Merge tag 'loongarch-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson

Pull LoongArch updates from Huacai Chen:

 - preliminary ClangBuiltLinux enablement

 - add support to clone a time namespace

 - add vector extensions support

 - add SMT (Simultaneous Multi-Threading) support

 - support dbar with different hints

 - introduce hardware page table walker

 - add jump-label implementation

 - add rethook and uprobes support

 - some bug fixes and other small changes

* tag 'loongarch-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson: (28 commits)
  LoongArch: Remove five DIE_* definitions in kdebug.h
  LoongArch: Add uprobes support
  LoongArch: Use larch_insn_gen_break() for kprobes
  LoongArch: Add larch_insn_gen_break() to generate break insns
  LoongArch: Check for AMO instructions in insns_not_supported()
  LoongArch: Move three functions from kprobes.c to inst.c
  LoongArch: Replace kretprobe with rethook
  LoongArch: Add jump-label implementation
  LoongArch: Select HAVE_DEBUG_KMEMLEAK to support kmemleak
  LoongArch: Export some arch-specific pm interfaces
  LoongArch: Introduce hardware page table walker
  LoongArch: Support dbar with different hints
  LoongArch: Add SMT (Simultaneous Multi-Threading) support
  LoongArch: Add vector extensions support
  LoongArch: Add support to clone a time namespace
  Makefile: Add loongarch target flag for Clang compilation
  LoongArch: Mark Clang LTO as working
  LoongArch: Include KBUILD_CPPFLAGS in CHECKFLAGS invocation
  LoongArch: vDSO: Use CLANG_FLAGS instead of filtering out '--target='
  LoongArch: Tweak CFLAGS for Clang compatibility
  ...

29 files changed, 972 insertions, 184 deletions

diff --git a/arch/loongarch/include/asm/Kbuild b/arch/loongarch/include/asm/Kbuild
index 77ad8e6f0906..6b222f227342 100644
--- a/arch/loongarch/include/asm/Kbuild
+++ b/arch/loongarch/include/asm/Kbuild
@@ -5,7 +5,6 @@ generic-y += mcs_spinlock.h
 generic-y += parport.h
 generic-y += early_ioremap.h
 generic-y += qrwlock.h
-generic-y += qspinlock.h
 generic-y += rwsem.h
 generic-y += segment.h
 generic-y += user.h
diff --git a/arch/loongarch/include/asm/acpi.h b/arch/loongarch/include/asm/acpi.h
index 976a810352c6..8de6c4b83a61 100644
--- a/arch/loongarch/include/asm/acpi.h
+++ b/arch/loongarch/include/asm/acpi.h
@@ -8,11 +8,14 @@
 #ifndef _ASM_LOONGARCH_ACPI_H
 #define _ASM_LOONGARCH_ACPI_H
 
+#include <asm/suspend.h>
+
 #ifdef CONFIG_ACPI
 extern int acpi_strict;
 extern int acpi_disabled;
 extern int acpi_pci_disabled;
 extern int acpi_noirq;
+extern int pptt_enabled;
 
 #define acpi_os_ioremap acpi_os_ioremap
 void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size);
@@ -30,6 +33,14 @@ static inline bool acpi_has_cpu_in_madt(void)
 }
 
 extern struct list_head acpi_wakeup_device_list;
+extern struct acpi_madt_core_pic acpi_core_pic[NR_CPUS];
+
+extern int __init parse_acpi_topology(void);
+
+static inline u32 get_acpi_id_for_cpu(unsigned int cpu)
+{
+	return acpi_core_pic[cpu_logical_map(cpu)].processor_id;
+}
 
 #endif /* !CONFIG_ACPI */
 
@@ -37,12 +48,10 @@ extern struct list_head acpi_wakeup_device_list;
 
 extern int loongarch_acpi_suspend(void);
 extern int (*acpi_suspend_lowlevel)(void);
-extern void loongarch_suspend_enter(void);
 
 static inline unsigned long acpi_get_wakeup_address(void)
 {
 #ifdef CONFIG_SUSPEND
-	extern void loongarch_wakeup_start(void);
 	return (unsigned long)loongarch_wakeup_start;
 #endif
 	return 0UL;
diff --git a/arch/loongarch/include/asm/asmmacro.h b/arch/loongarch/include/asm/asmmacro.h
index c51a1b43acb4..79e1d53fea89 100644
--- a/arch/loongarch/include/asm/asmmacro.h
+++ b/arch/loongarch/include/asm/asmmacro.h
@@ -270,6 +270,399 @@
 	fld.d	$f31, \tmp, THREAD_FPR31 - THREAD_FPR0
 	.endm
 
