadsegment(ds, __ds);						\
+	loadsegment(es, __es);						\
+									\
+	____s ^= (____s & BIT(31)) | (____s & BIT_ULL(31)) << 32;	\
+	____s;								\
+})
+
+/*
+ * Switch to the EFI page tables early so that we can access the 1:1
+ * runtime services mappings which are not mapped in any other page
+ * tables.
+ *
+ * Also, disable interrupts because the IDT points to 64-bit handlers,
+ * which aren't going to function correctly when we switch to 32-bit.
+ */
+#define efi_thunk(func...)						\
+({									\
+	efi_status_t __s;						\
+									\
+	arch_efi_call_virt_setup();					\
+									\
+	__s = __efi_thunk(func);					\
+									\
+	arch_efi_call_virt_teardown();					\
+									\
+	__s;								\
+})
+
+static efi_status_t __init __no_sanitize_address
+efi_thunk_set_virtual_address_map(unsigned long memory_map_size,
+				  unsigned long descriptor_size,
+				  u32 descriptor_version,
+				  efi_memory_desc_t *virtual_map)
+{
+	efi_status_t status;
+	unsigned long flags;
+
+	efi_sync_low_kernel_mappings();
+	local_irq_save(flags);
+
+	efi_enter_mm();
+
+	status = __efi_thunk(set_virtual_address_map, memory_map_size,
+			     descriptor_size, descriptor_version, virtual_map);
+
+	efi_leave_mm();
+	local_irq_restore(flags);
+
+	return status;
+}
+
+static efi_status_t efi_thunk_get_time(efi_time_t *tm, efi_time_cap_t *tc)
+{
+	return EFI_UNSUPPORTED;
+}
+
+static efi_status_t efi_thunk_set_time(efi_time_t *tm)
+{
+	return EFI_UNSUPPORTED;
+}
+
+static efi_status_t
+efi_thunk_get_wakeup_time(efi_bool_t *enabled, efi_bool_t *pending,
+			  efi_time_t *tm)
+{
+	return EFI_UNSUPPORTED;
+}
+
+static efi_status_t
+efi_thunk_set_wakeup_time(efi_bool_t enabled, efi_time_t *tm)
+{
+	return EFI_UNSUPPORTED;
+}
+
+static unsigned long efi_name_size(efi_char16_t *name)
+{
+	return ucs2_strsize(name, EFI_VAR_NAME_LEN) + 1;
+}
+
+static efi_status_t
+efi_thunk_get_variable(efi_char16_t *name, efi_guid_t *vendor,
+		       u32 *attr, unsigned long *data_size, void *data)
+{
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+	efi_status_t status;
+	u32 phys_name, phys_vendor, phys_attr;
+	u32 phys_data_size, phys_data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&efi_runtime_lock, flags);
+
+	*vnd = *vendor;
+
+	phys_data_size = virt_to_phys_or_null(data_size);
+	phys_vendor = virt_to_phys_or_null(vnd);
+	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
+	phys_attr = virt_to_phys_or_null(attr);
+	phys_data = virt_to_phys_or_null_size(data, *data_size);
+
+	if (!phys_name || (data && !phys_data))
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(get_variable, phys_name, phys_vendor,
+				   phys_attr, phys_data_size, phys_data);
+
+	spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+	return status;
+}
+
+static efi_status_t
+efi_thunk_set_variable(efi_char16_t *name, efi_guid_t *vendor,
+		       u32 attr, unsigned long data_size, void *data)
+{
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+	u32 phys_name, phys_vendor, phys_data;
+	efi_status_t status;
+	unsigned long flags;
+
+	spin_lock_irqsave(&efi_runtime_lock, flags);
+
+	*vnd = *vendor;
+
+	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
+	phys_vendor = virt_to_phys_or_null(vnd);
+	phys_data = virt_to_phys_or_null_size(data, data_size);
+
+	if (!phys_name || (data && !phys_data))
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(set_variable, phys_name, phys_vendor,
+				   attr, data_size, phys_data);
+
+	spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+	return status;
+}
+
+static efi_status_t
+efi_thunk_set_variable_nonblocking(efi_char16_t *name, efi_guid_t *vendor,
+				   u32 attr, unsigned long data_size,
+				   void *data)
+{
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+	u32 phys_name, phys_vendor, phys_data;
+	efi_status_t status;
+	unsigned long flags;
+
+	if (!spin_trylock_irqsave(&efi_runtime_lock, flags))
+		return EFI_NOT_READY;
+
+	*vnd = *vendor;
+
+	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
+	phys_vendor = virt_to_phys_or_null(vnd);
+	phys_data = virt_to_phys_or_null_size(data, data_size);
+
+	if (!phys_name || (data && !phys_data))
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(set_variable, phys_name, phys_vendor,
+				   attr, data_size, phys_data);
+
+	spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+	return status;
+}
+
+static efi_status_t
+efi_thunk_get_next_variable(unsigned long *name_size,
+			    efi_char16_t *name,
+			    efi_guid_t *vendor)
+{
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+	efi_status_t status;
+	u32 phys_name_size, phys_name, phys_vendor;
+	unsigned long flags;
+
+	spin_lock_irqsave(&efi_runtime_lock, flags);
+
+	*vnd = *vendor;
+
+	phys_name_size = virt_to_phys_or_null(name_size);
+	phys_vendor = virt_to_phys_or_null(vnd);
+	phys_name = virt_to_phys_or_null_size(name, *name_size);
+
+	if (!phys_name)
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(get_next_variable, phys_name_size,
+				   phys_name, phys_vendor);
+
+	spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+	*vendor = *vnd;
+	return status;
+}
+
+static efi_status_t
+efi_thunk_get_next_high_mono_count(u32 *count)
+{
+	return EFI_UNSUPPORTED;
+}
+
+static void
+efi_thunk_reset_system(int reset_type, efi_status_t status,
+		       unsigned long data_size, efi_char16_t *data)
+{
+	u32 phys_data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&efi_runtime_lock, flags);
+
+	phys_data = virt_to_phys_or_null_size(data, data_size);
+
+	efi_thunk(reset_system, reset_type, status, data_size, phys_data);
+
+	spin_unlock_irqrestore(&efi_runtime_lock, flags);
+}
+
+static efi_status_t
+efi_thunk_update_capsule(efi_capsule_header_t **capsules,
+			 unsigned long count, unsigned long sg_list)
+{
+	/*
+	 * To properly support this function we would need to repackage
+	 * 'capsules' because the firmware doesn't understand 64-bit
+	 * pointers.
+	 */
+	return EFI_UNSUPPORTED;
+}
+
+static efi_status_t
+efi_thunk_query_variable_info(u32 attr, u64 *storage_space,
+			      u64 *remaining_space,
+			      u64 *max_variable_size)
+{
+	efi_status_t status;
+	u32 phys_storage, phys_remaining, phys_max;
+	unsigned long flags;
+
+	if (efi.runtime_version < EFI_2_00_SYSTEM_TABLE_REVISION)
+		return EFI_UNSUPPORTED;
+
+	spin_lock_irqsave(&efi_runtime_lock, flags);
+
+	phys_storage = virt_to_phys_or_null(storage_space);
+	phys_remaining = virt_to_phys_or_null(remaining_space);
+	phys_max = virt_to_phys_or_null(max_variable_size);
+
+	status = efi_thunk(query_variable_info, attr, phys_storage,
+			   phys_remaining, phys_max);
+
+	spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+	return status;
+}
+
+static efi_status_t
+efi_thunk_query_variable_info_nonblocking(u32 attr, u64 *storage_space,
+					  u64 *remaining_space,
+					  u64 *max_variable_size)
+{
+	efi_status_t status;
+	u32 phys_storage, phys_remaining, phys_max;
+	unsigned long flags;
+
+	if (efi.runtime_version < EFI_2_00_SYSTEM_TABLE_REVISION)
+		return EFI_UNSUPPORTED;
+
+	if (!spin_trylock_irqsave(&efi_runtime_lock, flags))
+		return EFI_NOT_READY;
+
+	phys_storage = virt_to_phys_or_null(storage_space);
+	phys_remaining = virt_to_phys_or_null(remaining_space);
+	phys_max = virt_to_phys_or_null(max_variable_size);
+
+	status = efi_thunk(query_variable_info, attr, phys_storage,
+			   phys_remaining, phys_max);
+
+	spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+	return status;
+}
+
+static efi_status_t
+efi_thunk_query_capsule_caps(efi_capsule_header_t **capsules,
+			     unsigned long count, u64 *max_size,
+			     int *reset_type)
+{
+	/*
+	 * To properly support this function we would need to repackage
+	 * 'capsules' because the firmware doesn't understand 64-bit
+	 * pointers.
+	 */
+	return EFI_UNSUPPORTED;
+}
+
+void __init efi_thunk_runtime_setup(void)
+{
+	if (!IS_ENABLED(CONFIG_EFI_MIXED))
+		return;
+
+	efi.get_time = efi_thunk_get_time;
+	efi.set_time = efi_thunk_set_time;
+	efi.get_wakeup_time = efi_thunk_get_wakeup_time;
+	efi.set_wakeup_time = efi_thunk_set_wakeup_time;
+	efi.get_variable = efi_thunk_get_variable;
+	efi.get_next_variable = efi_thunk_get_next_variable;
+	efi.set_variable = efi_thunk_set_variable;
+	efi.set_variable_nonblocking = efi_thunk_set_variable_nonblocking;
+	efi.get_next_high_mono_count = efi_thunk_get_next_high_mono_count;
+	efi.reset_system = efi_thunk_reset_system;
+	efi.query_variable_info = efi_thunk_query_variable_info;
+	efi.query_variable_info_nonblocking = efi_thunk_query_variable_info_nonblocking;
+	efi.update_capsule = efi_thunk_update_capsule;
+	efi.query_capsule_caps = efi_thunk_query_capsule_caps;
+}
+
+efi_status_t __init __no_sanitize_address
+efi_set_virtual_address_map(unsigned long memory_map_size,
+			    unsigned long descriptor_size,
+			    u32 descriptor_version,
+			    efi_memory_desc_t *virtual_map,
+			    unsigned long systab_phys)
+{
+	const efi_system_table_t *systab = (efi_system_table_t *)systab_phys;
+	efi_status_t status;
+	unsigned long flags;
+
+	if (efi_is_mixed())
+		return efi_thunk_set_virtual_address_map(memory_map_size,
+							 descriptor_size,
+							 descriptor_version,
+							 virtual_map);
+	efi_enter_mm();
+
+	efi_fpu_begin();
+
+	/* Disable interrupts around EFI calls: */
+	local_irq_save(flags);
+	status = arch_efi_call_virt(efi.runtime, set_virtual_address_map,
+				    memory_map_size, descriptor_size,
+				    descriptor_version, virtual_map);
+	local_irq_restore(flags);
+
+	efi_fpu_end();
+
+	/* grab the virtually remapped EFI runtime services table pointer */
+	efi.runtime = READ_ONCE(systab->runtime);
+
+	efi_leave_mm();
+
+	return status;
+}

linux-7.2.3/arch/x86/platform/efi/efi_stub_32.S

@@ -0,0 +1,60 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * EFI call stub for IA32.
+ *
+ * This stub allows us to make EFI calls in physical mode with interrupts
+ * turned off.
+ */
+
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/asm-offsets.h>
+#include <asm/page_types.h>
+
+	__INIT
+SYM_FUNC_START(efi_call_svam)
+	push	%ebp
+	movl	%esp, %ebp
+	push	%ebx
+
+	push	16(%esp)
+	push	16(%esp)
+	push	%ecx
+	push	%edx
+	movl	%eax, %ebx		// &systab_phys->runtime
+
+	/*
+	 * Switch to the flat mapped alias of this routine, by jumping to the
+	 * address of label '1' after subtracting PAGE_OFFSET from it.
+	 */
+	movl	$1f, %edx
+	subl	$__PAGE_OFFSET, %edx
+	jmp	*%edx
+1:
+
+	/* disable paging */
+	movl	%cr0, %edx
+	andl	$0x7fffffff, %edx
+	movl	%edx, %cr0
+
+	/* convert the stack pointer to a flat mapped address */
+	subl	$__PAGE_OFFSET, %esp
+
+	/* call the EFI routine */
+	movl	(%eax), %eax
+	call	*EFI_svam(%eax)
+
+	/* grab the virtually remapped EFI runtime services table pointer */
+	movl	(%ebx), %ecx
+	movl	36(%esp), %edx		// &efi.runtime
+	movl	%ecx, (%edx)
+
+	/* re-enable paging */
+	movl	%cr0, %edx
+	orl	$0x80000000, %edx
+	movl	%edx, %cr0
+
+	movl	16(%esp), %ebx
+	leave
+	RET
+SYM_FUNC_END(efi_call_svam)

linux-7.2.3/arch/x86/platform/efi/efi_stub_64.S

@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Function calling ABI conversion from Linux to EFI for x86_64
+ *
+ * Copyright (C) 2007 Intel Corp
+ *	Bibo Mao <bibo.mao@intel.com>
+ *	Huang Ying <ying.huang@intel.com>
+ */
+
+#include <linux/linkage.h>
+#include <asm/nospec-branch.h>
+
+SYM_FUNC_START(__efi_call)
+	/*
+	 * The EFI code doesn't have any CFI, annotate away the CFI violation.
+	 */
+	ANNOTATE_NOCFI_SYM
+	pushq %rbp
+	movq %rsp, %rbp
+	and $~0xf, %rsp
+	mov 16(%rbp), %rax
+	subq $48, %rsp
+	mov %r9, 32(%rsp)
+	mov %rax, 40(%rsp)
+	mov %r8, %r9
+	mov %rcx, %r8
+	mov %rsi, %rcx
+	CALL_NOSPEC rdi
+	leave
+	RET
+SYM_FUNC_END(__efi_call)

linux-7.2.3/arch/x86/platform/efi/efi_thunk_64.S

@@ -0,0 +1,98 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2014 Intel Corporation; author Matt Fleming
+ *
+ * Support for invoking 32-bit EFI runtime services from a 64-bit
+ * kernel.
+ *
+ * The below thunking functions are only used after ExitBootServices()
+ * has been called. This simplifies things considerably as compared with
+ * the early EFI thunking because we can leave all the kernel state
+ * intact (GDT, IDT, etc) and simply invoke the 32-bit EFI runtime
+ * services from __KERNEL32_CS. This means we can continue to service
+ * interrupts across an EFI mixed mode call.
+ *
+ * We do however, need to handle the fact that we're running in a full
+ * 64-bit virtual address space. Things like the stack and instruction
+ * addresses need to be accessible by the 32-bit firmware, so we rely on
+ * using the identity mappings in the EFI page table to access the stack
+ * and kernel text (see efi_setup_page_tables()).
+ */
+
+#include <linux/linkage.h>
+#include <linux/objtool.h>
+#include <asm/page_types.h>
+#include <asm/segment.h>
+
+	.text
+	.code64
+SYM_FUNC_START(__efi64_thunk)
+STACK_FRAME_NON_STANDARD __efi64_thunk
+	push	%rbp
+	push	%rbx
+
+	/*
+	 * Switch to 1:1 mapped 32-bit stack pointer.
+	 */
+	movq	%rsp, %rax
+	movq	efi_mixed_mode_stack_pa(%rip), %rsp
+	push	%rax
+
+	/*
+	 * Copy args passed via the stack
+	 */
+	subq	$0x24, %rsp
+	movq	0x18(%rax), %rbp
+	movq	0x20(%rax), %rbx
+	movq	0x28(%rax), %rax
+	movl	%ebp, 0x18(%rsp)
+	movl	%ebx, 0x1c(%rsp)
+	movl	%eax, 0x20(%rsp)
+
+	/*
+	 * Calculate the physical address of the kernel text.
+	 */
+	movq	$__START_KERNEL_map, %rax
+	subq	phys_base(%rip), %rax
+
+	leaq	1f(%rip), %rbp
+	leaq	2f(%rip), %rbx
+	subq	%rax, %rbp
+	subq	%rax, %rbx
+
+	movl	%ebx, 0x0(%rsp)		/* return address */
+	movl	%esi, 0x4(%rsp)
+	movl	%edx, 0x8(%rsp)
+	movl	%ecx, 0xc(%rsp)
+	movl	%r8d, 0x10(%rsp)
+	movl	%r9d, 0x14(%rsp)
+
+	/* Switch to 32-bit descriptor */
+	pushq	$__KERNEL32_CS
+	pushq	%rdi			/* EFI runtime service address */
+	lretq
+
+	// This return instruction is not needed for correctness, as it will
+	// never be reached. It only exists to make objtool happy, which will
+	// otherwise complain about unreachable instructions in the callers.
+	RET
+SYM_FUNC_END(__efi64_thunk)
+
+	.section ".rodata", "a", @progbits
+	.balign	16
+SYM_DATA_START(__efi64_thunk_ret_tramp)
+1:	movq	0x20(%rsp), %rsp
+	pop	%rbx
+	pop	%rbp
+	ret
+	int3
+
+	.code32
+2:	pushl	$__KERNEL_CS
+	pushl	%ebp
+	lret
+SYM_DATA_END(__efi64_thunk_ret_tramp)
+
+	.bss
+	.balign 8
+SYM_DATA(efi_mixed_mode_stack_pa, .quad 0)

linux-7.2.3/arch/x86/platform/efi/memmap.c

@@ -0,0 +1,247 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Common EFI memory map functions.
+ */
+
+#define pr_fmt(fmt) "efi: " fmt
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/efi.h>
+#include <linux/io.h>
+#include <asm/early_ioremap.h>
+#include <asm/efi.h>
+#include <linux/memblock.h>
+#include <linux/slab.h>
+
+static phys_addr_t __init __efi_memmap_alloc_early(unsigned long size)
+{
+	return memblock_phys_alloc(size, SMP_CACHE_BYTES);
+}
+
+static phys_addr_t __init __efi_memmap_alloc_late(unsigned long size)
+{
+	unsigned int order = get_order(size);
+	struct page *p = alloc_pages(GFP_KERNEL, order);
+
+	if (!p)
+		return 0;
+
+	return PFN_PHYS(page_to_pfn(p));
+}
+
+static
+void __init __efi_memmap_free(u64 phys, unsigned long size, unsigned long flags)
+{
+	if (flags & EFI_MEMMAP_MEMBLOCK) {
+		memblock_phys_free(phys, size);
+	} else if (flags & EFI_MEMMAP_SLAB) {
+		struct page *p = pfn_to_page(PHYS_PFN(phys));
+		unsigned int order = get_order(size);
+
+		__free_pages(p, order);
+	}
+}
+
+/**
+ * efi_memmap_alloc - Allocate memory for the EFI memory map
+ * @num_entries: Number of entries in the allocated map.
+ * @data: efi memmap installation parameters
+ *
+ * Depending on whether mm_init() has already been invoked or not,
+ * either memblock or "normal" page allocation is used.
+ *
+ * Returns zero on success, a negative error code on failure.
+ */
+int __init efi_memmap_alloc(unsigned int num_entries,
+		struct efi_memory_map_data *data)
+{
+	/* Expect allocation parameters are zero initialized */
+	WARN_ON(data->phys_map || data->size);
+
+	data->size = num_entries * efi.memmap.desc_size;
+	data->desc_version = efi.memmap.desc_version;
+	data->desc_size = efi.memmap.desc_size;
+	data->flags &= ~(EFI_MEMMAP_SLAB | EFI_MEMMAP_MEMBLOCK);
+	data->flags |= efi.memmap.flags & EFI_MEMMAP_LATE;
+
+	if (slab_is_available()) {
+		data->flags |= EFI_MEMMAP_SLAB;
+		data->phys_map = __efi_memmap_alloc_late(data->size);
+	} else {
+		data->flags |= EFI_MEMMAP_MEMBLOCK;
+		data->phys_map = __efi_memmap_alloc_early(data->size);
+	}
+
+	if (!data->phys_map)
+		return -ENOMEM;
+	return 0;
+}
+
+/**
+ * efi_memmap_install - Install a new EFI memory map in efi.memmap
+ * @data: efi memmap installation parameters
+ *
+ * Unlike efi_memmap_init_*(), this function does not allow the caller
+ * to switch from early to late mappings. It simply uses the existing
+ * mapping function and installs the new memmap.
+ *
+ * Returns zero on success, a negative error code on failure.
+ */
+int __init efi_memmap_install(struct efi_memory_map_data *data)
+{
+	unsigned long size = efi.memmap.desc_size * efi.memmap.nr_map;
+	unsigned long flags = efi.memmap.flags;
+	u64 phys = efi.memmap.phys_map;
+	int ret;
+
+	efi_memmap_unmap();
+
+	if (efi_enabled(EFI_PARAVIRT))
+		return 0;
+
+	ret = __efi_memmap_init(data);
+	if (ret)
+		return ret;
+
+	__efi_memmap_free(phys, size, flags);
+	return 0;
+}
+
+/**
+ * efi_memmap_split_count - Count number of additional EFI memmap entries
+ * @md: EFI memory descriptor to split
+ * @range: Address range (start, end) to split around
+ *
+ * Returns the number of additional EFI memmap entries required to
+ * accommodate @range.
+ */
+int __init efi_memmap_split_count(efi_memory_desc_t *md, struct range *range)
+{
+	u64 m_start, m_end;
+	u64 start, end;
+	int count = 0;
+
+	start = md->phys_addr;
+	end = start + (md->num_pages << EFI_PAGE_SHIFT) - 1;
+
+	/* modifying range */
+	m_start = range->start;
+	m_end = range->end;
+
+	if (m_start <= start) {
+		/* split into 2 parts */
+		if (start < m_end && m_end < end)
+			count++;
+	}
+
+	if (start < m_start && m_start < end) {
+		/* split into 3 parts */
+		if (m_end < end)
+			count += 2;
+		/* split into 2 parts */
+		if (end <= m_end)
+			count++;
+	}
+
+	return count;
+}
+
+/**
+ * efi_memmap_insert - Insert a memory region in an EFI memmap
+ * @old_memmap: The existing EFI memory map structure
+ * @buf: Address of buffer to store new map
+ * @mem: Memory map entry to insert
+ *
+ * It is suggested that you call efi_memmap_split_count() first
+ * to see how large @buf needs to be.
+ */
+void __init efi_memmap_insert(struct efi_memory_map *old_memmap, void *buf,
+			      struct efi_mem_range *mem)
+{
+	u64 m_start, m_end, m_attr;
+	efi_memory_desc_t *md;
+	u64 start, end;
+	void *old, *new;
+
+	/* modifying range */
+	m_start = mem->range.start;
+	m_end = mem->range.end;
+	m_attr = mem->attribute;
+
+	/*
+	 * The EFI memory map deals with regions in EFI_PAGE_SIZE
+	 * units. Ensure that the region described by 'mem' is aligned
+	 * correctly.
+	 */
+	if (!IS_ALIGNED(m_start, EFI_PAGE_SIZE) ||
+	    !IS_ALIGNED(m_end + 1, EFI_PAGE_SIZE)) {
+		WARN_ON(1);
+		return;
+	}
+
+	for (old = old_memmap->map, new = buf;
+	     old < old_memmap->map_end;
+	     old += old_memmap->desc_size, new += old_memmap->desc_size) {
+
+		/* copy original EFI memory descriptor */
+		memcpy(new, old, old_memmap->desc_size);
+		md = new;
+		start = md->phys_addr;
+		end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) - 1;
+
+		if (m_start <= start && end <= m_end)
+			md->attribute |= m_attr;
+
+		if (m_start <= start &&
+		    (start < m_end && m_end < end)) {
+			/* first part */
+			md->attribute |= m_attr;
+			md->num_pages = (m_end - md->phys_addr + 1) >>
+				EFI_PAGE_SHIFT;
+			/* latter part */
+			new += old_memmap->desc_size;
+			memcpy(new, old, old_memmap->desc_size);
+			md = new;
+			md->phys_addr = m_end + 1;
+			md->num_pages = (end - md->phys_addr + 1) >>
+				EFI_PAGE_SHIFT;
+		}
+
+		if ((start < m_start && m_start < end) && m_end < end) {
+			/* first part */
+			md->num_pages = (m_start - md->phys_addr) >>
+				EFI_PAGE_SHIFT;
+			/* middle part */
+			new += old_memmap->desc_size;
+			memcpy(new, old, old_memmap->desc_size);
+			md = new;
+			md->attribute |= m_attr;
+			md->phys_addr = m_start;
+			md->num_pages = (m_end - m_start + 1) >>
+				EFI_PAGE_SHIFT;
+			/* last part */
+			new += old_memmap->desc_size;
+			memcpy(new, old, old_memmap->desc_size);
+			md = new;
+			md->phys_addr = m_end + 1;
+			md->num_pages = (end - m_end) >>
+				EFI_PAGE_SHIFT;
+		}
+
+		if ((start < m_start && m_start < end) &&
+		    (end <= m_end)) {
+			/* first part */
+			md->num_pages = (m_start - md->phys_addr) >>
+				EFI_PAGE_SHIFT;
+			/* latter part */
+			new += old_memmap->desc_size;
+			memcpy(new, old, old_memmap->desc_size);
+			md = new;
+			md->phys_addr = m_start;
+			md->num_pages = (end - md->phys_addr + 1) >>
+				EFI_PAGE_SHIFT;
+			md->attribute |= m_attr;
+		}
+	}
+}

