Merge branch 'loongarch-kvm' into loongarch-next

Some common KVM test cases are supported on LoongArch now as following:
  coalesced_io_test
  demand_paging_test
  dirty_log_perf_test
  dirty_log_test
  guest_print_test
  hardware_disable_test
  kvm_binary_stats_test
  kvm_create_max_vcpus
  kvm_page_table_test
  memslot_modification_stress_test
  memslot_perf_test
  set_memory_region_test

And other test cases are not supported by LoongArch such as rseq_test,
since it is not supported on LoongArch physical machine either.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>

Documentation/features/core/mseal_sys_mappings/arch-support.txt

@@ -12,7 +12,7 @@
     |       arm64: |  ok  |
     |        csky: |  N/A |
     |     hexagon: |  N/A |
-    |   loongarch: | TODO |
+    |   loongarch: |  ok  |
     |        m68k: |  N/A |
     |  microblaze: |  N/A |
     |        mips: | TODO |

Documentation/userspace-api/mseal.rst

@@ -144,7 +144,7 @@ Use cases
   architecture.
 
   The following architectures currently support this feature: x86-64, arm64,
-  and s390.
+  loongarch and s390.
 
   WARNING: This feature breaks programs which rely on relocating
   or unmapping system mappings. Known broken software at the time

MAINTAINERS

@@ -13051,6 +13051,8 @@ F:	Documentation/virt/kvm/loongarch/
 F:	arch/loongarch/include/asm/kvm*
 F:	arch/loongarch/include/uapi/asm/kvm*
 F:	arch/loongarch/kvm/
+F:	tools/testing/selftests/kvm/*/loongarch/
+F:	tools/testing/selftests/kvm/lib/loongarch/
 
 KERNEL VIRTUAL MACHINE FOR MIPS (KVM/mips)
 M:	Huacai Chen <chenhuacai@kernel.org>

arch/loongarch/Kconfig

@@ -69,6 +69,7 @@ config LOONGARCH
 	select ARCH_SUPPORTS_INT128 if CC_HAS_INT128
 	select ARCH_SUPPORTS_LTO_CLANG
 	select ARCH_SUPPORTS_LTO_CLANG_THIN
+	select ARCH_SUPPORTS_MSEAL_SYSTEM_MAPPINGS
 	select ARCH_SUPPORTS_NUMA_BALANCING
 	select ARCH_SUPPORTS_RT
 	select ARCH_USE_BUILTIN_BSWAP
@@ -123,6 +124,7 @@ config LOONGARCH
 	select HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET
 	select HAVE_ARCH_SECCOMP
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_STACKLEAK
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select HAVE_ARCH_USERFAULTFD_MINOR if USERFAULTFD
@@ -187,6 +189,7 @@ config LOONGARCH
 	select MODULES_USE_ELF_RELA if MODULES
 	select NEED_PER_CPU_EMBED_FIRST_CHUNK
 	select NEED_PER_CPU_PAGE_FIRST_CHUNK
+	select NUMA_MEMBLKS if NUMA
 	select OF
 	select OF_EARLY_FLATTREE
 	select PCI
@@ -456,6 +459,15 @@ config SCHED_SMT
 	  Improves scheduler's performance when there are multiple
 	  threads in one physical core.
 
+config SCHED_MC
+	bool "Multi-core scheduler support"
+	depends on SMP
+	default y
+	help
+	  Multi-core scheduler support improves the CPU scheduler's decision
+	  making when dealing with multi-core CPU chips at a cost of slightly
+	  increased overhead in some places.
+
 config SMP
 	bool "Multi-Processing support"
 	help
@@ -485,10 +497,10 @@ config HOTPLUG_CPU
 	  Say N if you want to disable CPU hotplug.
 
 config NR_CPUS
-	int "Maximum number of CPUs (2-256)"
-	range 2 256
+	int "Maximum number of CPUs (2-2048)"
+	range 2 2048
+	default "2048"
 	depends on SMP
-	default "64"
 	help
 	  This allows you to specify the maximum number of CPUs which this
 	  kernel will support.

arch/loongarch/Makefile

@@ -181,11 +181,14 @@ vmlinux.elf vmlinux.efi vmlinuz.efi: vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $(bootvars-y) $(boot)/$@
 
 install:
-	$(Q)install -D -m 755 $(KBUILD_IMAGE) $(INSTALL_PATH)/$(image-name-y)-$(KERNELRELEASE)
-	$(Q)install -D -m 644 .config $(INSTALL_PATH)/config-$(KERNELRELEASE)
-	$(Q)install -D -m 644 System.map $(INSTALL_PATH)/System.map-$(KERNELRELEASE)
+	$(call cmd,install)
 
 define archhelp
-	echo '  install              - install kernel into $(INSTALL_PATH)'
+	echo '  vmlinux.elf    - Uncompressed ELF kernel image (arch/loongarch/boot/vmlinux.elf)'
+	echo '  vmlinux.efi    - Uncompressed EFI kernel image (arch/loongarch/boot/vmlinux.efi)'
+	echo '  vmlinuz.efi    - GZIP/ZSTD-compressed EFI kernel image (arch/loongarch/boot/vmlinuz.efi)'
+	echo '                   Default when CONFIG_EFI_ZBOOT=y'
+	echo '  install        - Install kernel using (your) ~/bin/$(INSTALLKERNEL) or'
+	echo '                   (distribution) /sbin/$(INSTALLKERNEL) or install.sh to $$(INSTALL_PATH)'
 	echo
 endef

arch/loongarch/boot/dts/loongson-2k0500.dtsi

@@ -169,6 +169,166 @@ eiointc: interrupt-controller@1fe11600 {
 			interrupts = <3>;
 		};
 
+		pwm@1ff5c000 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c000 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <24 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c010 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c010 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <24 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c020 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c020 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <24 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c030 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c030 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <24 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c040 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c040 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <25 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c050 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c050 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <25 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c060 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c060 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <25 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c070 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c070 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <25 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c080 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c080 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <26 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c090 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c090 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <26 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c0a0 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c0a0 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <26 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c0b0 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c0b0 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <26 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c0c0 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c0c0 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <27 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c0d0 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c0d0 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <27 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c0e0 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c0e0 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <27 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1ff5c0f0 {
+			compatible = "loongson,ls2k0500-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1ff5c0f0 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <27 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
 		gmac0: ethernet@1f020000 {
 			compatible = "snps,dwmac-3.70a";
 			reg = <0x0 0x1f020000 0x0 0x10000>;

arch/loongarch/boot/dts/loongson-2k1000-ref.dts

@@ -5,6 +5,7 @@
 
 /dts-v1/;
 
+#include "dt-bindings/thermal/thermal.h"
 #include "loongson-2k1000.dtsi"
 
 / {
@@ -38,6 +39,13 @@ linux,cma {
 			linux,cma-default;
 		};
 	};
+
+	fan0: pwm-fan {
+		compatible = "pwm-fan";
+		cooling-levels = <255 153 85 25>;
+		pwms = <&pwm1 0 100000 0>;
+		#cooling-cells = <2>;
+	};
 };
 
 &gmac0 {
@@ -92,6 +100,22 @@ &spi0 {
 	#size-cells = <0>;
 };
 
+&pwm1 {
+	status = "okay";
+
+	pinctrl-0 = <&pwm1_pins_default>;
+	pinctrl-names = "default";
+};
+
+&cpu_thermal {
+	cooling-maps {
+		map0 {
+			trip = <&cpu_alert>;
+			cooling-device = <&fan0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
+		};
+	};
+};
+
 &ehci0 {
 	status = "okay";
 };

arch/loongarch/boot/dts/loongson-2k1000.dtsi

@@ -68,7 +68,7 @@ i2c-gpio-1 {
 	};
 
 	thermal-zones {
-		cpu-thermal {
+		cpu_thermal: cpu-thermal {
 			polling-delay-passive = <1000>;
 			polling-delay = <5000>;
 			thermal-sensors = <&tsensor 0>;
@@ -322,6 +322,46 @@ i2c3: i2c@1fe21800 {
 			status = "disabled";
 		};
 
+		pwm@1fe22000 {
+			compatible = "loongson,ls2k1000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1fe22000 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <24 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm1: pwm@1fe22010 {
+			compatible = "loongson,ls2k1000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1fe22010 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <25 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1fe22020 {
+			compatible = "loongson,ls2k1000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1fe22020 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <26 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@1fe22030 {
+			compatible = "loongson,ls2k1000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x1fe22030 0x0 0x10>;
+			interrupt-parent = <&liointc0>;
+			interrupts = <27 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_APB_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
 		pmc: power-management@1fe27000 {
 			compatible = "loongson,ls2k1000-pmc", "loongson,ls2k0500-pmc", "syscon";
 			reg = <0x0 0x1fe27000 0x0 0x58>;

arch/loongarch/boot/dts/loongson-2k2000.dtsi

@@ -165,6 +165,66 @@ msi: msi-controller@1fe01140 {
 			interrupt-parent = <&eiointc>;
 		};
 