+	.macro	lsx_save_data thread tmp
+	li.w	\tmp, THREAD_FPR0
+	PTR_ADD \tmp, \thread, \tmp
+	vst	$vr0, \tmp, THREAD_FPR0  - THREAD_FPR0
+	vst	$vr1, \tmp, THREAD_FPR1  - THREAD_FPR0
+	vst	$vr2, \tmp, THREAD_FPR2  - THREAD_FPR0
+	vst	$vr3, \tmp, THREAD_FPR3  - THREAD_FPR0
+	vst	$vr4, \tmp, THREAD_FPR4  - THREAD_FPR0
+	vst	$vr5, \tmp, THREAD_FPR5  - THREAD_FPR0
+	vst	$vr6, \tmp, THREAD_FPR6  - THREAD_FPR0
+	vst	$vr7, \tmp, THREAD_FPR7  - THREAD_FPR0
+	vst	$vr8, \tmp, THREAD_FPR8  - THREAD_FPR0
+	vst	$vr9, \tmp, THREAD_FPR9  - THREAD_FPR0
+	vst	$vr10, \tmp, THREAD_FPR10 - THREAD_FPR0
+	vst	$vr11, \tmp, THREAD_FPR11 - THREAD_FPR0
+	vst	$vr12, \tmp, THREAD_FPR12 - THREAD_FPR0
+	vst	$vr13, \tmp, THREAD_FPR13 - THREAD_FPR0
+	vst	$vr14, \tmp, THREAD_FPR14 - THREAD_FPR0
+	vst	$vr15, \tmp, THREAD_FPR15 - THREAD_FPR0
+	vst	$vr16, \tmp, THREAD_FPR16 - THREAD_FPR0
+	vst	$vr17, \tmp, THREAD_FPR17 - THREAD_FPR0
+	vst	$vr18, \tmp, THREAD_FPR18 - THREAD_FPR0
+	vst	$vr19, \tmp, THREAD_FPR19 - THREAD_FPR0
+	vst	$vr20, \tmp, THREAD_FPR20 - THREAD_FPR0
+	vst	$vr21, \tmp, THREAD_FPR21 - THREAD_FPR0
+	vst	$vr22, \tmp, THREAD_FPR22 - THREAD_FPR0
+	vst	$vr23, \tmp, THREAD_FPR23 - THREAD_FPR0
+	vst	$vr24, \tmp, THREAD_FPR24 - THREAD_FPR0
+	vst	$vr25, \tmp, THREAD_FPR25 - THREAD_FPR0
+	vst	$vr26, \tmp, THREAD_FPR26 - THREAD_FPR0
+	vst	$vr27, \tmp, THREAD_FPR27 - THREAD_FPR0
+	vst	$vr28, \tmp, THREAD_FPR28 - THREAD_FPR0
+	vst	$vr29, \tmp, THREAD_FPR29 - THREAD_FPR0
+	vst	$vr30, \tmp, THREAD_FPR30 - THREAD_FPR0
+	vst	$vr31, \tmp, THREAD_FPR31 - THREAD_FPR0
+	.endm
+
+	.macro	lsx_restore_data thread tmp
+	li.w	\tmp, THREAD_FPR0
+	PTR_ADD	\tmp, \thread, \tmp
+	vld	$vr0, \tmp, THREAD_FPR0  - THREAD_FPR0
+	vld	$vr1, \tmp, THREAD_FPR1  - THREAD_FPR0
+	vld	$vr2, \tmp, THREAD_FPR2  - THREAD_FPR0
+	vld	$vr3, \tmp, THREAD_FPR3  - THREAD_FPR0
+	vld	$vr4, \tmp, THREAD_FPR4  - THREAD_FPR0
+	vld	$vr5, \tmp, THREAD_FPR5  - THREAD_FPR0
+	vld	$vr6, \tmp, THREAD_FPR6  - THREAD_FPR0
+	vld	$vr7, \tmp, THREAD_FPR7  - THREAD_FPR0
+	vld	$vr8, \tmp, THREAD_FPR8  - THREAD_FPR0
+	vld	$vr9, \tmp, THREAD_FPR9  - THREAD_FPR0
+	vld	$vr10, \tmp, THREAD_FPR10 - THREAD_FPR0
+	vld	$vr11, \tmp, THREAD_FPR11 - THREAD_FPR0
+	vld	$vr12, \tmp, THREAD_FPR12 - THREAD_FPR0
+	vld	$vr13, \tmp, THREAD_FPR13 - THREAD_FPR0
+	vld	$vr14, \tmp, THREAD_FPR14 - THREAD_FPR0
+	vld	$vr15, \tmp, THREAD_FPR15 - THREAD_FPR0
+	vld	$vr16, \tmp, THREAD_FPR16 - THREAD_FPR0
+	vld	$vr17, \tmp, THREAD_FPR17 - THREAD_FPR0
+	vld	$vr18, \tmp, THREAD_FPR18 - THREAD_FPR0
+	vld	$vr19, \tmp, THREAD_FPR19 - THREAD_FPR0
+	vld	$vr20, \tmp, THREAD_FPR20 - THREAD_FPR0
+	vld	$vr21, \tmp, THREAD_FPR21 - THREAD_FPR0
+	vld	$vr22, \tmp, THREAD_FPR22 - THREAD_FPR0
+	vld	$vr23, \tmp, THREAD_FPR23 - THREAD_FPR0
+	vld	$vr24, \tmp, THREAD_FPR24 - THREAD_FPR0
+	vld	$vr25, \tmp, THREAD_FPR25 - THREAD_FPR0
+	vld	$vr26, \tmp, THREAD_FPR26 - THREAD_FPR0
+	vld	$vr27, \tmp, THREAD_FPR27 - THREAD_FPR0
+	vld	$vr28, \tmp, THREAD_FPR28 - THREAD_FPR0
+	vld	$vr29, \tmp, THREAD_FPR29 - THREAD_FPR0
+	vld	$vr30, \tmp, THREAD_FPR30 - THREAD_FPR0
+	vld	$vr31, \tmp, THREAD_FPR31 - THREAD_FPR0
+	.endm
+
+	.macro	lsx_save_all	thread tmp0 tmp1
+	fpu_save_cc		\thread, \tmp0, \tmp1
+	fpu_save_csr		\thread, \tmp0
+	lsx_save_data		\thread, \tmp0
+	.endm
+
+	.macro	lsx_restore_all	thread tmp0 tmp1
+	lsx_restore_data	\thread, \tmp0
+	fpu_restore_cc		\thread, \tmp0, \tmp1
+	fpu_restore_csr		\thread, \tmp0
+	.endm
+
+	.macro	lsx_save_upper vd base tmp off
+	vpickve2gr.d	\tmp, \vd, 1
+	st.d		\tmp, \base, (\off+8)
+	.endm
+
+	.macro	lsx_save_all_upper thread base tmp
+	li.w		\tmp, THREAD_FPR0
+	PTR_ADD		\base, \thread, \tmp
+	lsx_save_upper	$vr0,  \base, \tmp, (THREAD_FPR0-THREAD_FPR0)
+	lsx_save_upper	$vr1,  \base, \tmp, (THREAD_FPR1-THREAD_FPR0)
+	lsx_save_upper	$vr2,  \base, \tmp, (THREAD_FPR2-THREAD_FPR0)
+	lsx_save_upper	$vr3,  \base, \tmp, (THREAD_FPR3-THREAD_FPR0)
+	lsx_save_upper	$vr4,  \base, \tmp, (THREAD_FPR4-THREAD_FPR0)
+	lsx_save_upper	$vr5,  \base, \tmp, (THREAD_FPR5-THREAD_FPR0)
+	lsx_save_upper	$vr6,  \base, \tmp, (THREAD_FPR6-THREAD_FPR0)
+	lsx_save_upper	$vr7,  \base, \tmp, (THREAD_FPR7-THREAD_FPR0)
+	lsx_save_upper	$vr8,  \base, \tmp, (THREAD_FPR8-THREAD_FPR0)
+	lsx_save_upper	$vr9,  \base, \tmp, (THREAD_FPR9-THREAD_FPR0)
