Merge tag 'loongarch-fixes-6.0-1' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson

Pull LoongArch fixes from Huacai Chen:
 "Fix a bunch of build errors/warnings, a poweroff error and an
  unbalanced locking in do_page_fault()"

* tag 'loongarch-fixes-6.0-1' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson:
  LoongArch: mm: Avoid unnecessary page fault retires on shared memory types
  LoongArch: Add subword xchg/cmpxchg emulation
  LoongArch: Cleanup headers to avoid circular dependency
  LoongArch: Cleanup reset routines with new API
  LoongArch: Fix build warnings in VDSO
  LoongArch: Select PCI_QUIRKS to avoid build error

13 files changed, 164 insertions, 98 deletions

diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig
index 4abc9a28aba4..26aeb1408e56 100644
--- a/arch/loongarch/Kconfig
+++ b/arch/loongarch/Kconfig
@@ -111,6 +111,7 @@ config LOONGARCH
 	select PCI_ECAM if ACPI
 	select PCI_LOONGSON
 	select PCI_MSI_ARCH_FALLBACKS
+	select PCI_QUIRKS
 	select PERF_USE_VMALLOC
 	select RTC_LIB
 	select SMP
diff --git a/arch/loongarch/include/asm/addrspace.h b/arch/loongarch/include/asm/addrspace.h
index b91e0733b2e5..d342935e5a72 100644
--- a/arch/loongarch/include/asm/addrspace.h
+++ b/arch/loongarch/include/asm/addrspace.h
@@ -109,4 +109,20 @@ extern unsigned long vm_map_base;
  */
 #define PHYSADDR(a)		((_ACAST64_(a)) & TO_PHYS_MASK)
 
+/*
+ * On LoongArch, I/O ports mappring is following:
+ *
+ *              |         ....          |
+ *              |-----------------------|
+ *              | pci io ports(16K~32M) |
+ *              |-----------------------|
+ *              | isa io ports(0  ~16K) |
+ * PCI_IOBASE ->|-----------------------|
+ *              |         ....          |
+ */
+#define PCI_IOBASE	((void __iomem *)(vm_map_base + (2 * PAGE_SIZE)))
+#define PCI_IOSIZE	SZ_32M
+#define ISA_IOSIZE	SZ_16K
+#define IO_SPACE_LIMIT	(PCI_IOSIZE - 1)
+
 #endif /* _ASM_ADDRSPACE_H */
diff --git a/arch/loongarch/include/asm/cmpxchg.h b/arch/loongarch/include/asm/cmpxchg.h
index 0a9b0fac1eee..ae19e33c7754 100644
--- a/arch/loongarch/include/asm/cmpxchg.h
+++ b/arch/loongarch/include/asm/cmpxchg.h
@@ -5,8 +5,9 @@
 #ifndef __ASM_CMPXCHG_H
 #define __ASM_CMPXCHG_H
 
-#include <asm/barrier.h>
+#include <linux/bits.h>
 #include <linux/build_bug.h>
+#include <asm/barrier.h>
 
 #define __xchg_asm(amswap_db, m, val)		\
 ({						\
@@ -21,10 +22,53 @@
 		__ret;				\
 })
 
+static inline unsigned int __xchg_small(volatile void *ptr, unsigned int val,
+					unsigned int size)
+{
+	unsigned int shift;
+	u32 old32, mask, temp;
+	volatile u32 *ptr32;
+
+	/* Mask value to the correct size. */
+	mask = GENMASK((size * BITS_PER_BYTE) - 1, 0);
+	val &= mask;
+
+	/*
+	 * Calculate a shift & mask that correspond to the value we wish to
+	 * exchange within the naturally aligned 4 byte integerthat includes
+	 * it.
+	 */
+	shift = (unsigned long)ptr & 0x3;
+	shift *= BITS_PER_BYTE;
+	mask <<= shift;
+
+	/*
+	 * Calculate a pointer to the naturally aligned 4 byte integer that
+	 * includes our byte of interest, and load its value.
+	 */
+	ptr32 = (volatile u32 *)((unsigned long)ptr & ~0x3);
+
+	asm volatile (
+	"1:	ll.w		%0, %3		\n"
+	"	andn		%1, %0, %z4	\n"
+	"	or		%1, %1, %z5	\n"
+	"	sc.w		%1, %2		\n"
+	"	beqz		%1, 1b		\n"
+	: "=&r" (old32), "=&r" (temp), "=ZC" (*ptr32)
+	: "ZC" (*ptr32), "Jr" (mask), "Jr" (val << shift)
+	: "memory");
+
+	return (old32 & mask) >> shift;
+}
+
 static inline unsigned long __xchg(volatile void *ptr, unsigned long x,
 				   int size)
 {
 	switch (size) {
+	case 1:
+	case 2:
+		return __xchg_small(ptr, x, size);
+
 	case 4:
 		return __xchg_asm("amswap_db.w", (volatile u32 *)ptr, (u32)x);
 