+		pwm@100a0000 {
+			compatible = "loongson,ls2k2000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x100a0000 0x0 0x10>;
+			interrupt-parent = <&pic>;
+			interrupts = <24 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_MISC_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@100a0100 {
+			compatible = "loongson,ls2k2000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x100a0100 0x0 0x10>;
+			interrupt-parent = <&pic>;
+			interrupts = <25 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_MISC_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@100a0200 {
+			compatible = "loongson,ls2k2000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x100a0200 0x0 0x10>;
+			interrupt-parent = <&pic>;
+			interrupts = <26 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_MISC_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@100a0300 {
+			compatible = "loongson,ls2k2000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x100a0300 0x0 0x10>;
+			interrupt-parent = <&pic>;
+			interrupts = <27 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_MISC_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@100a0400 {
+			compatible = "loongson,ls2k2000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x100a0400 0x0 0x10>;
+			interrupt-parent = <&pic>;
+			interrupts = <38 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_MISC_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
+		pwm@100a0500 {
+			compatible = "loongson,ls2k2000-pwm", "loongson,ls7a-pwm";
+			reg = <0x0 0x100a0500 0x0 0x10>;
+			interrupt-parent = <&pic>;
+			interrupts = <39 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&clk LOONGSON2_MISC_CLK>;
+			#pwm-cells = <3>;
+			status = "disabled";
+		};
+
 		rtc0: rtc@100d0100 {
 			compatible = "loongson,ls2k2000-rtc", "loongson,ls7a-rtc";
 			reg = <0x0 0x100d0100 0x0 0x100>;

arch/loongarch/boot/install.sh

@@ -0,0 +1,56 @@
+#!/bin/sh
+#
+# 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.
+#
+# Copyright (C) 1995 by Linus Torvalds
+#
+# Adapted from code in arch/i386/boot/Makefile by H. Peter Anvin
+# Adapted from code in arch/i386/boot/install.sh by Russell King
+#
+# "make install" script for the LoongArch Linux port
+#
+# Arguments:
+#   $1 - kernel version
+#   $2 - kernel image file
+#   $3 - kernel map file
+#   $4 - default install path (blank if root directory)
+
+set -e
+
+case "${2##*/}" in
+vmlinux.elf)
+  echo "Installing uncompressed vmlinux.elf kernel"
+  base=vmlinux
+  ;;
+vmlinux.efi)
+  echo "Installing uncompressed vmlinux.efi kernel"
+  base=vmlinux
+  ;;
+vmlinuz.efi)
+  echo "Installing gzip/zstd compressed vmlinuz.efi kernel"
+  base=vmlinuz
+  ;;
+*)
+ echo "Warning: Unexpected kernel type"
+ exit 1
+ ;;
+esac
+
+if [ -f $4/$base-$1 ]; then
+  mv $4/$base-$1 $4/$base-$1.old
+fi
+cat $2 > $4/$base-$1
+
+# Install system map file
+if [ -f $4/System.map-$1 ]; then
+  mv $4/System.map-$1 $4/System.map-$1.old
+fi
+cp $3 $4/System.map-$1
+
+# Install kernel config file
+if [ -f $4/config-$1 ]; then
+  mv $4/config-$1 $4/config-$1.old
+fi
+cp .config $4/config-$1

arch/loongarch/include/asm/acpi.h

@@ -33,7 +33,7 @@ static inline bool acpi_has_cpu_in_madt(void)
 	return true;
 }
 
-#define MAX_CORE_PIC 256
+#define MAX_CORE_PIC 2048
 
 extern struct list_head acpi_wakeup_device_list;
 extern struct acpi_madt_core_pic acpi_core_pic[MAX_CORE_PIC];

arch/loongarch/include/asm/entry-common.h

@@ -2,12 +2,6 @@
 #ifndef ARCH_LOONGARCH_ENTRY_COMMON_H
 #define ARCH_LOONGARCH_ENTRY_COMMON_H
 
-#include <linux/sched.h>
-#include <linux/processor.h>
-
-static inline bool on_thread_stack(void)
-{
-	return !(((unsigned long)(current->stack) ^ current_stack_pointer) & ~(THREAD_SIZE - 1));
-}
+#include <asm/stacktrace.h> /* For on_thread_stack() */
 
 #endif

arch/loongarch/include/asm/kvm_host.h

@@ -301,7 +301,7 @@ int kvm_arch_vcpu_dump_regs(struct kvm_vcpu *vcpu);
 /* MMU handling */
 void kvm_flush_tlb_all(void);
 void kvm_flush_tlb_gpa(struct kvm_vcpu *vcpu, unsigned long gpa);
-int kvm_handle_mm_fault(struct kvm_vcpu *vcpu, unsigned long badv, bool write);
+int kvm_handle_mm_fault(struct kvm_vcpu *vcpu, unsigned long badv, bool write, int ecode);
 
 int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end, bool blockable);
 int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);

arch/loongarch/include/asm/kvm_vcpu.h

@@ -37,7 +37,7 @@
 #define KVM_LOONGSON_IRQ_NUM_MASK	0xffff
 
 typedef union loongarch_instruction  larch_inst;
-typedef int (*exit_handle_fn)(struct kvm_vcpu *);
+typedef int (*exit_handle_fn)(struct kvm_vcpu *, int);
 
 int  kvm_emu_mmio_read(struct kvm_vcpu *vcpu, larch_inst inst);
 int  kvm_emu_mmio_write(struct kvm_vcpu *vcpu, larch_inst inst);

arch/loongarch/include/asm/loongarch.h

@@ -411,8 +411,8 @@
 
 /* Config CSR registers */
 #define LOONGARCH_CSR_CPUID		0x20	/* CPU core id */
-#define  CSR_CPUID_COREID_WIDTH		9
-#define  CSR_CPUID_COREID		_ULCAST_(0x1ff)
+#define  CSR_CPUID_COREID_WIDTH		11
+#define  CSR_CPUID_COREID		_ULCAST_(0x7ff)
 
 #define LOONGARCH_CSR_PRCFG1		0x21	/* Config1 */
 #define  CSR_CONF1_VSMAX_SHIFT		12

arch/loongarch/include/asm/numa.h

@@ -22,20 +22,6 @@ extern int numa_off;
 extern s16 __cpuid_to_node[CONFIG_NR_CPUS];
 extern nodemask_t numa_nodes_parsed __initdata;
 
-struct numa_memblk {
-	u64			start;
-	u64			end;
-	int			nid;
-};
-
-#define NR_NODE_MEMBLKS		(MAX_NUMNODES*2)
-struct numa_meminfo {
-	int			nr_blks;
-	struct numa_memblk	blk[NR_NODE_MEMBLKS];
-};
-
-extern int __init numa_add_memblk(int nodeid, u64 start, u64 end);
-
 extern void __init early_numa_add_cpu(int cpuid, s16 node);
 extern void numa_add_cpu(unsigned int cpu);
 extern void numa_remove_cpu(unsigned int cpu);

arch/loongarch/include/asm/smp.h

@@ -25,6 +25,7 @@ extern int smp_num_siblings;
 extern int num_processors;
 extern int disabled_cpus;
 extern cpumask_t cpu_sibling_map[];
+extern cpumask_t cpu_llc_shared_map[];
 extern cpumask_t cpu_core_map[];
 extern cpumask_t cpu_foreign_map[];
 

arch/loongarch/include/asm/sparsemem.h

@@ -21,11 +21,6 @@
 #define VMEMMAP_SIZE	0	/* 1, For FLATMEM; 2, For SPARSEMEM without VMEMMAP. */
 #endif
 
-#ifdef CONFIG_MEMORY_HOTPLUG
-int memory_add_physaddr_to_nid(u64 addr);
-#define memory_add_physaddr_to_nid memory_add_physaddr_to_nid
-#endif
-
 #define INIT_MEMBLOCK_RESERVED_REGIONS	(INIT_MEMBLOCK_REGIONS + NR_CPUS)
 
 #endif /* _LOONGARCH_SPARSEMEM_H */

arch/loongarch/include/asm/stackframe.h

@@ -57,6 +57,12 @@
 	jirl	zero, \temp1, 0xc
 	.endm
 
+	.macro STACKLEAK_ERASE
+#ifdef CONFIG_GCC_PLUGIN_STACKLEAK
+	bl	stackleak_erase_on_task_stack
+#endif
+	.endm
+
 	.macro BACKUP_T0T1
 	csrwr	t0, EXCEPTION_KS0
 	csrwr	t1, EXCEPTION_KS1

arch/loongarch/include/asm/stacktrace.h

@@ -31,6 +31,11 @@ bool in_irq_stack(unsigned long stack, struct stack_info *info);
 bool in_task_stack(unsigned long stack, struct task_struct *task, struct stack_info *info);
 int get_stack_info(unsigned long stack, struct task_struct *task, struct stack_info *info);
 
+static __always_inline bool on_thread_stack(void)
+{
+	return !(((unsigned long)(current->stack) ^ current_stack_pointer) & ~(THREAD_SIZE - 1));
+}
+
 #define STR_LONG_L    __stringify(LONG_L)
 #define STR_LONG_S    __stringify(LONG_S)
 #define STR_LONGSIZE  __stringify(LONGSIZE)

arch/loongarch/include/asm/topology.h

@@ -19,17 +19,22 @@ extern int pcibus_to_node(struct pci_bus *);
 
 #define cpumask_of_pcibus(bus)	(cpu_online_mask)
 
-extern unsigned char node_distances[MAX_NUMNODES][MAX_NUMNODES];
-
-void numa_set_distance(int from, int to, int distance);
-
-#define node_distance(from, to)	(node_distances[(from)][(to)])
+int __node_distance(int from, int to);
+#define node_distance(from, to) __node_distance(from, to)
 