+	lsx_save_upper	$vr10, \base, \tmp, (THREAD_FPR10-THREAD_FPR0)
+	lsx_save_upper	$vr11, \base, \tmp, (THREAD_FPR11-THREAD_FPR0)
+	lsx_save_upper	$vr12, \base, \tmp, (THREAD_FPR12-THREAD_FPR0)
+	lsx_save_upper	$vr13, \base, \tmp, (THREAD_FPR13-THREAD_FPR0)
+	lsx_save_upper	$vr14, \base, \tmp, (THREAD_FPR14-THREAD_FPR0)
+	lsx_save_upper	$vr15, \base, \tmp, (THREAD_FPR15-THREAD_FPR0)
+	lsx_save_upper	$vr16, \base, \tmp, (THREAD_FPR16-THREAD_FPR0)
+	lsx_save_upper	$vr17, \base, \tmp, (THREAD_FPR17-THREAD_FPR0)
+	lsx_save_upper	$vr18, \base, \tmp, (THREAD_FPR18-THREAD_FPR0)
+	lsx_save_upper	$vr19, \base, \tmp, (THREAD_FPR19-THREAD_FPR0)
+	lsx_save_upper	$vr20, \base, \tmp, (THREAD_FPR20-THREAD_FPR0)
+	lsx_save_upper	$vr21, \base, \tmp, (THREAD_FPR21-THREAD_FPR0)
+	lsx_save_upper	$vr22, \base, \tmp, (THREAD_FPR22-THREAD_FPR0)
+	lsx_save_upper	$vr23, \base, \tmp, (THREAD_FPR23-THREAD_FPR0)
+	lsx_save_upper	$vr24, \base, \tmp, (THREAD_FPR24-THREAD_FPR0)
+	lsx_save_upper	$vr25, \base, \tmp, (THREAD_FPR25-THREAD_FPR0)
+	lsx_save_upper	$vr26, \base, \tmp, (THREAD_FPR26-THREAD_FPR0)
+	lsx_save_upper	$vr27, \base, \tmp, (THREAD_FPR27-THREAD_FPR0)
+	lsx_save_upper	$vr28, \base, \tmp, (THREAD_FPR28-THREAD_FPR0)
+	lsx_save_upper	$vr29, \base, \tmp, (THREAD_FPR29-THREAD_FPR0)
+	lsx_save_upper	$vr30, \base, \tmp, (THREAD_FPR30-THREAD_FPR0)
+	lsx_save_upper	$vr31, \base, \tmp, (THREAD_FPR31-THREAD_FPR0)
+	.endm
+
+	.macro	lsx_restore_upper vd base tmp off
+	ld.d		\tmp, \base, (\off+8)
+	vinsgr2vr.d	\vd,  \tmp, 1
+	.endm
+
+	.macro	lsx_restore_all_upper thread base tmp
+	li.w		  \tmp, THREAD_FPR0
+	PTR_ADD		  \base, \thread, \tmp
+	lsx_restore_upper $vr0,  \base, \tmp, (THREAD_FPR0-THREAD_FPR0)
+	lsx_restore_upper $vr1,  \base, \tmp, (THREAD_FPR1-THREAD_FPR0)
+	lsx_restore_upper $vr2,  \base, \tmp, (THREAD_FPR2-THREAD_FPR0)
+	lsx_restore_upper $vr3,  \base, \tmp, (THREAD_FPR3-THREAD_FPR0)
+	lsx_restore_upper $vr4,  \base, \tmp, (THREAD_FPR4-THREAD_FPR0)
+	lsx_restore_upper $vr5,  \base, \tmp, (THREAD_FPR5-THREAD_FPR0)
+	lsx_restore_upper $vr6,  \base, \tmp, (THREAD_FPR6-THREAD_FPR0)
+	lsx_restore_upper $vr7,  \base, \tmp, (THREAD_FPR7-THREAD_FPR0)
+	lsx_restore_upper $vr8,  \base, \tmp, (THREAD_FPR8-THREAD_FPR0)
+	lsx_restore_upper $vr9,  \base, \tmp, (THREAD_FPR9-THREAD_FPR0)
+	lsx_restore_upper $vr10, \base, \tmp, (THREAD_FPR10-THREAD_FPR0)
+	lsx_restore_upper $vr11, \base, \tmp, (THREAD_FPR11-THREAD_FPR0)
+	lsx_restore_upper $vr12, \base, \tmp, (THREAD_FPR12-THREAD_FPR0)
+	lsx_restore_upper $vr13, \base, \tmp, (THREAD_FPR13-THREAD_FPR0)
+	lsx_restore_upper $vr14, \base, \tmp, (THREAD_FPR14-THREAD_FPR0)
+	lsx_restore_upper $vr15, \base, \tmp, (THREAD_FPR15-THREAD_FPR0)
+	lsx_restore_upper $vr16, \base, \tmp, (THREAD_FPR16-THREAD_FPR0)
+	lsx_restore_upper $vr17, \base, \tmp, (THREAD_FPR17-THREAD_FPR0)
+	lsx_restore_upper $vr18, \base, \tmp, (THREAD_FPR18-THREAD_FPR0)
+	lsx_restore_upper $vr19, \base, \tmp, (THREAD_FPR19-THREAD_FPR0)
+	lsx_restore_upper $vr20, \base, \tmp, (THREAD_FPR20-THREAD_FPR0)
+	lsx_restore_upper $vr21, \base, \tmp, (THREAD_FPR21-THREAD_FPR0)
+	lsx_restore_upper $vr22, \base, \tmp, (THREAD_FPR22-THREAD_FPR0)
+	lsx_restore_upper $vr23, \base, \tmp, (THREAD_FPR23-THREAD_FPR0)
+	lsx_restore_upper $vr24, \base, \tmp, (THREAD_FPR24-THREAD_FPR0)
+	lsx_restore_upper $vr25, \base, \tmp, (THREAD_FPR25-THREAD_FPR0)
+	lsx_restore_upper $vr26, \base, \tmp, (THREAD_FPR26-THREAD_FPR0)
+	lsx_restore_upper $vr27, \base, \tmp, (THREAD_FPR27-THREAD_FPR0)
+	lsx_restore_upper $vr28, \base, \tmp, (THREAD_FPR28-THREAD_FPR0)
+	lsx_restore_upper $vr29, \base, \tmp, (THREAD_FPR29-THREAD_FPR0)
+	lsx_restore_upper $vr30, \base, \tmp, (THREAD_FPR30-THREAD_FPR0)
+	lsx_restore_upper $vr31, \base, \tmp, (THREAD_FPR31-THREAD_FPR0)
+	.endm
+
+	.macro	lsx_init_upper vd tmp
+	vinsgr2vr.d	\vd, \tmp, 1
+	.endm
+
+	.macro	lsx_init_all_upper tmp
+	not		\tmp, zero
+	lsx_init_upper	$vr0 \tmp
+	lsx_init_upper	$vr1 \tmp
+	lsx_init_upper	$vr2 \tmp
+	lsx_init_upper	$vr3 \tmp
+	lsx_init_upper	$vr4 \tmp
+	lsx_init_upper	$vr5 \tmp
+	lsx_init_upper	$vr6 \tmp
+	lsx_init_upper	$vr7 \tmp
+	lsx_init_upper	$vr8 \tmp
+	lsx_init_upper	$vr9 \tmp
+	lsx_init_upper	$vr10 \tmp
+	lsx_init_upper	$vr11 \tmp