@@ -67,10 +111,62 @@ static inline unsigned long __xchg(volatile void *ptr, unsigned long x,
 	__ret;								\
 })
 
+static inline unsigned int __cmpxchg_small(volatile void *ptr, unsigned int old,
+					   unsigned int new, unsigned int size)
+{
+	unsigned int shift;
+	u32 old32, mask, temp;
+	volatile u32 *ptr32;
+
+	/* Mask inputs to the correct size. */
+	mask = GENMASK((size * BITS_PER_BYTE) - 1, 0);
+	old &= mask;
+	new &= mask;
+
+	/*
+	 * Calculate a shift & mask that correspond to the value we wish to
+	 * compare & exchange within the naturally aligned 4 byte integer
+	 * that includes it.
+	 */
+	shift = (unsigned long)ptr & 0x3;
+	shift *= BITS_PER_BYTE;
+	old <<= shift;
+	new <<= shift;
+	mask <<= shift;
+
+	/*
+	 * Calculate a pointer to the naturally aligned 4 byte integer that
+	 * includes our byte of interest, and load its value.
+	 */
+	ptr32 = (volatile u32 *)((unsigned long)ptr & ~0x3);
+
+	asm volatile (
+	"1:	ll.w		%0, %3		\n"
+	"	and		%1, %0, %z4	\n"
+	"	bne		%1, %z5, 2f	\n"
+	"	andn		%1, %0, %z4	\n"
+	"	or		%1, %1, %z6	\n"
+	"	sc.w		%1, %2		\n"
+	"	beqz		%1, 1b		\n"
+	"	b		3f		\n"
+	"2:					\n"
+	__WEAK_LLSC_MB
+	"3:					\n"
+	: "=&r" (old32), "=&r" (temp), "=ZC" (*ptr32)
+	: "ZC" (*ptr32), "Jr" (mask), "Jr" (old), "Jr" (new)
+	: "memory");
+
+	return (old32 & mask) >> shift;
+}
+
 static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
 				      unsigned long new, unsigned int size)
 {
 	switch (size) {
+	case 1:
+	case 2:
+		return __cmpxchg_small(ptr, old, new, size);
+
 	case 4:
 		return __cmpxchg_asm("ll.w", "sc.w", (volatile u32 *)ptr,
 				     (u32)old, new);
diff --git a/arch/loongarch/include/asm/io.h b/arch/loongarch/include/asm/io.h
index 884599739b36..999944ea1cea 100644
--- a/arch/loongarch/include/asm/io.h
+++ b/arch/loongarch/include/asm/io.h
@@ -7,35 +7,16 @@
 
 #define ARCH_HAS_IOREMAP_WC
 
-#include <linux/compiler.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
 
 #include <asm/addrspace.h>
-#include <asm/bug.h>
-#include <asm/byteorder.h>
 #include <asm/cpu.h>
 #include <asm/page.h>
 #include <asm/pgtable-bits.h>
 #include <asm/string.h>
 