 #else
 #define pcibus_to_node(bus)	0
 #endif
 
 #ifdef CONFIG_SMP
+/*
+ * Return cpus that shares the last level cache.
+ */
+static inline const struct cpumask *cpu_coregroup_mask(int cpu)
+{
+	return &cpu_llc_shared_map[cpu];
+}
+
 #define topology_physical_package_id(cpu)	(cpu_data[cpu].package)
 #define topology_core_id(cpu)			(cpu_data[cpu].core)
 #define topology_core_cpumask(cpu)		(&cpu_core_map[cpu])

arch/loongarch/kernel/acpi.c

@@ -244,22 +244,6 @@ void __init acpi_boot_table_init(void)
 
 #ifdef CONFIG_ACPI_NUMA
 
-static __init int setup_node(int pxm)
-{
-	return acpi_map_pxm_to_node(pxm);
-}
-
-void __init numa_set_distance(int from, int to, int distance)
-{
-	if ((u8)distance != distance || (from == to && distance != LOCAL_DISTANCE)) {
-		pr_warn_once("Warning: invalid distance parameter, from=%d to=%d distance=%d\n",
-				from, to, distance);
-		return;
-	}
-
-	node_distances[from][to] = distance;
-}
-
 /* Callback for Proximity Domain -> CPUID mapping */
 void __init
 acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
@@ -280,7 +264,41 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 		pxm |= (pa->proximity_domain_hi[1] << 16);
 		pxm |= (pa->proximity_domain_hi[2] << 24);
 	}
-	node = setup_node(pxm);
+	node = acpi_map_pxm_to_node(pxm);
+	if (node < 0) {
+		pr_err("SRAT: Too many proximity domains %x\n", pxm);
+		bad_srat();
+		return;
+	}
+
+	if (pa->apic_id >= CONFIG_NR_CPUS) {
+		pr_info("SRAT: PXM %u -> CPU 0x%02x -> Node %u skipped apicid that is too big\n",
+				pxm, pa->apic_id, node);
+		return;
+	}
+
+	early_numa_add_cpu(pa->apic_id, node);
+
+	set_cpuid_to_node(pa->apic_id, node);
+	node_set(node, numa_nodes_parsed);
+	pr_info("SRAT: PXM %u -> CPU 0x%02x -> Node %u\n", pxm, pa->apic_id, node);
+}
+
+void __init
+acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa)
+{
+	int pxm, node;
+
+	if (srat_disabled())
+		return;
+	if (pa->header.length < sizeof(struct acpi_srat_x2apic_cpu_affinity)) {
+		bad_srat();
+		return;
+	}
+	if ((pa->flags & ACPI_SRAT_CPU_ENABLED) == 0)
+		return;
+	pxm = pa->proximity_domain;
+	node = acpi_map_pxm_to_node(pxm);
 	if (node < 0) {
 		pr_err("SRAT: Too many proximity domains %x\n", pxm);
 		bad_srat();

arch/loongarch/kernel/entry.S

@@ -73,6 +73,7 @@ SYM_CODE_START(handle_syscall)
 	move		a0, sp
 	bl		do_syscall
 
+	STACKLEAK_ERASE
 	RESTORE_ALL_AND_RET
 SYM_CODE_END(handle_syscall)
 _ASM_NOKPROBE(handle_syscall)
@@ -82,6 +83,7 @@ SYM_CODE_START(ret_from_fork)
 	bl		schedule_tail		# a0 = struct task_struct *prev
 	move		a0, sp
 	bl 		syscall_exit_to_user_mode
+	STACKLEAK_ERASE
 	RESTORE_STATIC
 	RESTORE_SOME
 	RESTORE_SP_AND_RET
@@ -94,6 +96,7 @@ SYM_CODE_START(ret_from_kernel_thread)
 	jirl		ra, s0, 0
 	move		a0, sp
 	bl		syscall_exit_to_user_mode
+	STACKLEAK_ERASE
 	RESTORE_STATIC
 	RESTORE_SOME
 	RESTORE_SP_AND_RET

arch/loongarch/kernel/numa.c

@@ -11,6 +11,7 @@
 #include <linux/mmzone.h>
 #include <linux/export.h>
 #include <linux/nodemask.h>
+#include <linux/numa_memblks.h>
 #include <linux/swap.h>
 #include <linux/memblock.h>
 #include <linux/pfn.h>
@@ -27,10 +28,6 @@
 #include <asm/time.h>
 
 int numa_off;
-unsigned char node_distances[MAX_NUMNODES][MAX_NUMNODES];
-EXPORT_SYMBOL(node_distances);
-
-static struct numa_meminfo numa_meminfo;
 cpumask_t cpus_on_node[MAX_NUMNODES];
 cpumask_t phys_cpus_on_node[MAX_NUMNODES];
 EXPORT_SYMBOL(cpus_on_node);
@@ -43,8 +40,6 @@ s16 __cpuid_to_node[CONFIG_NR_CPUS] = {
 };
 EXPORT_SYMBOL(__cpuid_to_node);
 
-nodemask_t numa_nodes_parsed __initdata;
-
 #ifdef CONFIG_HAVE_SETUP_PER_CPU_AREA
 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 EXPORT_SYMBOL(__per_cpu_offset);
@@ -145,48 +140,6 @@ void numa_remove_cpu(unsigned int cpu)
 	cpumask_clear_cpu(cpu, &cpus_on_node[nid]);
 }
 
-static int __init numa_add_memblk_to(int nid, u64 start, u64 end,
-				     struct numa_meminfo *mi)
-{
-	/* ignore zero length blks */
-	if (start == end)
-		return 0;
-
-	/* whine about and ignore invalid blks */
-	if (start > end || nid < 0 || nid >= MAX_NUMNODES) {
-		pr_warn("NUMA: Warning: invalid memblk node %d [mem %#010Lx-%#010Lx]\n",
-			   nid, start, end - 1);
-		return 0;
-	}
-
-	if (mi->nr_blks >= NR_NODE_MEMBLKS) {
-		pr_err("NUMA: too many memblk ranges\n");
-		return -EINVAL;
-	}
-
-	mi->blk[mi->nr_blks].start = PFN_ALIGN(start);
-	mi->blk[mi->nr_blks].end = PFN_ALIGN(end - PAGE_SIZE + 1);
-	mi->blk[mi->nr_blks].nid = nid;
-	mi->nr_blks++;
-	return 0;
-}
-
-/**
- * numa_add_memblk - Add one numa_memblk to numa_meminfo
- * @nid: NUMA node ID of the new memblk
- * @start: Start address of the new memblk
- * @end: End address of the new memblk
- *
- * Add a new memblk to the default numa_meminfo.
- *
- * RETURNS:
- * 0 on success, -errno on failure.
- */
-int __init numa_add_memblk(int nid, u64 start, u64 end)
-{
-	return numa_add_memblk_to(nid, start, end, &numa_meminfo);
-}
-
 static void __init node_mem_init(unsigned int node)
 {
 	unsigned long start_pfn, end_pfn;
@@ -205,18 +158,6 @@ static void __init node_mem_init(unsigned int node)
 
 #ifdef CONFIG_ACPI_NUMA
 
-static void __init add_node_intersection(u32 node, u64 start, u64 size, u32 type)
-{
-	static unsigned long num_physpages;
-
-	num_physpages += (size >> PAGE_SHIFT);
-	pr_info("Node%d: mem_type:%d, mem_start:0x%llx, mem_size:0x%llx Bytes\n",
-		node, type, start, size);
-	pr_info("       start_pfn:0x%llx, end_pfn:0x%llx, num_physpages:0x%lx\n",
-		start >> PAGE_SHIFT, (start + size) >> PAGE_SHIFT, num_physpages);
-	memblock_set_node(start, size, &memblock.memory, node);
-}
-
 /*
  * add_numamem_region
  *
@@ -228,28 +169,21 @@ static void __init add_node_intersection(u32 node, u64 start, u64 size, u32 type
  */
 static void __init add_numamem_region(u64 start, u64 end, u32 type)
 {
-	u32 i;
-	u64 ofs = start;
+	u32 node = pa_to_nid(start);
+	u64 size = end - start;
+	static unsigned long num_physpages;
 
 	if (start >= end) {
 		pr_debug("Invalid region: %016llx-%016llx\n", start, end);
 		return;
 	}
 
-	for (i = 0; i < numa_meminfo.nr_blks; i++) {
-		struct numa_memblk *mb = &numa_meminfo.blk[i];
-
-		if (ofs > mb->end)
-			continue;
-
-		if (end > mb->end) {
-			add_node_intersection(mb->nid, ofs, mb->end - ofs, type);
-			ofs = mb->end;
-		} else {
-			add_node_intersection(mb->nid, ofs, end - ofs, type);
-			break;
-		}
-	}
+	num_physpages += (size >> PAGE_SHIFT);
+	pr_info("Node%d: mem_type:%d, mem_start:0x%llx, mem_size:0x%llx Bytes\n",
+		node, type, start, size);
+	pr_info("       start_pfn:0x%llx, end_pfn:0x%llx, num_physpages:0x%lx\n",
+		start >> PAGE_SHIFT, end >> PAGE_SHIFT, num_physpages);
+	memblock_set_node(start, size, &memblock.memory, node);
 }
 
 static void __init init_node_memblock(void)
@@ -291,24 +225,6 @@ static void __init init_node_memblock(void)
 	}
 }
 