+	lsx_init_upper	$vr12 \tmp
+	lsx_init_upper	$vr13 \tmp
+	lsx_init_upper	$vr14 \tmp
+	lsx_init_upper	$vr15 \tmp
+	lsx_init_upper	$vr16 \tmp
+	lsx_init_upper	$vr17 \tmp
+	lsx_init_upper	$vr18 \tmp
+	lsx_init_upper	$vr19 \tmp
+	lsx_init_upper	$vr20 \tmp
+	lsx_init_upper	$vr21 \tmp
+	lsx_init_upper	$vr22 \tmp
+	lsx_init_upper	$vr23 \tmp
+	lsx_init_upper	$vr24 \tmp
+	lsx_init_upper	$vr25 \tmp
+	lsx_init_upper	$vr26 \tmp
+	lsx_init_upper	$vr27 \tmp
+	lsx_init_upper	$vr28 \tmp
+	lsx_init_upper	$vr29 \tmp
+	lsx_init_upper	$vr30 \tmp
+	lsx_init_upper	$vr31 \tmp
+	.endm
+
+	.macro	lasx_save_data thread tmp
+	li.w	\tmp, THREAD_FPR0
+	PTR_ADD	\tmp, \thread, \tmp
+	xvst	$xr0, \tmp, THREAD_FPR0  - THREAD_FPR0
+	xvst	$xr1, \tmp, THREAD_FPR1  - THREAD_FPR0
+	xvst	$xr2, \tmp, THREAD_FPR2  - THREAD_FPR0
+	xvst	$xr3, \tmp, THREAD_FPR3  - THREAD_FPR0
+	xvst	$xr4, \tmp, THREAD_FPR4  - THREAD_FPR0
+	xvst	$xr5, \tmp, THREAD_FPR5  - THREAD_FPR0
+	xvst	$xr6, \tmp, THREAD_FPR6  - THREAD_FPR0
+	xvst	$xr7, \tmp, THREAD_FPR7  - THREAD_FPR0
+	xvst	$xr8, \tmp, THREAD_FPR8  - THREAD_FPR0
+	xvst	$xr9, \tmp, THREAD_FPR9  - THREAD_FPR0
+	xvst	$xr10, \tmp, THREAD_FPR10 - THREAD_FPR0
+	xvst	$xr11, \tmp, THREAD_FPR11 - THREAD_FPR0
+	xvst	$xr12, \tmp, THREAD_FPR12 - THREAD_FPR0
+	xvst	$xr13, \tmp, THREAD_FPR13 - THREAD_FPR0
+	xvst	$xr14, \tmp, THREAD_FPR14 - THREAD_FPR0
+	xvst	$xr15, \tmp, THREAD_FPR15 - THREAD_FPR0
+	xvst	$xr16, \tmp, THREAD_FPR16 - THREAD_FPR0
+	xvst	$xr17, \tmp, THREAD_FPR17 - THREAD_FPR0
+	xvst	$xr18, \tmp, THREAD_FPR18 - THREAD_FPR0
+	xvst	$xr19, \tmp, THREAD_FPR19 - THREAD_FPR0
+	xvst	$xr20, \tmp, THREAD_FPR20 - THREAD_FPR0
+	xvst	$xr21, \tmp, THREAD_FPR21 - THREAD_FPR0
+	xvst	$xr22, \tmp, THREAD_FPR22 - THREAD_FPR0
+	xvst	$xr23, \tmp, THREAD_FPR23 - THREAD_FPR0
+	xvst	$xr24, \tmp, THREAD_FPR24 - THREAD_FPR0
+	xvst	$xr25, \tmp, THREAD_FPR25 - THREAD_FPR0
+	xvst	$xr26, \tmp, THREAD_FPR26 - THREAD_FPR0
+	xvst	$xr27, \tmp, THREAD_FPR27 - THREAD_FPR0
+	xvst	$xr28, \tmp, THREAD_FPR28 - THREAD_FPR0
+	xvst	$xr29, \tmp, THREAD_FPR29 - THREAD_FPR0
+	xvst	$xr30, \tmp, THREAD_FPR30 - THREAD_FPR0
+	xvst	$xr31, \tmp, THREAD_FPR31 - THREAD_FPR0
+	.endm
+
+	.macro	lasx_restore_data thread tmp
+	li.w	\tmp, THREAD_FPR0
+	PTR_ADD	\tmp, \thread, \tmp
+	xvld	$xr0, \tmp, THREAD_FPR0  - THREAD_FPR0
+	xvld	$xr1, \tmp, THREAD_FPR1  - THREAD_FPR0
+	xvld	$xr2, \tmp, THREAD_FPR2  - THREAD_FPR0
+	xvld	$xr3, \tmp, THREAD_FPR3  - THREAD_FPR0
+	xvld	$xr4, \tmp, THREAD_FPR4  - THREAD_FPR0
+	xvld	$xr5, \tmp, THREAD_FPR5  - THREAD_FPR0
+	xvld	$xr6, \tmp, THREAD_FPR6  - THREAD_FPR0
+	xvld	$xr7, \tmp, THREAD_FPR7  - THREAD_FPR0
+	xvld	$xr8, \tmp, THREAD_FPR8  - THREAD_FPR0
+	xvld	$xr9, \tmp, THREAD_FPR9  - THREAD_FPR0
+	xvld	$xr10, \tmp, THREAD_FPR10 - THREAD_FPR0
+	xvld	$xr11, \tmp, THREAD_FPR11 - THREAD_FPR0
+	xvld	$xr12, \tmp, THREAD_FPR12 - THREAD_FPR0
+	xvld	$xr13, \tmp, THREAD_FPR13 - THREAD_FPR0
+	xvld	$xr14, \tmp, THREAD_FPR14 - THREAD_FPR0
+	xvld	$xr15, \tmp, THREAD_FPR15 - THREAD_FPR0
+	xvld	$xr16, \tmp, THREAD_FPR16 - THREAD_FPR0
+	xvld	$xr17, \tmp, THREAD_FPR17 - THREAD_FPR0
+	xvld	$xr18, \tmp, THREAD_FPR18 - THREAD_FPR0
+	xvld	$xr19, \tmp, THREAD_FPR19 - THREAD_FPR0
+	xvld	$xr20, \tmp, THREAD_FPR20 - THREAD_FPR0
+	xvld	$xr21, \tmp, THREAD_FPR21 - THREAD_FPR0
+	xvld	$xr22, \tmp, THREAD_FPR22 - THREAD_FPR0
+	xvld	$xr23, \tmp, THREAD_FPR23 - THREAD_FPR0
+	xvld	$xr24, \tmp, THREAD_FPR24 - THREAD_FPR0
+	xvld	$xr25, \tmp, THREAD_FPR25 - THREAD_FPR0
+	xvld	$xr26, \tmp, THREAD_FPR26 - THREAD_FPR0
+	xvld	$xr27, \tmp, THREAD_FPR27 - THREAD_FPR0
+	xvld	$xr28, \tmp, THREAD_FPR28 - THREAD_FPR0
+	xvld	$xr29, \tmp, THREAD_FPR29 - THREAD_FPR0
+	xvld	$xr30, \tmp, THREAD_FPR30 - THREAD_FPR0
+	xvld	$xr31, \tmp, THREAD_FPR31 - THREAD_FPR0
+	.endm
+
+	.macro	lasx_save_all	thread tmp0 tmp1
+	fpu_save_cc		\thread, \tmp0, \tmp1
+	fpu_save_csr		\thread, \tmp0
+	lasx_save_data		\thread, \tmp0
+	.endm
+
+	.macro	lasx_restore_all thread tmp0 tmp1
+	lasx_restore_data	\thread, \tmp0
+	fpu_restore_cc		\thread, \tmp0, \tmp1
+	fpu_restore_csr		\thread, \tmp0
+	.endm
+
+	.macro	lasx_save_upper xd base tmp off
+	/* Nothing */
+	.endm
+
+	.macro	lasx_save_all_upper thread base tmp
+	/* Nothing */
+	.endm
+
+	.macro	lasx_restore_upper xd base tmp0 tmp1 off