linux-7.2.3/arch/x86/platform/efi/quirks.c

@@ -0,0 +1,843 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#define pr_fmt(fmt) "efi: " fmt
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/time.h>
+#include <linux/types.h>
+#include <linux/efi.h>
+#include <linux/slab.h>
+#include <linux/memblock.h>
+#include <linux/acpi.h>
+#include <linux/dmi.h>
+
+#include <asm/e820/api.h>
+#include <asm/efi.h>
+#include <asm/uv/uv.h>
+#include <asm/cpu_device_id.h>
+#include <asm/realmode.h>
+#include <asm/reboot.h>
+
+#define EFI_MIN_RESERVE 5120
+
+#define EFI_DUMMY_GUID \
+	EFI_GUID(0x4424ac57, 0xbe4b, 0x47dd, 0x9e, 0x97, 0xed, 0x50, 0xf0, 0x9f, 0x92, 0xa9)
+
+#define QUARK_CSH_SIGNATURE		0x5f435348	/* _CSH */
+#define QUARK_SECURITY_HEADER_SIZE	0x400
+
+/*
+ * Header prepended to the standard EFI capsule on Quark systems the are based
+ * on Intel firmware BSP.
+ * @csh_signature:	Unique identifier to sanity check signed module
+ * 			presence ("_CSH").
+ * @version:		Current version of CSH used. Should be one for Quark A0.
+ * @modulesize:		Size of the entire module including the module header
+ * 			and payload.
+ * @security_version_number_index: Index of SVN to use for validation of signed
+ * 			module.
+ * @security_version_number: Used to prevent against roll back of modules.
+ * @rsvd_module_id:	Currently unused for Clanton (Quark).
+ * @rsvd_module_vendor:	Vendor Identifier. For Intel products value is
+ * 			0x00008086.
+ * @rsvd_date:		BCD representation of build date as yyyymmdd, where
+ * 			yyyy=4 digit year, mm=1-12, dd=1-31.
+ * @headersize:		Total length of the header including including any
+ * 			padding optionally added by the signing tool.
+ * @hash_algo:		What Hash is used in the module signing.
+ * @cryp_algo:		What Crypto is used in the module signing.
+ * @keysize:		Total length of the key data including including any
+ * 			padding optionally added by the signing tool.
+ * @signaturesize:	Total length of the signature including including any
+ * 			padding optionally added by the signing tool.
+ * @rsvd_next_header:	32-bit pointer to the next Secure Boot Module in the
+ * 			chain, if there is a next header.
+ * @rsvd:		Reserved, padding structure to required size.
+ *
+ * See also QuartSecurityHeader_t in
+ * Quark_EDKII_v1.2.1.1/QuarkPlatformPkg/Include/QuarkBootRom.h
+ * from https://downloadcenter.intel.com/download/23197/Intel-Quark-SoC-X1000-Board-Support-Package-BSP
+ */
+struct quark_security_header {
+	u32 csh_signature;
+	u32 version;
+	u32 modulesize;
+	u32 security_version_number_index;
+	u32 security_version_number;
+	u32 rsvd_module_id;
+	u32 rsvd_module_vendor;
+	u32 rsvd_date;
+	u32 headersize;
+	u32 hash_algo;
+	u32 cryp_algo;
+	u32 keysize;
+	u32 signaturesize;
+	u32 rsvd_next_header;
+	u32 rsvd[2];
+};
+
+static const efi_char16_t efi_dummy_name[] = L"DUMMY";
+
+static bool efi_no_storage_paranoia;
+
+/*
+ * Some firmware implementations refuse to boot if there's insufficient
+ * space in the variable store. The implementation of garbage collection
+ * in some FW versions causes stale (deleted) variables to take up space
+ * longer than intended and space is only freed once the store becomes
+ * almost completely full.
+ *
+ * Enabling this option disables the space checks in
+ * efi_query_variable_store() and forces garbage collection.
+ *
+ * Only enable this option if deleting EFI variables does not free up
+ * space in your variable store, e.g. if despite deleting variables
+ * you're unable to create new ones.
+ */
+static int __init setup_storage_paranoia(char *arg)
+{
+	efi_no_storage_paranoia = true;
+	return 0;
+}
+early_param("efi_no_storage_paranoia", setup_storage_paranoia);
+
+/*
+ * Deleting the dummy variable which kicks off garbage collection
+*/
+void efi_delete_dummy_variable(void)
+{
+	efi.set_variable_nonblocking((efi_char16_t *)efi_dummy_name,
+				     &EFI_DUMMY_GUID,
+				     EFI_VARIABLE_NON_VOLATILE |
+				     EFI_VARIABLE_BOOTSERVICE_ACCESS |
+				     EFI_VARIABLE_RUNTIME_ACCESS, 0, NULL);
+}
+
+u64 efivar_reserved_space(void)
+{
+	if (efi_no_storage_paranoia)
+		return 0;
+	return EFI_MIN_RESERVE;
+}
+EXPORT_SYMBOL_GPL(efivar_reserved_space);
+
+/*
+ * In the nonblocking case we do not attempt to perform garbage
+ * collection if we do not have enough free space. Rather, we do the
+ * bare minimum check and give up immediately if the available space
+ * is below EFI_MIN_RESERVE.
+ *
+ * This function is intended to be small and simple because it is
+ * invoked from crash handler paths.
+ */
+static efi_status_t
+query_variable_store_nonblocking(u32 attributes, unsigned long size)
+{
+	efi_status_t status;
+	u64 storage_size, remaining_size, max_size;
+
+	status = efi.query_variable_info_nonblocking(attributes, &storage_size,
+						     &remaining_size,
+						     &max_size);
+	if (status != EFI_SUCCESS)
+		return status;
+
+	if (remaining_size - size < EFI_MIN_RESERVE)
+		return EFI_OUT_OF_RESOURCES;
+
+	return EFI_SUCCESS;
+}
+
+/*
+ * Some firmware implementations refuse to boot if there's insufficient space
+ * in the variable store. Ensure that we never use more than a safe limit.
+ *
+ * Return EFI_SUCCESS if it is safe to write 'size' bytes to the variable
+ * store.
+ */
+efi_status_t efi_query_variable_store(u32 attributes, unsigned long size,
+				      bool nonblocking)
+{
+	efi_status_t status;
+	u64 storage_size, remaining_size, max_size;
+
+	if (!(attributes & EFI_VARIABLE_NON_VOLATILE))
+		return 0;
+
+	if (nonblocking)
+		return query_variable_store_nonblocking(attributes, size);
+
+	status = efi.query_variable_info(attributes, &storage_size,
+					 &remaining_size, &max_size);
+	if (status != EFI_SUCCESS)
+		return status;
+
+	/*
+	 * We account for that by refusing the write if permitting it would
+	 * reduce the available space to under 5KB. This figure was provided by
+	 * Samsung, so should be safe.
+	 */
+	if ((remaining_size - size < EFI_MIN_RESERVE) &&
+		!efi_no_storage_paranoia) {
+
+		/*
+		 * Triggering garbage collection may require that the firmware
+		 * generate a real EFI_OUT_OF_RESOURCES error. We can force
+		 * that by attempting to use more space than is available.
+		 */
+		unsigned long dummy_size = remaining_size + 1024;
+		void *dummy = kzalloc(dummy_size, GFP_KERNEL);
+
+		if (!dummy)
+			return EFI_OUT_OF_RESOURCES;
+
+		status = efi.set_variable((efi_char16_t *)efi_dummy_name,
+					  &EFI_DUMMY_GUID,
+					  EFI_VARIABLE_NON_VOLATILE |
+					  EFI_VARIABLE_BOOTSERVICE_ACCESS |
+					  EFI_VARIABLE_RUNTIME_ACCESS,
+					  dummy_size, dummy);
+
+		if (status == EFI_SUCCESS) {
+			/*
+			 * This should have failed, so if it didn't make sure
+			 * that we delete it...
+			 */
+			efi_delete_dummy_variable();
+		}
+
+		kfree(dummy);
+
+		/*
+		 * The runtime code may now have triggered a garbage collection
+		 * run, so check the variable info again
+		 */
+		status = efi.query_variable_info(attributes, &storage_size,
+						 &remaining_size, &max_size);
+
+		if (status != EFI_SUCCESS)
+			return status;
+
+		/*
+		 * There still isn't enough room, so return an error
+		 */
+		if (remaining_size - size < EFI_MIN_RESERVE)
+			return EFI_OUT_OF_RESOURCES;
+	}
+
+	return EFI_SUCCESS;
+}
+EXPORT_SYMBOL_GPL(efi_query_variable_store);
+
+/*
+ * The UEFI specification makes it clear that the operating system is
+ * free to do whatever it wants with boot services code after
+ * ExitBootServices() has been called. Ignoring this recommendation a
+ * significant bunch of EFI implementations continue calling into boot
+ * services code (SetVirtualAddressMap). In order to work around such
+ * buggy implementations we reserve boot services region during EFI
+ * init and make sure it stays executable. Then, after
+ * SetVirtualAddressMap(), it is discarded.
+ *
+ * However, some boot services regions contain data that is required
+ * by drivers, so we need to track which memory ranges can never be
+ * freed. This is done by tagging those regions with the
+ * EFI_MEMORY_RUNTIME attribute.
+ *
+ * Any driver that wants to mark a region as reserved must use
+ * efi_mem_reserve() which will insert a new EFI memory descriptor
+ * into efi.memmap (splitting existing regions if necessary) and tag
+ * it with EFI_MEMORY_RUNTIME.
+ */
+void __init efi_arch_mem_reserve(phys_addr_t addr, u64 size)
+{
+	struct efi_memory_map_data data = { 0 };
+	struct efi_mem_range mr;
+	efi_memory_desc_t md;
+	int num_entries;
+	void *new;
+
+	if (efi_mem_desc_lookup(addr, &md) ||
+	    md.type != EFI_BOOT_SERVICES_DATA) {
+		pr_err("Failed to lookup EFI memory descriptor for %pa\n", &addr);
+		return;
+	}
+
+	if (addr + size > md.phys_addr + (md.num_pages << EFI_PAGE_SHIFT)) {
+		pr_err("Region spans EFI memory descriptors, %pa\n", &addr);
+		return;
+	}
+
+	size += addr % EFI_PAGE_SIZE;
+	size = round_up(size, EFI_PAGE_SIZE);
+	addr = round_down(addr, EFI_PAGE_SIZE);
+
+	mr.range.start = addr;
+	mr.range.end = addr + size - 1;
+	mr.attribute = md.attribute | EFI_MEMORY_RUNTIME;
+
+	num_entries = efi_memmap_split_count(&md, &mr.range);
+	num_entries += efi.memmap.nr_map;
+
+	if (efi_memmap_alloc(num_entries, &data) != 0) {
+		pr_err("Could not allocate boot services memmap\n");
+		return;
+	}
+
+	new = early_memremap_prot(data.phys_map, data.size,
+				  pgprot_val(pgprot_encrypted(FIXMAP_PAGE_NORMAL)));
+	if (!new) {
+		pr_err("Failed to map new boot services memmap\n");
+		return;
+	}
+
+	efi_memmap_insert(&efi.memmap, new, &mr);
+	early_memunmap(new, data.size);
+
+	efi_memmap_install(&data);
+	e820__range_update(addr, size, E820_TYPE_RAM, E820_TYPE_RESERVED);
+	e820__update_table(e820_table);
+}
+
+/*
+ * Helper function for efi_reserve_boot_services() to figure out if we
+ * can free regions in efi_free_boot_services().
+ *
+ * Use this function to ensure we do not free regions owned by somebody
+ * else. We must only reserve (and then free) regions:
+ *
+ * - Not within any part of the kernel
+ * - Not the BIOS reserved area (E820_TYPE_RESERVED, E820_TYPE_NVS, etc)
+ */
+static __init bool can_free_region(u64 start, u64 size)
+{
+	if (start + size > __pa_symbol(_text) && start <= __pa_symbol(_end))
+		return false;
+
+	if (!e820__mapped_all(start, start+size, E820_TYPE_RAM))
+		return false;
+
+	return true;
+}
+
+void __init efi_reserve_boot_services(void)
+{
+	efi_memory_desc_t *md;
+
+	if (!efi_enabled(EFI_MEMMAP))
+		return;
+
+	for_each_efi_memory_desc(md) {
+		u64 start = md->phys_addr;
+		u64 size = md->num_pages << EFI_PAGE_SHIFT;
+		bool already_reserved;
+
+		if (md->type != EFI_BOOT_SERVICES_CODE &&
+		    md->type != EFI_BOOT_SERVICES_DATA)
+			continue;
+
+		already_reserved = memblock_is_region_reserved(start, size);
+
+		/*
+		 * Because the following memblock_reserve() is paired
+		 * with free_reserved_area() for this region in
+		 * efi_free_boot_services(), we must be extremely
+		 * careful not to reserve, and subsequently free,
+		 * critical regions of memory (like the kernel image) or
+		 * those regions that somebody else has already
+		 * reserved.
+		 *
+		 * A good example of a critical region that must not be
+		 * freed is page zero (first 4Kb of memory), which may
+		 * contain boot services code/data but is marked
+		 * E820_TYPE_RESERVED by trim_bios_range().
+		 */
+		if (!already_reserved) {
+			memblock_reserve(start, size);
+
+			/*
+			 * If we are the first to reserve the region, no
+			 * one else cares about it. We own it and can
+			 * free it later.
+			 */
+			if (can_free_region(start, size))
+				continue;
+		}
+
+		/*
+		 * We don't own the region. We must not free it.
+		 *
+		 * Setting this bit for a boot services region really
+		 * doesn't make sense as far as the firmware is
+		 * concerned, but it does provide us with a way to tag
+		 * those regions that must not be paired with
+		 * memblock_phys_free().
+		 */
+		md->attribute |= EFI_MEMORY_RUNTIME;
+	}
+}
+
+/*
+ * Apart from having VA mappings for EFI boot services code/data regions,
+ * (duplicate) 1:1 mappings were also created as a quirk for buggy firmware. So,
+ * unmap both 1:1 and VA mappings.
+ */
+static void __init efi_unmap_pages(efi_memory_desc_t *md)
+{
+	pgd_t *pgd = efi_mm.pgd;
+	u64 pa = md->phys_addr;
+	u64 va = md->virt_addr;
+
+	/*
+	 * EFI mixed mode has all RAM mapped to access arguments while making
+	 * EFI runtime calls, hence don't unmap EFI boot services code/data
+	 * regions.
+	 */
+	if (efi_is_mixed())
+		return;
+
+	if (kernel_unmap_pages_in_pgd(pgd, pa, md->num_pages))
+		pr_err("Failed to unmap 1:1 mapping for 0x%llx\n", pa);
+
+	if (kernel_unmap_pages_in_pgd(pgd, va, md->num_pages))
+		pr_err("Failed to unmap VA mapping for 0x%llx\n", va);
+}
+
+struct efi_freeable_range {
+	u64 start;
+	u64 end;
+};
+
+static struct efi_freeable_range *ranges_to_free;
+
+void __init efi_unmap_boot_services(void)
+{
+	struct efi_memory_map_data data = { 0 };
+	efi_memory_desc_t *md;
+	int num_entries = 0;
+	int idx = 0;
+	size_t sz;
+	void *new, *new_md;
+
+	/* Keep all regions for /sys/kernel/debug/efi */
+	if (efi_enabled(EFI_DBG))
+		return;
+
+	sz = sizeof(*ranges_to_free) * (efi.memmap.nr_map + 1);
+	ranges_to_free = kzalloc(sz, GFP_KERNEL);
+	if (!ranges_to_free) {
+		pr_err("Failed to allocate storage for freeable EFI regions\n");
+		return;
+	}
+
+	for_each_efi_memory_desc(md) {
+		unsigned long long start = md->phys_addr;
+		unsigned long long size = md->num_pages << EFI_PAGE_SHIFT;
+		size_t rm_size;
+
+		if (md->type != EFI_BOOT_SERVICES_CODE &&
+		    md->type != EFI_BOOT_SERVICES_DATA) {
+			num_entries++;
+			continue;
+		}
+
+		/* Do not free, someone else owns it: */
+		if (md->attribute & EFI_MEMORY_RUNTIME) {
+			num_entries++;
+			continue;
+		}
+
+		/*
+		 * Before calling set_virtual_address_map(), EFI boot services
+		 * code/data regions were mapped as a quirk for buggy firmware.
+		 * Unmap them from efi_pgd before freeing them up.
+		 */
+		efi_unmap_pages(md);
+
+		/*
+		 * Nasty quirk: if all sub-1MB memory is used for boot
+		 * services, we can get here without having allocated the
+		 * real mode trampoline.  It's too late to hand boot services
+		 * memory back to the memblock allocator, so instead
+		 * try to manually allocate the trampoline if needed.
+		 *
+		 * I've seen this on a Dell XPS 13 9350 with firmware
+		 * 1.4.4 with SGX enabled booting Linux via Fedora 24's
+		 * grub2-efi on a hard disk.  (And no, I don't know why
+		 * this happened, but Linux should still try to boot rather
+		 * panicking early.)
+		 */
+		rm_size = real_mode_size_needed();
+		if (rm_size && (start + rm_size) < (1<<20) && size >= rm_size) {
+			set_real_mode_mem(start);
+			start += rm_size;
+			size -= rm_size;
+		}
+
+		/*
+		 * Don't free memory under 1M for two reasons:
+		 * - BIOS might clobber it
+		 * - Crash kernel needs it to be reserved
+		 */
+		if (start + size < SZ_1M)
+			continue;
+		if (start < SZ_1M) {
+			size -= (SZ_1M - start);
+			start = SZ_1M;
+		}
+
+		/*
+		 * With CONFIG_DEFERRED_STRUCT_PAGE_INIT parts of the memory
+		 * map are still not initialized and we can't reliably free
+		 * memory here.
+		 * Queue the ranges to free at a later point.
+		 */
+		ranges_to_free[idx].start = start;
+		ranges_to_free[idx].end = start + size;
+		idx++;
+	}
+
+	if (!num_entries)
+		return;
+
+	if (efi_memmap_alloc(num_entries, &data) != 0) {
+		pr_err("Failed to allocate new EFI memmap\n");
+		return;
+	}
+
+	new = memremap(data.phys_map, data.size, MEMREMAP_WB);
+	if (!new) {
+		pr_err("Failed to map new EFI memmap\n");
+		return;
+	}
+
+	/*
+	 * Build a new EFI memmap that excludes any boot services
+	 * regions that are not tagged EFI_MEMORY_RUNTIME, since those
+	 * regions have now been freed.
+	 */
+	new_md = new;
+	for_each_efi_memory_desc(md) {
+		if (!(md->attribute & EFI_MEMORY_RUNTIME) &&
+		    (md->type == EFI_BOOT_SERVICES_CODE ||
+		     md->type == EFI_BOOT_SERVICES_DATA))
+			continue;
+
+		memcpy(new_md, md, efi.memmap.desc_size);
+		new_md += efi.memmap.desc_size;
+	}
+
+	memunmap(new);
+
+	if (efi_memmap_install(&data) != 0) {
+		pr_err("Could not install new EFI memmap\n");
+		return;
+	}
+}
+
+static int __init efi_free_boot_services(void)
+{
+	struct efi_freeable_range *range = ranges_to_free;
+	unsigned long freed = 0;
+
+	if (!ranges_to_free)
+		return 0;
+
+	while (range->start) {
+		void *start = phys_to_virt(range->start);
+		void *end = phys_to_virt(range->end);
+
+		free_reserved_area(start, end, -1, NULL);
+		freed += (end - start);
+		range++;
+	}
+	kfree(ranges_to_free);
+
+	if (freed)
+		pr_info("Freeing EFI boot services memory: %ldK\n", freed / SZ_1K);
+
+	return 0;
+}
+arch_initcall(efi_free_boot_services);
+
+/*
+ * A number of config table entries get remapped to virtual addresses
+ * after entering EFI virtual mode. However, the kexec kernel requires
+ * their physical addresses therefore we pass them via setup_data and
+ * correct those entries to their respective physical addresses here.
+ *
+ * Currently only handles smbios which is necessary for some firmware
+ * implementation.
+ */
+int __init efi_reuse_config(u64 tables, int nr_tables)
+{
+	int i, sz, ret = 0;
+	void *p, *tablep;
+	struct efi_setup_data *data;
+
+	if (nr_tables == 0)
+		return 0;
+
+	if (!efi_setup)
+		return 0;
+
+	if (!efi_enabled(EFI_64BIT))
+		return 0;
+
+	data = early_memremap(efi_setup, sizeof(*data));
+	if (!data) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	if (!data->smbios)
+		goto out_memremap;
+
+	sz = sizeof(efi_config_table_64_t);
+
+	p = tablep = early_memremap(tables, nr_tables * sz);
+	if (!p) {
+		pr_err("Could not map Configuration table!\n");
+		ret = -ENOMEM;
+		goto out_memremap;
+	}
+
+	for (i = 0; i < nr_tables; i++) {
+		efi_guid_t guid;
+
+		guid = ((efi_config_table_64_t *)p)->guid;
+
+		if (!efi_guidcmp(guid, SMBIOS_TABLE_GUID))
+			((efi_config_table_64_t *)p)->table = data->smbios;
+
+		/* Do not bother to play with mem attr table across kexec */
+		if (!efi_guidcmp(guid, EFI_MEMORY_ATTRIBUTES_TABLE_GUID))
+			((efi_config_table_64_t *)p)->table = EFI_INVALID_TABLE_ADDR;
+
+		p += sz;
+	}
+	early_memunmap(tablep, nr_tables * sz);
+
+out_memremap:
+	early_memunmap(data, sizeof(*data));
+out:
+	return ret;
+}
+
+void __init efi_apply_memmap_quirks(void)
+{
+	/*
+	 * Once setup is done earlier, unmap the EFI memory map on mismatched
+	 * firmware/kernel architectures since there is no support for runtime
+	 * services.
+	 */
+	if (!efi_runtime_supported()) {
+		pr_info("Setup done, disabling due to 32/64-bit mismatch\n");
+		efi_memmap_unmap();
+	}
+}
+
+/*
+ * For most modern platforms the preferred method of powering off is via
+ * ACPI. However, there are some that are known to require the use of
+ * EFI runtime services and for which ACPI does not work at all.
+ *
+ * Using EFI is a last resort, to be used only if no other option
+ * exists.
+ */
+bool efi_reboot_required(void)
+{
+	if (!acpi_gbl_reduced_hardware)
+		return false;
+
+	efi_reboot_quirk_mode = EFI_RESET_WARM;
+	return true;
+}
+
+bool efi_poweroff_required(void)
+{
+	return acpi_gbl_reduced_hardware || acpi_no_s5;
+}
+
+#ifdef CONFIG_EFI_CAPSULE_QUIRK_QUARK_CSH
+
+static int qrk_capsule_setup_info(struct capsule_info *cap_info, void **pkbuff,
+				  size_t hdr_bytes)
+{
+	struct quark_security_header *csh = *pkbuff;
+
+	/* Only process data block that is larger than the security header */
+	if (hdr_bytes < sizeof(struct quark_security_header))
+		return 0;
+
+	if (csh->csh_signature != QUARK_CSH_SIGNATURE ||
+	    csh->headersize != QUARK_SECURITY_HEADER_SIZE)
+		return 1;
+
+	/* Only process data block if EFI header is included */
+	if (hdr_bytes < QUARK_SECURITY_HEADER_SIZE +
+			sizeof(efi_capsule_header_t))
+		return 0;
+
+	pr_debug("Quark security header detected\n");
+
+	if (csh->rsvd_next_header != 0) {
+		pr_err("multiple Quark security headers not supported\n");
+		return -EINVAL;
+	}
+
+	*pkbuff += csh->headersize;
+	cap_info->total_size = csh->headersize;
+
+	/*
+	 * Update the first page pointer to skip over the CSH header.
+	 */
+	cap_info->phys[0] += csh->headersize;
+
+	/*
+	 * cap_info->capsule should point at a virtual mapping of the entire
+	 * capsule, starting at the capsule header. Our image has the Quark
+	 * security header prepended, so we cannot rely on the default vmap()
+	 * mapping created by the generic capsule code.
+	 * Given that the Quark firmware does not appear to care about the
+	 * virtual mapping, let's just point cap_info->capsule at our copy
+	 * of the capsule header.
+	 */
+	cap_info->capsule = &cap_info->header;
+
+	return 1;
+}
+
+static const struct x86_cpu_id efi_capsule_quirk_ids[] = {
+	X86_MATCH_VFM(INTEL_QUARK_X1000, &qrk_capsule_setup_info),
+	{ }
+};
+
+int efi_capsule_setup_info(struct capsule_info *cap_info, void *kbuff,
+			   size_t hdr_bytes)
+{
+	int (*quirk_handler)(struct capsule_info *, void **, size_t);
+	const struct x86_cpu_id *id;
+	int ret;
+
+	if (hdr_bytes < sizeof(efi_capsule_header_t))
+		return 0;
+
+	cap_info->total_size = 0;
+
+	id = x86_match_cpu(efi_capsule_quirk_ids);
+	if (id) {
+		/*
+		 * The quirk handler is supposed to return
+		 *  - a value > 0 if the setup should continue, after advancing
+		 *    kbuff as needed
+		 *  - 0 if not enough hdr_bytes are available yet
+		 *  - a negative error code otherwise
+		 */
+		quirk_handler = (typeof(quirk_handler))id->driver_data;
+		ret = quirk_handler(cap_info, &kbuff, hdr_bytes);
+		if (ret <= 0)
+			return ret;
+	}
+
+	memcpy(&cap_info->header, kbuff, sizeof(cap_info->header));
+
+	cap_info->total_size += cap_info->header.imagesize;
+
+	return __efi_capsule_setup_info(cap_info);
+}
+
+#endif
+
+/*
+ * If any access by any efi runtime service causes a page fault, then,
+ * 1. If it's efi_reset_system(), reboot through BIOS.
+ * 2. If any other efi runtime service, then
+ *    a. Return error status to the efi caller process.
+ *    b. Disable EFI Runtime Services forever and
+ *    c. Freeze efi_rts_wq and schedule new process.
+ *
+ * @return: Returns, if the page fault is not handled. This function
+ * will never return if the page fault is handled successfully.
+ */
+void efi_crash_gracefully_on_page_fault(unsigned long phys_addr,
+					const struct pt_regs *regs)
+{
+	if (!IS_ENABLED(CONFIG_X86_64))
+		return;
+
+	/*
+	 * If we get an interrupt/NMI while processing an EFI runtime service
+	 * then this is a regular OOPS, not an EFI failure.
+	 */
+	if (!in_task())
+		return;
+
+	/*
+	 * Make sure that an efi runtime service caused the page fault.
+	 * READ_ONCE() because we might be OOPSing in a different thread,
+	 * and we don't want to trip KTSAN while trying to OOPS.
+	 */
+	if (READ_ONCE(efi_rts_work.efi_rts_id) == EFI_NONE ||
+	    current_work() != &efi_rts_work.work)
+		return;
+
+	/*
+	 * Address range 0x0000 - 0x0fff is always mapped in the efi_pgd, so
+	 * page faulting on these addresses isn't expected.
+	 */
+	if (phys_addr <= 0x0fff)
+		return;
+
+	/*
+	 * Print stack trace as it might be useful to know which EFI Runtime
+	 * Service is buggy.
+	 */
+	WARN(1, FW_BUG "Page fault caused by firmware at PA: 0x%lx\n",
+	     phys_addr);
+
+	/*
+	 * Buggy efi_reset_system() is handled differently from other EFI
+	 * Runtime Services as it doesn't use efi_rts_wq. Although,
+	 * native_machine_emergency_restart() says that machine_real_restart()
+	 * could fail, it's better not to complicate this fault handler
+	 * because this case occurs *very* rarely and hence could be improved
+	 * on a need by basis.
+	 */
+	if (efi_rts_work.efi_rts_id == EFI_RESET_SYSTEM) {
+		pr_info("efi_reset_system() buggy! Reboot through BIOS\n");
+		machine_real_restart(MRR_BIOS);
+		return;
+	}
+
+	/*
+	 * The API does not permit entering a kernel mode FPU section with
+	 * interrupts enabled and leaving it with interrupts disabled.  So
+	 * re-enable interrupts now if they were enabled when the page fault
+	 * occurred.
+	 */
+	local_irq_restore(regs->flags);
+
+	/*
+	 * Before calling EFI Runtime Service, the kernel has switched the
+	 * calling process to efi_mm. Hence, switch back to task_mm.
+	 */
+	arch_efi_call_virt_teardown();
+
+	/* Signal error status to the efi caller process */
+	efi_rts_work.status = EFI_ABORTED;
+	complete(&efi_rts_work.efi_rts_comp);
+
+	clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+	pr_info("Froze efi_rts_wq and disabled EFI Runtime Services\n");
+
+	/*
+	 * Call schedule() in an infinite loop, so that any spurious wake ups
+	 * will never run efi_rts_wq again.
+	 */
+	for (;;) {
+		set_current_state(TASK_IDLE);
+		schedule();
+	}
+}

linux-7.2.3/arch/x86/platform/efi/runtime-map.c

@@ -0,0 +1,194 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2013 Red Hat, Inc., Dave Young <dyoung@redhat.com>
+ */
+
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/efi.h>
+#include <linux/slab.h>
+
+#include <asm/efi.h>
+#include <asm/setup.h>
+
+struct efi_runtime_map_entry {
+	efi_memory_desc_t md;
+	struct kobject kobj;   /* kobject for each entry */
+};
+
+static struct efi_runtime_map_entry **map_entries;
+
+struct map_attribute {
+	struct attribute attr;
+	ssize_t (*show)(struct efi_runtime_map_entry *entry, char *buf);
+};
+
+static inline struct map_attribute *to_map_attr(struct attribute *attr)
+{
+	return container_of(attr, struct map_attribute, attr);
+}
+
+static ssize_t type_show(struct efi_runtime_map_entry *entry, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "0x%x\n", entry->md.type);
+}
+
+#define EFI_RUNTIME_FIELD(var) entry->md.var
+
+#define EFI_RUNTIME_U64_ATTR_SHOW(name) \
+static ssize_t name##_show(struct efi_runtime_map_entry *entry, char *buf) \
+{ \
+	return snprintf(buf, PAGE_SIZE, "0x%llx\n", EFI_RUNTIME_FIELD(name)); \
+}
+
+EFI_RUNTIME_U64_ATTR_SHOW(phys_addr);
+EFI_RUNTIME_U64_ATTR_SHOW(virt_addr);
+EFI_RUNTIME_U64_ATTR_SHOW(num_pages);
+EFI_RUNTIME_U64_ATTR_SHOW(attribute);
+
+static inline struct efi_runtime_map_entry *to_map_entry(struct kobject *kobj)
+{
+	return container_of(kobj, struct efi_runtime_map_entry, kobj);
+}
+
+static ssize_t map_attr_show(struct kobject *kobj, struct attribute *attr,
+			      char *buf)
+{
+	struct efi_runtime_map_entry *entry = to_map_entry(kobj);
+	struct map_attribute *map_attr = to_map_attr(attr);
+
+	return map_attr->show(entry, buf);
+}
+
+static struct map_attribute map_type_attr = __ATTR_RO_MODE(type, 0400);
+static struct map_attribute map_phys_addr_attr = __ATTR_RO_MODE(phys_addr, 0400);
+static struct map_attribute map_virt_addr_attr = __ATTR_RO_MODE(virt_addr, 0400);
+static struct map_attribute map_num_pages_attr = __ATTR_RO_MODE(num_pages, 0400);
+static struct map_attribute map_attribute_attr = __ATTR_RO_MODE(attribute, 0400);
+
+/*
+ * These are default attributes that are added for every memmap entry.
+ */
+static struct attribute *def_attrs[] = {
+	&map_type_attr.attr,
+	&map_phys_addr_attr.attr,
+	&map_virt_addr_attr.attr,
+	&map_num_pages_attr.attr,
+	&map_attribute_attr.attr,
+	NULL
+};
+ATTRIBUTE_GROUPS(def);
+
+static const struct sysfs_ops map_attr_ops = {
+	.show = map_attr_show,
+};
+
+static void map_release(struct kobject *kobj)
+{
+	struct efi_runtime_map_entry *entry;
+
+	entry = to_map_entry(kobj);
+	kfree(entry);
+}
+
+static const struct kobj_type __refconst map_ktype = {
+	.sysfs_ops	= &map_attr_ops,
+	.default_groups	= def_groups,
+	.release	= map_release,
+};
+
+static struct kset *map_kset;
+
+static struct efi_runtime_map_entry *
+add_sysfs_runtime_map_entry(struct kobject *kobj, int nr,
+			    efi_memory_desc_t *md)
+{
+	int ret;
+	struct efi_runtime_map_entry *entry;
+
+	if (!map_kset) {
+		map_kset = kset_create_and_add("runtime-map", NULL, kobj);
+		if (!map_kset)
+			return ERR_PTR(-ENOMEM);
+	}
+
+	entry = kzalloc_obj(*entry);
+	if (!entry) {
+		kset_unregister(map_kset);
+		map_kset = NULL;
+		return ERR_PTR(-ENOMEM);
+	}
+
+	memcpy(&entry->md, md, sizeof(efi_memory_desc_t));
+
+	kobject_init(&entry->kobj, &map_ktype);
+	entry->kobj.kset = map_kset;
+	ret = kobject_add(&entry->kobj, NULL, "%d", nr);
+	if (ret) {
+		kobject_put(&entry->kobj);
+		kset_unregister(map_kset);
+		map_kset = NULL;
+		return ERR_PTR(ret);
+	}
+
+	return entry;
+}
+
+int efi_get_runtime_map_size(void)
+{
+	return efi.memmap.nr_map * efi.memmap.desc_size;
+}
+
+int efi_get_runtime_map_desc_size(void)
+{
+	return efi.memmap.desc_size;
+}
+
+int efi_runtime_map_copy(void *buf, size_t bufsz)
+{
+	size_t sz = efi_get_runtime_map_size();
+
+	if (sz > bufsz)
+		sz = bufsz;
+
+	memcpy(buf, efi.memmap.map, sz);
+	return 0;
+}
+
+static int __init efi_runtime_map_init(void)
+{
+	int i, j, ret = 0;
+	struct efi_runtime_map_entry *entry;
+	efi_memory_desc_t *md;
+
+	if (!efi_enabled(EFI_MEMMAP) || !efi_kobj)
+		return 0;
+
+	map_entries = kzalloc_objs(entry, efi.memmap.nr_map);
+	if (!map_entries) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	i = 0;
+	for_each_efi_memory_desc(md) {
+		entry = add_sysfs_runtime_map_entry(efi_kobj, i, md);
+		if (IS_ERR(entry)) {
+			ret = PTR_ERR(entry);
+			goto out_add_entry;
+		}
+		*(map_entries + i++) = entry;
+	}
+
+	return 0;
+out_add_entry:
+	for (j = i - 1; j >= 0; j--) {
+		entry = *(map_entries + j);
+		kobject_put(&entry->kobj);
+	}
+out:
+	return ret;
+}
+subsys_initcall_sync(efi_runtime_map_init);

linux-7.2.3/arch/x86/platform/geode/Makefile

@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_GEODE_COMMON)	+= geode-common.o
+obj-$(CONFIG_ALIX)		+= alix.o
+obj-$(CONFIG_NET5501)		+= net5501.o
+obj-$(CONFIG_GEOS)		+= geos.o

linux-7.2.3/arch/x86/platform/geode/alix.c

@@ -0,0 +1,136 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * System Specific setup for PCEngines ALIX.
+ * At the moment this means setup of GPIO control of LEDs
+ * on Alix.2/3/6 boards.
+ *
+ * Copyright (C) 2008 Constantin Baranov <const@mimas.ru>
+ * Copyright (C) 2011 Ed Wildgoose <kernel@wildgooses.com>
+ *                and Philip Prindeville <philipp@redfish-solutions.com>
+ *
+ * TODO: There are large similarities with leds-net5501.c
+ * by Alessandro Zummo <a.zummo@towertech.it>
+ * In the future leds-net5501.c should be migrated over to platform
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/moduleparam.h>
+#include <linux/dmi.h>
+
+#include <asm/geode.h>
+
+#include "geode-common.h"
+
+#define BIOS_SIGNATURE_TINYBIOS		0xf0000
+#define BIOS_SIGNATURE_COREBOOT		0x500
+#define BIOS_REGION_SIZE		0x10000
+
+/*
+ * This driver is not modular, but to keep back compatibility
+ * with existing use cases, continuing with module_param is
+ * the easiest way forward.
+ */
+static bool force = 0;
+module_param(force, bool, 0444);
+/* FIXME: Award bios is not automatically detected as Alix platform */
+MODULE_PARM_DESC(force, "Force detection as ALIX.2/ALIX.3 platform");
+
+static const struct geode_led alix_leds[] __initconst = {
+	{ 6, true },
+	{ 25, false },
+	{ 27, false },
+};
+
+static void __init register_alix(void)
+{
+	geode_create_restart_key(24);
+	geode_create_leds("alix", alix_leds, ARRAY_SIZE(alix_leds));
+}
+
+static bool __init alix_present(unsigned long bios_phys,
+				const char *alix_sig,
+				size_t alix_sig_len)
+{
+	const size_t bios_len = BIOS_REGION_SIZE;
+	const char *bios_virt;
+	const char *scan_end;
+	const char *p;
+	char name[64];
+
+	if (force) {
+		printk(KERN_NOTICE "%s: forced to skip BIOS test, "
+		       "assume system is ALIX.2/ALIX.3\n",
+		       KBUILD_MODNAME);
+		return true;
+	}
+
+	bios_virt = phys_to_virt(bios_phys);
+	scan_end = bios_virt + bios_len - (alix_sig_len + 2);
+	for (p = bios_virt; p < scan_end; p++) {
+		const char *tail;
+		char *a;
+
+		if (memcmp(p, alix_sig, alix_sig_len) != 0)
+			continue;
+
+		memcpy(name, p, sizeof(name));
+
+		/* remove the first \0 character from string */
+		a = strchr(name, '\0');
+		if (a)
+			*a = ' ';
+
+		/* cut the string at a newline */
+		a = strchr(name, '\r');
+		if (a)
+			*a = '\0';
+
+		tail = p + alix_sig_len;
+		if ((tail[0] == '2' || tail[0] == '3' || tail[0] == '6')) {
+			printk(KERN_INFO
+			       "%s: system is recognized as \"%s\"\n",
+			       KBUILD_MODNAME, name);
+			return true;
+		}
+	}
+
+	return false;
+}
+
+static bool __init alix_present_dmi(void)
+{
+	const char *vendor, *product;
+
+	vendor = dmi_get_system_info(DMI_SYS_VENDOR);
+	if (!vendor || strcmp(vendor, "PC Engines"))
+		return false;
+
+	product = dmi_get_system_info(DMI_PRODUCT_NAME);
+	if (!product || (strcmp(product, "ALIX.2D") && strcmp(product, "ALIX.6")))
+		return false;
+
+	printk(KERN_INFO "%s: system is recognized as \"%s %s\"\n",
+	       KBUILD_MODNAME, vendor, product);
+
+	return true;
+}
+
+static int __init alix_init(void)
+{
+	const char tinybios_sig[] = "PC Engines ALIX.";
+	const char coreboot_sig[] = "PC Engines\0ALIX.";
+
+	if (!is_geode())
+		return 0;
+
+	if (alix_present(BIOS_SIGNATURE_TINYBIOS, tinybios_sig, sizeof(tinybios_sig) - 1) ||
+	    alix_present(BIOS_SIGNATURE_COREBOOT, coreboot_sig, sizeof(coreboot_sig) - 1) ||
+	    alix_present_dmi())
+		register_alix();
+
+	return 0;
+}
+device_initcall(alix_init);

linux-7.2.3/arch/x86/platform/geode/geode-common.c

@@ -0,0 +1,188 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Shared helpers to register GPIO-connected buttons and LEDs
+ * on AMD Geode boards.
+ */
+
+#include <linux/err.h>
+#include <linux/gpio/machine.h>
+#include <linux/gpio/property.h>
+#include <linux/input.h>
+#include <linux/leds.h>
+#include <linux/mfd/cs5535.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include "geode-common.h"
+
+static const struct property_entry geode_gpio_keys_props[] = {
+	PROPERTY_ENTRY_U32("poll-interval", 20),
+	{ }
+};
+
+static const struct software_node geode_gpio_keys_node = {
+	.name = "geode-gpio-keys",
+	.properties = geode_gpio_keys_props,
+};
+
+static struct software_node_ref_args geode_restart_gpio_ref;
+
+static const struct property_entry geode_restart_key_props[] = {
+	PROPERTY_ENTRY_REF_ARRAY_LEN("gpios", &geode_restart_gpio_ref, 1),
+	PROPERTY_ENTRY_U32("linux,code", KEY_RESTART),
+	PROPERTY_ENTRY_STRING("label", "Reset button"),
+	PROPERTY_ENTRY_U32("debounce-interval", 100),
+	{ }
+};
+
+static const struct software_node geode_restart_key_node = {
+	.parent = &geode_gpio_keys_node,
+	.properties = geode_restart_key_props,
+};
+
+static const struct software_node *geode_gpio_keys_swnodes[] __initconst = {
+	&geode_gpio_keys_node,
+	&geode_restart_key_node,
+	NULL
+};
+
+/*
+ * Creates gpio-keys-polled device for the restart key.
+ *
+ * Note that it needs to be called first, before geode_create_leds(),
+ * because it registers gpiochip software node used by both gpio-keys and
+ * leds-gpio devices.
+ */
+int __init geode_create_restart_key(unsigned int pin)
+{
+	struct platform_device_info keys_info = {
+		.name	= "gpio-keys-polled",
+		.id	= 1,
+	};
+	struct platform_device *pd;
+	int err;
+
+	geode_restart_gpio_ref = SOFTWARE_NODE_REFERENCE(&cs5535_gpio_swnode,
+							 pin, GPIO_ACTIVE_LOW);
+
+	err = software_node_register_node_group(geode_gpio_keys_swnodes);
+	if (err) {
+		pr_err("failed to register gpio-keys software nodes: %d\n", err);
+		return err;
+	}
+
+	keys_info.fwnode = software_node_fwnode(&geode_gpio_keys_node);
+
+	pd = platform_device_register_full(&keys_info);
+	err = PTR_ERR_OR_ZERO(pd);
+	if (err) {
+		pr_err("failed to create gpio-keys device: %d\n", err);
+		software_node_unregister_node_group(geode_gpio_keys_swnodes);
+		return err;
+	}
+
+	return 0;
+}
+
+static const struct software_node geode_gpio_leds_node = {
+	.name = "geode-leds",
+};
+
+#define MAX_LEDS	3
+
+int __init geode_create_leds(const char *label, const struct geode_led *leds,
+			      unsigned int n_leds)
+{
+	const struct software_node *group[MAX_LEDS + 2] = { 0 };
+	struct software_node *swnodes;
+	struct property_entry *props;
+	struct software_node_ref_args *gpio_refs;
+	struct platform_device_info led_info = {
+		.name	= "leds-gpio",
+		.id	= PLATFORM_DEVID_NONE,
+	};
+	struct platform_device *led_dev;
+	const char *node_name;
+	int err;
+	int i;
+
+	if (n_leds > MAX_LEDS) {
+		pr_err("%s: too many LEDs\n", __func__);
+		return -EINVAL;
+	}
+
+	swnodes = kzalloc_objs(*swnodes, n_leds);
+	if (!swnodes)
+		return -ENOMEM;
+
+	/*
+	 * Each LED is represented by 3 properties: "gpios",
+	 * "linux,default-trigger", and am empty terminator.
+	 */
+	props = kzalloc_objs(*props, n_leds * 3);
+	if (!props) {
+		err = -ENOMEM;
+		goto err_free_swnodes;
+	}
+
+	gpio_refs = kzalloc_objs(*gpio_refs, n_leds);
+	if (!gpio_refs) {
+		err = -ENOMEM;
+		goto err_free_props;
+	}
+
+	group[0] = &geode_gpio_leds_node;
+	for (i = 0; i < n_leds; i++) {
+		node_name = kasprintf(GFP_KERNEL, "%s:%d", label, i);
+		if (!node_name) {
+			err = -ENOMEM;
+			goto err_free_names;
+		}
+
+		gpio_refs[i] = SOFTWARE_NODE_REFERENCE(&cs5535_gpio_swnode,
+						       leds[i].pin,
+						       GPIO_ACTIVE_LOW);
+		props[i * 3 + 0] =
+			PROPERTY_ENTRY_REF_ARRAY_LEN("gpios", &gpio_refs[i], 1);
+		props[i * 3 + 1] =
+			PROPERTY_ENTRY_STRING("linux,default-trigger",
+					      leds[i].default_on ?
+					      "default-on" : "default-off");
+		/* props[i * 3 + 2] is an empty terminator */
+
+		swnodes[i] = SOFTWARE_NODE(node_name, &props[i * 3],
+					   &geode_gpio_leds_node);
+		group[i + 1] = &swnodes[i];
+	}
+
+	err = software_node_register_node_group(group);
+	if (err) {
+		pr_err("failed to register LED software nodes: %d\n", err);
+		goto err_free_names;
+	}
+
+	led_info.fwnode = software_node_fwnode(&geode_gpio_leds_node);
+
+	led_dev = platform_device_register_full(&led_info);
+	err = PTR_ERR_OR_ZERO(led_dev);
+	if (err) {
+		pr_err("failed to create LED device: %d\n", err);
+		goto err_unregister_group;
+	}
+
+	return 0;
+
+err_unregister_group:
+	software_node_unregister_node_group(group);
+err_free_names:
+	while (--i >= 0)
+		kfree(swnodes[i].name);
+	kfree(gpio_refs);
+err_free_props:
+	kfree(props);
+err_free_swnodes:
+	kfree(swnodes);
+	return err;
+}
+
+MODULE_IMPORT_NS("CS5535");