-static void __init numa_default_distance(void)
-{
-	int row, col;
-
-	for (row = 0; row < MAX_NUMNODES; row++)
-		for (col = 0; col < MAX_NUMNODES; col++) {
-			if (col == row)
-				node_distances[row][col] = LOCAL_DISTANCE;
-			else
-				/* We assume that one node per package here!
-				 *
-				 * A SLIT should be used for multiple nodes
-				 * per package to override default setting.
-				 */
-				node_distances[row][col] = REMOTE_DISTANCE;
-	}
-}
-
 /*
  * fake_numa_init() - For Non-ACPI systems
  * Return: 0 on success, -errno on failure.
@@ -333,11 +249,11 @@ int __init init_numa_memory(void)
 	for (i = 0; i < NR_CPUS; i++)
 		set_cpuid_to_node(i, NUMA_NO_NODE);
 
-	numa_default_distance();
+	numa_reset_distance();
 	nodes_clear(numa_nodes_parsed);
 	nodes_clear(node_possible_map);
 	nodes_clear(node_online_map);
-	memset(&numa_meminfo, 0, sizeof(numa_meminfo));
+	WARN_ON(memblock_clear_hotplug(0, PHYS_ADDR_MAX));
 
 	/* Parse SRAT and SLIT if provided by firmware. */
 	ret = acpi_disabled ? fake_numa_init() : acpi_numa_init();

arch/loongarch/kernel/smp.c

@@ -46,6 +46,10 @@ EXPORT_SYMBOL(__cpu_logical_map);
 cpumask_t cpu_sibling_map[NR_CPUS] __read_mostly;
 EXPORT_SYMBOL(cpu_sibling_map);
 
+/* Representing the last level cache shared map of each logical CPU */
+cpumask_t cpu_llc_shared_map[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(cpu_llc_shared_map);
+
 /* Representing the core map of multi-core chips of each logical CPU */
 cpumask_t cpu_core_map[NR_CPUS] __read_mostly;
 EXPORT_SYMBOL(cpu_core_map);
@@ -63,6 +67,9 @@ EXPORT_SYMBOL(cpu_foreign_map);
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
 
+/* representing cpus for which llc shared maps can be computed */
+static cpumask_t cpu_llc_shared_setup_map;
+
 /* representing cpus for which core maps can be computed */
 static cpumask_t cpu_core_setup_map;
 
@@ -102,6 +109,34 @@ static inline void set_cpu_core_map(int cpu)
 	}
 }
 
+static inline void set_cpu_llc_shared_map(int cpu)
+{
+	int i;
+
+	cpumask_set_cpu(cpu, &cpu_llc_shared_setup_map);
+
+	for_each_cpu(i, &cpu_llc_shared_setup_map) {
+		if (cpu_to_node(cpu) == cpu_to_node(i)) {
+			cpumask_set_cpu(i, &cpu_llc_shared_map[cpu]);
+			cpumask_set_cpu(cpu, &cpu_llc_shared_map[i]);
+		}
+	}
+}
+
+static inline void clear_cpu_llc_shared_map(int cpu)
+{
+	int i;
+
+	for_each_cpu(i, &cpu_llc_shared_setup_map) {
+		if (cpu_to_node(cpu) == cpu_to_node(i)) {
+			cpumask_clear_cpu(i, &cpu_llc_shared_map[cpu]);
+			cpumask_clear_cpu(cpu, &cpu_llc_shared_map[i]);
+		}
+	}
+
+	cpumask_clear_cpu(cpu, &cpu_llc_shared_setup_map);
+}
+
 static inline void set_cpu_sibling_map(int cpu)
 {
 	int i;
@@ -406,6 +441,7 @@ int loongson_cpu_disable(void)
 #endif
 	set_cpu_online(cpu, false);
 	clear_cpu_sibling_map(cpu);
+	clear_cpu_llc_shared_map(cpu);
 	calculate_cpu_foreign_map();
 	local_irq_save(flags);
 	irq_migrate_all_off_this_cpu();
@@ -572,6 +608,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	current_thread_info()->cpu = 0;
 	loongson_prepare_cpus(max_cpus);
 	set_cpu_sibling_map(0);
+	set_cpu_llc_shared_map(0);
 	set_cpu_core_map(0);
 	calculate_cpu_foreign_map();
 #ifndef CONFIG_HOTPLUG_CPU
@@ -613,6 +650,7 @@ asmlinkage void start_secondary(void)
 	loongson_init_secondary();
 
 	set_cpu_sibling_map(cpu);
+	set_cpu_llc_shared_map(cpu);
 	set_cpu_core_map(cpu);
 
 	notify_cpu_starting(cpu);

arch/loongarch/kernel/vdso.c

@@ -105,7 +105,9 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
 
 	vdso_addr = data_addr + VVAR_SIZE;
 	vma = _install_special_mapping(mm, vdso_addr, info->size,
-				       VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC,
+				       VM_READ | VM_EXEC |
+				       VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC |
+				       VM_SEALED_SYSMAP,
 				       &info->code_mapping);
 	if (IS_ERR(vma)) {
 		ret = PTR_ERR(vma);

arch/loongarch/kvm/exit.c

@@ -341,7 +341,7 @@ static int kvm_trap_handle_gspr(struct kvm_vcpu *vcpu)
  * 2) Execute CACOP/IDLE instructions;
  * 3) Access to unimplemented CSRs/IOCSRs.
  */
-static int kvm_handle_gspr(struct kvm_vcpu *vcpu)
+static int kvm_handle_gspr(struct kvm_vcpu *vcpu, int ecode)
 {
 	int ret = RESUME_GUEST;
 	enum emulation_result er = EMULATE_DONE;
@@ -661,7 +661,7 @@ int kvm_emu_mmio_write(struct kvm_vcpu *vcpu, larch_inst inst)
 	return ret;
 }
 
-static int kvm_handle_rdwr_fault(struct kvm_vcpu *vcpu, bool write)
+static int kvm_handle_rdwr_fault(struct kvm_vcpu *vcpu, bool write, int ecode)
 {
 	int ret;
 	larch_inst inst;
@@ -675,7 +675,7 @@ static int kvm_handle_rdwr_fault(struct kvm_vcpu *vcpu, bool write)
 		return RESUME_GUEST;
 	}
 
-	ret = kvm_handle_mm_fault(vcpu, badv, write);
+	ret = kvm_handle_mm_fault(vcpu, badv, write, ecode);
 	if (ret) {
 		/* Treat as MMIO */
 		inst.word = vcpu->arch.badi;
@@ -705,14 +705,14 @@ static int kvm_handle_rdwr_fault(struct kvm_vcpu *vcpu, bool write)
 	return ret;
 }
 
-static int kvm_handle_read_fault(struct kvm_vcpu *vcpu)
+static int kvm_handle_read_fault(struct kvm_vcpu *vcpu, int ecode)
 {
-	return kvm_handle_rdwr_fault(vcpu, false);
+	return kvm_handle_rdwr_fault(vcpu, false, ecode);
 }
 
-static int kvm_handle_write_fault(struct kvm_vcpu *vcpu)
+static int kvm_handle_write_fault(struct kvm_vcpu *vcpu, int ecode)
 {
-	return kvm_handle_rdwr_fault(vcpu, true);
+	return kvm_handle_rdwr_fault(vcpu, true, ecode);
 }
 
 int kvm_complete_user_service(struct kvm_vcpu *vcpu, struct kvm_run *run)
@@ -726,11 +726,12 @@ int kvm_complete_user_service(struct kvm_vcpu *vcpu, struct kvm_run *run)
 /**
  * kvm_handle_fpu_disabled() - Guest used fpu however it is disabled at host
  * @vcpu:	Virtual CPU context.
+ * @ecode:	Exception code.
  *
  * Handle when the guest attempts to use fpu which hasn't been allowed
  * by the root context.
  */
-static int kvm_handle_fpu_disabled(struct kvm_vcpu *vcpu)
+static int kvm_handle_fpu_disabled(struct kvm_vcpu *vcpu, int ecode)
 {
 	struct kvm_run *run = vcpu->run;
 
@@ -783,11 +784,12 @@ static long kvm_save_notify(struct kvm_vcpu *vcpu)
 /*
  * kvm_handle_lsx_disabled() - Guest used LSX while disabled in root.
  * @vcpu:      Virtual CPU context.
+ * @ecode:	Exception code.
  *
  * Handle when the guest attempts to use LSX when it is disabled in the root
  * context.
  */
-static int kvm_handle_lsx_disabled(struct kvm_vcpu *vcpu)
+static int kvm_handle_lsx_disabled(struct kvm_vcpu *vcpu, int ecode)
 {
 	if (kvm_own_lsx(vcpu))
 		kvm_queue_exception(vcpu, EXCCODE_INE, 0);
@@ -798,11 +800,12 @@ static int kvm_handle_lsx_disabled(struct kvm_vcpu *vcpu)
 /*
  * kvm_handle_lasx_disabled() - Guest used LASX while disabled in root.
  * @vcpu:	Virtual CPU context.
+ * @ecode:	Exception code.
  *
  * Handle when the guest attempts to use LASX when it is disabled in the root
  * context.
  */
-static int kvm_handle_lasx_disabled(struct kvm_vcpu *vcpu)
+static int kvm_handle_lasx_disabled(struct kvm_vcpu *vcpu, int ecode)
 {
 	if (kvm_own_lasx(vcpu))
 		kvm_queue_exception(vcpu, EXCCODE_INE, 0);
@@ -810,7 +813,7 @@ static int kvm_handle_lasx_disabled(struct kvm_vcpu *vcpu)
 	return RESUME_GUEST;
 }
 