+	vld		\tmp0, \base, (\off+16)
+	xvpermi.q 	\xd,   \tmp1, 0x2
+	.endm
+
+	.macro	lasx_restore_all_upper thread base tmp
+	li.w		\tmp, THREAD_FPR0
+	PTR_ADD		\base, \thread, \tmp
+	/* Save $vr31 ($xr31 lower bits) with xvpickve2gr */
+	xvpickve2gr.d	$r17, $xr31, 0
+	xvpickve2gr.d	$r18, $xr31, 1
+	lasx_restore_upper $xr0, \base, $vr31, $xr31, (THREAD_FPR0-THREAD_FPR0)
+	lasx_restore_upper $xr1, \base, $vr31, $xr31, (THREAD_FPR1-THREAD_FPR0)
+	lasx_restore_upper $xr2, \base, $vr31, $xr31, (THREAD_FPR2-THREAD_FPR0)
+	lasx_restore_upper $xr3, \base, $vr31, $xr31, (THREAD_FPR3-THREAD_FPR0)
+	lasx_restore_upper $xr4, \base, $vr31, $xr31, (THREAD_FPR4-THREAD_FPR0)
+	lasx_restore_upper $xr5, \base, $vr31, $xr31, (THREAD_FPR5-THREAD_FPR0)
+	lasx_restore_upper $xr6, \base, $vr31, $xr31, (THREAD_FPR6-THREAD_FPR0)
+	lasx_restore_upper $xr7, \base, $vr31, $xr31, (THREAD_FPR7-THREAD_FPR0)
+	lasx_restore_upper $xr8, \base, $vr31, $xr31, (THREAD_FPR8-THREAD_FPR0)
+	lasx_restore_upper $xr9, \base, $vr31, $xr31, (THREAD_FPR9-THREAD_FPR0)
+	lasx_restore_upper $xr10, \base, $vr31, $xr31, (THREAD_FPR10-THREAD_FPR0)
+	lasx_restore_upper $xr11, \base, $vr31, $xr31, (THREAD_FPR11-THREAD_FPR0)
+	lasx_restore_upper $xr12, \base, $vr31, $xr31, (THREAD_FPR12-THREAD_FPR0)
+	lasx_restore_upper $xr13, \base, $vr31, $xr31, (THREAD_FPR13-THREAD_FPR0)
+	lasx_restore_upper $xr14, \base, $vr31, $xr31, (THREAD_FPR14-THREAD_FPR0)
+	lasx_restore_upper $xr15, \base, $vr31, $xr31, (THREAD_FPR15-THREAD_FPR0)
+	lasx_restore_upper $xr16, \base, $vr31, $xr31, (THREAD_FPR16-THREAD_FPR0)
+	lasx_restore_upper $xr17, \base, $vr31, $xr31, (THREAD_FPR17-THREAD_FPR0)
+	lasx_restore_upper $xr18, \base, $vr31, $xr31, (THREAD_FPR18-THREAD_FPR0)
+	lasx_restore_upper $xr19, \base, $vr31, $xr31, (THREAD_FPR19-THREAD_FPR0)
+	lasx_restore_upper $xr20, \base, $vr31, $xr31, (THREAD_FPR20-THREAD_FPR0)
+	lasx_restore_upper $xr21, \base, $vr31, $xr31, (THREAD_FPR21-THREAD_FPR0)
+	lasx_restore_upper $xr22, \base, $vr31, $xr31, (THREAD_FPR22-THREAD_FPR0)
+	lasx_restore_upper $xr23, \base, $vr31, $xr31, (THREAD_FPR23-THREAD_FPR0)
+	lasx_restore_upper $xr24, \base, $vr31, $xr31, (THREAD_FPR24-THREAD_FPR0)
+	lasx_restore_upper $xr25, \base, $vr31, $xr31, (THREAD_FPR25-THREAD_FPR0)
+	lasx_restore_upper $xr26, \base, $vr31, $xr31, (THREAD_FPR26-THREAD_FPR0)
+	lasx_restore_upper $xr27, \base, $vr31, $xr31, (THREAD_FPR27-THREAD_FPR0)
+	lasx_restore_upper $xr28, \base, $vr31, $xr31, (THREAD_FPR28-THREAD_FPR0)
+	lasx_restore_upper $xr29, \base, $vr31, $xr31, (THREAD_FPR29-THREAD_FPR0)
+	lasx_restore_upper $xr30, \base, $vr31, $xr31, (THREAD_FPR30-THREAD_FPR0)
+	lasx_restore_upper $xr31, \base, $vr31, $xr31, (THREAD_FPR31-THREAD_FPR0)
+	/* Restore $vr31 ($xr31 lower bits) with xvinsgr2vr */
+	xvinsgr2vr.d	$xr31, $r17, 0
+	xvinsgr2vr.d	$xr31, $r18, 1
+	.endm
+
+	.macro	lasx_init_upper xd tmp
+	xvinsgr2vr.d	\xd, \tmp, 2
+	xvinsgr2vr.d	\xd, \tmp, 3
+	.endm
+
+	.macro	lasx_init_all_upper tmp
+	not		\tmp, zero
+	lasx_init_upper	$xr0 \tmp
+	lasx_init_upper	$xr1 \tmp
+	lasx_init_upper	$xr2 \tmp
+	lasx_init_upper	$xr3 \tmp
+	lasx_init_upper	$xr4 \tmp
+	lasx_init_upper	$xr5 \tmp
+	lasx_init_upper	$xr6 \tmp
+	lasx_init_upper	$xr7 \tmp
+	lasx_init_upper	$xr8 \tmp
+	lasx_init_upper	$xr9 \tmp
+	lasx_init_upper	$xr10 \tmp
+	lasx_init_upper	$xr11 \tmp
+	lasx_init_upper	$xr12 \tmp
+	lasx_init_upper	$xr13 \tmp
+	lasx_init_upper	$xr14 \tmp
+	lasx_init_upper	$xr15 \tmp
+	lasx_init_upper	$xr16 \tmp
+	lasx_init_upper	$xr17 \tmp
+	lasx_init_upper	$xr18 \tmp
+	lasx_init_upper	$xr19 \tmp
+	lasx_init_upper	$xr20 \tmp
+	lasx_init_upper	$xr21 \tmp
+	lasx_init_upper	$xr22 \tmp
+	lasx_init_upper	$xr23 \tmp
+	lasx_init_upper	$xr24 \tmp
+	lasx_init_upper	$xr25 \tmp
+	lasx_init_upper	$xr26 \tmp
+	lasx_init_upper	$xr27 \tmp
+	lasx_init_upper	$xr28 \tmp
+	lasx_init_upper	$xr29 \tmp
+	lasx_init_upper	$xr30 \tmp
+	lasx_init_upper	$xr31 \tmp
+	.endm
+
 .macro not dst src
 	nor	\dst, \src, zero
 .endm
diff --git a/arch/loongarch/include/asm/barrier.h b/arch/loongarch/include/asm/barrier.h
index cda977675854..4b663f197706 100644
--- a/arch/loongarch/include/asm/barrier.h
+++ b/arch/loongarch/include/asm/barrier.h
@@ -5,27 +5,56 @@
 #ifndef __ASM_BARRIER_H
 #define __ASM_BARRIER_H
 