linux-7.2.3/arch/x86/platform/geode/geode-common.h

@@ -0,0 +1,21 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Shared helpers to register GPIO-connected buttons and LEDs
+ * on AMD Geode boards.
+ */
+
+#ifndef __PLATFORM_GEODE_COMMON_H
+#define __PLATFORM_GEODE_COMMON_H
+
+#include <linux/property.h>
+
+struct geode_led {
+	unsigned int pin;
+	bool default_on;
+};
+
+int geode_create_restart_key(unsigned int pin);
+int geode_create_leds(const char *label, const struct geode_led *leds,
+		      unsigned int n_leds);
+
+#endif /* __PLATFORM_GEODE_COMMON_H */

linux-7.2.3/arch/x86/platform/geode/geos.c

@@ -0,0 +1,59 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * System Specific setup for Traverse Technologies GEOS.
+ * At the moment this means setup of GPIO control of LEDs.
+ *
+ * Copyright (C) 2008 Constantin Baranov <const@mimas.ru>
+ * Copyright (C) 2011 Ed Wildgoose <kernel@wildgooses.com>
+ *                and Philip Prindeville <philipp@redfish-solutions.com>
+ *
+ * TODO: There are large similarities with leds-net5501.c
+ * by Alessandro Zummo <a.zummo@towertech.it>
+ * In the future leds-net5501.c should be migrated over to platform
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/dmi.h>
+
+#include <asm/geode.h>
+
+#include "geode-common.h"
+
+static const struct geode_led geos_leds[] __initconst = {
+	{ 6, true },
+	{ 25, false },
+	{ 27, false },
+};
+
+static void __init register_geos(void)
+{
+	geode_create_restart_key(3);
+	geode_create_leds("geos", geos_leds, ARRAY_SIZE(geos_leds));
+}
+
+static int __init geos_init(void)
+{
+	const char *vendor, *product;
+
+	if (!is_geode())
+		return 0;
+
+	vendor = dmi_get_system_info(DMI_SYS_VENDOR);
+	if (!vendor || strcmp(vendor, "Traverse Technologies"))
+		return 0;
+
+	product = dmi_get_system_info(DMI_PRODUCT_NAME);
+	if (!product || strcmp(product, "Geos"))
+		return 0;
+
+	printk(KERN_INFO "%s: system is recognized as \"%s %s\"\n",
+	       KBUILD_MODNAME, vendor, product);
+
+	register_geos();
+
+	return 0;
+}
+device_initcall(geos_init);

linux-7.2.3/arch/x86/platform/geode/net5501.c

@@ -0,0 +1,97 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * System Specific setup for Soekris net5501
+ * At the moment this means setup of GPIO control of LEDs and buttons
+ * on net5501 boards.
+ *
+ * Copyright (C) 2008-2009 Tower Technologies
+ * Written by Alessandro Zummo <a.zummo@towertech.it>
+ *
+ * Copyright (C) 2008 Constantin Baranov <const@mimas.ru>
+ * Copyright (C) 2011 Ed Wildgoose <kernel@wildgooses.com>
+ *                and Philip Prindeville <philipp@redfish-solutions.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/input.h>
+#include <linux/gpio/machine.h>
+#include <linux/gpio/property.h>
+
+#include <asm/geode.h>
+
+#include "geode-common.h"
+
+#define BIOS_REGION_BASE		0xffff0000
+#define BIOS_REGION_SIZE		0x00010000
+
+static const struct geode_led net5501_leds[] __initconst = {
+	{ 6, true },
+};
+
+static void __init register_net5501(void)
+{
+	geode_create_restart_key(24);
+	geode_create_leds("net5501", net5501_leds, ARRAY_SIZE(net5501_leds));
+}
+
+struct net5501_board {
+	u16	offset;
+	u16	len;
+	char	*sig;
+};
+
+static struct net5501_board __initdata boards[] = {
+	{ 0xb7b, 7, "net5501" },	/* net5501 v1.33/1.33c */
+	{ 0xb1f, 7, "net5501" },	/* net5501 v1.32i */
+};
+
+static bool __init net5501_present(void)
+{
+	int i;
+	unsigned char *rombase, *bios;
+	bool found = false;
+
+	rombase = ioremap(BIOS_REGION_BASE, BIOS_REGION_SIZE - 1);
+	if (!rombase) {
+		printk(KERN_ERR "%s: failed to get rombase\n", KBUILD_MODNAME);
+		return found;
+	}
+
+	bios = rombase + 0x20;	/* null terminated */
+
+	if (memcmp(bios, "comBIOS", 7))
+		goto unmap;
+
+	for (i = 0; i < ARRAY_SIZE(boards); i++) {
+		unsigned char *model = rombase + boards[i].offset;
+
+		if (!memcmp(model, boards[i].sig, boards[i].len)) {
+			printk(KERN_INFO "%s: system is recognized as \"%s\"\n",
+			       KBUILD_MODNAME, model);
+
+			found = true;
+			break;
+		}
+	}
+
+unmap:
+	iounmap(rombase);
+	return found;
+}
+
+static int __init net5501_init(void)
+{
+	if (!is_geode())
+		return 0;
+
+	if (!net5501_present())
+		return 0;
+
+	register_net5501();
+
+	return 0;
+}
+device_initcall(net5501_init);

linux-7.2.3/arch/x86/platform/intel-mid/Makefile

@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_X86_INTEL_MID) += intel-mid.o pwr.o

linux-7.2.3/arch/x86/platform/intel-mid/intel-mid.c

@@ -0,0 +1,126 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Intel MID platform setup code
+ *
+ * (C) Copyright 2008, 2012, 2021 Intel Corporation
+ * Author: Jacob Pan (jacob.jun.pan@intel.com)
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ */
+
+#define pr_fmt(fmt) "intel_mid: " fmt
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/regulator/machine.h>
+#include <linux/scatterlist.h>
+#include <linux/irq.h>
+#include <linux/export.h>
+#include <linux/notifier.h>
+
+#include <asm/setup.h>
+#include <asm/mpspec_def.h>
+#include <asm/hw_irq.h>
+#include <asm/apic.h>
+#include <asm/cpu_device_id.h>
+#include <asm/io_apic.h>
+#include <asm/intel-mid.h>
+#include <asm/io.h>
+#include <asm/i8259.h>
+#include <asm/reboot.h>
+
+#include <linux/platform_data/x86/intel_scu_ipc.h>
+
+#define IPCMSG_COLD_OFF		0x80	/* Only for Tangier */
+#define IPCMSG_COLD_RESET	0xF1
+
+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_dev_simple_command(NULL, IPCMSG_COLD_OFF, 1);
+};
+
+static void intel_mid_reboot(void)
+{
+	intel_scu_ipc_dev_simple_command(NULL, IPCMSG_COLD_RESET, 0);
+}
+
+static void __init intel_mid_time_init(void)
+{
+	/* Lapic only, no apbt */
+	x86_init.timers.setup_percpu_clockev = setup_boot_APIC_clock;
+	x86_cpuinit.setup_percpu_clockev = setup_secondary_APIC_clock;
+}
+
+static void intel_mid_arch_setup(void)
+{
+	switch (boot_cpu_data.x86_vfm) {
+	case INTEL_ATOM_SILVERMONT_MID:
+		x86_platform.legacy.rtc = 1;
+		break;
+	default:
+		break;
+	}
+
+	/*
+	 * Intel MID platforms are using explicitly defined regulators.
+	 *
+	 * Let the regulator core know that we do not have any additional
+	 * regulators left. This lets it substitute unprovided regulators with
+	 * dummy ones:
+	 */
+	regulator_has_full_constraints();
+}
+
+/*
+ * Moorestown does not have external NMI source nor port 0x61 to report
+ * NMI status. The possible NMI sources are from pmu as a result of NMI
+ * watchdog or lock debug. Reading io port 0x61 results in 0xff which
+ * misled NMI handler.
+ */
+static unsigned char intel_mid_get_nmi_reason(void)
+{
+	return 0;
+}
+
+/*
+ * Moorestown specific x86_init function overrides and early setup
+ * calls.
+ */
+void __init x86_intel_mid_early_setup(void)
+{
+	x86_init.resources.probe_roms = x86_init_noop;
+	x86_init.resources.reserve_resources = x86_init_noop;
+
+	x86_init.timers.timer_init = intel_mid_time_init;
+	x86_init.timers.setup_percpu_clockev = x86_init_noop;
+
+	x86_init.irqs.pre_vector_init = x86_init_noop;
+
+	x86_init.oem.arch_setup = intel_mid_arch_setup;
+
+	x86_platform.get_nmi_reason = intel_mid_get_nmi_reason;
+
+	x86_init.pci.arch_init = intel_mid_pci_init;
+	x86_init.pci.fixup_irqs = x86_init_noop;
+
+	legacy_pic = &null_legacy_pic;
+
+	/*
+	 * Do nothing for now as everything needed done in
+	 * x86_intel_mid_early_setup() below.
+	 */
+	x86_init.acpi.reduced_hw_early_init = x86_init_noop;
+
+	pm_power_off = intel_mid_power_off;
+	machine_ops.emergency_restart  = intel_mid_reboot;
+
+	/* Avoid searching for BIOS MP tables */
+	x86_init.mpparse.find_mptable		= x86_init_noop;
+	x86_init.mpparse.early_parse_smp_cfg	= x86_init_noop;
+	x86_init.mpparse.parse_smp_cfg		= x86_init_noop;
+	set_bit(MP_BUS_ISA, mp_bus_not_pci);
+}

linux-7.2.3/arch/x86/platform/intel-mid/pwr.c

@@ -0,0 +1,485 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Intel MID Power Management Unit (PWRMU) device driver
+ *
+ * Copyright (C) 2016, Intel Corporation
+ *
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * Intel MID Power Management Unit device driver handles the South Complex PCI
+ * devices such as GPDMA, SPI, I2C, PWM, and so on. By default PCI core
+ * modifies bits in PMCSR register in the PCI configuration space. This is not
+ * enough on some SoCs like Intel Tangier. In such case PCI core sets a new
+ * power state of the device in question through a PM hook registered in struct
+ * pci_platform_pm_ops (see drivers/pci/pci-mid.c).
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/mutex.h>
+#include <linux/pci.h>
+
+#include <asm/intel-mid.h>
+
+/* Registers */
+#define PM_STS			0x00
+#define PM_CMD			0x04
+#define PM_ICS			0x08
+#define PM_WKC(x)		(0x10 + (x) * 4)
+#define PM_WKS(x)		(0x18 + (x) * 4)
+#define PM_SSC(x)		(0x20 + (x) * 4)
+#define PM_SSS(x)		(0x30 + (x) * 4)
+
+/* Bits in PM_STS */
+#define PM_STS_BUSY		(1 << 8)
+
+/* Bits in PM_CMD */
+#define PM_CMD_CMD(x)		((x) << 0)
+#define PM_CMD_IOC		(1 << 8)
+#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
+
+/* Bits in PM_ICS */
+#define PM_ICS_INT_STATUS(x)	((x) & 0xff)
+#define PM_ICS_IE		(1 << 8)
+#define PM_ICS_IP		(1 << 9)
+#define PM_ICS_SW_INT_STS	(1 << 10)
+
+/* List of interrupts */
+#define INT_INVALID		0
+#define INT_CMD_COMPLETE	1
+#define INT_CMD_ERR		2
+#define INT_WAKE_EVENT		3
+#define INT_LSS_POWER_ERR	4
+#define INT_S0iX_MSG_ERR	5
+#define INT_NO_C6		6
+#define INT_TRIGGER_ERR		7
+#define INT_INACTIVITY		8
+
+/* South Complex devices */
+#define LSS_MAX_SHARED_DEVS	4
+#define LSS_MAX_DEVS		64
+
+#define LSS_WS_BITS		1	/* wake state width */
+#define LSS_PWS_BITS		2	/* power state width */
+
+/* Supported device IDs */
+#define PCI_DEVICE_ID_PENWELL	0x0828
+#define PCI_DEVICE_ID_TANGIER	0x11a1
+
+struct mid_pwr_dev {
+	struct pci_dev *pdev;
+	pci_power_t state;
+};
+
+struct mid_pwr {
+	struct device *dev;
+	void __iomem *regs;
+	int irq;
+	bool available;
+
+	struct mutex lock;
+	struct mid_pwr_dev lss[LSS_MAX_DEVS][LSS_MAX_SHARED_DEVS];
+};
+
+static struct mid_pwr *midpwr;
+
+static u32 mid_pwr_get_state(struct mid_pwr *pwr, int reg)
+{
+	return readl(pwr->regs + PM_SSS(reg));
+}
+
+static void mid_pwr_set_state(struct mid_pwr *pwr, int reg, u32 value)
+{
+	writel(value, pwr->regs + PM_SSC(reg));
+}
+
+static void mid_pwr_set_wake(struct mid_pwr *pwr, int reg, u32 value)
+{
+	writel(value, pwr->regs + PM_WKC(reg));
+}
+
+static void mid_pwr_interrupt_disable(struct mid_pwr *pwr)
+{
+	writel(~PM_ICS_IE, pwr->regs + PM_ICS);
+}
+
+static bool mid_pwr_is_busy(struct mid_pwr *pwr)
+{
+	return !!(readl(pwr->regs + PM_STS) & PM_STS_BUSY);
+}
+
+/* Wait 500ms that the latest PWRMU command finished */
+static int mid_pwr_wait(struct mid_pwr *pwr)
+{
+	unsigned int count = 500000;
+	bool busy;
+
+	do {
+		busy = mid_pwr_is_busy(pwr);
+		if (!busy)
+			return 0;
+		udelay(1);
+	} while (--count);
+
+	return -EBUSY;
+}
+
+static int mid_pwr_wait_for_cmd(struct mid_pwr *pwr, u8 cmd)
+{
+	writel(PM_CMD_CMD(cmd) | PM_CMD_CM_IMMEDIATE, pwr->regs + PM_CMD);
+	return mid_pwr_wait(pwr);
+}
+
+static int __update_power_state(struct mid_pwr *pwr, int reg, int bit, int new)
+{
+	int curstate;
+	u32 power;
+	int ret;
+
+	/* Check if the device is already in desired state */
+	power = mid_pwr_get_state(pwr, reg);
+	curstate = (power >> bit) & 3;
+	if (curstate == new)
+		return 0;
+
+	/* Update the power state */
+	mid_pwr_set_state(pwr, reg, (power & ~(3 << bit)) | (new << bit));
+
+	/* Send command to SCU */
+	ret = mid_pwr_wait_for_cmd(pwr, CMD_SET_CFG);
+	if (ret)
+		return ret;
+
+	/* Check if the device is already in desired state */
+	power = mid_pwr_get_state(pwr, reg);
+	curstate = (power >> bit) & 3;
+	if (curstate != new)
+		return -EAGAIN;
+
+	return 0;
+}
+
+static pci_power_t __find_weakest_power_state(struct mid_pwr_dev *lss,
+					      struct pci_dev *pdev,
+					      pci_power_t state)
+{
+	pci_power_t weakest = PCI_D3hot;
+	unsigned int j;
+
+	/* Find device in cache or first free cell */
+	for (j = 0; j < LSS_MAX_SHARED_DEVS; j++) {
+		if (lss[j].pdev == pdev || !lss[j].pdev)
+			break;
+	}
+
+	/* Store the desired state in cache */
+	if (j < LSS_MAX_SHARED_DEVS) {
+		lss[j].pdev = pdev;
+		lss[j].state = state;
+	} else {
+		dev_WARN(&pdev->dev, "No room for device in PWRMU LSS cache\n");
+		weakest = state;
+	}
+
+	/* Find the power state we may use */
+	for (j = 0; j < LSS_MAX_SHARED_DEVS; j++) {
+		if (lss[j].state < weakest)
+			weakest = lss[j].state;
+	}
+
+	return weakest;
+}
+
+static int __set_power_state(struct mid_pwr *pwr, struct pci_dev *pdev,
+			     pci_power_t state, int id, int reg, int bit)
+{
+	const char *name;
+	int ret;
+
+	state = __find_weakest_power_state(pwr->lss[id], pdev, state);
+	name = pci_power_name(state);
+
+	ret = __update_power_state(pwr, reg, bit, (__force int)state);
+	if (ret) {
+		dev_warn(&pdev->dev, "Can't set power state %s: %d\n", name, ret);
+		return ret;
+	}
+
+	dev_vdbg(&pdev->dev, "Set power state %s\n", name);
+	return 0;
+}
+
+static int mid_pwr_set_power_state(struct mid_pwr *pwr, struct pci_dev *pdev,
+				   pci_power_t state)
+{
+	int id, reg, bit;
+	int ret;
+
+	id = intel_mid_pwr_get_lss_id(pdev);
+	if (id < 0)
+		return id;
+
+	reg = (id * LSS_PWS_BITS) / 32;
+	bit = (id * LSS_PWS_BITS) % 32;
+
+	/* We support states between PCI_D0 and PCI_D3hot */
+	if (state < PCI_D0)
+		state = PCI_D0;
+	if (state > PCI_D3hot)
+		state = PCI_D3hot;
+
+	mutex_lock(&pwr->lock);
+	ret = __set_power_state(pwr, pdev, state, id, reg, bit);
+	mutex_unlock(&pwr->lock);
+	return ret;
+}
+
+int intel_mid_pci_set_power_state(struct pci_dev *pdev, pci_power_t state)
+{
+	struct mid_pwr *pwr = midpwr;
+	int ret = 0;
+
+	might_sleep();
+
+	if (pwr && pwr->available)
+		ret = mid_pwr_set_power_state(pwr, pdev, state);
+	dev_vdbg(&pdev->dev, "set_power_state() returns %d\n", ret);
+
+	return 0;
+}
+
+pci_power_t intel_mid_pci_get_power_state(struct pci_dev *pdev)
+{
+	struct mid_pwr *pwr = midpwr;
+	int id, reg, bit;
+	u32 power;
+
+	if (!pwr || !pwr->available)
+		return PCI_UNKNOWN;
+
+	id = intel_mid_pwr_get_lss_id(pdev);
+	if (id < 0)
+		return PCI_UNKNOWN;
+
+	reg = (id * LSS_PWS_BITS) / 32;
+	bit = (id * LSS_PWS_BITS) % 32;
+	power = mid_pwr_get_state(pwr, reg);
+	return (__force pci_power_t)((power >> bit) & 3);
+}
+
+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;
+	u8 id;
+
+	/*
+	 * Mapping to PWRMU index is kept in the Logical SubSystem ID byte of
+	 * Vendor capability.
+	 */
+	vndr = pci_find_capability(pdev, PCI_CAP_ID_VNDR);
+	if (!vndr)
+		return -EINVAL;
+
+	/* Read the Logical SubSystem ID byte */
+	pci_read_config_byte(pdev, vndr + INTEL_MID_PWR_LSS_OFFSET, &id);
+	if (!(id & INTEL_MID_PWR_LSS_TYPE))
+		return -ENODEV;
+
+	id &= ~INTEL_MID_PWR_LSS_TYPE;
+	if (id >= LSS_MAX_DEVS)
+		return -ERANGE;
+
+	return id;
+}
+
+static irqreturn_t mid_pwr_irq_handler(int irq, void *dev_id)
+{
+	struct mid_pwr *pwr = dev_id;
+	u32 ics;
+
+	ics = readl(pwr->regs + PM_ICS);
+	if (!(ics & PM_ICS_IP))
+		return IRQ_NONE;
+
+	writel(ics | PM_ICS_IP, pwr->regs + PM_ICS);
+
+	dev_warn(pwr->dev, "Unexpected IRQ: %#x\n", PM_ICS_INT_STATUS(ics));
+	return IRQ_HANDLED;
+}
+
+struct mid_pwr_device_info {
+	int (*set_initial_state)(struct mid_pwr *pwr);
+};
+
+static int mid_pwr_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	struct mid_pwr_device_info *info = (void *)id->driver_data;
+	struct device *dev = &pdev->dev;
+	struct mid_pwr *pwr;
+	int ret;
+
+	ret = pcim_enable_device(pdev);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "error: could not enable device\n");
+		return ret;
+	}
+
+	pwr = devm_kzalloc(dev, sizeof(*pwr), GFP_KERNEL);
+	if (!pwr)
+		return -ENOMEM;
+
+	pwr->regs = pcim_iomap_region(pdev, 0, "intel_mid_pwr");
+	ret = PTR_ERR_OR_ZERO(pwr->regs);
+	if (ret) {
+		dev_err(&pdev->dev, "Could not request / ioremap I/O-Mem: %d\n", ret);
+		return ret;
+	}
+
+	pwr->dev = dev;
+	pwr->irq = pdev->irq;
+
+	mutex_init(&pwr->lock);
+
+	/* Disable interrupts */
+	mid_pwr_interrupt_disable(pwr);
+
+	if (info && info->set_initial_state) {
+		ret = info->set_initial_state(pwr);
+		if (ret)
+			dev_warn(dev, "Can't set initial state: %d\n", ret);
+	}
+
+	ret = devm_request_irq(dev, pdev->irq, mid_pwr_irq_handler,
+			       IRQF_NO_SUSPEND, pci_name(pdev), pwr);
+	if (ret)
+		return ret;
+
+	pwr->available = true;
+	midpwr = pwr;
+
+	pci_set_drvdata(pdev, pwr);
+	return 0;
+}
+
+static int mid_set_initial_state(struct mid_pwr *pwr, const u32 *states)
+{
+	unsigned int i, j;
+	int ret;
+
+	/*
+	 * Enable wake events.
+	 *
+	 * PWRMU supports up to 32 sources for wake up the system. Ungate them
+	 * all here.
+	 */
+	mid_pwr_set_wake(pwr, 0, 0xffffffff);
+	mid_pwr_set_wake(pwr, 1, 0xffffffff);
+
+	/*
+	 * Power off South Complex devices.
+	 *
+	 * There is a map (see a note below) of 64 devices with 2 bits per each
+	 * on 32-bit HW registers. The following calls set all devices to one
+	 * known initial state, i.e. PCI_D3hot. This is done in conjunction
+	 * with PMCSR setting in arch/x86/pci/intel_mid_pci.c.
+	 *
+	 * 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, 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);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < LSS_MAX_DEVS; i++) {
+		for (j = 0; j < LSS_MAX_SHARED_DEVS; j++)
+			pwr->lss[i][j].state = PCI_D3hot;
+	}
+
+	return 0;
+}
+
+static int pnw_set_initial_state(struct mid_pwr *pwr)
+{
+	/* On Penwell SRAM must stay powered on */
+	static 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)
+{
+	static 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)&pnw_info },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_TANGIER), (kernel_ulong_t)&tng_info },
+	{}
+};
+
+static struct pci_driver mid_pwr_pci_driver = {
+	.name		= "intel_mid_pwr",
+	.probe		= mid_pwr_probe,
+	.id_table	= mid_pwr_pci_ids,
+};
+
+builtin_pci_driver(mid_pwr_pci_driver);

linux-7.2.3/arch/x86/platform/intel-quark/Makefile

@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_INTEL_IMR) += imr.o
+obj-$(CONFIG_DEBUG_IMR_SELFTEST) += imr_selftest.o