-static int kvm_handle_lbt_disabled(struct kvm_vcpu *vcpu)
+static int kvm_handle_lbt_disabled(struct kvm_vcpu *vcpu, int ecode)
 {
 	if (kvm_own_lbt(vcpu))
 		kvm_queue_exception(vcpu, EXCCODE_INE, 0);
@@ -872,7 +875,7 @@ static void kvm_handle_service(struct kvm_vcpu *vcpu)
 	kvm_write_reg(vcpu, LOONGARCH_GPR_A0, ret);
 }
 
-static int kvm_handle_hypercall(struct kvm_vcpu *vcpu)
+static int kvm_handle_hypercall(struct kvm_vcpu *vcpu, int ecode)
 {
 	int ret;
 	larch_inst inst;
@@ -932,16 +935,14 @@ static int kvm_handle_hypercall(struct kvm_vcpu *vcpu)
 /*
  * LoongArch KVM callback handling for unimplemented guest exiting
  */
-static int kvm_fault_ni(struct kvm_vcpu *vcpu)
+static int kvm_fault_ni(struct kvm_vcpu *vcpu, int ecode)
 {
-	unsigned int ecode, inst;
-	unsigned long estat, badv;
+	unsigned int inst;
+	unsigned long badv;
 
 	/* Fetch the instruction */
 	inst = vcpu->arch.badi;
 	badv = vcpu->arch.badv;
-	estat = vcpu->arch.host_estat;
-	ecode = (estat & CSR_ESTAT_EXC) >> CSR_ESTAT_EXC_SHIFT;
 	kvm_err("ECode: %d PC=%#lx Inst=0x%08x BadVaddr=%#lx ESTAT=%#lx\n",
 			ecode, vcpu->arch.pc, inst, badv, read_gcsr_estat());
 	kvm_arch_vcpu_dump_regs(vcpu);
@@ -966,5 +967,5 @@ static exit_handle_fn kvm_fault_tables[EXCCODE_INT_START] = {
 
 int kvm_handle_fault(struct kvm_vcpu *vcpu, int fault)
 {
-	return kvm_fault_tables[fault](vcpu);
+	return kvm_fault_tables[fault](vcpu, fault);
 }

arch/loongarch/kvm/mmu.c

@@ -912,7 +912,7 @@ static int kvm_map_page(struct kvm_vcpu *vcpu, unsigned long gpa, bool write)
 	return err;
 }
 
-int kvm_handle_mm_fault(struct kvm_vcpu *vcpu, unsigned long gpa, bool write)
+int kvm_handle_mm_fault(struct kvm_vcpu *vcpu, unsigned long gpa, bool write, int ecode)
 {
 	int ret;
 
@@ -921,8 +921,17 @@ int kvm_handle_mm_fault(struct kvm_vcpu *vcpu, unsigned long gpa, bool write)
 		return ret;
 
 	/* Invalidate this entry in the TLB */
-	vcpu->arch.flush_gpa = gpa;
-	kvm_make_request(KVM_REQ_TLB_FLUSH_GPA, vcpu);
+	if (!cpu_has_ptw || (ecode == EXCCODE_TLBM)) {
+		/*
+		 * With HW PTW, invalid TLB is not added when page fault. But
+		 * for EXCCODE_TLBM exception, stale TLB may exist because of
+		 * the last read access.
+		 *
+		 * With SW PTW, invalid TLB is added in TLB refill exception.
+		 */
+		vcpu->arch.flush_gpa = gpa;
+		kvm_make_request(KVM_REQ_TLB_FLUSH_GPA, vcpu);
+	}
 
 	return 0;
 }

arch/loongarch/mm/init.c

@@ -106,14 +106,6 @@ void arch_remove_memory(u64 start, u64 size, struct vmem_altmap *altmap)
 		page += vmem_altmap_offset(altmap);
 	__remove_pages(start_pfn, nr_pages, altmap);
 }
-
-#ifdef CONFIG_NUMA
-int memory_add_physaddr_to_nid(u64 start)
-{
-	return pa_to_nid(start);
-}
-EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
-#endif
 #endif
 
 #ifdef CONFIG_SPARSEMEM_VMEMMAP

arch/loongarch/pci/acpi.c

@@ -194,6 +194,7 @@ struct pci_bus *pci_acpi_scan_root(struct acpi_pci_root *root)
 {
 	struct pci_bus *bus;
 	struct pci_root_info *info;
+	struct pci_host_bridge *host;
 	struct acpi_pci_root_ops *root_ops;
 	int domain = root->segment;
 	int busnum = root->secondary.start;
@@ -237,8 +238,17 @@ struct pci_bus *pci_acpi_scan_root(struct acpi_pci_root *root)
 			return NULL;
 		}
 
-		pci_bus_size_bridges(bus);
-		pci_bus_assign_resources(bus);
+		/* If we must preserve the resource configuration, claim now */
+		host = pci_find_host_bridge(bus);
+		if (host->preserve_config)
+			pci_bus_claim_resources(bus);
+
+		/*
+		 * Assign whatever was left unassigned. If we didn't claim above,
+		 * this will reassign everything.
+		 */
+		pci_assign_unassigned_root_bus_resources(bus);
+
 		list_for_each_entry(child, &bus->children, node)
 			pcie_bus_configure_settings(child);
 	}

drivers/firmware/efi/libstub/Makefile

@@ -31,7 +31,7 @@ cflags-$(CONFIG_ARM)		+= -DEFI_HAVE_STRLEN -DEFI_HAVE_STRNLEN \
 				   $(DISABLE_STACKLEAK_PLUGIN)
 cflags-$(CONFIG_RISCV)		+= -fpic -DNO_ALTERNATIVE -mno-relax \
 				   $(DISABLE_STACKLEAK_PLUGIN)
-cflags-$(CONFIG_LOONGARCH)	+= -fpie
+cflags-$(CONFIG_LOONGARCH)	+= -fpie $(DISABLE_STACKLEAK_PLUGIN)
 
 cflags-$(CONFIG_EFI_PARAMS_FROM_FDT)	+= -I$(srctree)/scripts/dtc/libfdt
 

mm/Kconfig

@@ -1317,6 +1317,7 @@ config NUMA_MEMBLKS
 config NUMA_EMU
 	bool "NUMA emulation"
 	depends on NUMA_MEMBLKS
+	depends on X86 || GENERIC_ARCH_NUMA
 	help
 	  Enable NUMA emulation. A flat machine will be split
 	  into virtual nodes when booted with "numa=fake=N", where N is the

tools/testing/selftests/kvm/Makefile

@@ -3,7 +3,7 @@ top_srcdir = ../../../..
 include $(top_srcdir)/scripts/subarch.include
 ARCH            ?= $(SUBARCH)
 
-ifeq ($(ARCH),$(filter $(ARCH),arm64 s390 riscv x86 x86_64))
+ifeq ($(ARCH),$(filter $(ARCH),arm64 s390 riscv x86 x86_64 loongarch))
 # Top-level selftests allows ARCH=x86_64 :-(
 ifeq ($(ARCH),x86_64)
 	ARCH := x86

tools/testing/selftests/kvm/Makefile.kvm

@@ -47,6 +47,10 @@ LIBKVM_riscv += lib/riscv/handlers.S
 LIBKVM_riscv += lib/riscv/processor.c
 LIBKVM_riscv += lib/riscv/ucall.c
 
+LIBKVM_loongarch += lib/loongarch/processor.c
+LIBKVM_loongarch += lib/loongarch/ucall.c
+LIBKVM_loongarch += lib/loongarch/exception.S
+
 # Non-compiled test targets
 TEST_PROGS_x86 += x86/nx_huge_pages_test.sh
 
@@ -190,6 +194,19 @@ TEST_GEN_PROGS_riscv += coalesced_io_test
 TEST_GEN_PROGS_riscv += get-reg-list
 TEST_GEN_PROGS_riscv += steal_time
 
+TEST_GEN_PROGS_loongarch += coalesced_io_test
+TEST_GEN_PROGS_loongarch += demand_paging_test
+TEST_GEN_PROGS_loongarch += dirty_log_perf_test
+TEST_GEN_PROGS_loongarch += dirty_log_test
+TEST_GEN_PROGS_loongarch += guest_print_test
+TEST_GEN_PROGS_loongarch += hardware_disable_test
+TEST_GEN_PROGS_loongarch += kvm_binary_stats_test
+TEST_GEN_PROGS_loongarch += kvm_create_max_vcpus
+TEST_GEN_PROGS_loongarch += kvm_page_table_test
+TEST_GEN_PROGS_loongarch += memslot_modification_stress_test
+TEST_GEN_PROGS_loongarch += memslot_perf_test
+TEST_GEN_PROGS_loongarch += set_memory_region_test
+
 SPLIT_TESTS += arch_timer
 SPLIT_TESTS += get-reg-list
 

tools/testing/selftests/kvm/include/kvm_util.h

@@ -177,6 +177,7 @@ enum vm_guest_mode {
 	VM_MODE_P36V48_4K,
 	VM_MODE_P36V48_16K,
 	VM_MODE_P36V48_64K,
+	VM_MODE_P47V47_16K,
 	VM_MODE_P36V47_16K,
 	NUM_VM_MODES,
 };
@@ -232,6 +233,11 @@ extern enum vm_guest_mode vm_mode_default;
 #define MIN_PAGE_SHIFT			12U
 #define ptes_per_page(page_size)	((page_size) / 8)
 