 /*
- * On LoongArch, I/O ports mappring is following:
- *
- *              |         ....          |
- *              |-----------------------|
- *              | pci io ports(64K~32M) |
- *              |-----------------------|
- *              | isa io ports(0  ~16K) |
- * PCI_IOBASE ->|-----------------------|
- *              |         ....          |
- */
-#define PCI_IOBASE	((void __iomem *)(vm_map_base + (2 * PAGE_SIZE)))
-#define PCI_IOSIZE	SZ_32M
-#define ISA_IOSIZE	SZ_16K
-#define IO_SPACE_LIMIT	(PCI_IOSIZE - 1)
-
-/*
  * Change "struct page" to physical address.
  */
 #define page_to_phys(page)	((phys_addr_t)page_to_pfn(page) << PAGE_SHIFT)
diff --git a/arch/loongarch/include/asm/page.h b/arch/loongarch/include/asm/page.h
index a37324ac460b..53f284a96182 100644
--- a/arch/loongarch/include/asm/page.h
+++ b/arch/loongarch/include/asm/page.h
@@ -95,7 +95,7 @@ static inline int pfn_valid(unsigned long pfn)
 
 #endif
 
-#define virt_to_pfn(kaddr)	PFN_DOWN(virt_to_phys((void *)(kaddr)))
+#define virt_to_pfn(kaddr)	PFN_DOWN(PHYSADDR(kaddr))
 #define virt_to_page(kaddr)	pfn_to_page(virt_to_pfn(kaddr))
 
 extern int __virt_addr_valid(volatile void *kaddr);
diff --git a/arch/loongarch/include/asm/percpu.h b/arch/loongarch/include/asm/percpu.h
index e6569f18c6dd..0bd6b0110198 100644
--- a/arch/loongarch/include/asm/percpu.h
+++ b/arch/loongarch/include/asm/percpu.h
@@ -123,6 +123,10 @@ static inline unsigned long __percpu_xchg(void *ptr, unsigned long val,
 						int size)
 {
 	switch (size) {
+	case 1:
+	case 2:
+		return __xchg_small((volatile void *)ptr, val, size);
+
 	case 4:
 		return __xchg_asm("amswap.w", (volatile u32 *)ptr, (u32)val);
 
@@ -204,9 +208,13 @@ do {									\
 #define this_cpu_write_4(pcp, val) _percpu_write(pcp, val)
 #define this_cpu_write_8(pcp, val) _percpu_write(pcp, val)
 
+#define this_cpu_xchg_1(pcp, val) _percpu_xchg(pcp, val)
+#define this_cpu_xchg_2(pcp, val) _percpu_xchg(pcp, val)
 #define this_cpu_xchg_4(pcp, val) _percpu_xchg(pcp, val)
 #define this_cpu_xchg_8(pcp, val) _percpu_xchg(pcp, val)
 
+#define this_cpu_cmpxchg_1(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
+#define this_cpu_cmpxchg_2(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
 #define this_cpu_cmpxchg_4(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
 #define this_cpu_cmpxchg_8(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
 
diff --git a/arch/loongarch/include/asm/pgtable.h b/arch/loongarch/include/asm/pgtable.h
index e03443abaf7d..8ea57e2f0e04 100644
--- a/arch/loongarch/include/asm/pgtable.h
+++ b/arch/loongarch/include/asm/pgtable.h
@@ -59,7 +59,6 @@
 #include <linux/mm_types.h>
 #include <linux/mmzone.h>
 #include <asm/fixmap.h>
-#include <asm/io.h>
 
 struct mm_struct;
 struct vm_area_struct;
@@ -145,7 +144,7 @@ static inline void set_p4d(p4d_t *p4d, p4d_t p4dval)
 	*p4d = p4dval;
 }
 
-#define p4d_phys(p4d)		virt_to_phys((void *)p4d_val(p4d))
+#define p4d_phys(p4d)		PHYSADDR(p4d_val(p4d))
 #define p4d_page(p4d)		(pfn_to_page(p4d_phys(p4d) >> PAGE_SHIFT))
 
 #endif
@@ -188,7 +187,7 @@ static inline pmd_t *pud_pgtable(pud_t pud)
 
 #define set_pud(pudptr, pudval) do { *(pudptr) = (pudval); } while (0)
 