linux-7.2.3/arch/x86/platform/intel-quark/imr.c

@@ -0,0 +1,597 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * imr.c -- Intel Isolated Memory Region driver
+ *
+ * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2015 Bryan O'Donoghue <pure.logic@nexus-software.ie>
+ *
+ * IMR registers define an isolated region of memory that can
+ * be masked to prohibit certain system agents from accessing memory.
+ * When a device behind a masked port performs an access - snooped or
+ * not, an IMR may optionally prevent that transaction from changing
+ * the state of memory or from getting correct data in response to the
+ * operation.
+ *
+ * Write data will be dropped and reads will return 0xFFFFFFFF, the
+ * system will reset and system BIOS will print out an error message to
+ * inform the user that an IMR has been violated.
+ *
+ * This code is based on the Linux MTRR code and reference code from
+ * Intel's Quark BSP EFI, Linux and grub code.
+ *
+ * See quark-x1000-datasheet.pdf for register definitions.
+ * http://www.intel.com/content/dam/www/public/us/en/documents/datasheets/quark-x1000-datasheet.pdf
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <asm-generic/sections.h>
+#include <asm/cpu_device_id.h>
+#include <asm/imr.h>
+#include <asm/iosf_mbi.h>
+#include <asm/io.h>
+
+#include <linux/debugfs.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+
+struct imr_device {
+	bool		init;
+	struct mutex	lock;
+	int		max_imr;
+	int		reg_base;
+};
+
+static struct imr_device imr_dev;
+
+/*
+ * IMR read/write mask control registers.
+ * See quark-x1000-datasheet.pdf sections 12.7.4.5 and 12.7.4.6 for
+ * bit definitions.
+ *
+ * addr_hi
+ * 31		Lock bit
+ * 30:24	Reserved
+ * 23:2		1 KiB aligned lo address
+ * 1:0		Reserved
+ *
+ * addr_hi
+ * 31:24	Reserved
+ * 23:2		1 KiB aligned hi address
+ * 1:0		Reserved
+ */
+#define IMR_LOCK	BIT(31)
+
+struct imr_regs {
+	u32 addr_lo;
+	u32 addr_hi;
+	u32 rmask;
+	u32 wmask;
+};
+
+#define IMR_NUM_REGS	(sizeof(struct imr_regs)/sizeof(u32))
+#define IMR_SHIFT	8
+#define imr_to_phys(x)	((x) << IMR_SHIFT)
+#define phys_to_imr(x)	((x) >> IMR_SHIFT)
+
+/**
+ * imr_is_enabled - true if an IMR is enabled false otherwise.
+ *
+ * Determines if an IMR is enabled based on address range and read/write
+ * mask. An IMR set with an address range set to zero and a read/write
+ * access mask set to all is considered to be disabled. An IMR in any
+ * other state - for example set to zero but without read/write access
+ * all is considered to be enabled. This definition of disabled is how
+ * firmware switches off an IMR and is maintained in kernel for
+ * consistency.
+ *
+ * @imr:	pointer to IMR descriptor.
+ * @return:	true if IMR enabled false if disabled.
+ */
+static inline int imr_is_enabled(struct imr_regs *imr)
+{
+	return !(imr->rmask == IMR_READ_ACCESS_ALL &&
+		 imr->wmask == IMR_WRITE_ACCESS_ALL &&
+		 imr_to_phys(imr->addr_lo) == 0 &&
+		 imr_to_phys(imr->addr_hi) == 0);
+}
+
+/**
+ * imr_read - read an IMR at a given index.
+ *
+ * Requires caller to hold imr mutex.
+ *
+ * @idev:	pointer to imr_device structure.
+ * @imr_id:	IMR entry to read.
+ * @imr:	IMR structure representing address and access masks.
+ * @return:	0 on success or error code passed from mbi_iosf on failure.
+ */
+static int imr_read(struct imr_device *idev, u32 imr_id, struct imr_regs *imr)
+{
+	u32 reg = imr_id * IMR_NUM_REGS + idev->reg_base;
+	int ret;
+
+	ret = iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->addr_lo);
+	if (ret)
+		return ret;
+
+	ret = iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->addr_hi);
+	if (ret)
+		return ret;
+
+	ret = iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->rmask);
+	if (ret)
+		return ret;
+
+	return iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->wmask);
+}
+
+/**
+ * imr_write - write an IMR at a given index.
+ *
+ * Requires caller to hold imr mutex.
+ * Note lock bits need to be written independently of address bits.
+ *
+ * @idev:	pointer to imr_device structure.
+ * @imr_id:	IMR entry to write.
+ * @imr:	IMR structure representing address and access masks.
+ * @return:	0 on success or error code passed from mbi_iosf on failure.
+ */
+static int imr_write(struct imr_device *idev, u32 imr_id, struct imr_regs *imr)
+{
+	unsigned long flags;
+	u32 reg = imr_id * IMR_NUM_REGS + idev->reg_base;
+	int ret;
+
+	local_irq_save(flags);
+
+	ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->addr_lo);
+	if (ret)
+		goto failed;
+
+	ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->addr_hi);
+	if (ret)
+		goto failed;
+
+	ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->rmask);
+	if (ret)
+		goto failed;
+
+	ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->wmask);
+	if (ret)
+		goto failed;
+
+	local_irq_restore(flags);
+	return 0;
+failed:
+	/*
+	 * If writing to the IOSF failed then we're in an unknown state,
+	 * likely a very bad state. An IMR in an invalid state will almost
+	 * certainly lead to a memory access violation.
+	 */
+	local_irq_restore(flags);
+	WARN(ret, "IOSF-MBI write fail range 0x%08x-0x%08x unreliable\n",
+	     imr_to_phys(imr->addr_lo), imr_to_phys(imr->addr_hi) + IMR_MASK);
+
+	return ret;
+}
+
+/**
+ * imr_dbgfs_state_show - print state of IMR registers.
+ *
+ * @s:		pointer to seq_file for output.
+ * @unused:	unused parameter.
+ * @return:	0 on success or error code passed from mbi_iosf on failure.
+ */
+static int imr_dbgfs_state_show(struct seq_file *s, void *unused)
+{
+	phys_addr_t base;
+	phys_addr_t end;
+	int i;
+	struct imr_device *idev = s->private;
+	struct imr_regs imr;
+	size_t size;
+	int ret = -ENODEV;
+
+	mutex_lock(&idev->lock);
+
+	for (i = 0; i < idev->max_imr; i++) {
+
+		ret = imr_read(idev, i, &imr);
+		if (ret)
+			break;
+
+		/*
+		 * Remember to add IMR_ALIGN bytes to size to indicate the
+		 * inherent IMR_ALIGN size bytes contained in the masked away
+		 * lower ten bits.
+		 */
+		if (imr_is_enabled(&imr)) {
+			base = imr_to_phys(imr.addr_lo);
+			end = imr_to_phys(imr.addr_hi) + IMR_MASK;
+			size = end - base + 1;
+		} else {
+			base = 0;
+			end = 0;
+			size = 0;
+		}
+		seq_printf(s, "imr%02i: base=%pa, end=%pa, size=0x%08zx "
+			   "rmask=0x%08x, wmask=0x%08x, %s, %s\n", i,
+			   &base, &end, size, imr.rmask, imr.wmask,
+			   imr_is_enabled(&imr) ? "enabled " : "disabled",
+			   imr.addr_lo & IMR_LOCK ? "locked" : "unlocked");
+	}
+
+	mutex_unlock(&idev->lock);
+	return ret;
+}
+DEFINE_SHOW_ATTRIBUTE(imr_dbgfs_state);
+
+/**
+ * imr_debugfs_register - register debugfs hooks.
+ *
+ * @idev:	pointer to imr_device structure.
+ */
+static void imr_debugfs_register(struct imr_device *idev)
+{
+	debugfs_create_file("imr_state", 0444, NULL, idev,
+			    &imr_dbgfs_state_fops);
+}
+
+/**
+ * imr_check_params - check passed address range IMR alignment and non-zero size
+ *
+ * @base:	base address of intended IMR.
+ * @size:	size of intended IMR.
+ * @return:	zero on valid range -EINVAL on unaligned base/size.
+ */
+static int imr_check_params(phys_addr_t base, size_t size)
+{
+	if ((base & IMR_MASK) || (size & IMR_MASK)) {
+		pr_err("base %pa size 0x%08zx must align to 1KiB\n",
+			&base, size);
+		return -EINVAL;
+	}
+	if (size == 0)
+		return -EINVAL;
+
+	return 0;
+}
+
+/**
+ * imr_raw_size - account for the IMR_ALIGN bytes that addr_hi appends.
+ *
+ * IMR addr_hi has a built in offset of plus IMR_ALIGN (0x400) bytes from the
+ * value in the register. We need to subtract IMR_ALIGN bytes from input sizes
+ * as a result.
+ *
+ * @size:	input size bytes.
+ * @return:	reduced size.
+ */
+static inline size_t imr_raw_size(size_t size)
+{
+	return size - IMR_ALIGN;
+}
+
+/**
+ * imr_address_overlap - detects an address overlap.
+ *
+ * @addr:	address to check against an existing IMR.
+ * @imr:	imr being checked.
+ * @return:	true for overlap false for no overlap.
+ */
+static inline int imr_address_overlap(phys_addr_t addr, struct imr_regs *imr)
+{
+	return addr >= imr_to_phys(imr->addr_lo) && addr <= imr_to_phys(imr->addr_hi);
+}
+
+/**
+ * imr_add_range - add an Isolated Memory Region.
+ *
+ * @base:	physical base address of region aligned to 1KiB.
+ * @size:	physical size of region in bytes must be aligned to 1KiB.
+ * @read_mask:	read access mask.
+ * @write_mask:	write access mask.
+ * @return:	zero on success or negative value indicating error.
+ */
+int imr_add_range(phys_addr_t base, size_t size,
+		  unsigned int rmask, unsigned int wmask)
+{
+	phys_addr_t end;
+	unsigned int i;
+	struct imr_device *idev = &imr_dev;
+	struct imr_regs imr;
+	size_t raw_size;
+	int reg;
+	int ret;
+
+	if (WARN_ONCE(idev->init == false, "driver not initialized"))
+		return -ENODEV;
+
+	ret = imr_check_params(base, size);
+	if (ret)
+		return ret;
+
+	/* Tweak the size value. */
+	raw_size = imr_raw_size(size);
+	end = base + raw_size;
+
+	/*
+	 * Check for reserved IMR value common to firmware, kernel and grub
+	 * indicating a disabled IMR.
+	 */
+	imr.addr_lo = phys_to_imr(base);
+	imr.addr_hi = phys_to_imr(end);
+	imr.rmask = rmask;
+	imr.wmask = wmask;
+	if (!imr_is_enabled(&imr))
+		return -ENOTSUPP;
+
+	mutex_lock(&idev->lock);
+
+	/*
+	 * Find a free IMR while checking for an existing overlapping range.
+	 * Note there's no restriction in silicon to prevent IMR overlaps.
+	 * For the sake of simplicity and ease in defining/debugging an IMR
+	 * memory map we exclude IMR overlaps.
+	 */
+	reg = -1;
+	for (i = 0; i < idev->max_imr; i++) {
+		ret = imr_read(idev, i, &imr);
+		if (ret)
+			goto failed;
+
+		/* Find overlap @ base or end of requested range. */
+		ret = -EINVAL;
+		if (imr_is_enabled(&imr)) {
+			if (imr_address_overlap(base, &imr))
+				goto failed;
+			if (imr_address_overlap(end, &imr))
+				goto failed;
+		} else {
+			reg = i;
+		}
+	}
+
+	/* Error out if we have no free IMR entries. */
+	if (reg == -1) {
+		ret = -ENOMEM;
+		goto failed;
+	}
+
+	pr_debug("add %d phys %pa-%pa size %zx mask 0x%08x wmask 0x%08x\n",
+		 reg, &base, &end, raw_size, rmask, wmask);
+
+	/* Enable IMR at specified range and access mask. */
+	imr.addr_lo = phys_to_imr(base);
+	imr.addr_hi = phys_to_imr(end);
+	imr.rmask = rmask;
+	imr.wmask = wmask;
+
+	ret = imr_write(idev, reg, &imr);
+	if (ret < 0) {
+		/*
+		 * In the highly unlikely event iosf_mbi_write failed
+		 * attempt to rollback the IMR setup skipping the trapping
+		 * of further IOSF write failures.
+		 */
+		imr.addr_lo = 0;
+		imr.addr_hi = 0;
+		imr.rmask = IMR_READ_ACCESS_ALL;
+		imr.wmask = IMR_WRITE_ACCESS_ALL;
+		imr_write(idev, reg, &imr);
+	}
+failed:
+	mutex_unlock(&idev->lock);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(imr_add_range);
+
+/**
+ * __imr_remove_range - delete an Isolated Memory Region.
+ *
+ * This function allows you to delete an IMR by its index specified by reg or
+ * by address range specified by base and size respectively. If you specify an
+ * index on its own the base and size parameters are ignored.
+ * imr_remove_range(0, base, size); delete IMR at index 0 base/size ignored.
+ * imr_remove_range(-1, base, size); delete IMR from base to base+size.
+ *
+ * @reg:	imr index to remove.
+ * @base:	physical base address of region aligned to 1 KiB.
+ * @size:	physical size of region in bytes aligned to 1 KiB.
+ * @return:	-EINVAL on invalid range or out or range id
+ *		-ENODEV if reg is valid but no IMR exists or is locked
+ *		0 on success.
+ */
+static int __imr_remove_range(int reg, phys_addr_t base, size_t size)
+{
+	phys_addr_t end;
+	bool found = false;
+	unsigned int i;
+	struct imr_device *idev = &imr_dev;
+	struct imr_regs imr;
+	size_t raw_size;
+	int ret = 0;
+
+	if (WARN_ONCE(idev->init == false, "driver not initialized"))
+		return -ENODEV;
+
+	/*
+	 * Validate address range if deleting by address, else we are
+	 * deleting by index where base and size will be ignored.
+	 */
+	if (reg == -1) {
+		ret = imr_check_params(base, size);
+		if (ret)
+			return ret;
+	}
+
+	/* Tweak the size value. */
+	raw_size = imr_raw_size(size);
+	end = base + raw_size;
+
+	mutex_lock(&idev->lock);
+
+	if (reg >= 0) {
+		/* If a specific IMR is given try to use it. */
+		ret = imr_read(idev, reg, &imr);
+		if (ret)
+			goto failed;
+
+		if (!imr_is_enabled(&imr) || imr.addr_lo & IMR_LOCK) {
+			ret = -ENODEV;
+			goto failed;
+		}
+		found = true;
+	} else {
+		/* Search for match based on address range. */
+		for (i = 0; i < idev->max_imr; i++) {
+			ret = imr_read(idev, i, &imr);
+			if (ret)
+				goto failed;
+
+			if (!imr_is_enabled(&imr) || imr.addr_lo & IMR_LOCK)
+				continue;
+
+			if ((imr_to_phys(imr.addr_lo) == base) &&
+			    (imr_to_phys(imr.addr_hi) == end)) {
+				found = true;
+				reg = i;
+				break;
+			}
+		}
+	}
+
+	if (!found) {
+		ret = -ENODEV;
+		goto failed;
+	}
+
+	pr_debug("remove %d phys %pa-%pa size %zx\n", reg, &base, &end, raw_size);
+
+	/* Tear down the IMR. */
+	imr.addr_lo = 0;
+	imr.addr_hi = 0;
+	imr.rmask = IMR_READ_ACCESS_ALL;
+	imr.wmask = IMR_WRITE_ACCESS_ALL;
+
+	ret = imr_write(idev, reg, &imr);
+
+failed:
+	mutex_unlock(&idev->lock);
+	return ret;
+}
+
+/**
+ * imr_remove_range - delete an Isolated Memory Region by address
+ *
+ * This function allows you to delete an IMR by an address range specified
+ * by base and size respectively.
+ * imr_remove_range(base, size); delete IMR from base to base+size.
+ *
+ * @base:	physical base address of region aligned to 1 KiB.
+ * @size:	physical size of region in bytes aligned to 1 KiB.
+ * @return:	-EINVAL on invalid range or out or range id
+ *		-ENODEV if reg is valid but no IMR exists or is locked
+ *		0 on success.
+ */
+int imr_remove_range(phys_addr_t base, size_t size)
+{
+	return __imr_remove_range(-1, base, size);
+}
+EXPORT_SYMBOL_GPL(imr_remove_range);
+
+/**
+ * imr_clear - delete an Isolated Memory Region by index
+ *
+ * This function allows you to delete an IMR by an address range specified
+ * by the index of the IMR. Useful for initial sanitization of the IMR
+ * address map.
+ * imr_ge(base, size); delete IMR from base to base+size.
+ *
+ * @reg:	imr index to remove.
+ * @return:	-EINVAL on invalid range or out or range id
+ *		-ENODEV if reg is valid but no IMR exists or is locked
+ *		0 on success.
+ */
+static inline int imr_clear(int reg)
+{
+	return __imr_remove_range(reg, 0, 0);
+}
+
+/**
+ * imr_fixup_memmap - Tear down IMRs used during bootup.
+ *
+ * BIOS and Grub both setup IMRs around compressed kernel, initrd memory
+ * that need to be removed before the kernel hands out one of the IMR
+ * encased addresses to a downstream DMA agent such as the SD or Ethernet.
+ * IMRs on Galileo are setup to immediately reset the system on violation.
+ * As a result if you're running a root filesystem from SD - you'll need
+ * the boot-time IMRs torn down or you'll find seemingly random resets when
+ * using your filesystem.
+ *
+ * @idev:	pointer to imr_device structure.
+ * @return:
+ */
+static void __init imr_fixup_memmap(struct imr_device *idev)
+{
+	phys_addr_t base = virt_to_phys(&_text);
+	size_t size = virt_to_phys(&__end_rodata) - base;
+	unsigned long start, end;
+	int i;
+	int ret;
+
+	/* Tear down all existing unlocked IMRs. */
+	for (i = 0; i < idev->max_imr; i++)
+		imr_clear(i);
+
+	start = (unsigned long)_text;
+	end = (unsigned long)__end_rodata - 1;
+
+	/*
+	 * Setup an unlocked IMR around the physical extent of the kernel
+	 * from the beginning of the .text section to the end of the
+	 * .rodata section as one physically contiguous block.
+	 *
+	 * We don't round up @size since it is already PAGE_SIZE aligned.
+	 * See vmlinux.lds.S for details.
+	 */
+	ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+	if (ret < 0) {
+		pr_err("unable to setup IMR for kernel: %zu KiB (%lx - %lx)\n",
+			size / 1024, start, end);
+	} else {
+		pr_info("protecting kernel .text - .rodata: %zu KiB (%lx - %lx)\n",
+			size / 1024, start, end);
+	}
+
+}
+
+static const struct x86_cpu_id imr_ids[] __initconst = {
+	X86_MATCH_VFM(INTEL_QUARK_X1000, NULL),
+	{}
+};
+
+/**
+ * imr_init - entry point for IMR driver.
+ *
+ * return: -ENODEV for no IMR support 0 if good to go.
+ */
+static int __init imr_init(void)
+{
+	struct imr_device *idev = &imr_dev;
+
+	if (!x86_match_cpu(imr_ids) || !iosf_mbi_available())
+		return -ENODEV;
+
+	idev->max_imr = QUARK_X1000_IMR_MAX;
+	idev->reg_base = QUARK_X1000_IMR_REGBASE;
+	idev->init = true;
+
+	mutex_init(&idev->lock);
+	imr_debugfs_register(idev);
+	imr_fixup_memmap(idev);
+	return 0;
+}
+device_initcall(imr_init);

linux-7.2.3/arch/x86/platform/intel-quark/imr_selftest.c

@@ -0,0 +1,129 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * imr_selftest.c -- Intel Isolated Memory Region self-test driver
+ *
+ * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2015 Bryan O'Donoghue <pure.logic@nexus-software.ie>
+ *
+ * IMR self test. The purpose of this module is to run a set of tests on the
+ * IMR API to validate its sanity. We check for overlapping, reserved
+ * addresses and setup/teardown sanity.
+ *
+ */
+
+#include <asm-generic/sections.h>
+#include <asm/cpu_device_id.h>
+#include <asm/imr.h>
+#include <asm/io.h>
+
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+
+#define SELFTEST KBUILD_MODNAME ": "
+/**
+ * imr_self_test_result - Print result string for self test.
+ *
+ * @res:	result code - true if test passed false otherwise.
+ * @fmt:	format string.
+ * ...		variadic argument list.
+ */
+static __printf(2, 3)
+void __init imr_self_test_result(int res, const char *fmt, ...)
+{
+	va_list vlist;
+
+	/* Print pass/fail. */
+	if (res)
+		pr_info(SELFTEST "pass ");
+	else
+		pr_info(SELFTEST "fail ");
+
+	/* Print variable string. */
+	va_start(vlist, fmt);
+	vprintk(fmt, vlist);
+	va_end(vlist);
+
+	/* Optional warning. */
+	WARN(res == 0, "test failed");
+}
+#undef SELFTEST
+
+/**
+ * imr_self_test
+ *
+ * Verify IMR self_test with some simple tests to verify overlap,
+ * zero sized allocations and 1 KiB sized areas.
+ *
+ */
+static void __init imr_self_test(void)
+{
+	phys_addr_t base  = virt_to_phys(&_text);
+	size_t size = virt_to_phys(&__end_rodata) - base;
+	const char *fmt_over = "overlapped IMR @ (0x%08lx - 0x%08lx)\n";
+	int ret;
+
+	/* Test zero zero. */
+	ret = imr_add_range(0, 0, 0, 0);
+	imr_self_test_result(ret < 0, "zero sized IMR\n");
+
+	/* Test exact overlap. */
+	ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+	imr_self_test_result(ret < 0, fmt_over, __va(base), __va(base + size));
+
+	/* Test overlap with base inside of existing. */
+	base += size - IMR_ALIGN;
+	ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+	imr_self_test_result(ret < 0, fmt_over, __va(base), __va(base + size));
+
+	/* Test overlap with end inside of existing. */
+	base -= size + IMR_ALIGN * 2;
+	ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+	imr_self_test_result(ret < 0, fmt_over, __va(base), __va(base + size));
+
+	/* Test that a 1 KiB IMR @ zero with read/write all will bomb out. */
+	ret = imr_add_range(0, IMR_ALIGN, IMR_READ_ACCESS_ALL,
+			    IMR_WRITE_ACCESS_ALL);
+	imr_self_test_result(ret < 0, "1KiB IMR @ 0x00000000 - access-all\n");
+
+	/* Test that a 1 KiB IMR @ zero with CPU only will work. */
+	ret = imr_add_range(0, IMR_ALIGN, IMR_CPU, IMR_CPU);
+	imr_self_test_result(ret >= 0, "1KiB IMR @ 0x00000000 - cpu-access\n");
+	if (ret >= 0) {
+		ret = imr_remove_range(0, IMR_ALIGN);
+		imr_self_test_result(ret == 0, "teardown - cpu-access\n");
+	}
+
+	/* Test 2 KiB works. */
+	size = IMR_ALIGN * 2;
+	ret = imr_add_range(0, size, IMR_READ_ACCESS_ALL, IMR_WRITE_ACCESS_ALL);
+	imr_self_test_result(ret >= 0, "2KiB IMR @ 0x00000000\n");
+	if (ret >= 0) {
+		ret = imr_remove_range(0, size);
+		imr_self_test_result(ret == 0, "teardown 2KiB\n");
+	}
+}
+
+static const struct x86_cpu_id imr_ids[] __initconst = {
+	X86_MATCH_VFM(INTEL_QUARK_X1000, NULL),
+	{}
+};
+
+/**
+ * imr_self_test_init - entry point for IMR driver.
+ *
+ * return: -ENODEV for no IMR support 0 if good to go.
+ */
+static int __init imr_self_test_init(void)
+{
+	if (x86_match_cpu(imr_ids))
+		imr_self_test();
+	return 0;
+}
+
+/**
+ * imr_self_test_exit - exit point for IMR code.
+ *
+ * return:
+ */
+device_initcall(imr_self_test_init);

linux-7.2.3/arch/x86/platform/intel/Makefile

@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_IOSF_MBI)			+= iosf_mbi.o

linux-7.2.3/arch/x86/platform/intel/iosf_mbi.c

@@ -0,0 +1,558 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * IOSF-SB MailBox Interface Driver
+ * Copyright (c) 2013, Intel Corporation.
+ *
+ * The IOSF-SB is a fabric bus available on Atom based SOC's that uses a
+ * mailbox interface (MBI) to communicate with multiple devices. This
+ * driver implements access to this interface for those platforms that can
+ * enumerate the device using PCI.
+ */
+
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/pci.h>
+#include <linux/debugfs.h>
+#include <linux/capability.h>
+#include <linux/pm_qos.h>
+#include <linux/wait.h>
+
+#include <asm/iosf_mbi.h>
+
+#define PCI_DEVICE_ID_INTEL_BAYTRAIL		0x0F00
+#define PCI_DEVICE_ID_INTEL_BRASWELL		0x2280
+#define PCI_DEVICE_ID_INTEL_QUARK_X1000		0x0958
+#define PCI_DEVICE_ID_INTEL_TANGIER		0x1170
+
+static struct pci_dev *mbi_pdev;
+static DEFINE_SPINLOCK(iosf_mbi_lock);
+
+/**************** Generic iosf_mbi access helpers ****************/
+
+static inline u32 iosf_mbi_form_mcr(u8 op, u8 port, u8 offset)
+{
+	return (op << 24) | (port << 16) | (offset << 8) | MBI_ENABLE;
+}
+
+static int iosf_mbi_pci_read_mdr(u32 mcrx, u32 mcr, u32 *mdr)
+{
+	int result;
+
+	if (!mbi_pdev)
+		return -ENODEV;
+
+	if (mcrx) {
+		result = pci_write_config_dword(mbi_pdev, MBI_MCRX_OFFSET,
+						mcrx);
+		if (result < 0)
+			goto fail_read;
+	}
+
+	result = pci_write_config_dword(mbi_pdev, MBI_MCR_OFFSET, mcr);
+	if (result < 0)
+		goto fail_read;
+
+	result = pci_read_config_dword(mbi_pdev, MBI_MDR_OFFSET, mdr);
+	if (result < 0)
+		goto fail_read;
+
+	return 0;
+
+fail_read:
+	dev_err(&mbi_pdev->dev, "PCI config access failed with %d\n", result);
+	return pcibios_err_to_errno(result);
+}
+
+static int iosf_mbi_pci_write_mdr(u32 mcrx, u32 mcr, u32 mdr)
+{
+	int result;
+
+	if (!mbi_pdev)
+		return -ENODEV;
+
+	result = pci_write_config_dword(mbi_pdev, MBI_MDR_OFFSET, mdr);
+	if (result < 0)
+		goto fail_write;
+
+	if (mcrx) {
+		result = pci_write_config_dword(mbi_pdev, MBI_MCRX_OFFSET,
+						mcrx);
+		if (result < 0)
+			goto fail_write;
+	}
+
+	result = pci_write_config_dword(mbi_pdev, MBI_MCR_OFFSET, mcr);
+	if (result < 0)
+		goto fail_write;
+
+	return 0;
+
+fail_write:
+	dev_err(&mbi_pdev->dev, "PCI config access failed with %d\n", result);
+	return pcibios_err_to_errno(result);
+}
+
+int iosf_mbi_read(u8 port, u8 opcode, u32 offset, u32 *mdr)
+{
+	u32 mcr, mcrx;
+	unsigned long flags;
+	int ret;
+
+	/* Access to the GFX unit is handled by GPU code */
+	if (port == BT_MBI_UNIT_GFX) {
+		WARN_ON(1);
+		return -EPERM;
+	}
+
+	mcr = iosf_mbi_form_mcr(opcode, port, offset & MBI_MASK_LO);
+	mcrx = offset & MBI_MASK_HI;
+
+	spin_lock_irqsave(&iosf_mbi_lock, flags);
+	ret = iosf_mbi_pci_read_mdr(mcrx, mcr, mdr);
+	spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_read);
+
+int iosf_mbi_write(u8 port, u8 opcode, u32 offset, u32 mdr)
+{
+	u32 mcr, mcrx;
+	unsigned long flags;
+	int ret;
+
+	/* Access to the GFX unit is handled by GPU code */
+	if (port == BT_MBI_UNIT_GFX) {
+		WARN_ON(1);
+		return -EPERM;
+	}
+
+	mcr = iosf_mbi_form_mcr(opcode, port, offset & MBI_MASK_LO);
+	mcrx = offset & MBI_MASK_HI;
+
+	spin_lock_irqsave(&iosf_mbi_lock, flags);
+	ret = iosf_mbi_pci_write_mdr(mcrx, mcr, mdr);
+	spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_write);
+
+int iosf_mbi_modify(u8 port, u8 opcode, u32 offset, u32 mdr, u32 mask)
+{
+	u32 mcr, mcrx;
+	u32 value;
+	unsigned long flags;
+	int ret;
+
+	/* Access to the GFX unit is handled by GPU code */
+	if (port == BT_MBI_UNIT_GFX) {
+		WARN_ON(1);
+		return -EPERM;
+	}
+
+	mcr = iosf_mbi_form_mcr(opcode, port, offset & MBI_MASK_LO);
+	mcrx = offset & MBI_MASK_HI;
+
+	spin_lock_irqsave(&iosf_mbi_lock, flags);
+
+	/* Read current mdr value */
+	ret = iosf_mbi_pci_read_mdr(mcrx, mcr & MBI_RD_MASK, &value);
+	if (ret < 0) {
+		spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+		return ret;
+	}
+
+	/* Apply mask */
+	value &= ~mask;
+	mdr &= mask;
+	value |= mdr;
+
+	/* Write back */
+	ret = iosf_mbi_pci_write_mdr(mcrx, mcr | MBI_WR_MASK, value);
+
+	spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_modify);
+
+bool iosf_mbi_available(void)
+{
+	/* Mbi isn't hot-pluggable. No remove routine is provided */
+	return mbi_pdev;
+}
+EXPORT_SYMBOL(iosf_mbi_available);
+
+/*
+ **************** P-Unit/kernel shared I2C bus arbitration ****************
+ *
+ * Some Bay Trail and Cherry Trail devices have the P-Unit and us (the kernel)
+ * share a single I2C bus to the PMIC. Below are helpers to arbitrate the
+ * accesses between the kernel and the P-Unit.
+ *
+ * See arch/x86/include/asm/iosf_mbi.h for kernel-doc text for each function.
+ */
+
+#define SEMAPHORE_TIMEOUT		500
+#define PUNIT_SEMAPHORE_BYT		0x7
+#define PUNIT_SEMAPHORE_CHT		0x10e
+#define PUNIT_SEMAPHORE_BIT		BIT(0)
+#define PUNIT_SEMAPHORE_ACQUIRE		BIT(1)
+
+static DEFINE_MUTEX(iosf_mbi_pmic_access_mutex);
+static BLOCKING_NOTIFIER_HEAD(iosf_mbi_pmic_bus_access_notifier);
+static DECLARE_WAIT_QUEUE_HEAD(iosf_mbi_pmic_access_waitq);
+static u32 iosf_mbi_pmic_punit_access_count;
+static u32 iosf_mbi_pmic_i2c_access_count;
+static u32 iosf_mbi_sem_address;
+static unsigned long iosf_mbi_sem_acquired;
+static struct pm_qos_request iosf_mbi_pm_qos;
+
+void iosf_mbi_punit_acquire(void)
+{
+	/* Wait for any I2C PMIC accesses from in kernel drivers to finish. */
+	mutex_lock(&iosf_mbi_pmic_access_mutex);
+	while (iosf_mbi_pmic_i2c_access_count != 0) {
+		mutex_unlock(&iosf_mbi_pmic_access_mutex);
+		wait_event(iosf_mbi_pmic_access_waitq,
+			   iosf_mbi_pmic_i2c_access_count == 0);
+		mutex_lock(&iosf_mbi_pmic_access_mutex);
+	}
+	/*
+	 * We do not need to do anything to allow the PUNIT to safely access
+	 * the PMIC, other then block in kernel accesses to the PMIC.
+	 */
+	iosf_mbi_pmic_punit_access_count++;
+	mutex_unlock(&iosf_mbi_pmic_access_mutex);
+}
+EXPORT_SYMBOL(iosf_mbi_punit_acquire);
+
+void iosf_mbi_punit_release(void)
+{
+	bool do_wakeup;
+
+	mutex_lock(&iosf_mbi_pmic_access_mutex);
+	iosf_mbi_pmic_punit_access_count--;
+	do_wakeup = iosf_mbi_pmic_punit_access_count == 0;
+	mutex_unlock(&iosf_mbi_pmic_access_mutex);
+
+	if (do_wakeup)
+		wake_up(&iosf_mbi_pmic_access_waitq);
+}
+EXPORT_SYMBOL(iosf_mbi_punit_release);
+
+static int iosf_mbi_get_sem(u32 *sem)
+{
+	int ret;
+
+	ret = iosf_mbi_read(BT_MBI_UNIT_PMC, MBI_REG_READ,
+			    iosf_mbi_sem_address, sem);
+	if (ret) {
+		dev_err(&mbi_pdev->dev, "Error P-Unit semaphore read failed\n");
+		return ret;
+	}
+
+	*sem &= PUNIT_SEMAPHORE_BIT;
+	return 0;
+}
+
+static void iosf_mbi_reset_semaphore(void)
+{
+	if (iosf_mbi_modify(BT_MBI_UNIT_PMC, MBI_REG_READ,
+			    iosf_mbi_sem_address, 0, PUNIT_SEMAPHORE_BIT))
+		dev_err(&mbi_pdev->dev, "Error P-Unit semaphore reset failed\n");
+
+	cpu_latency_qos_update_request(&iosf_mbi_pm_qos, PM_QOS_DEFAULT_VALUE);
+
+	blocking_notifier_call_chain(&iosf_mbi_pmic_bus_access_notifier,
+				     MBI_PMIC_BUS_ACCESS_END, NULL);
+}
+
+/*
+ * This function blocks P-Unit accesses to the PMIC I2C bus, so that kernel
+ * I2C code, such as e.g. a fuel-gauge driver, can access it safely.
+ *
+ * This function may be called by I2C controller code while an I2C driver has
+ * already blocked P-Unit accesses because it wants them blocked over multiple
+ * i2c-transfers, for e.g. read-modify-write of an I2C client register.
+ *
+ * To allow safe PMIC i2c bus accesses this function takes the following steps:
+ *
+ * 1) Some code sends request to the P-Unit which make it access the PMIC
+ *    I2C bus. Testing has shown that the P-Unit does not check its internal
+ *    PMIC bus semaphore for these requests. Callers of these requests call
+ *    iosf_mbi_punit_acquire()/_release() around their P-Unit accesses, these
+ *    functions increase/decrease iosf_mbi_pmic_punit_access_count, so first
+ *    we wait for iosf_mbi_pmic_punit_access_count to become 0.
+ *
+ * 2) Check iosf_mbi_pmic_i2c_access_count, if access has already
+ *    been blocked by another caller, we only need to increment
+ *    iosf_mbi_pmic_i2c_access_count and we can skip the other steps.
+ *
+ * 3) Some code makes such P-Unit requests from atomic contexts where it
+ *    cannot call iosf_mbi_punit_acquire() as that may sleep.
+ *    As the second step we call a notifier chain which allows any code
+ *    needing P-Unit resources from atomic context to acquire them before
+ *    we take control over the PMIC I2C bus.
+ *
+ * 4) When CPU cores enter C6 or C7 the P-Unit needs to talk to the PMIC
+ *    if this happens while the kernel itself is accessing the PMIC I2C bus
+ *    the SoC hangs.
+ *    As the third step we call cpu_latency_qos_update_request() to disallow the
+ *    CPU to enter C6 or C7.
+ *
+ * 5) The P-Unit has a PMIC bus semaphore which we can request to stop
+ *    autonomous P-Unit tasks from accessing the PMIC I2C bus while we hold it.
+ *    As the fourth and final step we request this semaphore and wait for our
+ *    request to be acknowledged.
+ */
+int iosf_mbi_block_punit_i2c_access(void)
+{
+	unsigned long start, end;
+	int ret = 0;
+	u32 sem;
+
+	if (WARN_ON(!mbi_pdev || !iosf_mbi_sem_address))
+		return -ENXIO;
+
+	mutex_lock(&iosf_mbi_pmic_access_mutex);
+
+	while (iosf_mbi_pmic_punit_access_count != 0) {
+		mutex_unlock(&iosf_mbi_pmic_access_mutex);
+		wait_event(iosf_mbi_pmic_access_waitq,
+			   iosf_mbi_pmic_punit_access_count == 0);
+		mutex_lock(&iosf_mbi_pmic_access_mutex);
+	}
+
+	if (iosf_mbi_pmic_i2c_access_count > 0)
+		goto success;
+
+	blocking_notifier_call_chain(&iosf_mbi_pmic_bus_access_notifier,
+				     MBI_PMIC_BUS_ACCESS_BEGIN, NULL);
+
+	/*
+	 * Disallow the CPU to enter C6 or C7 state, entering these states
+	 * requires the P-Unit to talk to the PMIC and if this happens while
+	 * we're holding the semaphore, the SoC hangs.
+	 */
+	cpu_latency_qos_update_request(&iosf_mbi_pm_qos, 0);
+
+	/* host driver writes to side band semaphore register */
+	ret = iosf_mbi_write(BT_MBI_UNIT_PMC, MBI_REG_WRITE,
+			     iosf_mbi_sem_address, PUNIT_SEMAPHORE_ACQUIRE);
+	if (ret) {
+		dev_err(&mbi_pdev->dev, "Error P-Unit semaphore request failed\n");
+		goto error;
+	}
+
+	/* host driver waits for bit 0 to be set in semaphore register */
+	start = jiffies;
+	end = start + msecs_to_jiffies(SEMAPHORE_TIMEOUT);
+	do {
+		ret = iosf_mbi_get_sem(&sem);
+		if (!ret && sem) {
+			iosf_mbi_sem_acquired = jiffies;
+			dev_dbg(&mbi_pdev->dev, "P-Unit semaphore acquired after %ums\n",
+				jiffies_to_msecs(jiffies - start));
+			goto success;
+		}
+
+		usleep_range(1000, 2000);
+	} while (time_before(jiffies, end));
+
+	ret = -ETIMEDOUT;
+	dev_err(&mbi_pdev->dev, "Error P-Unit semaphore timed out, resetting\n");
+error:
+	iosf_mbi_reset_semaphore();
+	if (!iosf_mbi_get_sem(&sem))
+		dev_err(&mbi_pdev->dev, "P-Unit semaphore: %d\n", sem);
+success:
+	if (!WARN_ON(ret))
+		iosf_mbi_pmic_i2c_access_count++;
+
+	mutex_unlock(&iosf_mbi_pmic_access_mutex);
+
+	return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_block_punit_i2c_access);
+
+void iosf_mbi_unblock_punit_i2c_access(void)
+{
+	bool do_wakeup = false;
+
+	mutex_lock(&iosf_mbi_pmic_access_mutex);
+	iosf_mbi_pmic_i2c_access_count--;
+	if (iosf_mbi_pmic_i2c_access_count == 0) {
+		iosf_mbi_reset_semaphore();
+		dev_dbg(&mbi_pdev->dev, "punit semaphore held for %ums\n",
+			jiffies_to_msecs(jiffies - iosf_mbi_sem_acquired));
+		do_wakeup = true;
+	}
+	mutex_unlock(&iosf_mbi_pmic_access_mutex);
+
+	if (do_wakeup)
+		wake_up(&iosf_mbi_pmic_access_waitq);
+}
+EXPORT_SYMBOL(iosf_mbi_unblock_punit_i2c_access);
+
+int iosf_mbi_register_pmic_bus_access_notifier(struct notifier_block *nb)
+{
+	int ret;
+
+	/* Wait for the bus to go inactive before registering */
+	iosf_mbi_punit_acquire();
+	ret = blocking_notifier_chain_register(
+				&iosf_mbi_pmic_bus_access_notifier, nb);
+	iosf_mbi_punit_release();
+
+	return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_register_pmic_bus_access_notifier);
+
+int iosf_mbi_unregister_pmic_bus_access_notifier_unlocked(
+	struct notifier_block *nb)
+{
+	iosf_mbi_assert_punit_acquired();
+
+	return blocking_notifier_chain_unregister(
+				&iosf_mbi_pmic_bus_access_notifier, nb);
+}
+EXPORT_SYMBOL(iosf_mbi_unregister_pmic_bus_access_notifier_unlocked);
+
+void iosf_mbi_assert_punit_acquired(void)
+{
+	WARN_ON(iosf_mbi_pmic_punit_access_count == 0);
+}
+EXPORT_SYMBOL(iosf_mbi_assert_punit_acquired);
+
+/**************** iosf_mbi debug code ****************/
+
+#ifdef CONFIG_IOSF_MBI_DEBUG
+static u32	dbg_mdr;
+static u32	dbg_mcr;
+static u32	dbg_mcrx;
+
+static int mcr_get(void *data, u64 *val)
+{
+	*val = *(u32 *)data;
+	return 0;
+}
+
+static int mcr_set(void *data, u64 val)
+{
+	u8 command = ((u32)val & 0xFF000000) >> 24,
+	   port	   = ((u32)val & 0x00FF0000) >> 16,
+	   offset  = ((u32)val & 0x0000FF00) >> 8;
+	int err;
+
+	*(u32 *)data = val;
+
+	if (!capable(CAP_SYS_RAWIO))
+		return -EACCES;
+
+	if (command & 1u)
+		err = iosf_mbi_write(port,
+			       command,
+			       dbg_mcrx | offset,
+			       dbg_mdr);
+	else
+		err = iosf_mbi_read(port,
+			      command,
+			      dbg_mcrx | offset,
+			      &dbg_mdr);
+
+	return err;
+}
+DEFINE_SIMPLE_ATTRIBUTE(iosf_mcr_fops, mcr_get, mcr_set , "%llx\n");
+
+static struct dentry *iosf_dbg;
+
+static void iosf_sideband_debug_init(void)
+{
+	iosf_dbg = debugfs_create_dir("iosf_sb", NULL);
+
+	/* mdr */
+	debugfs_create_x32("mdr", 0660, iosf_dbg, &dbg_mdr);
+
+	/* mcrx */
+	debugfs_create_x32("mcrx", 0660, iosf_dbg, &dbg_mcrx);
+
+	/* mcr - initiates mailbox transaction */
+	debugfs_create_file("mcr", 0660, iosf_dbg, &dbg_mcr, &iosf_mcr_fops);
+}
+
+static void iosf_debugfs_init(void)
+{
+	iosf_sideband_debug_init();
+}
+
+static void iosf_debugfs_remove(void)
+{
+	debugfs_remove_recursive(iosf_dbg);
+}
+#else
+static inline void iosf_debugfs_init(void) { }
+static inline void iosf_debugfs_remove(void) { }
+#endif /* CONFIG_IOSF_MBI_DEBUG */
+
+static int iosf_mbi_probe(struct pci_dev *pdev,
+			  const struct pci_device_id *dev_id)
+{
+	int ret;
+
+	ret = pci_enable_device(pdev);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "error: could not enable device\n");
+		return ret;
+	}
+
+	mbi_pdev = pci_dev_get(pdev);
+	iosf_mbi_sem_address = dev_id->driver_data;
+
+	return 0;
+}
+
+static const struct pci_device_id iosf_mbi_pci_ids[] = {
+	{ PCI_DEVICE_DATA(INTEL, BAYTRAIL, PUNIT_SEMAPHORE_BYT) },
+	{ PCI_DEVICE_DATA(INTEL, BRASWELL, PUNIT_SEMAPHORE_CHT) },
+	{ PCI_DEVICE_DATA(INTEL, QUARK_X1000, 0) },
+	{ PCI_DEVICE_DATA(INTEL, TANGIER, 0) },
+	{ 0, },
+};
+MODULE_DEVICE_TABLE(pci, iosf_mbi_pci_ids);
+
+static struct pci_driver iosf_mbi_pci_driver = {
+	.name		= "iosf_mbi_pci",
+	.probe		= iosf_mbi_probe,
+	.id_table	= iosf_mbi_pci_ids,
+};
+
+static int __init iosf_mbi_init(void)
+{
+	iosf_debugfs_init();
+
+	cpu_latency_qos_add_request(&iosf_mbi_pm_qos, PM_QOS_DEFAULT_VALUE);
+
+	return pci_register_driver(&iosf_mbi_pci_driver);
+}
+
+static void __exit iosf_mbi_exit(void)
+{
+	iosf_debugfs_remove();
+
+	pci_unregister_driver(&iosf_mbi_pci_driver);
+	pci_dev_put(mbi_pdev);
+	mbi_pdev = NULL;
+
+	cpu_latency_qos_remove_request(&iosf_mbi_pm_qos);
+}
+
+module_init(iosf_mbi_init);
+module_exit(iosf_mbi_exit);
+
+MODULE_AUTHOR("David E. Box <david.e.box@linux.intel.com>");
+MODULE_DESCRIPTION("IOSF Mailbox Interface accessor");
+MODULE_LICENSE("GPL v2");

