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
  ...

Author: torvalds

Documentation/features/core/jump-labels/arch-support.txt

@@ -13,7 +13,7 @@
     |        csky: |  ok  |
     |     hexagon: | TODO |
     |        ia64: | TODO |
-    |   loongarch: | TODO |
+    |   loongarch: |  ok  |
     |        m68k: | TODO |
     |  microblaze: | TODO |
     |        mips: |  ok  |

Documentation/features/debug/kmemleak/arch-support.txt

@@ -13,7 +13,7 @@
     |        csky: |  ok  |
     |     hexagon: | TODO |
     |        ia64: | TODO |
-    |   loongarch: | TODO |
+    |   loongarch: |  ok  |
     |        m68k: | TODO |
     |  microblaze: |  ok  |
     |        mips: |  ok  |

arch/loongarch/Kconfig

@@ -5,6 +5,7 @@ config LOONGARCH
 	select ACPI
 	select ACPI_GENERIC_GSI if ACPI
 	select ACPI_MCFG if ACPI
+	select ACPI_PPTT if ACPI
 	select ACPI_SYSTEM_POWER_STATES_SUPPORT	if ACPI
 	select ARCH_BINFMT_ELF_STATE
 	select ARCH_ENABLE_MEMORY_HOTPLUG
@@ -49,6 +50,8 @@ config LOONGARCH
 	select ARCH_SUPPORTS_ACPI
 	select ARCH_SUPPORTS_ATOMIC_RMW
 	select ARCH_SUPPORTS_HUGETLBFS
+	select ARCH_SUPPORTS_LTO_CLANG
+	select ARCH_SUPPORTS_LTO_CLANG_THIN
 	select ARCH_SUPPORTS_NUMA_BALANCING
 	select ARCH_USE_BUILTIN_BSWAP
 	select ARCH_USE_CMPXCHG_LOCKREF
@@ -81,16 +84,20 @@ config LOONGARCH
 	select GENERIC_SCHED_CLOCK
 	select GENERIC_SMP_IDLE_THREAD
 	select GENERIC_TIME_VSYSCALL
+	select GENERIC_VDSO_TIME_NS
 	select GPIOLIB
 	select HAS_IOPORT
 	select HAVE_ARCH_AUDITSYSCALL
+	select HAVE_ARCH_JUMP_LABEL
+	select HAVE_ARCH_JUMP_LABEL_RELATIVE
 	select HAVE_ARCH_MMAP_RND_BITS if MMU
 	select HAVE_ARCH_SECCOMP_FILTER
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select HAVE_ASM_MODVERSIONS
 	select HAVE_CONTEXT_TRACKING_USER
 	select HAVE_C_RECORDMCOUNT
+	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_DEBUG_STACKOVERFLOW
 	select HAVE_DMA_CONTIGUOUS
 	select HAVE_DYNAMIC_FTRACE
@@ -121,6 +128,7 @@ config LOONGARCH
 	select HAVE_PERF_REGS
 	select HAVE_PERF_USER_STACK_DUMP
 	select HAVE_REGS_AND_STACK_ACCESS_API
+	select HAVE_RETHOOK
 	select HAVE_RSEQ
 	select HAVE_SAMPLE_FTRACE_DIRECT
 	select HAVE_SAMPLE_FTRACE_DIRECT_MULTI
@@ -163,14 +171,6 @@ config 32BIT
 config 64BIT
 	def_bool y
 
-config CPU_HAS_FPU
-	bool
-	default y
-
-config CPU_HAS_PREFETCH
-	bool
-	default y
-
 config GENERIC_BUG
 	def_bool y
 	depends on BUG
@@ -243,6 +243,15 @@ config SCHED_OMIT_FRAME_POINTER
 config AS_HAS_EXPLICIT_RELOCS
 	def_bool $(as-instr,x:pcalau12i \$t0$(comma)%pc_hi20(x))
 
+config AS_HAS_FCSR_CLASS
+	def_bool $(as-instr,movfcsr2gr \$t0$(comma)\$fcsr0)
+
+config AS_HAS_LSX_EXTENSION
+	def_bool $(as-instr,vld \$vr0$(comma)\$a0$(comma)0)
+
+config AS_HAS_LASX_EXTENSION
+	def_bool $(as-instr,xvld \$xr0$(comma)\$a0$(comma)0)
+
 menu "Kernel type and options"
 
 source "kernel/Kconfig.hz"
@@ -374,6 +383,13 @@ config EFI_STUB
 	  This kernel feature allows the kernel to be loaded directly by
 	  EFI firmware without the use of a bootloader.
 
+config SCHED_SMT
+	bool "SMT scheduler support"
+	default y
+	help
+	  Improves scheduler's performance when there are multiple
+	  threads in one physical core.
+
 config SMP
 	bool "Multi-Processing support"
 	help
@@ -483,6 +499,43 @@ config ARCH_STRICT_ALIGN
 	  to run kernel only on systems with h/w unaligned access support in
 	  order to optimise for performance.
 