-#define pud_phys(pud)		virt_to_phys((void *)pud_val(pud))
+#define pud_phys(pud)		PHYSADDR(pud_val(pud))
 #define pud_page(pud)		(pfn_to_page(pud_phys(pud) >> PAGE_SHIFT))
 
 #endif
@@ -221,7 +220,7 @@ static inline void pmd_clear(pmd_t *pmdp)
 
 #define set_pmd(pmdptr, pmdval) do { *(pmdptr) = (pmdval); } while (0)
 
-#define pmd_phys(pmd)		virt_to_phys((void *)pmd_val(pmd))
+#define pmd_phys(pmd)		PHYSADDR(pmd_val(pmd))
 
 #ifndef CONFIG_TRANSPARENT_HUGEPAGE
 #define pmd_page(pmd)		(pfn_to_page(pmd_phys(pmd) >> PAGE_SHIFT))
diff --git a/arch/loongarch/include/asm/reboot.h b/arch/loongarch/include/asm/reboot.h
deleted file mode 100644
index 51151749d8f0..000000000000
--- a/arch/loongarch/include/asm/reboot.h
+++ /dev/null
@@ -1,10 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
- */
-#ifndef _ASM_REBOOT_H
-#define _ASM_REBOOT_H
-
-extern void (*pm_restart)(void);
-
-#endif /* _ASM_REBOOT_H */
diff --git a/arch/loongarch/kernel/reset.c b/arch/loongarch/kernel/reset.c
index 800c965a17ea..8c82021eb2f4 100644
--- a/arch/loongarch/kernel/reset.c
+++ b/arch/loongarch/kernel/reset.c
@@ -15,10 +15,16 @@
 #include <acpi/reboot.h>
 #include <asm/idle.h>
 #include <asm/loongarch.h>
-#include <asm/reboot.h>
 
-static void default_halt(void)
+void (*pm_power_off)(void);
+EXPORT_SYMBOL(pm_power_off);
+
+void machine_halt(void)
 {
+#ifdef CONFIG_SMP
+	preempt_disable();
+	smp_send_stop();
+#endif
 	local_irq_disable();
 	clear_csr_ecfg(ECFG0_IM);
 
@@ -30,18 +36,29 @@ static void default_halt(void)
 	}
 }
 
-static void default_poweroff(void)
+void machine_power_off(void)
 {
+#ifdef CONFIG_SMP
+	preempt_disable();
+	smp_send_stop();
+#endif
+	do_kernel_power_off();
 #ifdef CONFIG_EFI
 	efi.reset_system(EFI_RESET_SHUTDOWN, EFI_SUCCESS, 0, NULL);
 #endif
+
 	while (true) {
 		__arch_cpu_idle();
 	}
 }
 
-static void default_restart(void)
+void machine_restart(char *command)
 {
+#ifdef CONFIG_SMP
+	preempt_disable();
+	smp_send_stop();
+#endif
+	do_kernel_restart(command);
 #ifdef CONFIG_EFI
 	if (efi_capsule_pending(NULL))
 		efi_reboot(REBOOT_WARM, NULL);
@@ -55,47 +72,3 @@ static void default_restart(void)
 		__arch_cpu_idle();
 	}
 }
-
-void (*pm_restart)(void);
-EXPORT_SYMBOL(pm_restart);
-
-void (*pm_power_off)(void);
-EXPORT_SYMBOL(pm_power_off);
-
-void machine_halt(void)
-{
-#ifdef CONFIG_SMP
-	preempt_disable();
-	smp_send_stop();
-#endif
-	default_halt();
-}
-
-void machine_power_off(void)
-{
-#ifdef CONFIG_SMP
-	preempt_disable();
-	smp_send_stop();
-#endif
-	pm_power_off();
-}
-
-void machine_restart(char *command)
-{
-#ifdef CONFIG_SMP
-	preempt_disable();
-	smp_send_stop();
-#endif
-	do_kernel_restart(command);
-	pm_restart();
-}
-
-static int __init loongarch_reboot_setup(void)
-{
-	pm_restart = default_restart;
-	pm_power_off = default_poweroff;
-
-	return 0;
-}
-
-arch_initcall(loongarch_reboot_setup);
diff --git a/arch/loongarch/mm/fault.c b/arch/loongarch/mm/fault.c
index 605579b19a00..1ccd53655cab 100644
--- a/arch/loongarch/mm/fault.c
+++ b/arch/loongarch/mm/fault.c
@@ -216,6 +216,10 @@ good_area:
 		return;
 	}
 