linux-7.2.3/arch/x86/platform/iris/Makefile

@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_X86_32_IRIS)		+= iris.o

linux-7.2.3/arch/x86/platform/iris/iris.c

@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Eurobraille/Iris power off support.
+ *
+ * Eurobraille's Iris machine is a PC with no APM or ACPI support.
+ * It is shutdown by a special I/O sequence which this module provides.
+ *
+ *  Copyright (C) Shérab <Sebastien.Hinderer@ens-lyon.org>
+ */
+
+#include <linux/moduleparam.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/delay.h>
+#include <linux/pm.h>
+#include <asm/io.h>
+
+#define IRIS_GIO_BASE		0x340
+#define IRIS_GIO_INPUT		IRIS_GIO_BASE
+#define IRIS_GIO_OUTPUT		(IRIS_GIO_BASE + 1)
+#define IRIS_GIO_PULSE		0x80 /* First byte to send */
+#define IRIS_GIO_REST		0x00 /* Second byte to send */
+#define IRIS_GIO_NODEV		0xff /* Likely not an Iris */
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Sébastien Hinderer <Sebastien.Hinderer@ens-lyon.org>");
+MODULE_DESCRIPTION("A power_off handler for Iris devices from EuroBraille");
+
+static bool force;
+
+module_param(force, bool, 0);
+MODULE_PARM_DESC(force, "Set to one to force poweroff handler installation.");
+
+static void (*old_pm_power_off)(void);
+
+static void iris_power_off(void)
+{
+	outb(IRIS_GIO_PULSE, IRIS_GIO_OUTPUT);
+	msleep(850);
+	outb(IRIS_GIO_REST, IRIS_GIO_OUTPUT);
+}
+
+/*
+ * Before installing the power_off handler, try to make sure the OS is
+ * running on an Iris.  Since Iris does not support DMI, this is done
+ * by reading its input port and seeing whether the read value is
+ * meaningful.
+ */
+static int iris_probe(struct platform_device *pdev)
+{
+	unsigned char status = inb(IRIS_GIO_INPUT);
+	if (status == IRIS_GIO_NODEV) {
+		printk(KERN_ERR "This machine does not seem to be an Iris. "
+			"Power off handler not installed.\n");
+		return -ENODEV;
+	}
+	old_pm_power_off = pm_power_off;
+	pm_power_off = &iris_power_off;
+	printk(KERN_INFO "Iris power_off handler installed.\n");
+	return 0;
+}
+
+static void iris_remove(struct platform_device *pdev)
+{
+	pm_power_off = old_pm_power_off;
+	printk(KERN_INFO "Iris power_off handler uninstalled.\n");
+}
+
+static struct platform_driver iris_driver = {
+	.driver		= {
+		.name   = "iris",
+	},
+	.probe          = iris_probe,
+	.remove         = iris_remove,
+};
+
+static struct resource iris_resources[] = {
+	{
+		.start  = IRIS_GIO_BASE,
+		.end    = IRIS_GIO_OUTPUT,
+		.flags  = IORESOURCE_IO,
+		.name   = "address"
+	}
+};
+
+static struct platform_device *iris_device;
+
+static int iris_init(void)
+{
+	int ret;
+	if (force != 1) {
+		printk(KERN_ERR "The force parameter has not been set to 1."
+			" The Iris poweroff handler will not be installed.\n");
+		return -ENODEV;
+	}
+	ret = platform_driver_register(&iris_driver);
+	if (ret < 0) {
+		printk(KERN_ERR "Failed to register iris platform driver: %d\n",
+			ret);
+		return ret;
+	}
+	iris_device = platform_device_register_simple("iris", (-1),
+				iris_resources, ARRAY_SIZE(iris_resources));
+	if (IS_ERR(iris_device)) {
+		printk(KERN_ERR "Failed to register iris platform device\n");
+		platform_driver_unregister(&iris_driver);
+		return PTR_ERR(iris_device);
+	}
+	return 0;
+}
+
+static void iris_exit(void)
+{
+	platform_device_unregister(iris_device);
+	platform_driver_unregister(&iris_driver);
+}
+
+module_init(iris_init);
+module_exit(iris_exit);

linux-7.2.3/arch/x86/platform/olpc/Makefile

@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_OLPC)		+= olpc.o olpc_ofw.o olpc_dt.o
+obj-$(CONFIG_OLPC_XO1_PM)		+= olpc-xo1-pm.o xo1-wakeup.o
+obj-$(CONFIG_OLPC_XO1_RTC)	+= olpc-xo1-rtc.o
+obj-$(CONFIG_OLPC_XO1_SCI)	+= olpc-xo1-sci.o
+obj-$(CONFIG_OLPC_XO15_SCI)	+= olpc-xo15-sci.o

linux-7.2.3/arch/x86/platform/olpc/olpc-xo1-pm.c

@@ -0,0 +1,187 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Support for power management features of the OLPC XO-1 laptop
+ *
+ * Copyright (C) 2010 Andres Salomon <dilinger@queued.net>
+ * Copyright (C) 2010 One Laptop per Child
+ * Copyright (C) 2006 Red Hat, Inc.
+ * Copyright (C) 2006 Advanced Micro Devices, Inc.
+ */
+
+#include <linux/cs5535.h>
+#include <linux/platform_device.h>
+#include <linux/export.h>
+#include <linux/pm.h>
+#include <linux/suspend.h>
+#include <linux/olpc-ec.h>
+
+#include <asm/io.h>
+#include <asm/olpc.h>
+
+#define DRV_NAME "olpc-xo1-pm"
+
+static unsigned long acpi_base;
+static unsigned long pms_base;
+
+static u16 wakeup_mask = CS5536_PM_PWRBTN;
+
+static struct {
+	unsigned long address;
+	unsigned short segment;
+} ofw_bios_entry = { 0xF0000 + PAGE_OFFSET, __KERNEL_CS };
+
+/* Set bits in the wakeup mask */
+void olpc_xo1_pm_wakeup_set(u16 value)
+{
+	wakeup_mask |= value;
+}
+EXPORT_SYMBOL_GPL(olpc_xo1_pm_wakeup_set);
+
+/* Clear bits in the wakeup mask */
+void olpc_xo1_pm_wakeup_clear(u16 value)
+{
+	wakeup_mask &= ~value;
+}
+EXPORT_SYMBOL_GPL(olpc_xo1_pm_wakeup_clear);
+
+static int xo1_power_state_enter(suspend_state_t pm_state)
+{
+	unsigned long saved_sci_mask;
+
+	/* Only STR is supported */
+	if (pm_state != PM_SUSPEND_MEM)
+		return -EINVAL;
+
+	/*
+	 * Save SCI mask (this gets lost since PM1_EN is used as a mask for
+	 * wakeup events, which is not necessarily the same event set)
+	 */
+	saved_sci_mask = inl(acpi_base + CS5536_PM1_STS);
+	saved_sci_mask &= 0xffff0000;
+
+	/* Save CPU state */
+	do_olpc_suspend_lowlevel();
+
+	/* Resume path starts here */
+
+	/* Restore SCI mask (using dword access to CS5536_PM1_EN) */
+	outl(saved_sci_mask, acpi_base + CS5536_PM1_STS);
+
+	return 0;
+}
+
+asmlinkage __visible int xo1_do_sleep(u8 sleep_state)
+{
+	void *pgd_addr = __va(read_cr3_pa());
+
+	/* Program wakeup mask (using dword access to CS5536_PM1_EN) */
+	outl(wakeup_mask << 16, acpi_base + CS5536_PM1_STS);
+
+	__asm__("movl %0,%%eax" : : "r" (pgd_addr));
+	__asm__("call *(%%edi); cld"
+		: : "D" (&ofw_bios_entry));
+	__asm__("movb $0x34, %al\n\t"
+		"outb %al, $0x70\n\t"
+		"movb $0x30, %al\n\t"
+		"outb %al, $0x71\n\t");
+	return 0;
+}
+
+static void xo1_power_off(void)
+{
+	printk(KERN_INFO "OLPC XO-1 power off sequence...\n");
+
+	/* Enable all of these controls with 0 delay */
+	outl(0x40000000, pms_base + CS5536_PM_SCLK);
+	outl(0x40000000, pms_base + CS5536_PM_IN_SLPCTL);
+	outl(0x40000000, pms_base + CS5536_PM_WKXD);
+	outl(0x40000000, pms_base + CS5536_PM_WKD);
+
+	/* Clear status bits (possibly unnecessary) */
+	outl(0x0002ffff, pms_base  + CS5536_PM_SSC);
+	outl(0xffffffff, acpi_base + CS5536_PM_GPE0_STS);
+
+	/* Write SLP_EN bit to start the machinery */
+	outl(0x00002000, acpi_base + CS5536_PM1_CNT);
+}
+
+static int xo1_power_state_valid(suspend_state_t pm_state)
+{
+	/* suspend-to-RAM only */
+	return pm_state == PM_SUSPEND_MEM;
+}
+
+static const struct platform_suspend_ops xo1_suspend_ops = {
+	.valid = xo1_power_state_valid,
+	.enter = xo1_power_state_enter,
+};
+
+static int xo1_pm_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+
+	/* don't run on non-XOs */
+	if (!machine_is_olpc())
+		return -ENODEV;
+
+	res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+	if (!res) {
+		dev_err(&pdev->dev, "can't fetch device resource info\n");
+		return -EIO;
+	}
+	if (strcmp(pdev->name, "cs5535-pms") == 0)
+		pms_base = res->start;
+	else if (strcmp(pdev->name, "olpc-xo1-pm-acpi") == 0)
+		acpi_base = res->start;
+
+	/* If we have both addresses, we can override the poweroff hook */
+	if (pms_base && acpi_base) {
+		suspend_set_ops(&xo1_suspend_ops);
+		pm_power_off = xo1_power_off;
+		printk(KERN_INFO "OLPC XO-1 support registered\n");
+	}
+
+	return 0;
+}
+
+static void xo1_pm_remove(struct platform_device *pdev)
+{
+	if (strcmp(pdev->name, "cs5535-pms") == 0)
+		pms_base = 0;
+	else if (strcmp(pdev->name, "olpc-xo1-pm-acpi") == 0)
+		acpi_base = 0;
+
+	pm_power_off = NULL;
+}
+
+static struct platform_driver cs5535_pms_driver = {
+	.driver = {
+		.name = "cs5535-pms",
+	},
+	.probe = xo1_pm_probe,
+	.remove = xo1_pm_remove,
+};
+
+static struct platform_driver cs5535_acpi_driver = {
+	.driver = {
+		.name = "olpc-xo1-pm-acpi",
+	},
+	.probe = xo1_pm_probe,
+	.remove = xo1_pm_remove,
+};
+
+static int __init xo1_pm_init(void)
+{
+	int r;
+
+	r = platform_driver_register(&cs5535_pms_driver);
+	if (r)
+		return r;
+
+	r = platform_driver_register(&cs5535_acpi_driver);
+	if (r)
+		platform_driver_unregister(&cs5535_pms_driver);
+
+	return r;
+}
+arch_initcall(xo1_pm_init);

linux-7.2.3/arch/x86/platform/olpc/olpc-xo1-rtc.c

@@ -0,0 +1,80 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Support for OLPC XO-1 Real Time Clock (RTC)
+ *
+ * Copyright (C) 2011 One Laptop per Child
+ */
+
+#include <linux/mc146818rtc.h>
+#include <linux/platform_device.h>
+#include <linux/rtc.h>
+#include <linux/of.h>
+
+#include <asm/msr.h>
+#include <asm/olpc.h>
+#include <asm/x86_init.h>
+
+static void rtc_wake_on(struct device *dev)
+{
+	olpc_xo1_pm_wakeup_set(CS5536_PM_RTC);
+}
+
+static void rtc_wake_off(struct device *dev)
+{
+	olpc_xo1_pm_wakeup_clear(CS5536_PM_RTC);
+}
+
+static struct resource rtc_platform_resource[] = {
+	[0] = {
+		.start	= RTC_PORT(0),
+		.end	= RTC_PORT(1),
+		.flags	= IORESOURCE_IO,
+	},
+	[1] = {
+		.start	= RTC_IRQ,
+		.end	= RTC_IRQ,
+		.flags	= IORESOURCE_IRQ,
+	}
+};
+
+static struct cmos_rtc_board_info rtc_info = {
+	.rtc_day_alarm = 0,
+	.rtc_mon_alarm = 0,
+	.rtc_century = 0,
+	.wake_on = rtc_wake_on,
+	.wake_off = rtc_wake_off,
+};
+
+static struct platform_device xo1_rtc_device = {
+	.name = "rtc_cmos",
+	.id = -1,
+	.num_resources = ARRAY_SIZE(rtc_platform_resource),
+	.dev.platform_data = &rtc_info,
+	.resource = rtc_platform_resource,
+};
+
+static int __init xo1_rtc_init(void)
+{
+	int r;
+	struct device_node *node;
+
+	node = of_find_compatible_node(NULL, NULL, "olpc,xo1-rtc");
+	if (!node)
+		return 0;
+	of_node_put(node);
+
+	pr_info("olpc-xo1-rtc: Initializing OLPC XO-1 RTC\n");
+	rdmsrq(MSR_RTC_DOMA_OFFSET, rtc_info.rtc_day_alarm);
+	rdmsrq(MSR_RTC_MONA_OFFSET, rtc_info.rtc_mon_alarm);
+	rdmsrq(MSR_RTC_CEN_OFFSET, rtc_info.rtc_century);
+
+	r = platform_device_register(&xo1_rtc_device);
+	if (r)
+		return r;
+
+	x86_platform.legacy.rtc = 0;
+
+	device_init_wakeup(&xo1_rtc_device.dev, 1);
+	return 0;
+}
+arch_initcall(xo1_rtc_init);

linux-7.2.3/arch/x86/platform/olpc/olpc-xo1-sci.c

@@ -0,0 +1,627 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Support for OLPC XO-1 System Control Interrupts (SCI)
+ *
+ * Copyright (C) 2010 One Laptop per Child
+ * Copyright (C) 2006 Red Hat, Inc.
+ * Copyright (C) 2006 Advanced Micro Devices, Inc.
+ */
+
+#include <linux/cs5535.h>
+#include <linux/device.h>
+#include <linux/gpio.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/pm.h>
+#include <linux/power_supply.h>
+#include <linux/suspend.h>
+#include <linux/workqueue.h>
+#include <linux/olpc-ec.h>
+
+#include <asm/io.h>
+#include <asm/msr.h>
+#include <asm/olpc.h>
+
+#define DRV_NAME	"olpc-xo1-sci"
+#define PFX		DRV_NAME ": "
+
+static unsigned long acpi_base;
+static struct input_dev *power_button_idev;
+static struct input_dev *ebook_switch_idev;
+static struct input_dev *lid_switch_idev;
+
+static int sci_irq;
+
+static bool lid_open;
+static bool lid_inverted;
+static int lid_wake_mode;
+
+enum lid_wake_modes {
+	LID_WAKE_ALWAYS,
+	LID_WAKE_OPEN,
+	LID_WAKE_CLOSE,
+};
+
+static const char * const lid_wake_mode_names[] = {
+	[LID_WAKE_ALWAYS] = "always",
+	[LID_WAKE_OPEN] = "open",
+	[LID_WAKE_CLOSE] = "close",
+};
+
+static void battery_status_changed(void)
+{
+	struct power_supply *psy = power_supply_get_by_name("olpc_battery");
+
+	if (psy) {
+		power_supply_changed(psy);
+		power_supply_put(psy);
+	}
+}
+
+static void ac_status_changed(void)
+{
+	struct power_supply *psy = power_supply_get_by_name("olpc_ac");
+
+	if (psy) {
+		power_supply_changed(psy);
+		power_supply_put(psy);
+	}
+}
+
+/* Report current ebook switch state through input layer */
+static void send_ebook_state(void)
+{
+	unsigned char state;
+
+	if (olpc_ec_cmd(EC_READ_EB_MODE, NULL, 0, &state, 1)) {
+		pr_err(PFX "failed to get ebook state\n");
+		return;
+	}
+
+	if (test_bit(SW_TABLET_MODE, ebook_switch_idev->sw) == !!state)
+		return; /* Nothing new to report. */
+
+	input_report_switch(ebook_switch_idev, SW_TABLET_MODE, state);
+	input_sync(ebook_switch_idev);
+	pm_wakeup_event(&ebook_switch_idev->dev, 0);
+}
+
+static void flip_lid_inverter(void)
+{
+	/* gpio is high; invert so we'll get l->h event interrupt */
+	if (lid_inverted)
+		cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_INPUT_INVERT);
+	else
+		cs5535_gpio_set(OLPC_GPIO_LID, GPIO_INPUT_INVERT);
+	lid_inverted = !lid_inverted;
+}
+
+static void detect_lid_state(void)
+{
+	/*
+	 * the edge detector hookup on the gpio inputs on the geode is
+	 * odd, to say the least.  See http://dev.laptop.org/ticket/5703
+	 * for details, but in a nutshell:  we don't use the edge
+	 * detectors.  instead, we make use of an anomaly:  with the both
+	 * edge detectors turned off, we still get an edge event on a
+	 * positive edge transition.  to take advantage of this, we use the
+	 * front-end inverter to ensure that that's the edge we're always
+	 * going to see next.
+	 */
+
+	int state;
+
+	state = cs5535_gpio_isset(OLPC_GPIO_LID, GPIO_READ_BACK);
+	lid_open = !state ^ !lid_inverted; /* x ^^ y */
+	if (!state)
+		return;
+
+	flip_lid_inverter();
+}
+
+/* Report current lid switch state through input layer */
+static void send_lid_state(void)
+{
+	if (!!test_bit(SW_LID, lid_switch_idev->sw) == !lid_open)
+		return; /* Nothing new to report. */
+
+	input_report_switch(lid_switch_idev, SW_LID, !lid_open);
+	input_sync(lid_switch_idev);
+	pm_wakeup_event(&lid_switch_idev->dev, 0);
+}
+
+static ssize_t lid_wake_mode_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	const char *mode = lid_wake_mode_names[lid_wake_mode];
+	return sprintf(buf, "%s\n", mode);
+}
+static ssize_t lid_wake_mode_set(struct device *dev,
+				 struct device_attribute *attr,
+				 const char *buf, size_t count)
+{
+	int i;
+	for (i = 0; i < ARRAY_SIZE(lid_wake_mode_names); i++) {
+		const char *mode = lid_wake_mode_names[i];
+		if (strlen(mode) != count || strncasecmp(mode, buf, count))
+			continue;
+
+		lid_wake_mode = i;
+		return count;
+	}
+	return -EINVAL;
+}
+static DEVICE_ATTR(lid_wake_mode, S_IWUSR | S_IRUGO, lid_wake_mode_show,
+		   lid_wake_mode_set);
+
+static struct attribute *lid_attrs[] = {
+	&dev_attr_lid_wake_mode.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(lid);
+
+/*
+ * Process all items in the EC's SCI queue.
+ *
+ * This is handled in a workqueue because olpc_ec_cmd can be slow (and
+ * can even timeout).
+ *
+ * If propagate_events is false, the queue is drained without events being
+ * generated for the interrupts.
+ */
+static void process_sci_queue(bool propagate_events)
+{
+	int r;
+	u16 data;
+
+	do {
+		r = olpc_ec_sci_query(&data);
+		if (r || !data)
+			break;
+
+		pr_debug(PFX "SCI 0x%x received\n", data);
+
+		switch (data) {
+		case EC_SCI_SRC_BATERR:
+		case EC_SCI_SRC_BATSOC:
+		case EC_SCI_SRC_BATTERY:
+		case EC_SCI_SRC_BATCRIT:
+			battery_status_changed();
+			break;
+		case EC_SCI_SRC_ACPWR:
+			ac_status_changed();
+			break;
+		}
+
+		if (data == EC_SCI_SRC_EBOOK && propagate_events)
+			send_ebook_state();
+	} while (data);
+
+	if (r)
+		pr_err(PFX "Failed to clear SCI queue");
+}
+
+static void process_sci_queue_work(struct work_struct *work)
+{
+	process_sci_queue(true);
+}
+
+static DECLARE_WORK(sci_work, process_sci_queue_work);
+
+static irqreturn_t xo1_sci_intr(int irq, void *dev_id)
+{
+	struct platform_device *pdev = dev_id;
+	u32 sts;
+	u32 gpe;
+
+	sts = inl(acpi_base + CS5536_PM1_STS);
+	outl(sts | 0xffff, acpi_base + CS5536_PM1_STS);
+
+	gpe = inl(acpi_base + CS5536_PM_GPE0_STS);
+	outl(0xffffffff, acpi_base + CS5536_PM_GPE0_STS);
+
+	dev_dbg(&pdev->dev, "sts %x gpe %x\n", sts, gpe);
+
+	if (sts & CS5536_PWRBTN_FLAG) {
+		if (!(sts & CS5536_WAK_FLAG)) {
+			/* Only report power button input when it was pressed
+			 * during regular operation (as opposed to when it
+			 * was used to wake the system). */
+			input_report_key(power_button_idev, KEY_POWER, 1);
+			input_sync(power_button_idev);
+			input_report_key(power_button_idev, KEY_POWER, 0);
+			input_sync(power_button_idev);
+		}
+		/* Report the wakeup event in all cases. */
+		pm_wakeup_event(&power_button_idev->dev, 0);
+	}
+
+	if ((sts & (CS5536_RTC_FLAG | CS5536_WAK_FLAG)) ==
+			(CS5536_RTC_FLAG | CS5536_WAK_FLAG)) {
+		/* When the system is woken by the RTC alarm, report the
+		 * event on the rtc device. */
+		struct device *rtc = bus_find_device_by_name(
+			&platform_bus_type, NULL, "rtc_cmos");
+		if (rtc) {
+			pm_wakeup_event(rtc, 0);
+			put_device(rtc);
+		}
+	}
+
+	if (gpe & CS5536_GPIOM7_PME_FLAG) { /* EC GPIO */
+		cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_NEGATIVE_EDGE_STS);
+		schedule_work(&sci_work);
+	}
+
+	cs5535_gpio_set(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_STS);
+	cs5535_gpio_set(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_STS);
+	detect_lid_state();
+	send_lid_state();
+
+	return IRQ_HANDLED;
+}
+
+static int xo1_sci_suspend(struct platform_device *pdev, pm_message_t state)
+{
+	if (device_may_wakeup(&power_button_idev->dev))
+		olpc_xo1_pm_wakeup_set(CS5536_PM_PWRBTN);
+	else
+		olpc_xo1_pm_wakeup_clear(CS5536_PM_PWRBTN);
+
+	if (device_may_wakeup(&ebook_switch_idev->dev))
+		olpc_ec_wakeup_set(EC_SCI_SRC_EBOOK);
+	else
+		olpc_ec_wakeup_clear(EC_SCI_SRC_EBOOK);
+
+	if (!device_may_wakeup(&lid_switch_idev->dev)) {
+		cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+	} else if ((lid_open && lid_wake_mode == LID_WAKE_OPEN) ||
+		   (!lid_open && lid_wake_mode == LID_WAKE_CLOSE)) {
+		flip_lid_inverter();
+
+		/* we may have just caused an event */
+		cs5535_gpio_set(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_STS);
+		cs5535_gpio_set(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_STS);
+
+		cs5535_gpio_set(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+	}
+
+	return 0;
+}
+
+static int xo1_sci_resume(struct platform_device *pdev)
+{
+	/*
+	 * We don't know what may have happened while we were asleep.
+	 * Reestablish our lid setup so we're sure to catch all transitions.
+	 */
+	detect_lid_state();
+	send_lid_state();
+	cs5535_gpio_set(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+
+	/* Enable all EC events */
+	olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+	/* Power/battery status might have changed too */
+	battery_status_changed();
+	ac_status_changed();
+	return 0;
+}
+
+static int setup_sci_interrupt(struct platform_device *pdev)
+{
+	u32 lo, hi;
+	u32 sts;
+	int r;
+
+	rdmsr(0x51400020, lo, hi);
+	sci_irq = (lo >> 20) & 15;
+
+	if (sci_irq) {
+		dev_info(&pdev->dev, "SCI is mapped to IRQ %d\n", sci_irq);
+	} else {
+		/* Zero means masked */
+		dev_info(&pdev->dev, "SCI unmapped. Mapping to IRQ 3\n");
+		sci_irq = 3;
+		lo |= 0x00300000;
+		wrmsrq(0x51400020, lo);
+	}
+
+	/* Select level triggered in PIC */
+	if (sci_irq < 8) {
+		lo = inb(CS5536_PIC_INT_SEL1);
+		lo |= 1 << sci_irq;
+		outb(lo, CS5536_PIC_INT_SEL1);
+	} else {
+		lo = inb(CS5536_PIC_INT_SEL2);
+		lo |= 1 << (sci_irq - 8);
+		outb(lo, CS5536_PIC_INT_SEL2);
+	}
+
+	/* Enable interesting SCI events, and clear pending interrupts */
+	sts = inl(acpi_base + CS5536_PM1_STS);
+	outl(((CS5536_PM_PWRBTN | CS5536_PM_RTC) << 16) | 0xffff,
+	     acpi_base + CS5536_PM1_STS);
+
+	r = request_irq(sci_irq, xo1_sci_intr, 0, DRV_NAME, pdev);
+	if (r)
+		dev_err(&pdev->dev, "can't request interrupt\n");
+
+	return r;
+}
+
+static int setup_ec_sci(void)
+{
+	int r;
+
+	r = gpio_request(OLPC_GPIO_ECSCI, "OLPC-ECSCI");
+	if (r)
+		return r;
+
+	gpio_direction_input(OLPC_GPIO_ECSCI);
+
+	/* Clear pending EC SCI events */
+	cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_NEGATIVE_EDGE_STS);
+	cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_POSITIVE_EDGE_STS);
+
+	/*
+	 * Enable EC SCI events, and map them to both a PME and the SCI
+	 * interrupt.
+	 *
+	 * Ordinarily, in addition to functioning as GPIOs, Geode GPIOs can
+	 * be mapped to regular interrupts *or* Geode-specific Power
+	 * Management Events (PMEs) - events that bring the system out of
+	 * suspend. In this case, we want both of those things - the system
+	 * wakeup, *and* the ability to get an interrupt when an event occurs.
+	 *
+	 * To achieve this, we map the GPIO to a PME, and then we use one
+	 * of the many generic knobs on the CS5535 PIC to additionally map the
+	 * PME to the regular SCI interrupt line.
+	 */
+	cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_EVENTS_ENABLE);
+
+	/* Set the SCI to cause a PME event on group 7 */
+	cs5535_gpio_setup_event(OLPC_GPIO_ECSCI, 7, 1);
+
+	/* And have group 7 also fire the SCI interrupt */
+	cs5535_pic_unreqz_select_high(7, sci_irq);
+
+	return 0;
+}
+
+static void free_ec_sci(void)
+{
+	gpio_free(OLPC_GPIO_ECSCI);
+}
+
+static int setup_lid_events(void)
+{
+	int r;
+
+	r = gpio_request(OLPC_GPIO_LID, "OLPC-LID");
+	if (r)
+		return r;
+
+	gpio_direction_input(OLPC_GPIO_LID);
+
+	cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_INPUT_INVERT);
+	lid_inverted = 0;
+
+	/* Clear edge detection and event enable for now */
+	cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+	cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_EN);
+	cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_EN);
+	cs5535_gpio_set(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_STS);
+	cs5535_gpio_set(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_STS);
+
+	/* Set the LID to cause an PME event on group 6 */
+	cs5535_gpio_setup_event(OLPC_GPIO_LID, 6, 1);
+
+	/* Set PME group 6 to fire the SCI interrupt */
+	cs5535_gpio_set_irq(6, sci_irq);
+
+	/* Enable the event */
+	cs5535_gpio_set(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+
+	return 0;
+}
+
+static void free_lid_events(void)
+{
+	gpio_free(OLPC_GPIO_LID);
+}
+
+static int setup_power_button(struct platform_device *pdev)
+{
+	int r;
+
+	power_button_idev = input_allocate_device();
+	if (!power_button_idev)
+		return -ENOMEM;
+
+	power_button_idev->name = "Power Button";
+	power_button_idev->phys = DRV_NAME "/input0";
+	set_bit(EV_KEY, power_button_idev->evbit);
+	set_bit(KEY_POWER, power_button_idev->keybit);
+
+	power_button_idev->dev.parent = &pdev->dev;
+	device_init_wakeup(&power_button_idev->dev, 1);
+
+	r = input_register_device(power_button_idev);
+	if (r) {
+		dev_err(&pdev->dev, "failed to register power button: %d\n", r);
+		input_free_device(power_button_idev);
+	}
+
+	return r;
+}
+
+static void free_power_button(void)
+{
+	input_unregister_device(power_button_idev);
+}
+
+static int setup_ebook_switch(struct platform_device *pdev)
+{
+	int r;
+
+	ebook_switch_idev = input_allocate_device();
+	if (!ebook_switch_idev)
+		return -ENOMEM;
+
+	ebook_switch_idev->name = "EBook Switch";
+	ebook_switch_idev->phys = DRV_NAME "/input1";
+	set_bit(EV_SW, ebook_switch_idev->evbit);
+	set_bit(SW_TABLET_MODE, ebook_switch_idev->swbit);
+
+	ebook_switch_idev->dev.parent = &pdev->dev;
+	device_set_wakeup_capable(&ebook_switch_idev->dev, true);
+
+	r = input_register_device(ebook_switch_idev);
+	if (r) {
+		dev_err(&pdev->dev, "failed to register ebook switch: %d\n", r);
+		input_free_device(ebook_switch_idev);
+	}
+
+	return r;
+}
+
+static void free_ebook_switch(void)
+{
+	input_unregister_device(ebook_switch_idev);
+}
+
+static int setup_lid_switch(struct platform_device *pdev)
+{
+	int r;
+
+	lid_switch_idev = input_allocate_device();
+	if (!lid_switch_idev)
+		return -ENOMEM;
+
+	lid_switch_idev->name = "Lid Switch";
+	lid_switch_idev->phys = DRV_NAME "/input2";
+	set_bit(EV_SW, lid_switch_idev->evbit);
+	set_bit(SW_LID, lid_switch_idev->swbit);
+
+	lid_switch_idev->dev.parent = &pdev->dev;
+	device_set_wakeup_capable(&lid_switch_idev->dev, true);
+
+	r = input_register_device(lid_switch_idev);
+	if (r) {
+		dev_err(&pdev->dev, "failed to register lid switch: %d\n", r);
+		goto err_register;
+	}
+
+	return 0;
+
+err_register:
+	input_free_device(lid_switch_idev);
+	return r;
+}
+
+static void free_lid_switch(void)
+{
+	input_unregister_device(lid_switch_idev);
+}
+
+static int xo1_sci_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	int r;
+
+	/* don't run on non-XOs */
+	if (!machine_is_olpc())
+		return -ENODEV;
+
+	res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+	if (!res) {
+		dev_err(&pdev->dev, "can't fetch device resource info\n");
+		return -EIO;
+	}
+	acpi_base = res->start;
+
+	r = setup_power_button(pdev);
+	if (r)
+		return r;
+
+	r = setup_ebook_switch(pdev);
+	if (r)
+		goto err_ebook;
+
+	r = setup_lid_switch(pdev);
+	if (r)
+		goto err_lid;
+
+	r = setup_lid_events();
+	if (r)
+		goto err_lidevt;
+
+	r = setup_ec_sci();
+	if (r)
+		goto err_ecsci;
+
+	/* Enable PME generation for EC-generated events */
+	outl(CS5536_GPIOM6_PME_EN | CS5536_GPIOM7_PME_EN,
+		acpi_base + CS5536_PM_GPE0_EN);
+
+	/* Clear pending events */
+	outl(0xffffffff, acpi_base + CS5536_PM_GPE0_STS);
+	process_sci_queue(false);
+
+	/* Initial sync */
+	send_ebook_state();
+	detect_lid_state();
+	send_lid_state();
+
+	r = setup_sci_interrupt(pdev);
+	if (r)
+		goto err_sci;
+
+	/* Enable all EC events */
+	olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+	return r;
+
+err_sci:
+	free_ec_sci();
+err_ecsci:
+	free_lid_events();
+err_lidevt:
+	free_lid_switch();
+err_lid:
+	free_ebook_switch();
+err_ebook:
+	free_power_button();
+	return r;
+}
+
+static void xo1_sci_remove(struct platform_device *pdev)
+{
+	free_irq(sci_irq, pdev);
+	cancel_work_sync(&sci_work);
+	free_ec_sci();
+	free_lid_events();
+	free_lid_switch();
+	free_ebook_switch();
+	free_power_button();
+	acpi_base = 0;
+}
+
+static struct platform_driver xo1_sci_driver = {
+	.driver = {
+		.name = "olpc-xo1-sci-acpi",
+		.dev_groups = lid_groups,
+	},
+	.probe = xo1_sci_probe,
+	.remove = xo1_sci_remove,
+	.suspend = xo1_sci_suspend,
+	.resume = xo1_sci_resume,
+};
+
+static int __init xo1_sci_init(void)
+{
+	return platform_driver_register(&xo1_sci_driver);
+}
+arch_initcall(xo1_sci_init);