-#define __sync()	__asm__ __volatile__("dbar 0" : : : "memory")
+/*
+ * Hint encoding:
+ *
+ * Bit4: ordering or completion (0: completion, 1: ordering)
+ * Bit3: barrier for previous read (0: true, 1: false)
+ * Bit2: barrier for previous write (0: true, 1: false)
+ * Bit1: barrier for succeeding read (0: true, 1: false)
+ * Bit0: barrier for succeeding write (0: true, 1: false)
+ *
+ * Hint 0x700: barrier for "read after read" from the same address
+ */
+
+#define DBAR(hint) __asm__ __volatile__("dbar %0 " : : "I"(hint) : "memory")
+
+#define crwrw		0b00000
+#define cr_r_		0b00101
+#define c_w_w		0b01010
 
-#define fast_wmb()	__sync()
-#define fast_rmb()	__sync()
-#define fast_mb()	__sync()
-#define fast_iob()	__sync()
-#define wbflush()	__sync()
+#define orwrw		0b10000
+#define or_r_		0b10101
+#define o_w_w		0b11010
 
-#define wmb()		fast_wmb()
-#define rmb()		fast_rmb()
-#define mb()		fast_mb()
-#define iob()		fast_iob()
+#define orw_w		0b10010
+#define or_rw		0b10100
 