+config CPU_HAS_FPU
+	bool
+	default y
+
+config CPU_HAS_LSX
+	bool "Support for the Loongson SIMD Extension"
+	depends on AS_HAS_LSX_EXTENSION
+	help
+	  Loongson SIMD Extension (LSX) introduces 128 bit wide vector registers
+	  and a set of SIMD instructions to operate on them. When this option
+	  is enabled the kernel will support allocating & switching LSX
+	  vector register contexts. If you know that your kernel will only be
+	  running on CPUs which do not support LSX or that your userland will
+	  not be making use of it then you may wish to say N here to reduce
+	  the size & complexity of your kernel.
+
+	  If unsure, say Y.
+
+config CPU_HAS_LASX
+	bool "Support for the Loongson Advanced SIMD Extension"
+	depends on CPU_HAS_LSX
+	depends on AS_HAS_LASX_EXTENSION
+	help
+	  Loongson Advanced SIMD Extension (LASX) introduces 256 bit wide vector
+	  registers and a set of SIMD instructions to operate on them. When this
+	  option is enabled the kernel will support allocating & switching LASX
+	  vector register contexts. If you know that your kernel will only be
+	  running on CPUs which do not support LASX or that your userland will
+	  not be making use of it then you may wish to say N here to reduce
+	  the size & complexity of your kernel.
+
+	  If unsure, say Y.
+
+config CPU_HAS_PREFETCH
+	bool
+	default y
+
 config KEXEC
 	bool "Kexec system call"
 	select KEXEC_CORE
@@ -592,6 +645,9 @@ config ARCH_MMAP_RND_BITS_MIN
 config ARCH_MMAP_RND_BITS_MAX
 	default 18
 
+config ARCH_SUPPORTS_UPROBES
+	def_bool y
+
 menu "Power management options"
 
 config ARCH_SUSPEND_POSSIBLE

arch/loongarch/Makefile

@@ -46,8 +46,8 @@ ld-emul			= $(64bit-emul)
 cflags-y		+= -mabi=lp64s
 endif
 
-cflags-y			+= -G0 -pipe -msoft-float
-LDFLAGS_vmlinux			+= -G0 -static -n -nostdlib
+cflags-y			+= -pipe -msoft-float
+LDFLAGS_vmlinux			+= -static -n -nostdlib
 
 # When the assembler supports explicit relocation hint, we must use it.
 # GCC may have -mexplicit-relocs off by default if it was built with an old
@@ -56,13 +56,18 @@ LDFLAGS_vmlinux			+= -G0 -static -n -nostdlib
 # When the assembler does not supports explicit relocation hint, we can't use
 # it.  Disable it if the compiler supports it.
 #
-# If you've seen "unknown reloc hint" message building the kernel and you are
-# now wondering why "-mexplicit-relocs" is not wrapped with cc-option: the
-# combination of a "new" assembler and "old" compiler is not supported.  Either
-# upgrade the compiler or downgrade the assembler.
+# The combination of a "new" assembler and "old" GCC is not supported, given
+# the rarity of this combo and the extra complexity needed to make it work.
+# Either upgrade the compiler or downgrade the assembler; the build will error
+# out if it is the case (by probing for the model attribute; all supported
+# compilers in this case would have support).
+#
+# Also, -mdirect-extern-access is useful in case of building with explicit
+# relocs, for avoiding unnecessary GOT accesses. It is harmless to not have
+# support though.
 ifdef CONFIG_AS_HAS_EXPLICIT_RELOCS
-cflags-y			+= -mexplicit-relocs
-KBUILD_CFLAGS_KERNEL		+= -mdirect-extern-access
+cflags-y			+= $(call cc-option,-mexplicit-relocs)
+KBUILD_CFLAGS_KERNEL		+= $(call cc-option,-mdirect-extern-access)
 else
 cflags-y			+= $(call cc-option,-mno-explicit-relocs)
 KBUILD_AFLAGS_KERNEL		+= -Wa,-mla-global-with-pcrel
@@ -107,7 +112,7 @@ KBUILD_CFLAGS += -isystem $(shell $(CC) -print-file-name=include)
 KBUILD_LDFLAGS	+= -m $(ld-emul)
 
 ifdef CONFIG_LOONGARCH
-CHECKFLAGS += $(shell $(CC) $(KBUILD_CFLAGS) -dM -E -x c /dev/null | \
+CHECKFLAGS += $(shell $(CC) $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) -dM -E -x c /dev/null | \
 	grep -E -vw '__GNUC_(MINOR_|PATCHLEVEL_)?_' | \
 	sed -e "s/^\#define /-D'/" -e "s/ /'='/" -e "s/$$/'/" -e 's/\$$/&&/g')
 endif

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

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);
@@ -30,19 +33,25 @@ 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 */
 
 #define ACPI_TABLE_UPGRADE_MAX_PHYS ARCH_LOW_ADDRESS_LIMIT
 
 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;