linux-7.2.3/arch/x86/platform/olpc/olpc-xo15-sci.c

@@ -0,0 +1,236 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Support for OLPC XO-1.5 System Control Interrupts (SCI)
+ *
+ * Copyright (C) 2009-2010 One Laptop per Child
+ */
+
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/workqueue.h>
+#include <linux/platform_device.h>
+#include <linux/power_supply.h>
+#include <linux/olpc-ec.h>
+
+#include <linux/acpi.h>
+#include <asm/olpc.h>
+
+#define DRV_NAME			"olpc-xo15-sci"
+#define PFX				DRV_NAME ": "
+#define XO15_SCI_CLASS			DRV_NAME
+#define XO15_SCI_DEVICE_NAME		"OLPC XO-1.5 SCI"
+
+static unsigned long			xo15_sci_gpe;
+static bool				lid_wake_on_close;
+
+/*
+ * The normal ACPI LID wakeup behavior is wake-on-open, but not
+ * wake-on-close. This is implemented as standard by the XO-1.5 DSDT.
+ *
+ * We provide here a sysfs attribute that will additionally enable
+ * wake-on-close behavior. This is useful (e.g.) when we opportunistically
+ * suspend with the display running; if the lid is then closed, we want to
+ * wake up to turn the display off.
+ *
+ * This is controlled through a custom method in the XO-1.5 DSDT.
+ */
+static int set_lid_wake_behavior(bool wake_on_close)
+{
+	acpi_status status;
+
+	status = acpi_execute_simple_method(NULL, "\\_SB.PCI0.LID.LIDW", wake_on_close);
+	if (ACPI_FAILURE(status)) {
+		pr_warn(PFX "failed to set lid behavior\n");
+		return 1;
+	}
+
+	lid_wake_on_close = wake_on_close;
+
+	return 0;
+}
+
+static ssize_t
+lid_wake_on_close_show(struct kobject *s, struct kobj_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%u\n", lid_wake_on_close);
+}
+
+static ssize_t lid_wake_on_close_store(struct kobject *s,
+				       struct kobj_attribute *attr,
+				       const char *buf, size_t n)
+{
+	unsigned int val;
+
+	if (sscanf(buf, "%u", &val) != 1)
+		return -EINVAL;
+
+	set_lid_wake_behavior(!!val);
+
+	return n;
+}
+
+static struct kobj_attribute lid_wake_on_close_attr =
+	__ATTR(lid_wake_on_close, 0644,
+	       lid_wake_on_close_show,
+	       lid_wake_on_close_store);
+
+static void battery_status_changed(void)
+{
+	struct power_supply *psy = power_supply_get_by_name("olpc_battery");
+
+	if (psy) {
+		power_supply_changed(psy);
+		power_supply_put(psy);
+	}
+}
+
+static void ac_status_changed(void)
+{
+	struct power_supply *psy = power_supply_get_by_name("olpc_ac");
+
+	if (psy) {
+		power_supply_changed(psy);
+		power_supply_put(psy);
+	}
+}
+
+static void process_sci_queue(void)
+{
+	u16 data;
+	int r;
+
+	do {
+		r = olpc_ec_sci_query(&data);
+		if (r || !data)
+			break;
+
+		pr_debug(PFX "SCI 0x%x received\n", data);
+
+		switch (data) {
+		case EC_SCI_SRC_BATERR:
+		case EC_SCI_SRC_BATSOC:
+		case EC_SCI_SRC_BATTERY:
+		case EC_SCI_SRC_BATCRIT:
+			battery_status_changed();
+			break;
+		case EC_SCI_SRC_ACPWR:
+			ac_status_changed();
+			break;
+		}
+	} while (data);
+
+	if (r)
+		pr_err(PFX "Failed to clear SCI queue");
+}
+
+static void process_sci_queue_work(struct work_struct *work)
+{
+	process_sci_queue();
+}
+
+static DECLARE_WORK(sci_work, process_sci_queue_work);
+
+static u32 xo15_sci_gpe_handler(acpi_handle gpe_device, u32 gpe, void *context)
+{
+	schedule_work(&sci_work);
+	return ACPI_INTERRUPT_HANDLED | ACPI_REENABLE_GPE;
+}
+
+static int xo15_sci_probe(struct platform_device *pdev)
+{
+	struct acpi_device *device;
+	unsigned long long tmp;
+	acpi_status status;
+	int r;
+
+	device = ACPI_COMPANION(&pdev->dev);
+	if (!device)
+		return -ENODEV;
+
+	strscpy(acpi_device_name(device), XO15_SCI_DEVICE_NAME);
+	strscpy(acpi_device_class(device), XO15_SCI_CLASS);
+
+	/* Get GPE bit assignment (EC events). */
+	status = acpi_evaluate_integer(device->handle, "_GPE", NULL, &tmp);
+	if (ACPI_FAILURE(status))
+		return -EINVAL;
+
+	xo15_sci_gpe = tmp;
+	status = acpi_install_gpe_handler(NULL, xo15_sci_gpe,
+					  ACPI_GPE_EDGE_TRIGGERED,
+					  xo15_sci_gpe_handler, device);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	dev_info(&pdev->dev, "Initialized, GPE = 0x%lx\n", xo15_sci_gpe);
+
+	r = sysfs_create_file(&device->dev.kobj, &lid_wake_on_close_attr.attr);
+	if (r)
+		goto err_sysfs;
+
+	/* Flush queue, and enable all SCI events */
+	process_sci_queue();
+	olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+	acpi_enable_gpe(NULL, xo15_sci_gpe);
+
+	/* Enable wake-on-EC */
+	if (device->wakeup.flags.valid)
+		device_init_wakeup(&pdev->dev, true);
+
+	return 0;
+
+err_sysfs:
+	acpi_remove_gpe_handler(NULL, xo15_sci_gpe, xo15_sci_gpe_handler);
+	cancel_work_sync(&sci_work);
+	return r;
+}
+
+static void xo15_sci_remove(struct platform_device *pdev)
+{
+	struct acpi_device *device = ACPI_COMPANION(&pdev->dev);
+
+	device_init_wakeup(&pdev->dev, false);
+	acpi_disable_gpe(NULL, xo15_sci_gpe);
+	acpi_remove_gpe_handler(NULL, xo15_sci_gpe, xo15_sci_gpe_handler);
+	cancel_work_sync(&sci_work);
+	sysfs_remove_file(&device->dev.kobj, &lid_wake_on_close_attr.attr);
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int xo15_sci_resume(struct device *dev)
+{
+	/* Enable all EC events */
+	olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+	/* Power/battery status might have changed */
+	battery_status_changed();
+	ac_status_changed();
+
+	return 0;
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(xo15_sci_pm, NULL, xo15_sci_resume);
+
+static const struct acpi_device_id xo15_sci_device_ids[] = {
+	{"XO15EC", 0},
+	{"", 0},
+};
+
+static struct platform_driver xo15_sci_drv = {
+	.probe = xo15_sci_probe,
+	.remove = xo15_sci_remove,
+	.driver = {
+		.name = DRV_NAME,
+		.acpi_match_table = xo15_sci_device_ids,
+		.pm = &xo15_sci_pm,
+	},
+};
+
+static int __init xo15_sci_init(void)
+{
+	return platform_driver_register(&xo15_sci_drv);
+}
+device_initcall(xo15_sci_init);

linux-7.2.3/arch/x86/platform/olpc/olpc.c

@@ -0,0 +1,321 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Support for the OLPC DCON and OLPC EC access
+ *
+ * Copyright © 2006  Advanced Micro Devices, Inc.
+ * Copyright © 2007-2008  Andres Salomon <dilinger@debian.org>
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/export.h>
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/platform_device.h>
+#include <linux/of.h>
+#include <linux/syscore_ops.h>
+#include <linux/mutex.h>
+#include <linux/olpc-ec.h>
+
+#include <asm/geode.h>
+#include <asm/setup.h>
+#include <asm/olpc.h>
+#include <asm/olpc_ofw.h>
+
+struct olpc_platform_t olpc_platform_info;
+EXPORT_SYMBOL_GPL(olpc_platform_info);
+
+/* what the timeout *should* be (in ms) */
+#define EC_BASE_TIMEOUT 20
+
+/* the timeout that bugs in the EC might force us to actually use */
+static int ec_timeout = EC_BASE_TIMEOUT;
+
+static int __init olpc_ec_timeout_set(char *str)
+{
+	if (get_option(&str, &ec_timeout) != 1) {
+		ec_timeout = EC_BASE_TIMEOUT;
+		printk(KERN_ERR "olpc-ec:  invalid argument to "
+				"'olpc_ec_timeout=', ignoring!\n");
+	}
+	printk(KERN_DEBUG "olpc-ec:  using %d ms delay for EC commands.\n",
+			ec_timeout);
+	return 1;
+}
+__setup("olpc_ec_timeout=", olpc_ec_timeout_set);
+
+/*
+ * These {i,o}bf_status functions return whether the buffers are full or not.
+ */
+
+static inline unsigned int ibf_status(unsigned int port)
+{
+	return !!(inb(port) & 0x02);
+}
+
+static inline unsigned int obf_status(unsigned int port)
+{
+	return inb(port) & 0x01;
+}
+
+#define wait_on_ibf(p, d) __wait_on_ibf(__LINE__, (p), (d))
+static int __wait_on_ibf(unsigned int line, unsigned int port, int desired)
+{
+	unsigned int timeo;
+	int state = ibf_status(port);
+
+	for (timeo = ec_timeout; state != desired && timeo; timeo--) {
+		mdelay(1);
+		state = ibf_status(port);
+	}
+
+	if ((state == desired) && (ec_timeout > EC_BASE_TIMEOUT) &&
+			timeo < (ec_timeout - EC_BASE_TIMEOUT)) {
+		printk(KERN_WARNING "olpc-ec:  %d: waited %u ms for IBF!\n",
+				line, ec_timeout - timeo);
+	}
+
+	return !(state == desired);
+}
+
+#define wait_on_obf(p, d) __wait_on_obf(__LINE__, (p), (d))
+static int __wait_on_obf(unsigned int line, unsigned int port, int desired)
+{
+	unsigned int timeo;
+	int state = obf_status(port);
+
+	for (timeo = ec_timeout; state != desired && timeo; timeo--) {
+		mdelay(1);
+		state = obf_status(port);
+	}
+
+	if ((state == desired) && (ec_timeout > EC_BASE_TIMEOUT) &&
+			timeo < (ec_timeout - EC_BASE_TIMEOUT)) {
+		printk(KERN_WARNING "olpc-ec:  %d: waited %u ms for OBF!\n",
+				line, ec_timeout - timeo);
+	}
+
+	return !(state == desired);
+}
+
+/*
+ * This allows the kernel to run Embedded Controller commands.  The EC is
+ * documented at <http://wiki.laptop.org/go/Embedded_controller>, and the
+ * available EC commands are here:
+ * <http://wiki.laptop.org/go/Ec_specification>.  Unfortunately, while
+ * OpenFirmware's source is available, the EC's is not.
+ */
+static int olpc_xo1_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf,
+		size_t outlen, void *arg)
+{
+	int ret = -EIO;
+	int i;
+	int restarts = 0;
+
+	/* Clear OBF */
+	for (i = 0; i < 10 && (obf_status(0x6c) == 1); i++)
+		inb(0x68);
+	if (i == 10) {
+		printk(KERN_ERR "olpc-ec:  timeout while attempting to "
+				"clear OBF flag!\n");
+		goto err;
+	}
+
+	if (wait_on_ibf(0x6c, 0)) {
+		printk(KERN_ERR "olpc-ec:  timeout waiting for EC to "
+				"quiesce!\n");
+		goto err;
+	}
+
+restart:
+	/*
+	 * Note that if we time out during any IBF checks, that's a failure;
+	 * we have to return.  There's no way for the kernel to clear that.
+	 *
+	 * If we time out during an OBF check, we can restart the command;
+	 * reissuing it will clear the OBF flag, and we should be alright.
+	 * The OBF flag will sometimes misbehave due to what we believe
+	 * is a hardware quirk..
+	 */
+	pr_devel("olpc-ec:  running cmd 0x%x\n", cmd);
+	outb(cmd, 0x6c);
+
+	if (wait_on_ibf(0x6c, 0)) {
+		printk(KERN_ERR "olpc-ec:  timeout waiting for EC to read "
+				"command!\n");
+		goto err;
+	}
+
+	if (inbuf && inlen) {
+		/* write data to EC */
+		for (i = 0; i < inlen; i++) {
+			pr_devel("olpc-ec:  sending cmd arg 0x%x\n", inbuf[i]);
+			outb(inbuf[i], 0x68);
+			if (wait_on_ibf(0x6c, 0)) {
+				printk(KERN_ERR "olpc-ec:  timeout waiting for"
+						" EC accept data!\n");
+				goto err;
+			}
+		}
+	}
+	if (outbuf && outlen) {
+		/* read data from EC */
+		for (i = 0; i < outlen; i++) {
+			if (wait_on_obf(0x6c, 1)) {
+				printk(KERN_ERR "olpc-ec:  timeout waiting for"
+						" EC to provide data!\n");
+				if (restarts++ < 10)
+					goto restart;
+				goto err;
+			}
+			outbuf[i] = inb(0x68);
+			pr_devel("olpc-ec:  received 0x%x\n", outbuf[i]);
+		}
+	}
+
+	ret = 0;
+err:
+	return ret;
+}
+
+static bool __init check_ofw_architecture(struct device_node *root)
+{
+	const char *olpc_arch;
+	int propsize;
+
+	olpc_arch = of_get_property(root, "architecture", &propsize);
+	return propsize == 5 && strncmp("OLPC", olpc_arch, 5) == 0;
+}
+
+static u32 __init get_board_revision(struct device_node *root)
+{
+	int propsize;
+	const __be32 *rev;
+
+	rev = of_get_property(root, "board-revision-int", &propsize);
+	if (propsize != 4)
+		return 0;
+
+	return be32_to_cpu(*rev);
+}
+
+static bool __init platform_detect(void)
+{
+	struct device_node *root = of_find_node_by_path("/");
+	bool success;
+
+	if (!root)
+		return false;
+
+	success = check_ofw_architecture(root);
+	if (success) {
+		olpc_platform_info.boardrev = get_board_revision(root);
+		olpc_platform_info.flags |= OLPC_F_PRESENT;
+
+		pr_info("OLPC board revision %s%X\n",
+			((olpc_platform_info.boardrev & 0xf) < 8) ? "pre" : "",
+			olpc_platform_info.boardrev >> 4);
+	}
+
+	of_node_put(root);
+	return success;
+}
+
+static int __init add_xo1_platform_devices(void)
+{
+	struct platform_device *pdev;
+
+	pdev = platform_device_register_simple("xo1-rfkill", -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	pdev = platform_device_register_simple("olpc-xo1", -1, NULL, 0);
+
+	return PTR_ERR_OR_ZERO(pdev);
+}
+
+static int olpc_xo1_ec_suspend(struct platform_device *pdev)
+{
+	/*
+	 * Squelch SCIs while suspended.  This is a fix for
+	 * <http://dev.laptop.org/ticket/1835>.
+	 */
+	return olpc_ec_cmd(EC_SET_SCI_INHIBIT, NULL, 0, NULL, 0);
+}
+
+static int olpc_xo1_ec_resume(struct platform_device *pdev)
+{
+	/* Tell the EC to stop inhibiting SCIs */
+	olpc_ec_cmd(EC_SET_SCI_INHIBIT_RELEASE, NULL, 0, NULL, 0);
+
+	/*
+	 * Tell the wireless module to restart USB communication.
+	 * Must be done twice.
+	 */
+	olpc_ec_cmd(EC_WAKE_UP_WLAN, NULL, 0, NULL, 0);
+	olpc_ec_cmd(EC_WAKE_UP_WLAN, NULL, 0, NULL, 0);
+
+	return 0;
+}
+
+static struct olpc_ec_driver ec_xo1_driver = {
+	.suspend = olpc_xo1_ec_suspend,
+	.resume = olpc_xo1_ec_resume,
+	.ec_cmd = olpc_xo1_ec_cmd,
+#ifdef CONFIG_OLPC_XO1_SCI
+	/*
+	 * XO-1 EC wakeups are available when olpc-xo1-sci driver is
+	 * compiled in
+	 */
+	.wakeup_available = true,
+#endif
+};
+
+static struct olpc_ec_driver ec_xo1_5_driver = {
+	.ec_cmd = olpc_xo1_ec_cmd,
+#ifdef CONFIG_OLPC_XO15_SCI
+	/*
+	 * XO-1.5 EC wakeups are available when olpc-xo15-sci driver is
+	 * compiled in
+	 */
+	.wakeup_available = true,
+#endif
+};
+
+static int __init olpc_init(void)
+{
+	int r = 0;
+
+	if (!olpc_ofw_present() || !platform_detect())
+		return 0;
+
+	/* register the XO-1 and 1.5-specific EC handler */
+	if (olpc_platform_info.boardrev < olpc_board_pre(0xd0))	/* XO-1 */
+		olpc_ec_driver_register(&ec_xo1_driver, NULL);
+	else
+		olpc_ec_driver_register(&ec_xo1_5_driver, NULL);
+	platform_device_register_simple("olpc-ec", -1, NULL, 0);
+
+	/* assume B1 and above models always have a DCON */
+	if (olpc_board_at_least(olpc_board(0xb1)))
+		olpc_platform_info.flags |= OLPC_F_DCON;
+
+#ifdef CONFIG_PCI_OLPC
+	/* If the VSA exists let it emulate PCI, if not emulate in kernel.
+	 * XO-1 only. */
+	if (olpc_platform_info.boardrev < olpc_board_pre(0xd0) &&
+			!cs5535_has_vsa2())
+		x86_init.pci.arch_init = pci_olpc_init;
+#endif
+
+	if (olpc_platform_info.boardrev < olpc_board_pre(0xd0)) { /* XO-1 */
+		r = add_xo1_platform_devices();
+		if (r)
+			return r;
+	}
+
+	return 0;
+}
+
+postcore_initcall(olpc_init);

linux-7.2.3/arch/x86/platform/olpc/olpc_dt.c

@@ -0,0 +1,321 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * OLPC-specific OFW device tree support code.
+ *
+ * Paul Mackerras	August 1996.
+ * Copyright (C) 1996-2005 Paul Mackerras.
+ *
+ *  Adapted for 64bit PowerPC by Dave Engebretsen and Peter Bergner.
+ *    {engebret|bergner}@us.ibm.com
+ *
+ *  Adapted for sparc by David S. Miller davem@davemloft.net
+ *  Adapted for x86/OLPC by Andres Salomon <dilinger@queued.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/memblock.h>
+#include <linux/of.h>
+#include <linux/of_pdt.h>
+#include <asm/olpc.h>
+#include <asm/olpc_ofw.h>
+
+static phandle __init olpc_dt_getsibling(phandle node)
+{
+	const void *args[] = { (void *)node };
+	void *res[] = { &node };
+
+	if ((s32)node == -1)
+		return 0;
+
+	if (olpc_ofw("peer", args, res) || (s32)node == -1)
+		return 0;
+
+	return node;
+}
+
+static phandle __init olpc_dt_getchild(phandle node)
+{
+	const void *args[] = { (void *)node };
+	void *res[] = { &node };
+
+	if ((s32)node == -1)
+		return 0;
+
+	if (olpc_ofw("child", args, res) || (s32)node == -1) {
+		pr_err("PROM: %s: fetching child failed!\n", __func__);
+		return 0;
+	}
+
+	return node;
+}
+
+static int __init olpc_dt_getproplen(phandle node, const char *prop)
+{
+	const void *args[] = { (void *)node, prop };
+	int len;
+	void *res[] = { &len };
+
+	if ((s32)node == -1)
+		return -1;
+
+	if (olpc_ofw("getproplen", args, res)) {
+		pr_err("PROM: %s: getproplen failed!\n", __func__);
+		return -1;
+	}
+
+	return len;
+}
+
+static int __init olpc_dt_getproperty(phandle node, const char *prop,
+		char *buf, int bufsize)
+{
+	int plen;
+
+	plen = olpc_dt_getproplen(node, prop);
+	if (plen > bufsize || plen < 1) {
+		return -1;
+	} else {
+		const void *args[] = { (void *)node, prop, buf, (void *)plen };
+		void *res[] = { &plen };
+
+		if (olpc_ofw("getprop", args, res)) {
+			pr_err("PROM: %s: getprop failed!\n", __func__);
+			return -1;
+		}
+	}
+
+	return plen;
+}
+
+static int __init olpc_dt_nextprop(phandle node, char *prev, char *buf)
+{
+	const void *args[] = { (void *)node, prev, buf };
+	int success;
+	void *res[] = { &success };
+
+	buf[0] = '\0';
+
+	if ((s32)node == -1)
+		return -1;
+
+	if (olpc_ofw("nextprop", args, res) || success != 1)
+		return -1;
+
+	return 0;
+}
+
+static int __init olpc_dt_pkg2path(phandle node, char *buf,
+		const int buflen, int *len)
+{
+	const void *args[] = { (void *)node, buf, (void *)buflen };
+	void *res[] = { len };
+
+	if ((s32)node == -1)
+		return -1;
+
+	if (olpc_ofw("package-to-path", args, res) || *len < 1)
+		return -1;
+
+	return 0;
+}
+
+static unsigned int prom_early_allocated __initdata;
+
+void * __init prom_early_alloc(unsigned long size)
+{
+	static u8 *mem;
+	static size_t free_mem;
+	void *res;
+
+	if (free_mem < size) {
+		const size_t chunk_size = max(PAGE_SIZE, size);
+
+		/*
+		 * To minimize the number of allocations, grab at least
+		 * PAGE_SIZE of memory (that's an arbitrary choice that's
+		 * fast enough on the platforms we care about while minimizing
+		 * wasted bootmem) and hand off chunks of it to callers.
+		 */
+		res = memblock_alloc_or_panic(chunk_size, SMP_CACHE_BYTES);
+		prom_early_allocated += chunk_size;
+		memset(res, 0, chunk_size);
+		free_mem = chunk_size;
+		mem = res;
+	}
+
+	/* allocate from the local cache */
+	free_mem -= size;
+	res = mem;
+	mem += size;
+	return res;
+}
+
+static struct of_pdt_ops prom_olpc_ops __initdata = {
+	.nextprop = olpc_dt_nextprop,
+	.getproplen = olpc_dt_getproplen,
+	.getproperty = olpc_dt_getproperty,
+	.getchild = olpc_dt_getchild,
+	.getsibling = olpc_dt_getsibling,
+	.pkg2path = olpc_dt_pkg2path,
+};
+
+static phandle __init olpc_dt_finddevice(const char *path)
+{
+	phandle node;
+	const void *args[] = { path };
+	void *res[] = { &node };
+
+	if (olpc_ofw("finddevice", args, res)) {
+		pr_err("olpc_dt: finddevice failed!\n");
+		return 0;
+	}
+
+	if ((s32) node == -1)
+		return 0;
+
+	return node;
+}
+
+static int __init olpc_dt_interpret(const char *words)
+{
+	int result;
+	const void *args[] = { words };
+	void *res[] = { &result };
+
+	if (olpc_ofw("interpret", args, res)) {
+		pr_err("olpc_dt: interpret failed!\n");
+		return -1;
+	}
+
+	return result;
+}
+
+/*
+ * Extract board revision directly from OFW device tree.
+ * We can't use olpc_platform_info because that hasn't been set up yet.
+ */
+static u32 __init olpc_dt_get_board_revision(void)
+{
+	phandle node;
+	__be32 rev;
+	int r;
+
+	node = olpc_dt_finddevice("/");
+	if (!node)
+		return 0;
+
+	r = olpc_dt_getproperty(node, "board-revision-int",
+				(char *) &rev, sizeof(rev));
+	if (r < 0)
+		return 0;
+
+	return be32_to_cpu(rev);
+}
+
+static int __init olpc_dt_compatible_match(phandle node, const char *compat)
+{
+	char buf[64], *p;
+	int plen;
+
+	plen = olpc_dt_getproperty(node, "compatible", buf, sizeof(buf));
+	if (plen <= 0)
+		return 0;
+
+	for (p = buf; p < buf + plen; p += strlen(p) + 1) {
+		if (strcmp(p, compat) == 0)
+			return 1;
+	}
+
+	return 0;
+}
+
+static void __init olpc_dt_fixup(void)
+{
+	phandle node;
+	u32 board_rev;
+
+	node = olpc_dt_finddevice("/battery@0");
+	if (!node)
+		return;
+
+	board_rev = olpc_dt_get_board_revision();
+	if (!board_rev)
+		return;
+
+	if (board_rev >= olpc_board_pre(0xd0)) {
+		/* XO-1.5 */
+
+		if (olpc_dt_compatible_match(node, "olpc,xo1.5-battery"))
+			return;
+
+		/* Add olpc,xo1.5-battery compatible marker to battery node */
+		olpc_dt_interpret("\" /battery@0\" find-device");
+		olpc_dt_interpret("  \" olpc,xo1.5-battery\" +compatible");
+		olpc_dt_interpret("device-end");
+
+		if (olpc_dt_compatible_match(node, "olpc,xo1-battery")) {
+			/*
+			 * If we have a olpc,xo1-battery compatible, then we're
+			 * running a new enough firmware that already has
+			 * the dcon node.
+			 */
+			return;
+		}
+
+		/* Add dcon device */
+		olpc_dt_interpret("\" /pci/display@1\" find-device");
+		olpc_dt_interpret("  new-device");
+		olpc_dt_interpret("    \" dcon\" device-name");
+		olpc_dt_interpret("    \" olpc,xo1-dcon\" +compatible");
+		olpc_dt_interpret("  finish-device");
+		olpc_dt_interpret("device-end");
+	} else {
+		/* XO-1 */
+
+		if (olpc_dt_compatible_match(node, "olpc,xo1-battery")) {
+			/*
+			 * If we have a olpc,xo1-battery compatible, then we're
+			 * running a new enough firmware that already has
+			 * the dcon and RTC nodes.
+			 */
+			return;
+		}
+
+		/* Add dcon device, mark RTC as olpc,xo1-rtc */
+		olpc_dt_interpret("\" /pci/display@1,1\" find-device");
+		olpc_dt_interpret("  new-device");
+		olpc_dt_interpret("    \" dcon\" device-name");
+		olpc_dt_interpret("    \" olpc,xo1-dcon\" +compatible");
+		olpc_dt_interpret("  finish-device");
+		olpc_dt_interpret("device-end");
+
+		olpc_dt_interpret("\" /rtc\" find-device");
+		olpc_dt_interpret(" \" olpc,xo1-rtc\" +compatible");
+		olpc_dt_interpret("device-end");
+	}
+
+	/* Add olpc,xo1-battery compatible marker to battery node */
+	olpc_dt_interpret("\" /battery@0\" find-device");
+	olpc_dt_interpret("  \" olpc,xo1-battery\" +compatible");
+	olpc_dt_interpret("device-end");
+}
+
+void __init olpc_dt_build_devicetree(void)
+{
+	phandle root;
+
+	if (!olpc_ofw_is_installed())
+		return;
+
+	olpc_dt_fixup();
+
+	root = olpc_dt_getsibling(0);
+	if (!root) {
+		pr_err("PROM: unable to get root node from OFW!\n");
+		return;
+	}
+	of_pdt_build_devicetree(root, &prom_olpc_ops);
+
+	pr_info("PROM DT: Built device tree with %u bytes of memory.\n",
+			prom_early_allocated);
+}

linux-7.2.3/arch/x86/platform/olpc/olpc_ofw.c

@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/spinlock_types.h>
+#include <linux/init.h>
+#include <linux/pgtable.h>
+#include <asm/page.h>
+#include <asm/setup.h>
+#include <asm/io.h>
+#include <asm/cpufeature.h>
+#include <asm/special_insns.h>
+#include <asm/olpc_ofw.h>
+
+/* address of OFW callback interface; will be NULL if OFW isn't found */
+static int (*olpc_ofw_cif)(int *);
+
+/* page dir entry containing OFW's pgdir table; filled in by head_32.S */
+u32 olpc_ofw_pgd __initdata;
+
+static DEFINE_SPINLOCK(ofw_lock);
+
+#define MAXARGS 10
+
+void __init setup_olpc_ofw_pgd(void)
+{
+	pgd_t *base, *ofw_pde;
+
+	if (!olpc_ofw_cif)
+		return;
+
+	/* fetch OFW's PDE */
+	base = early_ioremap(olpc_ofw_pgd, sizeof(olpc_ofw_pgd) * PTRS_PER_PGD);
+	if (!base) {
+		printk(KERN_ERR "failed to remap OFW's pgd - disabling OFW!\n");
+		olpc_ofw_cif = NULL;
+		return;
+	}
+	ofw_pde = &base[OLPC_OFW_PDE_NR];
+
+	/* install OFW's PDE permanently into the kernel's pgtable */
+	set_pgd(&swapper_pg_dir[OLPC_OFW_PDE_NR], *ofw_pde);
+	/* implicit optimization barrier here due to uninline function return */
+
+	early_iounmap(base, sizeof(olpc_ofw_pgd) * PTRS_PER_PGD);
+}
+
+int __olpc_ofw(const char *name, int nr_args, const void **args, int nr_res,
+		void **res)
+{
+	int ofw_args[MAXARGS + 3];
+	unsigned long flags;
+	int ret, i, *p;
+
+	BUG_ON(nr_args + nr_res > MAXARGS);
+
+	if (!olpc_ofw_cif)
+		return -EIO;
+
+	ofw_args[0] = (int)name;
+	ofw_args[1] = nr_args;
+	ofw_args[2] = nr_res;
+
+	p = &ofw_args[3];
+	for (i = 0; i < nr_args; i++, p++)
+		*p = (int)args[i];
+
+	/* call into ofw */
+	spin_lock_irqsave(&ofw_lock, flags);
+	ret = olpc_ofw_cif(ofw_args);
+	spin_unlock_irqrestore(&ofw_lock, flags);
+
+	if (!ret) {
+		for (i = 0; i < nr_res; i++, p++)
+			*((int *)res[i]) = *p;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(__olpc_ofw);
+
+bool olpc_ofw_present(void)
+{
+	return olpc_ofw_cif != NULL;
+}
+EXPORT_SYMBOL_GPL(olpc_ofw_present);
+
+/* OFW cif _should_ be above this address */
+#define OFW_MIN 0xff000000
+
+/* OFW starts on a 1MB boundary */
+#define OFW_BOUND (1<<20)
+
+void __init olpc_ofw_detect(void)
+{
+	struct olpc_ofw_header *hdr = &boot_params.olpc_ofw_header;
+	unsigned long start;
+
+	/* ensure OFW booted us by checking for "OFW " string */
+	if (hdr->ofw_magic != OLPC_OFW_SIG)
+		return;
+
+	olpc_ofw_cif = (int (*)(int *))hdr->cif_handler;
+
+	if ((unsigned long)olpc_ofw_cif < OFW_MIN) {
+		printk(KERN_ERR "OFW detected, but cif has invalid address 0x%lx - disabling.\n",
+				(unsigned long)olpc_ofw_cif);
+		olpc_ofw_cif = NULL;
+		return;
+	}
+
+	/* determine where OFW starts in memory */
+	start = round_down((unsigned long)olpc_ofw_cif, OFW_BOUND);
+	printk(KERN_INFO "OFW detected in memory, cif @ 0x%lx (reserving top %ldMB)\n",
+			(unsigned long)olpc_ofw_cif, (-start) >> 20);
+	reserve_top_address(-start);
+}
+
+bool __init olpc_ofw_is_installed(void)
+{
+	return olpc_ofw_cif != NULL;
+}