+#elif defined(__loongarch__)
+#define VM_MODE_DEFAULT			VM_MODE_P47V47_16K
+#define MIN_PAGE_SHIFT			12U
+#define ptes_per_page(page_size)	((page_size) / 8)
+
 #endif
 
 #define VM_SHAPE_DEFAULT	VM_SHAPE(VM_MODE_DEFAULT)

tools/testing/selftests/kvm/include/loongarch/kvm_util_arch.h

@@ -0,0 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef SELFTEST_KVM_UTIL_ARCH_H
+#define SELFTEST_KVM_UTIL_ARCH_H
+
+struct kvm_vm_arch {};
+
+#endif  // SELFTEST_KVM_UTIL_ARCH_H

tools/testing/selftests/kvm/include/loongarch/processor.h

@@ -0,0 +1,141 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef SELFTEST_KVM_PROCESSOR_H
+#define SELFTEST_KVM_PROCESSOR_H
+
+#ifndef __ASSEMBLER__
+#include "ucall_common.h"
+
+#else
+/* general registers */
+#define zero				$r0
+#define ra				$r1
+#define tp				$r2
+#define sp				$r3
+#define a0				$r4
+#define a1				$r5
+#define a2				$r6
+#define a3				$r7
+#define a4				$r8
+#define a5				$r9
+#define a6				$r10
+#define a7				$r11
+#define t0				$r12
+#define t1				$r13
+#define t2				$r14
+#define t3				$r15
+#define t4				$r16
+#define t5				$r17
+#define t6				$r18
+#define t7				$r19
+#define t8				$r20
+#define u0				$r21
+#define fp				$r22
+#define s0				$r23
+#define s1				$r24
+#define s2				$r25
+#define s3				$r26
+#define s4				$r27
+#define s5				$r28
+#define s6				$r29
+#define s7				$r30
+#define s8				$r31
+#endif
+
+/*
+ * LoongArch page table entry definition
+ * Original header file arch/loongarch/include/asm/loongarch.h
+ */
+#define _PAGE_VALID_SHIFT		0
+#define _PAGE_DIRTY_SHIFT		1
+#define _PAGE_PLV_SHIFT			2  /* 2~3, two bits */
+#define  PLV_KERN			0
+#define  PLV_USER			3
+#define  PLV_MASK			0x3
+#define _CACHE_SHIFT			4  /* 4~5, two bits */
+#define _PAGE_PRESENT_SHIFT		7
+#define _PAGE_WRITE_SHIFT		8
+
+#define _PAGE_VALID			BIT_ULL(_PAGE_VALID_SHIFT)
+#define _PAGE_PRESENT			BIT_ULL(_PAGE_PRESENT_SHIFT)
+#define _PAGE_WRITE			BIT_ULL(_PAGE_WRITE_SHIFT)
+#define _PAGE_DIRTY			BIT_ULL(_PAGE_DIRTY_SHIFT)
+#define _PAGE_USER			(PLV_USER << _PAGE_PLV_SHIFT)
+#define   __READABLE			(_PAGE_VALID)
+#define   __WRITEABLE			(_PAGE_DIRTY | _PAGE_WRITE)
+/* Coherent Cached */
+#define _CACHE_CC			BIT_ULL(_CACHE_SHIFT)
+#define PS_4K				0x0000000c
+#define PS_16K				0x0000000e
+#define PS_64K				0x00000010
+#define PS_DEFAULT_SIZE			PS_16K
+
+/* LoongArch Basic CSR registers */
+#define LOONGARCH_CSR_CRMD		0x0 /* Current mode info */
+#define  CSR_CRMD_PG_SHIFT		4
+#define  CSR_CRMD_PG			BIT_ULL(CSR_CRMD_PG_SHIFT)
+#define  CSR_CRMD_IE_SHIFT		2
+#define  CSR_CRMD_IE			BIT_ULL(CSR_CRMD_IE_SHIFT)
+#define  CSR_CRMD_PLV_SHIFT		0
+#define  CSR_CRMD_PLV_WIDTH		2
+#define  CSR_CRMD_PLV			(0x3UL << CSR_CRMD_PLV_SHIFT)
+#define  PLV_MASK			0x3
+#define LOONGARCH_CSR_PRMD		0x1
+#define LOONGARCH_CSR_EUEN		0x2
+#define LOONGARCH_CSR_ECFG		0x4
+#define LOONGARCH_CSR_ESTAT		0x5  /* Exception status */
+#define LOONGARCH_CSR_ERA		0x6  /* ERA */
+#define LOONGARCH_CSR_BADV		0x7  /* Bad virtual address */
+#define LOONGARCH_CSR_EENTRY		0xc
+#define LOONGARCH_CSR_TLBIDX		0x10 /* TLB Index, EHINV, PageSize */
+#define  CSR_TLBIDX_PS_SHIFT		24
+#define  CSR_TLBIDX_PS_WIDTH		6
+#define  CSR_TLBIDX_PS			(0x3fUL << CSR_TLBIDX_PS_SHIFT)
+#define  CSR_TLBIDX_SIZEM		0x3f000000
+#define  CSR_TLBIDX_SIZE		CSR_TLBIDX_PS_SHIFT
+#define LOONGARCH_CSR_ASID		0x18 /* ASID */
+#define LOONGARCH_CSR_PGDL		0x19
+#define LOONGARCH_CSR_PGDH		0x1a
+/* Page table base */
+#define LOONGARCH_CSR_PGD		0x1b
+#define LOONGARCH_CSR_PWCTL0		0x1c
+#define LOONGARCH_CSR_PWCTL1		0x1d
+#define LOONGARCH_CSR_STLBPGSIZE	0x1e
+#define LOONGARCH_CSR_CPUID		0x20
+#define LOONGARCH_CSR_KS0		0x30
+#define LOONGARCH_CSR_KS1		0x31
+#define LOONGARCH_CSR_TMID		0x40
+#define LOONGARCH_CSR_TCFG		0x41
+/* TLB refill exception entry */
+#define LOONGARCH_CSR_TLBRENTRY		0x88
+#define LOONGARCH_CSR_TLBRSAVE		0x8b
+#define LOONGARCH_CSR_TLBREHI		0x8e
+#define  CSR_TLBREHI_PS_SHIFT		0
+#define  CSR_TLBREHI_PS			(0x3fUL << CSR_TLBREHI_PS_SHIFT)
+
+#define EXREGS_GPRS			(32)
+
+#ifndef __ASSEMBLER__
+void handle_tlb_refill(void);
+void handle_exception(void);
+
+struct ex_regs {
+	unsigned long regs[EXREGS_GPRS];
+	unsigned long pc;
+	unsigned long estat;
+	unsigned long badv;
+};
+
+#define PC_OFFSET_EXREGS		offsetof(struct ex_regs, pc)
+#define ESTAT_OFFSET_EXREGS		offsetof(struct ex_regs, estat)
+#define BADV_OFFSET_EXREGS		offsetof(struct ex_regs, badv)
+#define EXREGS_SIZE			sizeof(struct ex_regs)
+
+#else
+#define PC_OFFSET_EXREGS		((EXREGS_GPRS + 0) * 8)
+#define ESTAT_OFFSET_EXREGS		((EXREGS_GPRS + 1) * 8)
+#define BADV_OFFSET_EXREGS		((EXREGS_GPRS + 2) * 8)
+#define EXREGS_SIZE			((EXREGS_GPRS + 3) * 8)
+#endif
+
+#endif /* SELFTEST_KVM_PROCESSOR_H */

tools/testing/selftests/kvm/include/loongarch/ucall.h

@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef SELFTEST_KVM_UCALL_H
+#define SELFTEST_KVM_UCALL_H
+
+#include "kvm_util.h"
+
+#define UCALL_EXIT_REASON       KVM_EXIT_MMIO
+
+/*
+ * ucall_exit_mmio_addr holds per-VM values (global data is duplicated by each
+ * VM), it must not be accessed from host code.
+ */
+extern vm_vaddr_t *ucall_exit_mmio_addr;
+
+static inline void ucall_arch_do_ucall(vm_vaddr_t uc)
+{
+	WRITE_ONCE(*ucall_exit_mmio_addr, uc);
+}
+
+#endif