-#define __smp_mb()	__asm__ __volatile__("dbar 0" : : : "memory")
-#define __smp_rmb()	__asm__ __volatile__("dbar 0" : : : "memory")
-#define __smp_wmb()	__asm__ __volatile__("dbar 0" : : : "memory")
+#define c_sync()	DBAR(crwrw)
+#define c_rsync()	DBAR(cr_r_)
+#define c_wsync()	DBAR(c_w_w)
+
+#define o_sync()	DBAR(orwrw)
+#define o_rsync()	DBAR(or_r_)
+#define o_wsync()	DBAR(o_w_w)
+
+#define ldacq_mb()	DBAR(or_rw)
+#define strel_mb()	DBAR(orw_w)
+
+#define mb()		c_sync()
+#define rmb()		c_rsync()
+#define wmb()		c_wsync()
+#define iob()		c_sync()
+#define wbflush()	c_sync()
+
+#define __smp_mb()	o_sync()
+#define __smp_rmb()	o_rsync()
+#define __smp_wmb()	o_wsync()
 
 #ifdef CONFIG_SMP
-#define __WEAK_LLSC_MB		"	dbar 0  \n"
+#define __WEAK_LLSC_MB		"	dbar 0x700	\n"
 #else
-#define __WEAK_LLSC_MB		"		\n"
+#define __WEAK_LLSC_MB		"			\n"
 #endif
 
 #define __smp_mb__before_atomic()	barrier()
@@ -59,68 +88,19 @@ static inline unsigned long array_index_mask_nospec(unsigned long index,
 	return mask;
 }
 
-#define __smp_load_acquire(p)							\
-({										\
-	union { typeof(*p) __val; char __c[1]; } __u;				\
-	unsigned long __tmp = 0;							\
-	compiletime_assert_atomic_type(*p);					\
-	switch (sizeof(*p)) {							\
-	case 1:									\
-		*(__u8 *)__u.__c = *(volatile __u8 *)p;				\
-		__smp_mb();							\
-		break;								\
-	case 2:									\
-		*(__u16 *)__u.__c = *(volatile __u16 *)p;			\
-		__smp_mb();							\
-		break;								\
-	case 4:									\
-		__asm__ __volatile__(						\
-		"amor_db.w %[val], %[tmp], %[mem]	\n"				\
-		: [val] "=&r" (*(__u32 *)__u.__c)				\
-		: [mem] "ZB" (*(u32 *) p), [tmp] "r" (__tmp)			\
-		: "memory");							\
-		break;								\
-	case 8:									\
-		__asm__ __volatile__(						\
-		"amor_db.d %[val], %[tmp], %[mem]	\n"				\
-		: [val] "=&r" (*(__u64 *)__u.__c)				\
-		: [mem] "ZB" (*(u64 *) p), [tmp] "r" (__tmp)			\
-		: "memory");							\
-		break;								\
-	}									\
-	(typeof(*p))__u.__val;								\
+#define __smp_load_acquire(p)				\
+({							\
+	typeof(*p) ___p1 = READ_ONCE(*p);		\
+	compiletime_assert_atomic_type(*p);		\
+	ldacq_mb();					\
+	___p1;						\
 })
 