linux-7.2.3/arch/x86/platform/olpc/xo1-wakeup.S

@@ -0,0 +1,126 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+.text
+#include <linux/linkage.h>
+#include <asm/segment.h>
+#include <asm/page.h>
+#include <asm/pgtable_32.h>
+
+	.macro writepost,value
+		movb $0x34, %al
+		outb %al, $0x70
+		movb $\value, %al
+		outb %al, $0x71
+	.endm
+
+wakeup_start:
+	# OFW lands us here, running in protected mode, with a
+	# kernel-compatible GDT already setup.
+
+	# Clear any dangerous flags
+	pushl $0
+	popfl
+
+	writepost 0x31
+
+	# Set up %cr3
+	movl $initial_page_table - __PAGE_OFFSET, %eax
+	movl %eax, %cr3
+
+	movl saved_cr4, %eax
+	movl %eax, %cr4
+
+	movl saved_cr0, %eax
+	movl %eax, %cr0
+
+	# Control registers were modified, pipeline resync is needed
+	jmp 1f
+1:
+
+	movw    $__KERNEL_DS, %ax
+	movw    %ax, %ss
+	movw    %ax, %ds
+	movw    %ax, %es
+	movw    %ax, %fs
+	movw    %ax, %gs
+
+	lgdt    saved_gdt
+	lidt    saved_idt
+	lldt    saved_ldt
+	ljmp    $(__KERNEL_CS),$1f
+1:
+	movl    %cr3, %eax
+	movl    %eax, %cr3
+	wbinvd
+
+	# Go back to the return point
+	jmp ret_point
+
+save_registers:
+	sgdt  saved_gdt
+	sidt  saved_idt
+	sldt  saved_ldt
+
+	pushl %edx
+	movl %cr4, %edx
+	movl %edx, saved_cr4
+
+	movl %cr0, %edx
+	movl %edx, saved_cr0
+
+	popl %edx
+
+	movl %ebx, saved_context_ebx
+	movl %ebp, saved_context_ebp
+	movl %esi, saved_context_esi
+	movl %edi, saved_context_edi
+
+	pushfl
+	popl saved_context_eflags
+
+	RET
+
+restore_registers:
+	movl saved_context_ebp, %ebp
+	movl saved_context_ebx, %ebx
+	movl saved_context_esi, %esi
+	movl saved_context_edi, %edi
+
+	pushl saved_context_eflags
+	popfl
+
+	RET
+
+SYM_CODE_START(do_olpc_suspend_lowlevel)
+	call	save_processor_state
+	call	save_registers
+
+	# This is the stack context we want to remember
+	movl %esp, saved_context_esp
+
+	pushl	$3
+	call	xo1_do_sleep
+
+	jmp	wakeup_start
+	.p2align 4,,7
+ret_point:
+	movl    saved_context_esp, %esp
+
+	writepost 0x32
+
+	call	restore_registers
+	call	restore_processor_state
+	RET
+SYM_CODE_END(do_olpc_suspend_lowlevel)
+
+.data
+saved_gdt:             .long   0,0
+saved_idt:             .long   0,0
+saved_ldt:             .long   0
+saved_cr4:             .long   0
+saved_cr0:             .long   0
+saved_context_esp:     .long   0
+saved_context_edi:     .long   0
+saved_context_esi:     .long   0
+saved_context_ebx:     .long   0
+saved_context_ebp:     .long   0
+saved_context_eflags:  .long   0

linux-7.2.3/arch/x86/platform/pvh/Makefile

@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0
+OBJECT_FILES_NON_STANDARD_head.o := y
+KASAN_SANITIZE := n
+
+obj-$(CONFIG_PVH) += enlighten.o
+obj-$(CONFIG_PVH) += head.o

linux-7.2.3/arch/x86/platform/pvh/enlighten.c

@@ -0,0 +1,137 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/acpi.h>
+
+#include <xen/hvc-console.h>
+
+#include <asm/bootparam.h>
+#include <asm/io_apic.h>
+#include <asm/hypervisor.h>
+#include <asm/e820/api.h>
+#include <asm/x86_init.h>
+
+#include <asm/xen/interface.h>
+
+#include <xen/xen.h>
+#include <xen/interface/hvm/start_info.h>
+
+/*
+ * PVH variables.
+ *
+ * pvh_bootparams and pvh_start_info need to live in a data segment since
+ * they are used after startup_{32|64}, which clear .bss, are invoked.
+ */
+struct boot_params __initdata pvh_bootparams;
+struct hvm_start_info __initdata pvh_start_info;
+
+const unsigned int __initconst pvh_start_info_sz = sizeof(pvh_start_info);
+
+/*
+ * Xen guests are able to obtain the memory map from the hypervisor via the
+ * HYPERVISOR_memory_op hypercall.
+ * If we are trying to boot a Xen PVH guest, it is expected that the kernel
+ * will have been configured to provide an override for this routine to do
+ * just that.
+ */
+void __init __weak mem_map_via_hcall(struct boot_params *ptr __maybe_unused)
+{
+	xen_raw_printk("Error: Could not find memory map\n");
+	BUG();
+}
+
+static void __init init_pvh_bootparams(bool xen_guest)
+{
+	if ((pvh_start_info.version > 0) && (pvh_start_info.memmap_entries)) {
+		struct hvm_memmap_table_entry *ep;
+		int i;
+
+		ep = __va(pvh_start_info.memmap_paddr);
+		pvh_bootparams.e820_entries = pvh_start_info.memmap_entries;
+
+		for (i = 0; i < pvh_bootparams.e820_entries ; i++, ep++) {
+			pvh_bootparams.e820_table[i].addr = ep->addr;
+			pvh_bootparams.e820_table[i].size = ep->size;
+			pvh_bootparams.e820_table[i].type = ep->type;
+		}
+	} else if (xen_guest) {
+		mem_map_via_hcall(&pvh_bootparams);
+	} else {
+		/* Non-xen guests are not supported by version 0 */
+		BUG();
+	}
+
+	if (pvh_bootparams.e820_entries < E820_MAX_ENTRIES_ZEROPAGE - 1) {
+		pvh_bootparams.e820_table[pvh_bootparams.e820_entries].addr =
+			ISA_START_ADDRESS;
+		pvh_bootparams.e820_table[pvh_bootparams.e820_entries].size =
+			ISA_END_ADDRESS - ISA_START_ADDRESS;
+		pvh_bootparams.e820_table[pvh_bootparams.e820_entries].type =
+			E820_TYPE_RESERVED;
+		pvh_bootparams.e820_entries++;
+	} else
+		xen_raw_printk("Warning: Can fit ISA range into e820\n");
+
+	pvh_bootparams.hdr.cmd_line_ptr =
+		pvh_start_info.cmdline_paddr;
+
+	/* The first module is always ramdisk. */
+	if (pvh_start_info.nr_modules) {
+		struct hvm_modlist_entry *modaddr =
+			__va(pvh_start_info.modlist_paddr);
+		pvh_bootparams.hdr.ramdisk_image = modaddr->paddr;
+		pvh_bootparams.hdr.ramdisk_size = modaddr->size;
+	}
+
+	/*
+	 * See Documentation/arch/x86/boot.rst.
+	 *
+	 * Version 2.12 supports Xen entry point but we will use default x86/PC
+	 * environment (i.e. hardware_subarch 0).
+	 */
+	pvh_bootparams.hdr.version = (2 << 8) | 12;
+	pvh_bootparams.hdr.type_of_loader = ((xen_guest ? 0x9 : 0xb) << 4) | 0;
+
+	pvh_bootparams.acpi_rsdp_addr = pvh_start_info.rsdp_paddr;
+}
+
+/*
+ * If we are trying to boot a Xen PVH guest, it is expected that the kernel
+ * will have been configured to provide the required override for this routine.
+ */
+void __init __weak xen_pvh_init(struct boot_params *boot_params)
+{
+	xen_raw_printk("Error: Missing xen PVH initialization\n");
+	BUG();
+}
+
+static void __init hypervisor_specific_init(bool xen_guest)
+{
+	if (xen_guest)
+		xen_pvh_init(&pvh_bootparams);
+}
+
+/*
+ * This routine (and those that it might call) should not use
+ * anything that lives in .bss since that segment will be cleared later.
+ */
+void __init xen_prepare_pvh(void)
+{
+
+	u32 msr = xen_cpuid_base();
+	bool xen_guest = !!msr;
+
+	if (pvh_start_info.magic != XEN_HVM_START_MAGIC_VALUE) {
+		xen_raw_printk("Error: Unexpected magic value (0x%08x)\n",
+				pvh_start_info.magic);
+		BUG();
+	}
+
+	/*
+	 * This must not compile to "call memset" because memset() may be
+	 * instrumented.
+	 */
+	__builtin_memset(&pvh_bootparams, 0, sizeof(pvh_bootparams));
+
+	hypervisor_specific_init(xen_guest);
+
+	init_pvh_bootparams(xen_guest);
+}

linux-7.2.3/arch/x86/platform/pvh/head.S

@@ -0,0 +1,313 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+/*
+ * Copyright C 2016, Oracle and/or its affiliates. All rights reserved.
+ */
+
+	.code32
+	.text
+#ifdef CONFIG_X86_32
+#define _pa(x)          ((x) - __START_KERNEL_map)
+#endif
+#define rva(x)          ((x) - pvh_start_xen)
+
+#include <linux/elfnote.h>
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <asm/desc_defs.h>
+#include <asm/segment.h>
+#include <asm/asm.h>
+#include <asm/boot.h>
+#include <asm/pgtable.h>
+#include <asm/processor-flags.h>
+#include <asm/msr.h>
+#include <asm/nospec-branch.h>
+#include <xen/interface/elfnote.h>
+
+	__INIT
+
+/*
+ * Entry point for PVH guests.
+ *
+ * Xen ABI specifies the following register state when we come here:
+ *
+ * - `ebx`: contains the physical memory address where the loader has placed
+ *          the boot start info structure.
+ * - `cr0`: bit 0 (PE) must be set. All the other writeable bits are cleared.
+ * - `cr4`: all bits are cleared.
+ * - `cs `: must be a 32-bit read/execute code segment with a base of `0`
+ *          and a limit of `0xFFFFFFFF`. The selector value is unspecified.
+ * - `ds`, `es`: must be a 32-bit read/write data segment with a base of
+ *               `0` and a limit of `0xFFFFFFFF`. The selector values are all
+ *               unspecified.
+ * - `tr`: must be a 32-bit TSS (active) with a base of '0' and a limit
+ *         of '0x67'.
+ * - `eflags`: bit 17 (VM) must be cleared. Bit 9 (IF) must be cleared.
+ *             Bit 8 (TF) must be cleared. Other bits are all unspecified.
+ *
+ * All other processor registers and flag bits are unspecified. The OS is in
+ * charge of setting up its own stack, GDT and IDT.
+ */
+
+#define PVH_GDT_ENTRY_CS	1
+#define PVH_GDT_ENTRY_DS	2
+#define PVH_CS_SEL		(PVH_GDT_ENTRY_CS * 8)
+#define PVH_DS_SEL		(PVH_GDT_ENTRY_DS * 8)
+
+SYM_CODE_START(pvh_start_xen)
+	UNWIND_HINT_END_OF_STACK
+	cld
+
+	/*
+	 * See the comment for startup_32 for more details.  We need to
+	 * execute a call to get the execution address to be position
+	 * independent, but we don't have a stack.  Save and restore the
+	 * magic field of start_info in ebx, and use that as the stack.
+	 */
+	mov  (%ebx), %eax
+	leal 4(%ebx), %esp
+	ANNOTATE_INTRA_FUNCTION_CALL
+	call 1f
+1:	popl %ebp
+	mov  %eax, (%ebx)
+	subl $rva(1b), %ebp
+	movl $0, %esp
+
+	leal rva(gdt)(%ebp), %eax
+	addl %eax, 2(%eax)
+	lgdt (%eax)
+
+	mov $PVH_DS_SEL,%eax
+	mov %eax,%ds
+	mov %eax,%es
+	mov %eax,%ss
+
+	/* Stash hvm_start_info. */
+	leal rva(pvh_start_info)(%ebp), %edi
+	mov %ebx, %esi
+	movl rva(pvh_start_info_sz)(%ebp), %ecx
+	shr $2,%ecx
+	rep movsl
+
+	leal rva(early_stack_end)(%ebp), %esp
+
+#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
+	/* Enable PAE mode. */
+	mov %cr4, %eax
+	orl $X86_CR4_PAE, %eax
+	mov %eax, %cr4
+#endif
+
+#ifdef CONFIG_X86_64
+	/* Enable Long mode. */
+	mov $MSR_EFER, %ecx
+	rdmsr
+	btsl $_EFER_LME, %eax
+	wrmsr
+
+	/*
+	 * Reuse the non-relocatable symbol emitted for the ELF note to
+	 * subtract the build time physical address of pvh_start_xen() from
+	 * its actual runtime address, without relying on absolute 32-bit ELF
+	 * relocations, as these are not supported by the linker when running
+	 * in -pie mode, and should be avoided in .head.text in general.
+	 */
+	mov %ebp, %ebx
+	subl rva(xen_elfnote_phys32_entry)(%ebp), %ebx
+	jz .Lpagetable_done
+
+	/*
+	 * Store the resulting load offset in phys_base.  __pa() needs
+	 * phys_base set to calculate the hypercall page in xen_pvh_init().
+	 */
+	movl %ebx, rva(phys_base)(%ebp)
+
+	/* Fixup page-tables for relocation. */
+	leal rva(pvh_init_top_pgt)(%ebp), %edi
+	movl $PTRS_PER_PGD, %ecx
+2:
+	testl $_PAGE_PRESENT, 0x00(%edi)
+	jz 1f
+	addl %ebx, 0x00(%edi)
+1:
+	addl $8, %edi
+	decl %ecx
+	jnz 2b
+
+	/* L3 ident has a single entry. */
+	leal rva(pvh_level3_ident_pgt)(%ebp), %edi
+	addl %ebx, 0x00(%edi)
+
+	leal rva(pvh_level3_kernel_pgt)(%ebp), %edi
+	addl %ebx, (PAGE_SIZE - 16)(%edi)
+	addl %ebx, (PAGE_SIZE - 8)(%edi)
+
+	/* pvh_level2_ident_pgt is fine - large pages */
+
+	/* pvh_level2_kernel_pgt needs adjustment - large pages */
+	leal rva(pvh_level2_kernel_pgt)(%ebp), %edi
+	movl $PTRS_PER_PMD, %ecx
+2:
+	testl $_PAGE_PRESENT, 0x00(%edi)
+	jz 1f
+	addl %ebx, 0x00(%edi)
+1:
+	addl $8, %edi
+	decl %ecx
+	jnz 2b
+
+.Lpagetable_done:
+	/* Enable pre-constructed page tables. */
+	leal rva(pvh_init_top_pgt)(%ebp), %eax
+	mov %eax, %cr3
+	mov $(X86_CR0_PG | X86_CR0_PE), %eax
+	mov %eax, %cr0
+
+	/* Jump to 64-bit mode. */
+	pushl $PVH_CS_SEL
+	leal  rva(1f)(%ebp), %eax
+	pushl %eax
+	lretl
+
+	/* 64-bit entry point. */
+	.code64
+1:
+	UNWIND_HINT_END_OF_STACK
+
+	/*
+	 * Set up GSBASE.
+	 * Note that on SMP the boot CPU uses the init data section until
+	 * the per-CPU areas are set up.
+	 */
+	movl $MSR_GS_BASE,%ecx
+	xorl %eax, %eax
+	xorl %edx, %edx
+	wrmsr
+
+	/* Call xen_prepare_pvh() via the kernel virtual mapping */
+	leaq xen_prepare_pvh(%rip), %rax
+	subq phys_base(%rip), %rax
+	addq $__START_KERNEL_map, %rax
+	ANNOTATE_RETPOLINE_SAFE
+	call *%rax
+
+	/* startup_64 expects boot_params in %rsi. */
+	lea pvh_bootparams(%rip), %rsi
+	jmp startup_64
+
+#else /* CONFIG_X86_64 */
+
+	call mk_early_pgtbl_32
+
+	mov $_pa(initial_page_table), %eax
+	mov %eax, %cr3
+
+	mov %cr0, %eax
+	or $(X86_CR0_PG | X86_CR0_PE), %eax
+	mov %eax, %cr0
+
+	ljmp $PVH_CS_SEL, $1f
+1:
+	call xen_prepare_pvh
+	mov $_pa(pvh_bootparams), %esi
+
+	/* startup_32 doesn't expect paging and PAE to be on. */
+	ljmp $PVH_CS_SEL, $_pa(2f)
+2:
+	mov %cr0, %eax
+	and $~X86_CR0_PG, %eax
+	mov %eax, %cr0
+	mov %cr4, %eax
+	and $~X86_CR4_PAE, %eax
+	mov %eax, %cr4
+
+	ljmp $PVH_CS_SEL, $_pa(startup_32)
+#endif
+SYM_CODE_END(pvh_start_xen)
+
+	.section ".init.data","aw"
+	.balign 8
+SYM_DATA_START_LOCAL(gdt)
+	.word gdt_end - gdt_start - 1
+	.long gdt_start - gdt
+	.word 0
+SYM_DATA_END(gdt)
+SYM_DATA_START_LOCAL(gdt_start)
+	.quad 0x0000000000000000            /* NULL descriptor */
+#ifdef CONFIG_X86_64
+	.quad GDT_ENTRY(DESC_CODE64, 0, 0xfffff) /* PVH_CS_SEL */
+#else
+	.quad GDT_ENTRY(DESC_CODE32, 0, 0xfffff) /* PVH_CS_SEL */
+#endif
+	.quad GDT_ENTRY(DESC_DATA32, 0, 0xfffff) /* PVH_DS_SEL */
+SYM_DATA_END_LABEL(gdt_start, SYM_L_LOCAL, gdt_end)
+
+	.balign 16
+SYM_DATA_START_LOCAL(early_stack)
+	.fill BOOT_STACK_SIZE, 1, 0
+SYM_DATA_END_LABEL(early_stack, SYM_L_LOCAL, early_stack_end)
+
+#ifdef CONFIG_X86_64
+/*
+ * Xen PVH needs a set of identity mapped and kernel high mapping
+ * page tables.  pvh_start_xen starts running on the identity mapped
+ * page tables, but xen_prepare_pvh calls into the high mapping.
+ * These page tables need to be relocatable and are only used until
+ * startup_64 transitions to init_top_pgt.
+ */
+SYM_DATA_START_PAGE_ALIGNED(pvh_init_top_pgt)
+	.quad   pvh_level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE_NOENC
+	.org    pvh_init_top_pgt + L4_PAGE_OFFSET * 8, 0
+	.quad   pvh_level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE_NOENC
+	.org    pvh_init_top_pgt + L4_START_KERNEL * 8, 0
+	/* (2^48-(2*1024*1024*1024))/(2^39) = 511 */
+	.quad   pvh_level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE_NOENC
+SYM_DATA_END(pvh_init_top_pgt)
+
+SYM_DATA_START_PAGE_ALIGNED(pvh_level3_ident_pgt)
+	.quad	pvh_level2_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE_NOENC
+	.fill	511, 8, 0
+SYM_DATA_END(pvh_level3_ident_pgt)
+SYM_DATA_START_PAGE_ALIGNED(pvh_level2_ident_pgt)
+	/*
+	 * Since I easily can, map the first 1G.
+	 * Don't set NX because code runs from these pages.
+	 *
+	 * Note: This sets _PAGE_GLOBAL despite whether
+	 * the CPU supports it or it is enabled.  But,
+	 * the CPU should ignore the bit.
+	 */
+	PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD)
+SYM_DATA_END(pvh_level2_ident_pgt)
+SYM_DATA_START_PAGE_ALIGNED(pvh_level3_kernel_pgt)
+	.fill	L3_START_KERNEL, 8, 0
+	/* (2^48-(2*1024*1024*1024)-((2^39)*511))/(2^30) = 510 */
+	.quad	pvh_level2_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE_NOENC
+	.quad	0 /* no fixmap */
+SYM_DATA_END(pvh_level3_kernel_pgt)
+
+SYM_DATA_START_PAGE_ALIGNED(pvh_level2_kernel_pgt)
+	/*
+	 * Kernel high mapping.
+	 *
+	 * The kernel code+data+bss must be located below KERNEL_IMAGE_SIZE in
+	 * virtual address space, which is 1 GiB if RANDOMIZE_BASE is enabled,
+	 * 512 MiB otherwise.
+	 *
+	 * (NOTE: after that starts the module area, see MODULES_VADDR.)
+	 *
+	 * This table is eventually used by the kernel during normal runtime.
+	 * Care must be taken to clear out undesired bits later, like _PAGE_RW
+	 * or _PAGE_GLOBAL in some cases.
+	 */
+	PMDS(0, __PAGE_KERNEL_LARGE_EXEC, KERNEL_IMAGE_SIZE / PMD_SIZE)
+SYM_DATA_END(pvh_level2_kernel_pgt)
+
+	ELFNOTE(Xen, XEN_ELFNOTE_PHYS32_RELOC,
+		     .long CONFIG_PHYSICAL_ALIGN;
+		     .long LOAD_PHYSICAL_ADDR;
+		     .long KERNEL_IMAGE_SIZE - 1)
+#endif
+
+	ELFNOTE(Xen, XEN_ELFNOTE_PHYS32_ENTRY, .global xen_elfnote_phys32_entry;
+		xen_elfnote_phys32_entry: _ASM_PTR xen_elfnote_phys32_entry_value - .)

linux-7.2.3/arch/x86/platform/scx200/Makefile

@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_SCx200)		+= scx200.o
+scx200-y			+= scx200_32.o

linux-7.2.3/arch/x86/platform/scx200/scx200_32.c

@@ -0,0 +1,130 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Copyright (c) 2001,2002 Christer Weinigel <wingel@nano-system.com>
+ *
+ *  National Semiconductor SCx200 support.
+ */
+
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/mutex.h>
+#include <linux/pci.h>
+
+#include <linux/scx200.h>
+#include <linux/scx200_gpio.h>
+
+/* Verify that the configuration block really is there */
+#define scx200_cb_probe(base) (inw((base) + SCx200_CBA) == (base))
+
+MODULE_AUTHOR("Christer Weinigel <wingel@nano-system.com>");
+MODULE_DESCRIPTION("NatSemi SCx200 Driver");
+MODULE_LICENSE("GPL");
+
+unsigned scx200_gpio_base = 0;
+unsigned long scx200_gpio_shadow[2];
+
+unsigned scx200_cb_base = 0;
+
+static struct pci_device_id scx200_tbl[] = {
+	{ PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SCx200_BRIDGE) },
+	{ PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SC1100_BRIDGE) },
+	{ PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SCx200_XBUS)   },
+	{ PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SC1100_XBUS)   },
+	{ },
+};
+MODULE_DEVICE_TABLE(pci,scx200_tbl);
+
+static int scx200_probe(struct pci_dev *, const struct pci_device_id *);
+
+static struct pci_driver scx200_pci_driver = {
+	.name = "scx200",
+	.id_table = scx200_tbl,
+	.probe = scx200_probe,
+};
+
+static DEFINE_MUTEX(scx200_gpio_config_lock);
+
+static void scx200_init_shadow(void)
+{
+	int bank;
+
+	/* read the current values driven on the GPIO signals */
+	for (bank = 0; bank < 2; ++bank)
+		scx200_gpio_shadow[bank] = inl(scx200_gpio_base + 0x10 * bank);
+}
+
+static int scx200_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	unsigned base;
+
+	if (pdev->device == PCI_DEVICE_ID_NS_SCx200_BRIDGE ||
+	    pdev->device == PCI_DEVICE_ID_NS_SC1100_BRIDGE) {
+		base = pci_resource_start(pdev, 0);
+		pr_info("GPIO base 0x%x\n", base);
+
+		if (!request_region(base, SCx200_GPIO_SIZE,
+				    "NatSemi SCx200 GPIO")) {
+			pr_err("can't allocate I/O for GPIOs\n");
+			return -EBUSY;
+		}
+
+		scx200_gpio_base = base;
+		scx200_init_shadow();
+
+	} else {
+		/* find the base of the Configuration Block */
+		if (scx200_cb_probe(SCx200_CB_BASE_FIXED)) {
+			scx200_cb_base = SCx200_CB_BASE_FIXED;
+		} else {
+			pci_read_config_dword(pdev, SCx200_CBA_SCRATCH, &base);
+			if (scx200_cb_probe(base)) {
+				scx200_cb_base = base;
+			} else {
+				pr_warn("Configuration Block not found\n");
+				return -ENODEV;
+			}
+		}
+		pr_info("Configuration Block base 0x%x\n", scx200_cb_base);
+	}
+
+	return 0;
+}
+
+u32 scx200_gpio_configure(unsigned index, u32 mask, u32 bits)
+{
+	u32 config, new_config;
+
+	mutex_lock(&scx200_gpio_config_lock);
+
+	outl(index, scx200_gpio_base + 0x20);
+	config = inl(scx200_gpio_base + 0x24);
+
+	new_config = (config & mask) | bits;
+	outl(new_config, scx200_gpio_base + 0x24);
+
+	mutex_unlock(&scx200_gpio_config_lock);
+
+	return config;
+}
+
+static int __init scx200_init(void)
+{
+	pr_info("NatSemi SCx200 Driver\n");
+	return pci_register_driver(&scx200_pci_driver);
+}
+
+static void __exit scx200_cleanup(void)
+{
+	pci_unregister_driver(&scx200_pci_driver);
+	release_region(scx200_gpio_base, SCx200_GPIO_SIZE);
+}
+
+module_init(scx200_init);
+module_exit(scx200_cleanup);
+
+EXPORT_SYMBOL(scx200_gpio_base);
+EXPORT_SYMBOL(scx200_gpio_shadow);
+EXPORT_SYMBOL(scx200_gpio_configure);
+EXPORT_SYMBOL(scx200_cb_base);

linux-7.2.3/arch/x86/platform/ts5500/Makefile

@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_TS5500)	+= ts5500.o

linux-7.2.3/arch/x86/platform/ts5500/ts5500.c

@@ -0,0 +1,341 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Technologic Systems TS-5500 Single Board Computer support
+ *
+ * Copyright (C) 2013-2014 Savoir-faire Linux Inc.
+ *	Vivien Didelot <vivien.didelot@savoirfairelinux.com>
+ *
+ * This driver registers the Technologic Systems TS-5500 Single Board Computer
+ * (SBC) and its devices, and exposes information to userspace such as jumpers'
+ * state or available options. For further information about sysfs entries, see
+ * Documentation/ABI/testing/sysfs-platform-ts5500.
+ *
+ * This code may be extended to support similar x86-based platforms.
+ * Actually, the TS-5500 and TS-5400 are supported.
+ */
+
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/leds.h>
+#include <linux/init.h>
+#include <linux/platform_data/max197.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+/* Product code register */
+#define TS5500_PRODUCT_CODE_ADDR	0x74
+#define TS5500_PRODUCT_CODE		0x60	/* TS-5500 product code */
+#define TS5400_PRODUCT_CODE		0x40	/* TS-5400 product code */
+
+/* SRAM/RS-485/ADC options, and RS-485 RTS/Automatic RS-485 flags register */
+#define TS5500_SRAM_RS485_ADC_ADDR	0x75
+#define TS5500_SRAM			BIT(0)	/* SRAM option */
+#define TS5500_RS485			BIT(1)	/* RS-485 option */
+#define TS5500_ADC			BIT(2)	/* A/D converter option */
+#define TS5500_RS485_RTS		BIT(6)	/* RTS for RS-485 */
+#define TS5500_RS485_AUTO		BIT(7)	/* Automatic RS-485 */
+
+/* External Reset/Industrial Temperature Range options register */
+#define TS5500_ERESET_ITR_ADDR		0x76
+#define TS5500_ERESET			BIT(0)	/* External Reset option */
+#define TS5500_ITR			BIT(1)	/* Indust. Temp. Range option */
+
+/* LED/Jumpers register */
+#define TS5500_LED_JP_ADDR		0x77
+#define TS5500_LED			BIT(0)	/* LED flag */
+#define TS5500_JP1			BIT(1)	/* Automatic CMOS */
+#define TS5500_JP2			BIT(2)	/* Enable Serial Console */
+#define TS5500_JP3			BIT(3)	/* Write Enable Drive A */
+#define TS5500_JP4			BIT(4)	/* Fast Console (115K baud) */
+#define TS5500_JP5			BIT(5)	/* User Jumper */
+#define TS5500_JP6			BIT(6)	/* Console on COM1 (req. JP2) */
+#define TS5500_JP7			BIT(7)	/* Undocumented (Unused) */
+
+/* A/D Converter registers */
+#define TS5500_ADC_CONV_BUSY_ADDR	0x195	/* Conversion state register */
+#define TS5500_ADC_CONV_BUSY		BIT(0)
+#define TS5500_ADC_CONV_INIT_LSB_ADDR	0x196	/* Start conv. / LSB register */
+#define TS5500_ADC_CONV_MSB_ADDR	0x197	/* MSB register */
+#define TS5500_ADC_CONV_DELAY		12	/* usec */
+
+/**
+ * struct ts5500_sbc - TS-5500 board description
+ * @name:	Board model name.
+ * @id:		Board product ID.
+ * @sram:	Flag for SRAM option.
+ * @rs485:	Flag for RS-485 option.
+ * @adc:	Flag for Analog/Digital converter option.
+ * @ereset:	Flag for External Reset option.
+ * @itr:	Flag for Industrial Temperature Range option.
+ * @jumpers:	Bitfield for jumpers' state.
+ */
+struct ts5500_sbc {
+	const char *name;
+	int	id;
+	bool	sram;
+	bool	rs485;
+	bool	adc;
+	bool	ereset;
+	bool	itr;
+	u8	jumpers;
+};
+
+/* Board signatures in BIOS shadow RAM */
+static const struct {
+	const char * const string;
+	const ssize_t offset;
+} ts5500_signatures[] __initconst = {
+	{ "TS-5x00 AMD Elan", 0xb14 },
+};
+
+static int __init ts5500_check_signature(void)
+{
+	void __iomem *bios;
+	int i, ret = -ENODEV;
+
+	bios = ioremap(0xf0000, 0x10000);
+	if (!bios)
+		return -ENOMEM;
+
+	for (i = 0; i < ARRAY_SIZE(ts5500_signatures); i++) {
+		if (check_signature(bios + ts5500_signatures[i].offset,
+				    ts5500_signatures[i].string,
+				    strlen(ts5500_signatures[i].string))) {
+			ret = 0;
+			break;
+		}
+	}
+
+	iounmap(bios);
+	return ret;
+}
+
+static int __init ts5500_detect_config(struct ts5500_sbc *sbc)
+{
+	u8 tmp;
+	int ret = 0;
+
+	if (!request_region(TS5500_PRODUCT_CODE_ADDR, 4, "ts5500"))
+		return -EBUSY;
+
+	sbc->id = inb(TS5500_PRODUCT_CODE_ADDR);
+	if (sbc->id == TS5500_PRODUCT_CODE) {
+		sbc->name = "TS-5500";
+	} else if (sbc->id == TS5400_PRODUCT_CODE) {
+		sbc->name = "TS-5400";
+	} else {
+		pr_err("ts5500: unknown product code 0x%x\n", sbc->id);
+		ret = -ENODEV;
+		goto cleanup;
+	}
+
+	tmp = inb(TS5500_SRAM_RS485_ADC_ADDR);
+	sbc->sram = tmp & TS5500_SRAM;
+	sbc->rs485 = tmp & TS5500_RS485;
+	sbc->adc = tmp & TS5500_ADC;
+
+	tmp = inb(TS5500_ERESET_ITR_ADDR);
+	sbc->ereset = tmp & TS5500_ERESET;
+	sbc->itr = tmp & TS5500_ITR;
+
+	tmp = inb(TS5500_LED_JP_ADDR);
+	sbc->jumpers = tmp & ~TS5500_LED;
+
+cleanup:
+	release_region(TS5500_PRODUCT_CODE_ADDR, 4);
+	return ret;
+}
+
+static ssize_t name_show(struct device *dev, struct device_attribute *attr,
+		char *buf)
+{
+	struct ts5500_sbc *sbc = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%s\n", sbc->name);
+}
+static DEVICE_ATTR_RO(name);
+
+static ssize_t id_show(struct device *dev, struct device_attribute *attr,
+		char *buf)
+{
+	struct ts5500_sbc *sbc = dev_get_drvdata(dev);
+
+	return sprintf(buf, "0x%.2x\n", sbc->id);
+}
+static DEVICE_ATTR_RO(id);
+
+static ssize_t jumpers_show(struct device *dev, struct device_attribute *attr,
+		char *buf)
+{
+	struct ts5500_sbc *sbc = dev_get_drvdata(dev);
+
+	return sprintf(buf, "0x%.2x\n", sbc->jumpers >> 1);
+}
+static DEVICE_ATTR_RO(jumpers);
+
+#define TS5500_ATTR_BOOL(_field)					\
+	static ssize_t _field##_show(struct device *dev,		\
+			struct device_attribute *attr, char *buf)	\
+	{								\
+		struct ts5500_sbc *sbc = dev_get_drvdata(dev);		\
+									\
+		return sprintf(buf, "%d\n", sbc->_field);		\
+	}								\
+	static DEVICE_ATTR_RO(_field)
+
+TS5500_ATTR_BOOL(sram);
+TS5500_ATTR_BOOL(rs485);
+TS5500_ATTR_BOOL(adc);
+TS5500_ATTR_BOOL(ereset);
+TS5500_ATTR_BOOL(itr);
+
+static struct attribute *ts5500_attributes[] = {
+	&dev_attr_id.attr,
+	&dev_attr_name.attr,
+	&dev_attr_jumpers.attr,
+	&dev_attr_sram.attr,
+	&dev_attr_rs485.attr,
+	&dev_attr_adc.attr,
+	&dev_attr_ereset.attr,
+	&dev_attr_itr.attr,
+	NULL
+};
+
+static const struct attribute_group ts5500_attr_group = {
+	.attrs = ts5500_attributes,
+};
+
+static struct resource ts5500_dio1_resource[] = {
+	DEFINE_RES_IRQ_NAMED(7, "DIO1 interrupt"),
+};
+
+static struct platform_device ts5500_dio1_pdev = {
+	.name = "ts5500-dio1",
+	.id = -1,
+	.resource = ts5500_dio1_resource,
+	.num_resources = 1,
+};
+
+static struct resource ts5500_dio2_resource[] = {
+	DEFINE_RES_IRQ_NAMED(6, "DIO2 interrupt"),
+};
+
+static struct platform_device ts5500_dio2_pdev = {
+	.name = "ts5500-dio2",
+	.id = -1,
+	.resource = ts5500_dio2_resource,
+	.num_resources = 1,
+};
+
+static void ts5500_led_set(struct led_classdev *led_cdev,
+			   enum led_brightness brightness)
+{
+	outb(!!brightness, TS5500_LED_JP_ADDR);
+}
+
+static enum led_brightness ts5500_led_get(struct led_classdev *led_cdev)
+{
+	return (inb(TS5500_LED_JP_ADDR) & TS5500_LED) ? LED_FULL : LED_OFF;
+}
+
+static struct led_classdev ts5500_led_cdev = {
+	.name = "ts5500:green:",
+	.brightness_set = ts5500_led_set,
+	.brightness_get = ts5500_led_get,
+};
+
+static int ts5500_adc_convert(u8 ctrl)
+{
+	u8 lsb, msb;
+
+	/* Start conversion (ensure the 3 MSB are set to 0) */
+	outb(ctrl & 0x1f, TS5500_ADC_CONV_INIT_LSB_ADDR);
+
+	/*
+	 * The platform has CPLD logic driving the A/D converter.
+	 * The conversion must complete within 11 microseconds,
+	 * otherwise we have to re-initiate a conversion.
+	 */
+	udelay(TS5500_ADC_CONV_DELAY);
+	if (inb(TS5500_ADC_CONV_BUSY_ADDR) & TS5500_ADC_CONV_BUSY)
+		return -EBUSY;
+
+	/* Read the raw data */
+	lsb = inb(TS5500_ADC_CONV_INIT_LSB_ADDR);
+	msb = inb(TS5500_ADC_CONV_MSB_ADDR);
+
+	return (msb << 8) | lsb;
+}
+
+static struct max197_platform_data ts5500_adc_pdata = {
+	.convert = ts5500_adc_convert,
+};
+
+static struct platform_device ts5500_adc_pdev = {
+	.name = "max197",
+	.id = -1,
+	.dev = {
+		.platform_data = &ts5500_adc_pdata,
+	},
+};
+
+static int __init ts5500_init(void)
+{
+	struct platform_device *pdev;
+	struct ts5500_sbc *sbc;
+	int err;
+
+	/*
+	 * There is no DMI available or PCI bridge subvendor info,
+	 * only the BIOS provides a 16-bit identification call.
+	 * It is safer to find a signature in the BIOS shadow RAM.
+	 */
+	err = ts5500_check_signature();
+	if (err)
+		return err;
+
+	pdev = platform_device_register_simple("ts5500", -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	sbc = devm_kzalloc(&pdev->dev, sizeof(struct ts5500_sbc), GFP_KERNEL);
+	if (!sbc) {
+		err = -ENOMEM;
+		goto error;
+	}
+
+	err = ts5500_detect_config(sbc);
+	if (err)
+		goto error;
+
+	platform_set_drvdata(pdev, sbc);
+
+	err = sysfs_create_group(&pdev->dev.kobj, &ts5500_attr_group);
+	if (err)
+		goto error;
+
+	if (sbc->id == TS5500_PRODUCT_CODE) {
+		ts5500_dio1_pdev.dev.parent = &pdev->dev;
+		if (platform_device_register(&ts5500_dio1_pdev))
+			dev_warn(&pdev->dev, "DIO1 block registration failed\n");
+		ts5500_dio2_pdev.dev.parent = &pdev->dev;
+		if (platform_device_register(&ts5500_dio2_pdev))
+			dev_warn(&pdev->dev, "DIO2 block registration failed\n");
+	}
+
+	if (led_classdev_register(&pdev->dev, &ts5500_led_cdev))
+		dev_warn(&pdev->dev, "LED registration failed\n");
+
+	if (sbc->adc) {
+		ts5500_adc_pdev.dev.parent = &pdev->dev;
+		if (platform_device_register(&ts5500_adc_pdev))
+			dev_warn(&pdev->dev, "ADC registration failed\n");
+	}
+
+	return 0;
+error:
+	platform_device_unregister(pdev);
+	return err;
+}
+device_initcall(ts5500_init);