tools/testing/selftests/kvm/lib/kvm_util.c

@@ -222,6 +222,7 @@ const char *vm_guest_mode_string(uint32_t i)
 		[VM_MODE_P36V48_4K]	= "PA-bits:36,  VA-bits:48,  4K pages",
 		[VM_MODE_P36V48_16K]	= "PA-bits:36,  VA-bits:48, 16K pages",
 		[VM_MODE_P36V48_64K]	= "PA-bits:36,  VA-bits:48, 64K pages",
+		[VM_MODE_P47V47_16K]	= "PA-bits:47,  VA-bits:47, 16K pages",
 		[VM_MODE_P36V47_16K]	= "PA-bits:36,  VA-bits:47, 16K pages",
 	};
 	_Static_assert(sizeof(strings)/sizeof(char *) == NUM_VM_MODES,
@@ -248,6 +249,7 @@ const struct vm_guest_mode_params vm_guest_mode_params[] = {
 	[VM_MODE_P36V48_4K]	= { 36, 48,  0x1000, 12 },
 	[VM_MODE_P36V48_16K]	= { 36, 48,  0x4000, 14 },
 	[VM_MODE_P36V48_64K]	= { 36, 48, 0x10000, 16 },
+	[VM_MODE_P47V47_16K]	= { 47, 47,  0x4000, 14 },
 	[VM_MODE_P36V47_16K]	= { 36, 47,  0x4000, 14 },
 };
 _Static_assert(sizeof(vm_guest_mode_params)/sizeof(struct vm_guest_mode_params) == NUM_VM_MODES,
@@ -319,6 +321,7 @@ struct kvm_vm *____vm_create(struct vm_shape shape)
 	case VM_MODE_P36V48_16K:
 		vm->pgtable_levels = 4;
 		break;
+	case VM_MODE_P47V47_16K:
 	case VM_MODE_P36V47_16K:
 		vm->pgtable_levels = 3;
 		break;

tools/testing/selftests/kvm/lib/loongarch/exception.S

@@ -0,0 +1,59 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#include "processor.h"
+
+/* address of refill exception should be 4K aligned */
+.balign	4096
+.global handle_tlb_refill
+handle_tlb_refill:
+	csrwr	t0, LOONGARCH_CSR_TLBRSAVE
+	csrrd	t0, LOONGARCH_CSR_PGD
+	lddir	t0, t0, 3
+	lddir	t0, t0, 1
+	ldpte	t0, 0
+	ldpte	t0, 1
+	tlbfill
+	csrrd	t0, LOONGARCH_CSR_TLBRSAVE
+	ertn
+
+	/*
+	 * save and restore all gprs except base register,
+	 * and default value of base register is sp ($r3).
+	 */
+.macro save_gprs base
+	.irp n,1,2,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+	st.d    $r\n, \base, 8 * \n
+	.endr
+.endm
+
+.macro restore_gprs base
+	.irp n,1,2,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+	ld.d    $r\n, \base, 8 * \n
+	.endr
+.endm
+
+/* address of general exception should be 4K aligned */
+.balign	4096
+.global handle_exception
+handle_exception:
+	csrwr  sp, LOONGARCH_CSR_KS0
+	csrrd  sp, LOONGARCH_CSR_KS1
+	addi.d sp, sp, -EXREGS_SIZE
+
+	save_gprs sp
+	/* save sp register to stack */
+	csrrd  t0, LOONGARCH_CSR_KS0
+	st.d   t0, sp, 3 * 8
+
+	csrrd  t0, LOONGARCH_CSR_ERA
+	st.d   t0, sp, PC_OFFSET_EXREGS
+	csrrd  t0, LOONGARCH_CSR_ESTAT
+	st.d   t0, sp, ESTAT_OFFSET_EXREGS
+	csrrd  t0, LOONGARCH_CSR_BADV
+	st.d   t0, sp, BADV_OFFSET_EXREGS
+
+	or     a0, sp, zero
+	bl route_exception
+	restore_gprs sp
+	csrrd  sp, LOONGARCH_CSR_KS0
+	ertn