+	/* The fault is fully completed (including releasing mmap lock) */
+	if (fault & VM_FAULT_COMPLETED)
+		return;
+
 	if (unlikely(fault & VM_FAULT_RETRY)) {
 		flags |= FAULT_FLAG_TRIED;
 
diff --git a/arch/loongarch/mm/mmap.c b/arch/loongarch/mm/mmap.c
index 52e40f0ba732..381a569635a9 100644
--- a/arch/loongarch/mm/mmap.c
+++ b/arch/loongarch/mm/mmap.c
@@ -2,16 +2,9 @@
 /*
  * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
  */
-#include <linux/compiler.h>
-#include <linux/elf-randomize.h>
-#include <linux/errno.h>
+#include <linux/export.h>
 #include <linux/mm.h>
 #include <linux/mman.h>
-#include <linux/export.h>
-#include <linux/personality.h>
-#include <linux/random.h>
-#include <linux/sched/signal.h>
-#include <linux/sched/mm.h>
 
 unsigned long shm_align_mask = PAGE_SIZE - 1;	/* Sane caches */
 EXPORT_SYMBOL(shm_align_mask);
@@ -120,6 +113,6 @@ int __virt_addr_valid(volatile void *kaddr)
 	if ((vaddr < PAGE_OFFSET) || (vaddr >= vm_map_base))
 		return 0;
 
-	return pfn_valid(PFN_DOWN(virt_to_phys(kaddr)));
+	return pfn_valid(PFN_DOWN(PHYSADDR(kaddr)));
 }
 EXPORT_SYMBOL_GPL(__virt_addr_valid);
diff --git a/arch/loongarch/vdso/vgetcpu.c b/arch/loongarch/vdso/vgetcpu.c
index 43a0078e4418..e02e775f5360 100644
--- a/arch/loongarch/vdso/vgetcpu.c
+++ b/arch/loongarch/vdso/vgetcpu.c
@@ -24,6 +24,8 @@ static __always_inline const struct vdso_pcpu_data *get_pcpu_data(void)
 	return (struct vdso_pcpu_data *)(get_vdso_base() - VDSO_DATA_SIZE);
 }
 
+extern
+int __vdso_getcpu(unsigned int *cpu, unsigned int *node, struct getcpu_cache *unused);
 int __vdso_getcpu(unsigned int *cpu, unsigned int *node, struct getcpu_cache *unused)
 {
 	int cpu_id;
diff --git a/arch/loongarch/vdso/vgettimeofday.c b/arch/loongarch/vdso/vgettimeofday.c
index b1f4548dae92..8f22863bd7ea 100644
--- a/arch/loongarch/vdso/vgettimeofday.c
+++ b/arch/loongarch/vdso/vgettimeofday.c
@@ -6,20 +6,23 @@
  */
 #include <linux/types.h>
 
-int __vdso_clock_gettime(clockid_t clock,
-			 struct __kernel_timespec *ts)
+extern
+int __vdso_clock_gettime(clockid_t clock, struct __kernel_timespec *ts);
+int __vdso_clock_gettime(clockid_t clock, struct __kernel_timespec *ts)
 {
 	return __cvdso_clock_gettime(clock, ts);
 }
 
-int __vdso_gettimeofday(struct __kernel_old_timeval *tv,
-			struct timezone *tz)
+extern
+int __vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz);
+int __vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz)
 {
 	return __cvdso_gettimeofday(tv, tz);
 }
 
-int __vdso_clock_getres(clockid_t clock_id,
-			struct __kernel_timespec *res)
+extern
+int __vdso_clock_getres(clockid_t clock_id, struct __kernel_timespec *res);
+int __vdso_clock_getres(clockid_t clock_id, struct __kernel_timespec *res)
 {
 	return __cvdso_clock_getres(clock_id, res);
 }