-#define __smp_store_release(p, v)						\
-do {										\
-	union { typeof(*p) __val; char __c[1]; } __u =				\
-		{ .__val = (__force typeof(*p)) (v) };				\
-	unsigned long __tmp;							\
-	compiletime_assert_atomic_type(*p);					\
-	switch (sizeof(*p)) {							\
-	case 1:									\
-		__smp_mb();							\
-		*(volatile __u8 *)p = *(__u8 *)__u.__c;				\
-		break;								\
-	case 2:									\
-		__smp_mb();							\
-		*(volatile __u16 *)p = *(__u16 *)__u.__c;			\
-		break;								\
-	case 4:									\
-		__asm__ __volatile__(						\
-		"amswap_db.w %[tmp], %[val], %[mem]	\n"			\
-		: [mem] "+ZB" (*(u32 *)p), [tmp] "=&r" (__tmp)			\
-		: [val] "r" (*(__u32 *)__u.__c)					\
-		: );								\
-		break;								\
-	case 8:									\
-		__asm__ __volatile__(						\
-		"amswap_db.d %[tmp], %[val], %[mem]	\n"			\
-		: [mem] "+ZB" (*(u64 *)p), [tmp] "=&r" (__tmp)			\
-		: [val] "r" (*(__u64 *)__u.__c)					\
-		: );								\
-		break;								\
-	}									\
+#define __smp_store_release(p, v)			\
+do {							\
+	compiletime_assert_atomic_type(*p);		\
+	strel_mb();					\
+	WRITE_ONCE(*p, v);				\
 } while (0)
 
 #define __smp_store_mb(p, v)							\
diff --git a/arch/loongarch/include/asm/cpu-features.h b/arch/loongarch/include/asm/cpu-features.h
index f6177f133477..2eafe6a6aca8 100644
--- a/arch/loongarch/include/asm/cpu-features.h
+++ b/arch/loongarch/include/asm/cpu-features.h
@@ -64,6 +64,6 @@
 #define cpu_has_eiodecode	cpu_opt(LOONGARCH_CPU_EIODECODE)
 #define cpu_has_guestid		cpu_opt(LOONGARCH_CPU_GUESTID)
 #define cpu_has_hypervisor	cpu_opt(LOONGARCH_CPU_HYPERVISOR)
-
+#define cpu_has_ptw		cpu_opt(LOONGARCH_CPU_PTW)
 
 #endif /* __ASM_CPU_FEATURES_H */
diff --git a/arch/loongarch/include/asm/cpu-info.h b/arch/loongarch/include/asm/cpu-info.h
index cd73a6f57fe3..900589cb159d 100644
--- a/arch/loongarch/include/asm/cpu-info.h
+++ b/arch/loongarch/include/asm/cpu-info.h
@@ -54,6 +54,7 @@ struct cpuinfo_loongarch {
 	struct cache_desc	cache_leaves[CACHE_LEAVES_MAX];
 	int			core;   /* physical core number in package */
 	int			package;/* physical package number */
+	int			global_id; /* physical global thread number */
 	int			vabits; /* Virtual Address size in bits */
 	int			pabits; /* Physical Address size in bits */
 	unsigned int		ksave_mask; /* Usable KSave mask. */
diff --git a/arch/loongarch/include/asm/cpu.h b/arch/loongarch/include/asm/cpu.h
index 88773d849e33..48b9f7168bcc 100644
--- a/arch/loongarch/include/asm/cpu.h
+++ b/arch/loongarch/include/asm/cpu.h
@@ -98,6 +98,7 @@ enum cpu_type_enum {
 #define CPU_FEATURE_EIODECODE		23	/* CPU has EXTIOI interrupt pin decode mode */
 #define CPU_FEATURE_GUESTID		24	/* CPU has GuestID feature */
 #define CPU_FEATURE_HYPERVISOR		25	/* CPU has hypervisor (running in VM) */
+#define CPU_FEATURE_PTW			26	/* CPU has hardware page table walker */
 
 #define LOONGARCH_CPU_CPUCFG		BIT_ULL(CPU_FEATURE_CPUCFG)
 #define LOONGARCH_CPU_LAM		BIT_ULL(CPU_FEATURE_LAM)
@@ -125,5 +126,6 @@ enum cpu_type_enum {
 #define LOONGARCH_CPU_EIODECODE		BIT_ULL(CPU_FEATURE_EIODECODE)
 #define LOONGARCH_CPU_GUESTID		BIT_ULL(CPU_FEATURE_GUESTID)
 #define LOONGARCH_CPU_HYPERVISOR	BIT_ULL(CPU_FEATURE_HYPERVISOR)
+#define LOONGARCH_CPU_PTW		BIT_ULL(CPU_FEATURE_PTW)
 
 #endif /* _ASM_CPU_H */
diff --git a/arch/loongarch/include/asm/fpregdef.h b/arch/loongarch/include/asm/fpregdef.h
index b6be527831dd..e56610ae8592 100644
--- a/arch/loongarch/include/asm/fpregdef.h
+++ b/arch/loongarch/include/asm/fpregdef.h
@@ -40,6 +40,7 @@
 #define fs6	$f30
 #define fs7	$f31
 
+#ifndef CONFIG_AS_HAS_FCSR_CLASS
 /*
  * Current binutils expects *GPRs* at FCSR position for the FCSR
  * operation instructions, so define aliases for those used.
@@ -48,5 +49,11 @@
 #define fcsr1	$r1
 #define fcsr2	$r2
 #define fcsr3	$r3
+#else
+#define fcsr0	$fcsr0
+#define fcsr1	$fcsr1
+#define fcsr2	$fcsr2
+#define fcsr3	$fcsr3
+#endif
 
 #endif /* _ASM_FPREGDEF_H */
diff --git a/arch/loongarch/include/asm/fpu.h b/arch/loongarch/include/asm/fpu.h
index 192f8e35d912..e4193d637f66 100644
--- a/arch/loongarch/include/asm/fpu.h
+++ b/arch/loongarch/inclu