tools/testing/selftests/kvm/lib/loongarch/processor.c

@@ -0,0 +1,346 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <assert.h>
+#include <linux/compiler.h>
+
+#include "kvm_util.h"
+#include "processor.h"
+#include "ucall_common.h"
+
+#define LOONGARCH_PAGE_TABLE_PHYS_MIN		0x200000
+#define LOONGARCH_GUEST_STACK_VADDR_MIN		0x200000
+
+static vm_paddr_t invalid_pgtable[4];
+
+static uint64_t virt_pte_index(struct kvm_vm *vm, vm_vaddr_t gva, int level)
+{
+	unsigned int shift;
+	uint64_t mask;
+
+	shift = level * (vm->page_shift - 3) + vm->page_shift;
+	mask = (1UL << (vm->page_shift - 3)) - 1;
+	return (gva >> shift) & mask;
+}
+
+static uint64_t pte_addr(struct kvm_vm *vm, uint64_t entry)
+{
+	return entry &  ~((0x1UL << vm->page_shift) - 1);
+}
+
+static uint64_t ptrs_per_pte(struct kvm_vm *vm)
+{
+	return 1 << (vm->page_shift - 3);
+}
+
+static void virt_set_pgtable(struct kvm_vm *vm, vm_paddr_t table, vm_paddr_t child)
+{
+	uint64_t *ptep;
+	int i, ptrs_per_pte;
+
+	ptep = addr_gpa2hva(vm, table);
+	ptrs_per_pte = 1 << (vm->page_shift - 3);
+	for (i = 0; i < ptrs_per_pte; i++)
+		WRITE_ONCE(*(ptep + i), child);
+}
+
+void virt_arch_pgd_alloc(struct kvm_vm *vm)
+{
+	int i;
+	vm_paddr_t child, table;
+
+	if (vm->pgd_created)
+		return;
+
+	child = table = 0;
+	for (i = 0; i < vm->pgtable_levels; i++) {
+		invalid_pgtable[i] = child;
+		table = vm_phy_page_alloc(vm, LOONGARCH_PAGE_TABLE_PHYS_MIN,
+				vm->memslots[MEM_REGION_PT]);
+		TEST_ASSERT(table, "Fail to allocate page tale at level %d\n", i);
+		virt_set_pgtable(vm, table, child);
+		child = table;
+	}
+	vm->pgd = table;
+	vm->pgd_created = true;
+}
+
+static int virt_pte_none(uint64_t *ptep, int level)
+{
+	return *ptep == invalid_pgtable[level];
+}
+
+static uint64_t *virt_populate_pte(struct kvm_vm *vm, vm_vaddr_t gva, int alloc)
+{
+	int level;
+	uint64_t *ptep;
+	vm_paddr_t child;
+
+	if (!vm->pgd_created)
+		goto unmapped_gva;
+
+	child = vm->pgd;
+	level = vm->pgtable_levels - 1;
+	while (level > 0) {
+		ptep = addr_gpa2hva(vm, child) + virt_pte_index(vm, gva, level) * 8;
+		if (virt_pte_none(ptep, level)) {
+			if (alloc) {
+				child = vm_alloc_page_table(vm);
+				virt_set_pgtable(vm, child, invalid_pgtable[level - 1]);
+				WRITE_ONCE(*ptep, child);
+			} else
+				goto unmapped_gva;
+
+		} else
+			child = pte_addr(vm, *ptep);
+		level--;
+	}
+
+	ptep = addr_gpa2hva(vm, child) + virt_pte_index(vm, gva, level) * 8;
+	return ptep;
+
+unmapped_gva:
+	TEST_FAIL("No mapping for vm virtual address, gva: 0x%lx", gva);
+	exit(EXIT_FAILURE);
+}
+
+vm_paddr_t addr_arch_gva2gpa(struct kvm_vm *vm, vm_vaddr_t gva)
+{
+	uint64_t *ptep;
+
+	ptep = virt_populate_pte(vm, gva, 0);
+	TEST_ASSERT(*ptep != 0, "Virtual address vaddr: 0x%lx not mapped\n", gva);
+
+	return pte_addr(vm, *ptep) + (gva & (vm->page_size - 1));
+}
+
+void virt_arch_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr)
+{
+	uint32_t prot_bits;
+	uint64_t *ptep;
+
+	TEST_ASSERT((vaddr % vm->page_size) == 0,
+			"Virtual address not on page boundary,\n"
+			"vaddr: 0x%lx vm->page_size: 0x%x", vaddr, vm->page_size);
+	TEST_ASSERT(sparsebit_is_set(vm->vpages_valid,
+			(vaddr >> vm->page_shift)),
+			"Invalid virtual address, vaddr: 0x%lx", vaddr);
+	TEST_ASSERT((paddr % vm->page_size) == 0,
+			"Physical address not on page boundary,\n"
+			"paddr: 0x%lx vm->page_size: 0x%x", paddr, vm->page_size);
+	TEST_ASSERT((paddr >> vm->page_shift) <= vm->max_gfn,
+			"Physical address beyond maximum supported,\n"
+			"paddr: 0x%lx vm->max_gfn: 0x%lx vm->page_size: 0x%x",
+			paddr, vm->max_gfn, vm->page_size);
+
+	ptep = virt_populate_pte(vm, vaddr, 1);
+	prot_bits = _PAGE_PRESENT | __READABLE | __WRITEABLE | _CACHE_CC | _PAGE_USER;
+	WRITE_ONCE(*ptep, paddr | prot_bits);
+}
+
+static void pte_dump(FILE *stream, struct kvm_vm *vm, uint8_t indent, uint64_t page, int level)
+{
+	uint64_t pte, *ptep;
+	static const char * const type[] = { "pte", "pmd", "pud", "pgd"};
+
+	if (level < 0)
+		return;
+
+	for (pte = page; pte < page + ptrs_per_pte(vm) * 8; pte += 8) {
+		ptep = addr_gpa2hva(vm, pte);
+		if (virt_pte_none(ptep, level))
+			continue;
+		fprintf(stream, "%*s%s: %lx: %lx at %p\n",
+				indent, "", type[level], pte, *ptep, ptep);
+		pte_dump(stream, vm, indent + 1, pte_addr(vm, *ptep), level--);
+	}
+}
+
+void virt_arch_dump(FILE *stream, struct kvm_vm *vm, uint8_t indent)
+{
+	int level;
+
+	if (!vm->pgd_created)
+		return;
+
+	level = vm->pgtable_levels - 1;
+	pte_dump(stream, vm, indent, vm->pgd, level);
+}
+
+void vcpu_arch_dump(FILE *stream, struct kvm_vcpu *vcpu, uint8_t indent)
+{
+}
+
+void assert_on_unhandled_exception(struct kvm_vcpu *vcpu)
+{
+	struct ucall uc;
+
+	if (get_ucall(vcpu, &uc) != UCALL_UNHANDLED)
+		return;
+
+	TEST_FAIL("Unexpected exception (pc:0x%lx, estat:0x%lx, badv:0x%lx)",
+			uc.args[0], uc.args[1], uc.args[2]);
+}
+
+void route_exception(struct ex_regs *regs)
+{
+	unsigned long pc, estat, badv;
+
+	pc = regs->pc;
+	badv  = regs->badv;
+	estat = regs->estat;
+	ucall(UCALL_UNHANDLED, 3, pc, estat, badv);
+	while (1) ;
+}
+
+void vcpu_args_set(struct kvm_vcpu *vcpu, unsigned int num, ...)
+{
+	int i;
+	va_list ap;
+	struct kvm_regs regs;
+
+	TEST_ASSERT(num >= 1 && num <= 8, "Unsupported number of args,\n"
+		    "num: %u\n", num);
+
+	vcpu_regs_get(vcpu, &regs);
+
+	va_start(ap, num);
+	for (i = 0; i < num; i++)
+		regs.gpr[i + 4] = va_arg(ap, uint64_t);
+	va_end(ap);
+
+	vcpu_regs_set(vcpu, &regs);
+}
+
+static void loongarch_get_csr(struct kvm_vcpu *vcpu, uint64_t id, void *addr)
+{
+	uint64_t csrid;
+
+	csrid = KVM_REG_LOONGARCH_CSR | KVM_REG_SIZE_U64 | 8 * id;
+	__vcpu_get_reg(vcpu, csrid, addr);
+}
+
+static void loongarch_set_csr(struct kvm_vcpu *vcpu, uint64_t id, uint64_t val)
+{
+	uint64_t csrid;
+
+	csrid = KVM_REG_LOONGARCH_CSR | KVM_REG_SIZE_U64 | 8 * id;
+	__vcpu_set_reg(vcpu, csrid, val);
+}
+
+static void loongarch_vcpu_setup(struct kvm_vcpu *vcpu)
+{
+	int width;
+	unsigned long val;
+	struct kvm_vm *vm = vcpu->vm;
+
+	switch (vm->mode) {
+	case VM_MODE_P36V47_16K:
+	case VM_MODE_P47V47_16K:
+		break;
+
+	default:
+		TEST_FAIL("Unknown guest mode, mode: 0x%x", vm->mode);
+	}
+
+	/* user mode and page enable mode */
+	val = PLV_USER | CSR_CRMD_PG;
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_CRMD, val);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_PRMD, val);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_EUEN, 1);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_ECFG, 0);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_TCFG, 0);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_ASID, 1);
+
+	val = 0;
+	width = vm->page_shift - 3;
+
+	switch (vm->pgtable_levels) {
+	case 4:
+		/* pud page shift and width */
+		val = (vm->page_shift + width * 2) << 20 | (width << 25);
+		/* fall throuth */
+	case 3:
+		/* pmd page shift and width */
+		val |= (vm->page_shift + width) << 10 | (width << 15);
+		/* pte page shift and width */
+		val |= vm->page_shift | width << 5;
+		break;
+	default:
+		TEST_FAIL("Got %u page table levels, expected 3 or 4", vm->pgtable_levels);
+	}
+
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_PWCTL0, val);
+
+	/* PGD page shift and width */
+	val = (vm->page_shift + width * (vm->pgtable_levels - 1)) | width << 6;
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_PWCTL1, val);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_PGDL, vm->pgd);
+
+	/*
+	 * Refill exception runs on real mode
+	 * Entry address should be physical address
+	 */
+	val = addr_gva2gpa(vm, (unsigned long)handle_tlb_refill);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_TLBRENTRY, val);
+
+	/*
+	 * General exception runs on page-enabled mode
+	 * Entry address should be virtual address
+	 */
+	val = (unsigned long)handle_exception;
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_EENTRY, val);
+
+	loongarch_get_csr(vcpu, LOONGARCH_CSR_TLBIDX, &val);
+	val &= ~CSR_TLBIDX_SIZEM;
+	val |= PS_DEFAULT_SIZE << CSR_TLBIDX_SIZE;
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_TLBIDX, val);
+
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_STLBPGSIZE, PS_DEFAULT_SIZE);
+
+	/* LOONGARCH_CSR_KS1 is used for exception stack */
+	val = __vm_vaddr_alloc(vm, vm->page_size,
+			LOONGARCH_GUEST_STACK_VADDR_MIN, MEM_REGION_DATA);
+	TEST_ASSERT(val != 0,  "No memory for exception stack");
+	val = val + vm->page_size;
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_KS1, val);
+
+	loongarch_get_csr(vcpu, LOONGARCH_CSR_TLBREHI, &val);
+	val &= ~CSR_TLBREHI_PS;
+	val |= PS_DEFAULT_SIZE << CSR_TLBREHI_PS_SHIFT;
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_TLBREHI, val);
+
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_CPUID, vcpu->id);
+	loongarch_set_csr(vcpu, LOONGARCH_CSR_TMID,  vcpu->id);
+}
+
+struct kvm_vcpu *vm_arch_vcpu_add(struct kvm_vm *vm, uint32_t vcpu_id)
+{
+	size_t stack_size;
+	uint64_t stack_vaddr;
+	struct kvm_regs regs;
+	struct kvm_vcpu *vcpu;
+
+	vcpu = __vm_vcpu_add(vm, vcpu_id);
+	stack_size = vm->page_size;
+	stack_vaddr = __vm_vaddr_alloc(vm, stack_size,
+			LOONGARCH_GUEST_STACK_VADDR_MIN, MEM_REGION_DATA);
+	TEST_ASSERT(stack_vaddr != 0,  "No memory for vm stack");
+
+	loongarch_vcpu_setup(vcpu);
+	/* Setup guest general purpose registers */
+	vcpu_regs_get(vcpu, &regs);
+	regs.gpr[3] = stack_vaddr + stack_size;
+	vcpu_regs_set(vcpu, &regs);
+
+	return vcpu;
+}
+
+void vcpu_arch_set_entry_point(struct kvm_vcpu *vcpu, void *guest_code)
+{
+	struct kvm_regs regs;
+
+	/* Setup guest PC register */
+	vcpu_regs_get(vcpu, &regs);
+	regs.pc = (uint64_t)guest_code;
+	vcpu_regs_set(vcpu, &regs);
+}

tools/testing/selftests/kvm/lib/loongarch/ucall.c

@@ -0,0 +1,38 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * ucall support. A ucall is a "hypercall to userspace".
+ *
+ */
+#include "kvm_util.h"
+
+/*
+ * ucall_exit_mmio_addr holds per-VM values (global data is duplicated by each
+ * VM), it must not be accessed from host code.
+ */
+vm_vaddr_t *ucall_exit_mmio_addr;
+
+void ucall_arch_init(struct kvm_vm *vm, vm_paddr_t mmio_gpa)
+{
+	vm_vaddr_t mmio_gva = vm_vaddr_unused_gap(vm, vm->page_size, KVM_UTIL_MIN_VADDR);
+
+	virt_map(vm, mmio_gva, mmio_gpa, 1);
+
+	vm->ucall_mmio_addr = mmio_gpa;
+
+	write_guest_global(vm, ucall_exit_mmio_addr, (vm_vaddr_t *)mmio_gva);
+}
+
+void *ucall_arch_get_ucall(struct kvm_vcpu *vcpu)
+{
+	struct kvm_run *run = vcpu->run;
+
+	if (run->exit_reason == KVM_EXIT_MMIO &&
+	    run->mmio.phys_addr == vcpu->vm->ucall_mmio_addr) {
+		TEST_ASSERT(run->mmio.is_write && run->mmio.len == sizeof(uint64_t),
+			    "Unexpected ucall exit mmio address access");
+
+		return (void *)(*((uint64_t *)run->mmio.data));
+	}
+
+	return NULL;
+}

tools/testing/selftests/kvm/set_memory_region_test.c

@@ -350,7 +350,7 @@ static void test_invalid_memory_region_flags(void)
 	struct kvm_vm *vm;
 	int r, i;
 
-#if defined __aarch64__ || defined __riscv || defined __x86_64__
+#if defined __aarch64__ || defined __riscv || defined __x86_64__ || defined __loongarch__
 	supported_flags |= KVM_MEM_READONLY;
 #endif