linux-7.2.3/arch/x86/platform/uv/Makefile

@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_X86_UV)		+= bios_uv.o uv_irq.o uv_time.o uv_nmi.o

linux-7.2.3/arch/x86/platform/uv/bios_uv.c

@@ -0,0 +1,269 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * BIOS run time interface routines.
+ *
+ * (C) Copyright 2020 Hewlett Packard Enterprise Development LP
+ * Copyright (C) 2007-2017 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (c) Russ Anderson <rja@sgi.com>
+ */
+
+#include <linux/efi.h>
+#include <linux/export.h>
+#include <linux/slab.h>
+#include <asm/efi.h>
+#include <linux/io.h>
+#include <asm/pgalloc.h>
+#include <asm/uv/bios.h>
+#include <asm/uv/uv_hub.h>
+
+unsigned long uv_systab_phys __ro_after_init = EFI_INVALID_TABLE_ADDR;
+
+struct uv_systab *uv_systab;
+
+static s64 __uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3,
+			u64 a4, u64 a5)
+{
+	struct uv_systab *tab = uv_systab;
+	s64 ret;
+
+	if (!tab || !tab->function)
+		/*
+		 * BIOS does not support UV systab
+		 */
+		return BIOS_STATUS_UNIMPLEMENTED;
+
+	ret = efi_call_virt_pointer(tab, function, (u64)which, a1, a2, a3, a4, a5);
+
+	return ret;
+}
+
+static s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4,
+		u64 a5)
+{
+	s64 ret;
+
+	if (down_interruptible(&__efi_uv_runtime_lock))
+		return BIOS_STATUS_ABORT;
+
+	ret = __uv_bios_call(which, a1, a2, a3, a4, a5);
+	up(&__efi_uv_runtime_lock);
+
+	return ret;
+}
+
+static s64 uv_bios_call_irqsave(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3,
+		u64 a4, u64 a5)
+{
+	unsigned long bios_flags;
+	s64 ret;
+
+	if (down_interruptible(&__efi_uv_runtime_lock))
+		return BIOS_STATUS_ABORT;
+
+	local_irq_save(bios_flags);
+	ret = __uv_bios_call(which, a1, a2, a3, a4, a5);
+	local_irq_restore(bios_flags);
+
+	up(&__efi_uv_runtime_lock);
+
+	return ret;
+}
+
+long sn_partition_id;
+EXPORT_SYMBOL_GPL(sn_partition_id);
+long sn_coherency_id;
+EXPORT_SYMBOL_GPL(sn_coherency_id);
+long sn_region_size;
+EXPORT_SYMBOL_GPL(sn_region_size);
+long system_serial_number;
+int uv_type;
+
+s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher,
+		long *region, long *ssn)
+{
+	s64 ret;
+	u64 v0, v1;
+	union partition_info_u part;
+
+	ret = uv_bios_call_irqsave(UV_BIOS_GET_SN_INFO, fc,
+				(u64)(&v0), (u64)(&v1), 0, 0);
+	if (ret != BIOS_STATUS_SUCCESS)
+		return ret;
+
+	part.val = v0;
+	if (uvtype)
+		*uvtype = part.hub_version;
+	if (partid)
+		*partid = part.partition_id;
+	if (coher)
+		*coher = part.coherence_id;
+	if (region)
+		*region = part.region_size;
+	if (ssn)
+		*ssn = v1;
+	return ret;
+}
+
+int
+uv_bios_mq_watchlist_alloc(unsigned long addr, unsigned int mq_size,
+			   unsigned long *intr_mmr_offset)
+{
+	u64 watchlist;
+	s64 ret;
+
+	/*
+	 * bios returns watchlist number or negative error number.
+	 */
+	ret = (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_ALLOC, addr,
+			mq_size, (u64)intr_mmr_offset,
+			(u64)&watchlist, 0);
+	if (ret < BIOS_STATUS_SUCCESS)
+		return ret;
+
+	return watchlist;
+}
+EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_alloc);
+
+int
+uv_bios_mq_watchlist_free(int blade, int watchlist_num)
+{
+	return (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_FREE,
+				blade, watchlist_num, 0, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_free);
+
+s64
+uv_bios_change_memprotect(u64 paddr, u64 len, enum uv_memprotect perms)
+{
+	return uv_bios_call_irqsave(UV_BIOS_MEMPROTECT, paddr, len,
+					perms, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_change_memprotect);
+
+s64
+uv_bios_reserved_page_pa(u64 buf, u64 *cookie, u64 *addr, u64 *len)
+{
+	return uv_bios_call_irqsave(UV_BIOS_GET_PARTITION_ADDR, (u64)cookie,
+				    (u64)addr, buf, (u64)len, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_reserved_page_pa);
+
+s64 uv_bios_freq_base(u64 clock_type, u64 *ticks_per_second)
+{
+	return uv_bios_call(UV_BIOS_FREQ_BASE, clock_type,
+			   (u64)ticks_per_second, 0, 0, 0);
+}
+
+/*
+ * uv_bios_set_legacy_vga_target - Set Legacy VGA I/O Target
+ * @decode: true to enable target, false to disable target
+ * @domain: PCI domain number
+ * @bus: PCI bus number
+ *
+ * Returns:
+ *    0: Success
+ *    -EINVAL: Invalid domain or bus number
+ *    -ENOSYS: Capability not available
+ *    -EBUSY: Legacy VGA I/O cannot be retargeted at this time
+ */
+int uv_bios_set_legacy_vga_target(bool decode, int domain, int bus)
+{
+	return uv_bios_call(UV_BIOS_SET_LEGACY_VGA_TARGET,
+				(u64)decode, (u64)domain, (u64)bus, 0, 0);
+}
+
+extern s64 uv_bios_get_master_nasid(u64 size, u64 *master_nasid)
+{
+	return uv_bios_call(UV_BIOS_EXTRA, 0, UV_BIOS_EXTRA_MASTER_NASID, 0,
+				size, (u64)master_nasid);
+}
+EXPORT_SYMBOL_GPL(uv_bios_get_master_nasid);
+
+extern s64 uv_bios_get_heapsize(u64 nasid, u64 size, u64 *heap_size)
+{
+	return uv_bios_call(UV_BIOS_EXTRA, nasid, UV_BIOS_EXTRA_GET_HEAPSIZE,
+				0, size, (u64)heap_size);
+}
+EXPORT_SYMBOL_GPL(uv_bios_get_heapsize);
+
+extern s64 uv_bios_install_heap(u64 nasid, u64 heap_size, u64 *bios_heap)
+{
+	return uv_bios_call(UV_BIOS_EXTRA, nasid, UV_BIOS_EXTRA_INSTALL_HEAP,
+				0, heap_size, (u64)bios_heap);
+}
+EXPORT_SYMBOL_GPL(uv_bios_install_heap);
+
+extern s64 uv_bios_obj_count(u64 nasid, u64 size, u64 *objcnt)
+{
+	return uv_bios_call(UV_BIOS_EXTRA, nasid, UV_BIOS_EXTRA_OBJECT_COUNT,
+				0, size, (u64)objcnt);
+}
+EXPORT_SYMBOL_GPL(uv_bios_obj_count);
+
+extern s64 uv_bios_enum_objs(u64 nasid, u64 size, u64 *objbuf)
+{
+	return uv_bios_call(UV_BIOS_EXTRA, nasid, UV_BIOS_EXTRA_ENUM_OBJECTS,
+				0, size, (u64)objbuf);
+}
+EXPORT_SYMBOL_GPL(uv_bios_enum_objs);
+
+extern s64 uv_bios_enum_ports(u64 nasid, u64 obj_id, u64 size, u64 *portbuf)
+{
+	return uv_bios_call(UV_BIOS_EXTRA, nasid, UV_BIOS_EXTRA_ENUM_PORTS,
+				obj_id, size, (u64)portbuf);
+}
+EXPORT_SYMBOL_GPL(uv_bios_enum_ports);
+
+extern s64 uv_bios_get_geoinfo(u64 nasid, u64 size, u64 *buf)
+{
+	return uv_bios_call(UV_BIOS_GET_GEOINFO, nasid, (u64)buf, size, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_get_geoinfo);
+
+extern s64 uv_bios_get_pci_topology(u64 size, u64 *buf)
+{
+	return uv_bios_call(UV_BIOS_GET_PCI_TOPOLOGY, (u64)buf, size, 0, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_get_pci_topology);
+
+unsigned long get_uv_systab_phys(bool msg)
+{
+	if ((uv_systab_phys == EFI_INVALID_TABLE_ADDR) ||
+	    !uv_systab_phys || efi_runtime_disabled()) {
+		if (msg)
+			pr_crit("UV: UVsystab: missing\n");
+		return 0;
+	}
+	return uv_systab_phys;
+}
+
+int uv_bios_init(void)
+{
+	unsigned long uv_systab_phys_addr;
+
+	uv_systab = NULL;
+	uv_systab_phys_addr = get_uv_systab_phys(1);
+	if (!uv_systab_phys_addr)
+		return -EEXIST;
+
+	uv_systab = ioremap(uv_systab_phys_addr, sizeof(struct uv_systab));
+	if (!uv_systab || strncmp(uv_systab->signature, UV_SYSTAB_SIG, 4)) {
+		pr_err("UV: UVsystab: bad signature!\n");
+		iounmap(uv_systab);
+		return -EINVAL;
+	}
+
+	/* Starting with UV4 the UV systab size is variable */
+	if (uv_systab->revision >= UV_SYSTAB_VERSION_UV4) {
+		int size = uv_systab->size;
+
+		iounmap(uv_systab);
+		uv_systab = ioremap(uv_systab_phys_addr, size);
+		if (!uv_systab) {
+			pr_err("UV: UVsystab: ioremap(%d) failed!\n", size);
+			return -EFAULT;
+		}
+	}
+	pr_info("UV: UVsystab: Revision:%x\n", uv_systab->revision);
+	return 0;
+}

linux-7.2.3/arch/x86/platform/uv/uv_irq.c

@@ -0,0 +1,213 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * SGI UV IRQ functions
+ *
+ * Copyright (C) 2008 Silicon Graphics, Inc. All rights reserved.
+ */
+
+#include <linux/export.h>
+#include <linux/rbtree.h>
+#include <linux/slab.h>
+#include <linux/irq.h>
+
+#include <asm/irqdomain.h>
+#include <asm/apic.h>
+#include <asm/uv/uv_irq.h>
+#include <asm/uv/uv_hub.h>
+
+/* MMR offset and pnode of hub sourcing interrupts for a given irq */
+struct uv_irq_2_mmr_pnode {
+	unsigned long		offset;
+	int			pnode;
+};
+
+static void uv_program_mmr(struct irq_cfg *cfg, struct uv_irq_2_mmr_pnode *info)
+{
+	unsigned long mmr_value;
+	struct uv_IO_APIC_route_entry *entry;
+
+	BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) !=
+		     sizeof(unsigned long));
+
+	mmr_value = 0;
+	entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
+	entry->vector		= cfg->vector;
+	entry->delivery_mode	= APIC_DELIVERY_MODE_FIXED;
+	entry->dest_mode	= apic->dest_mode_logical;
+	entry->polarity		= 0;
+	entry->trigger		= 0;
+	entry->mask		= 0;
+	entry->dest		= cfg->dest_apicid;
+
+	uv_write_global_mmr64(info->pnode, info->offset, mmr_value);
+}
+
+static void uv_noop(struct irq_data *data) { }
+
+static int
+uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask,
+		    bool force)
+{
+	struct irq_data *parent = data->parent_data;
+	struct irq_cfg *cfg = irqd_cfg(data);
+	int ret;
+
+	ret = parent->chip->irq_set_affinity(parent, mask, force);
+	if (ret >= 0) {
+		uv_program_mmr(cfg, data->chip_data);
+		vector_schedule_cleanup(cfg);
+	}
+
+	return ret;
+}
+
+static struct irq_chip uv_irq_chip = {
+	.name			= "UV-CORE",
+	.irq_mask		= uv_noop,
+	.irq_unmask		= uv_noop,
+	.irq_eoi		= apic_ack_irq,
+	.irq_set_affinity	= uv_set_irq_affinity,
+};
+
+static int uv_domain_alloc(struct irq_domain *domain, unsigned int virq,
+			   unsigned int nr_irqs, void *arg)
+{
+	struct uv_irq_2_mmr_pnode *chip_data;
+	struct irq_alloc_info *info = arg;
+	struct irq_data *irq_data = irq_domain_get_irq_data(domain, virq);
+	int ret;
+
+	if (nr_irqs > 1 || !info || info->type != X86_IRQ_ALLOC_TYPE_UV)
+		return -EINVAL;
+
+	chip_data = kmalloc_node(sizeof(*chip_data), GFP_KERNEL,
+				 irq_data_get_node(irq_data));
+	if (!chip_data)
+		return -ENOMEM;
+
+	ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, arg);
+	if (ret >= 0) {
+		if (info->uv.limit == UV_AFFINITY_CPU)
+			irq_set_status_flags(virq, IRQ_NO_BALANCING);
+
+		chip_data->pnode = uv_blade_to_pnode(info->uv.blade);
+		chip_data->offset = info->uv.offset;
+		irq_domain_set_info(domain, virq, virq, &uv_irq_chip, chip_data,
+				    handle_percpu_irq, NULL, info->uv.name);
+	} else {
+		kfree(chip_data);
+	}
+
+	return ret;
+}
+
+static void uv_domain_free(struct irq_domain *domain, unsigned int virq,
+			   unsigned int nr_irqs)
+{
+	struct irq_data *irq_data = irq_domain_get_irq_data(domain, virq);
+
+	BUG_ON(nr_irqs != 1);
+	kfree(irq_data->chip_data);
+	irq_clear_status_flags(virq, IRQ_NO_BALANCING);
+	irq_domain_free_irqs_top(domain, virq, nr_irqs);
+}
+
+/*
+ * Re-target the irq to the specified CPU and enable the specified MMR located
+ * on the specified blade to allow the sending of MSIs to the specified CPU.
+ */
+static int uv_domain_activate(struct irq_domain *domain,
+			      struct irq_data *irq_data, bool reserve)
+{
+	uv_program_mmr(irqd_cfg(irq_data), irq_data->chip_data);
+	return 0;
+}
+
+/*
+ * Disable the specified MMR located on the specified blade so that MSIs are
+ * longer allowed to be sent.
+ */
+static void uv_domain_deactivate(struct irq_domain *domain,
+				 struct irq_data *irq_data)
+{
+	unsigned long mmr_value;
+	struct uv_IO_APIC_route_entry *entry;
+
+	mmr_value = 0;
+	entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
+	entry->mask = 1;
+	uv_program_mmr(irqd_cfg(irq_data), irq_data->chip_data);
+}
+
+static const struct irq_domain_ops uv_domain_ops = {
+	.alloc		= uv_domain_alloc,
+	.free		= uv_domain_free,
+	.activate	= uv_domain_activate,
+	.deactivate	= uv_domain_deactivate,
+};
+
+static struct irq_domain *uv_get_irq_domain(void)
+{
+	static struct irq_domain *uv_domain;
+	static DEFINE_MUTEX(uv_lock);
+	struct fwnode_handle *fn;
+
+	mutex_lock(&uv_lock);
+	if (uv_domain)
+		goto out;
+
+	fn = irq_domain_alloc_named_fwnode("UV-CORE");
+	if (!fn)
+		goto out;
+
+	uv_domain = irq_domain_create_hierarchy(x86_vector_domain, 0, 0, fn,
+						&uv_domain_ops, NULL);
+	if (!uv_domain)
+		irq_domain_free_fwnode(fn);
+out:
+	mutex_unlock(&uv_lock);
+
+	return uv_domain;
+}
+
+/*
+ * Set up a mapping of an available irq and vector, and enable the specified
+ * MMR that defines the MSI that is to be sent to the specified CPU when an
+ * interrupt is raised.
+ */
+int uv_setup_irq(char *irq_name, int cpu, int mmr_blade,
+		 unsigned long mmr_offset, int limit)
+{
+	struct irq_alloc_info info;
+	struct irq_domain *domain = uv_get_irq_domain();
+
+	if (!domain)
+		return -ENOMEM;
+
+	init_irq_alloc_info(&info, cpumask_of(cpu));
+	info.type = X86_IRQ_ALLOC_TYPE_UV;
+	info.uv.limit = limit;
+	info.uv.blade = mmr_blade;
+	info.uv.offset = mmr_offset;
+	info.uv.name = irq_name;
+
+	return irq_domain_alloc_irqs(domain, 1,
+				     uv_blade_to_memory_nid(mmr_blade), &info);
+}
+EXPORT_SYMBOL_GPL(uv_setup_irq);
+
+/*
+ * Tear down a mapping of an irq and vector, and disable the specified MMR that
+ * defined the MSI that was to be sent to the specified CPU when an interrupt
+ * was raised.
+ *
+ * Set mmr_blade and mmr_offset to what was passed in on uv_setup_irq().
+ */
+void uv_teardown_irq(unsigned int irq)
+{
+	irq_domain_free_irqs(irq, 1);
+}
+EXPORT_SYMBOL_GPL(uv_teardown_irq);

linux-7.2.3/arch/x86/platform/uv/uv_time.c

@@ -0,0 +1,393 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * SGI RTC clock/timer routines.
+ *
+ *  (C) Copyright 2020 Hewlett Packard Enterprise Development LP
+ *  Copyright (c) 2009-2013 Silicon Graphics, Inc.  All Rights Reserved.
+ *  Copyright (c) Dimitri Sivanich
+ */
+#include <linux/clockchips.h>
+#include <linux/slab.h>
+
+#include <asm/uv/uv_mmrs.h>
+#include <asm/uv/uv_hub.h>
+#include <asm/uv/bios.h>
+#include <asm/uv/uv.h>
+#include <asm/apic.h>
+#include <asm/cpu.h>
+
+#define RTC_NAME		"sgi_rtc"
+
+static u64 uv_read_rtc(struct clocksource *cs);
+static int uv_rtc_next_event(unsigned long, struct clock_event_device *);
+static int uv_rtc_shutdown(struct clock_event_device *evt);
+
+static struct clocksource clocksource_uv = {
+	.name		= RTC_NAME,
+	.rating		= 299,
+	.read		= uv_read_rtc,
+	.mask		= (u64)UVH_RTC_REAL_TIME_CLOCK_MASK,
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static struct clock_event_device clock_event_device_uv = {
+	.name			= RTC_NAME,
+	.features		= CLOCK_EVT_FEAT_ONESHOT,
+	.shift			= 20,
+	.rating			= 400,
+	.irq			= -1,
+	.set_next_event		= uv_rtc_next_event,
+	.set_state_shutdown	= uv_rtc_shutdown,
+	.event_handler		= NULL,
+};
+
+static DEFINE_PER_CPU(struct clock_event_device, cpu_ced);
+
+/* There is one of these allocated per node */
+struct uv_rtc_timer_head {
+	spinlock_t	lock;
+	/* next cpu waiting for timer, local node relative: */
+	int		next_cpu;
+	/* number of cpus on this node: */
+	int		ncpus;
+	struct {
+		int	lcpu;		/* systemwide logical cpu number */
+		u64	expires;	/* next timer expiration for this cpu */
+	} cpu[] __counted_by(ncpus);
+};
+
+/*
+ * Access to uv_rtc_timer_head via blade id.
+ */
+static struct uv_rtc_timer_head		**blade_info __read_mostly;
+
+static int				uv_rtc_evt_enable;
+
+/*
+ * Hardware interface routines
+ */
+
+/* Send IPIs to another node */
+static void uv_rtc_send_IPI(int cpu)
+{
+	unsigned long apicid, val;
+	int pnode;
+
+	apicid = cpu_physical_id(cpu);
+	pnode = uv_apicid_to_pnode(apicid);
+	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
+	      (apicid << UVH_IPI_INT_APIC_ID_SHFT) |
+	      (X86_PLATFORM_IPI_VECTOR << UVH_IPI_INT_VECTOR_SHFT);
+
+	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
+}
+
+/* Check for an RTC interrupt pending */
+static int uv_intr_pending(int pnode)
+{
+	return uv_read_global_mmr64(pnode, UVH_EVENT_OCCURRED2) &
+		UVH_EVENT_OCCURRED2_RTC_1_MASK;
+}
+
+/* Setup interrupt and return non-zero if early expiration occurred. */
+static int uv_setup_intr(int cpu, u64 expires)
+{
+	u64 val;
+	unsigned long apicid = cpu_physical_id(cpu);
+	int pnode = uv_cpu_to_pnode(cpu);
+
+	uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG,
+		UVH_RTC1_INT_CONFIG_M_MASK);
+	uv_write_global_mmr64(pnode, UVH_INT_CMPB, -1L);
+
+	uv_write_global_mmr64(pnode, UVH_EVENT_OCCURRED2_ALIAS,
+			      UVH_EVENT_OCCURRED2_RTC_1_MASK);
+
+	val = (X86_PLATFORM_IPI_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) |
+		((u64)apicid << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT);
+
+	/* Set configuration */
+	uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, val);
+	/* Initialize comparator value */
+	uv_write_global_mmr64(pnode, UVH_INT_CMPB, expires);
+
+	if (uv_read_rtc(NULL) <= expires)
+		return 0;
+
+	return !uv_intr_pending(pnode);
+}
+
+/*
+ * Per-cpu timer tracking routines
+ */
+
+static __init void uv_rtc_deallocate_timers(void)
+{
+	int bid;
+
+	for_each_possible_blade(bid) {
+		kfree(blade_info[bid]);
+	}
+	kfree(blade_info);
+}
+
+/* Allocate per-node list of cpu timer expiration times. */
+static __init int uv_rtc_allocate_timers(void)
+{
+	int cpu;
+
+	blade_info = kcalloc(uv_possible_blades, sizeof(void *), GFP_KERNEL);
+	if (!blade_info)
+		return -ENOMEM;
+
+	for_each_present_cpu(cpu) {
+		int nid = cpu_to_node(cpu);
+		int bid = uv_cpu_to_blade_id(cpu);
+		int bcpu = uv_cpu_blade_processor_id(cpu);
+		struct uv_rtc_timer_head *head = blade_info[bid];
+
+		if (!head) {
+			head = kmalloc_node(struct_size(head, cpu,
+				uv_blade_nr_possible_cpus(bid)),
+				GFP_KERNEL, nid);
+			if (!head) {
+				uv_rtc_deallocate_timers();
+				return -ENOMEM;
+			}
+			spin_lock_init(&head->lock);
+			head->ncpus = uv_blade_nr_possible_cpus(bid);
+			head->next_cpu = -1;
+			blade_info[bid] = head;
+		}
+
+		head->cpu[bcpu].lcpu = cpu;
+		head->cpu[bcpu].expires = ULLONG_MAX;
+	}
+
+	return 0;
+}
+
+/* Find and set the next expiring timer.  */
+static void uv_rtc_find_next_timer(struct uv_rtc_timer_head *head, int pnode)
+{
+	u64 lowest = ULLONG_MAX;
+	int c, bcpu = -1;
+
+	head->next_cpu = -1;
+	for (c = 0; c < head->ncpus; c++) {
+		u64 exp = head->cpu[c].expires;
+		if (exp < lowest) {
+			bcpu = c;
+			lowest = exp;
+		}
+	}
+	if (bcpu >= 0) {
+		head->next_cpu = bcpu;
+		c = head->cpu[bcpu].lcpu;
+		if (uv_setup_intr(c, lowest))
+			/* If we didn't set it up in time, trigger */
+			uv_rtc_send_IPI(c);
+	} else {
+		uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG,
+			UVH_RTC1_INT_CONFIG_M_MASK);
+	}
+}
+
+/*
+ * Set expiration time for current cpu.
+ *
+ * Returns 1 if we missed the expiration time.
+ */
+static int uv_rtc_set_timer(int cpu, u64 expires)
+{
+	int pnode = uv_cpu_to_pnode(cpu);
+	int bid = uv_cpu_to_blade_id(cpu);
+	struct uv_rtc_timer_head *head = blade_info[bid];
+	int bcpu = uv_cpu_blade_processor_id(cpu);
+	u64 *t = &head->cpu[bcpu].expires;
+	unsigned long flags;
+	int next_cpu;
+
+	spin_lock_irqsave(&head->lock, flags);
+
+	next_cpu = head->next_cpu;
+	*t = expires;
+
+	/* Will this one be next to go off? */
+	if (next_cpu < 0 || bcpu == next_cpu ||
+			expires < head->cpu[next_cpu].expires) {
+		head->next_cpu = bcpu;
+		if (uv_setup_intr(cpu, expires)) {
+			*t = ULLONG_MAX;
+			uv_rtc_find_next_timer(head, pnode);
+			spin_unlock_irqrestore(&head->lock, flags);
+			return -ETIME;
+		}
+	}
+
+	spin_unlock_irqrestore(&head->lock, flags);
+	return 0;
+}
+
+/*
+ * Unset expiration time for current cpu.
+ *
+ * Returns 1 if this timer was pending.
+ */
+static int uv_rtc_unset_timer(int cpu, int force)
+{
+	int pnode = uv_cpu_to_pnode(cpu);
+	int bid = uv_cpu_to_blade_id(cpu);
+	struct uv_rtc_timer_head *head = blade_info[bid];
+	int bcpu = uv_cpu_blade_processor_id(cpu);
+	u64 *t = &head->cpu[bcpu].expires;
+	unsigned long flags;
+	int rc = 0;
+
+	spin_lock_irqsave(&head->lock, flags);
+
+	if ((head->next_cpu == bcpu && uv_read_rtc(NULL) >= *t) || force)
+		rc = 1;
+
+	if (rc) {
+		*t = ULLONG_MAX;
+		/* Was the hardware setup for this timer? */
+		if (head->next_cpu == bcpu)
+			uv_rtc_find_next_timer(head, pnode);
+	}
+
+	spin_unlock_irqrestore(&head->lock, flags);
+
+	return rc;
+}
+
+
+/*
+ * Kernel interface routines.
+ */
+
+/*
+ * Read the RTC.
+ *
+ * Starting with HUB rev 2.0, the UV RTC register is replicated across all
+ * cachelines of its own page.  This allows faster simultaneous reads
+ * from a given socket.
+ */
+static u64 uv_read_rtc(struct clocksource *cs)
+{
+	unsigned long offset;
+
+	if (uv_get_min_hub_revision_id() == 1)
+		offset = 0;
+	else
+		offset = (uv_blade_processor_id() * L1_CACHE_BYTES) % PAGE_SIZE;
+
+	return (u64)uv_read_local_mmr(UVH_RTC | offset);
+}
+
+/*
+ * Program the next event, relative to now
+ */
+static int uv_rtc_next_event(unsigned long delta,
+			     struct clock_event_device *ced)
+{
+	int ced_cpu = cpumask_first(ced->cpumask);
+
+	return uv_rtc_set_timer(ced_cpu, delta + uv_read_rtc(NULL));
+}
+
+/*
+ * Shutdown the RTC timer
+ */
+static int uv_rtc_shutdown(struct clock_event_device *evt)
+{
+	int ced_cpu = cpumask_first(evt->cpumask);
+
+	uv_rtc_unset_timer(ced_cpu, 1);
+	return 0;
+}
+
+static void uv_rtc_interrupt(void)
+{
+	int cpu = smp_processor_id();
+	struct clock_event_device *ced = &per_cpu(cpu_ced, cpu);
+
+	if (!ced || !ced->event_handler)
+		return;
+
+	if (uv_rtc_unset_timer(cpu, 0) != 1)
+		return;
+
+	ced->event_handler(ced);
+}
+
+static int __init uv_enable_evt_rtc(char *str)
+{
+	uv_rtc_evt_enable = 1;
+
+	return 1;
+}
+__setup("uvrtcevt", uv_enable_evt_rtc);
+
+static __init void uv_rtc_register_clockevents(struct work_struct *dummy)
+{
+	struct clock_event_device *ced = this_cpu_ptr(&cpu_ced);
+
+	*ced = clock_event_device_uv;
+	ced->cpumask = cpumask_of(smp_processor_id());
+	clockevents_register_device(ced);
+}
+
+static __init int uv_rtc_setup_clock(void)
+{
+	int rc;
+
+	if (!is_uv_system())
+		return -ENODEV;
+
+	rc = clocksource_register_hz(&clocksource_uv, sn_rtc_cycles_per_second);
+	if (rc)
+		printk(KERN_INFO "UV RTC clocksource failed rc %d\n", rc);
+	else
+		printk(KERN_INFO "UV RTC clocksource registered freq %lu MHz\n",
+			sn_rtc_cycles_per_second/(unsigned long)1E6);
+
+	if (rc || !uv_rtc_evt_enable || x86_platform_ipi_callback)
+		return rc;
+
+	/* Setup and register clockevents */
+	rc = uv_rtc_allocate_timers();
+	if (rc)
+		goto error;
+
+	x86_platform_ipi_callback = uv_rtc_interrupt;
+
+	clock_event_device_uv.mult = div_sc(sn_rtc_cycles_per_second,
+				NSEC_PER_SEC, clock_event_device_uv.shift);
+
+	clock_event_device_uv.min_delta_ns = NSEC_PER_SEC /
+						sn_rtc_cycles_per_second;
+	clock_event_device_uv.min_delta_ticks = 1;
+
+	clock_event_device_uv.max_delta_ns = clocksource_uv.mask *
+				(NSEC_PER_SEC / sn_rtc_cycles_per_second);
+	clock_event_device_uv.max_delta_ticks = clocksource_uv.mask;
+
+	rc = schedule_on_each_cpu(uv_rtc_register_clockevents);
+	if (rc) {
+		x86_platform_ipi_callback = NULL;
+		uv_rtc_deallocate_timers();
+		goto error;
+	}
+
+	printk(KERN_INFO "UV RTC clockevents registered\n");
+
+	return 0;
+
+error:
+	clocksource_unregister(&clocksource_uv);
+	printk(KERN_INFO "UV RTC clockevents failed rc %d\n", rc);
+
+	return rc;
+}
+arch_initcall(uv_rtc_setup_clock);