Merge tag 'riscv-for-linus-4.15-rc2_cleanups' of...

Merge tag 'riscv-for-linus-4.15-rc2_cleanups' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/linux

Pull RISC-V cleanups and ABI fixes from Palmer Dabbelt:
 "This contains a handful of small cleanups that are a result of
  feedback that didn't make it into our original patch set, either
  because the feedback hadn't been given yet, I missed the original
  emails, or we weren't ready to submit the changes yet.

  I've been maintaining the various cleanup patch sets I have as their
  own branches, which I then merged together and signed. Each merge
  commit has a short summary of the changes, and each branch is based on
  your latest tag (4.15-rc1, in this case). If this isn't the right way
  to do this then feel free to suggest something else, but it seems sane
  to me.

  Here's a short summary of the changes, roughly in order of how
  interesting they are.

   - libgcc.h has been moved from include/lib, where it's the only
     member, to include/linux. This is meant to avoid tab completion
     conflicts.

   - VDSO entries for clock_get/gettimeofday/getcpu have been added.
     These are simple syscalls now, but we want to let glibc use them
     from the start so we can make them faster later.

   - A VDSO entry for instruction cache flushing has been added so
     userspace can flush the instruction cache.

   - The VDSO symbol versions for __vdso_cmpxchg{32,64} have been
     removed, as those VDSO entries don't actually exist.

   - __io_writes has been corrected to respect the given type.

   - A new READ_ONCE in arch_spin_is_locked().

   - __test_and_op_bit_ord() is now actually ordered.

   - Various small fixes throughout the tree to enable allmodconfig to
     build cleanly.

   - Removal of some dead code in our atomic support headers.

   - Improvements to various comments in our atomic support headers"

* tag 'riscv-for-linus-4.15-rc2_cleanups' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/linux: (23 commits)
  RISC-V: __io_writes should respect the length argument
  move libgcc.h to include/linux
  RISC-V: Clean up an unused include
  RISC-V: Allow userspace to flush the instruction cache
  RISC-V: Flush I$ when making a dirty page executable
  RISC-V: Add missing include
  RISC-V: Use define for get_cycles like other architectures
  RISC-V: Provide stub of setup_profiling_timer()
  RISC-V: Export some expected symbols for modules
  RISC-V: move empty_zero_page definition to C and export it
  RISC-V: io.h: type fixes for warnings
  RISC-V: use RISCV_{INT,SHORT} instead of {INT,SHORT} for asm macros
  RISC-V: use generic serial.h
  RISC-V: remove spin_unlock_wait()
  RISC-V: `sfence.vma` orderes the instruction cache
  RISC-V: Add READ_ONCE in arch_spin_is_locked()
  RISC-V: __test_and_op_bit_ord should be strongly ordered
  RISC-V: Remove smb_mb__{before,after}_spinlock()
  RISC-V: Remove __smp_bp__{before,after}_atomic
  RISC-V: Comment on why {,cmp}xchg is ordered how it is
  ...

arch/riscv/include/asm/Kbuild

@@ -40,6 +40,7 @@ generic-y += resource.h
 generic-y += scatterlist.h
 generic-y += sections.h
 generic-y += sembuf.h
+generic-y += serial.h
 generic-y += setup.h
 generic-y += shmbuf.h
 generic-y += shmparam.h

arch/riscv/include/asm/asm.h

@@ -58,17 +58,17 @@
 #endif
 
 #if (__SIZEOF_INT__ == 4)
-#define INT		__ASM_STR(.word)
-#define SZINT		__ASM_STR(4)
-#define LGINT		__ASM_STR(2)
+#define RISCV_INT		__ASM_STR(.word)
+#define RISCV_SZINT		__ASM_STR(4)
+#define RISCV_LGINT		__ASM_STR(2)
 #else
 #error "Unexpected __SIZEOF_INT__"
 #endif
 
 #if (__SIZEOF_SHORT__ == 2)
-#define SHORT		__ASM_STR(.half)
-#define SZSHORT		__ASM_STR(2)
-#define LGSHORT		__ASM_STR(1)
+#define RISCV_SHORT		__ASM_STR(.half)
+#define RISCV_SZSHORT		__ASM_STR(2)
+#define RISCV_LGSHORT		__ASM_STR(1)
 #else
 #error "Unexpected __SIZEOF_SHORT__"
 #endif

arch/riscv/include/asm/atomic.h

@@ -50,30 +50,30 @@ static __always_inline void atomic64_set(atomic64_t *v, long i)
  * have the AQ or RL bits set.  These don't return anything, so there's only
  * one version to worry about.
  */
-#define ATOMIC_OP(op, asm_op, c_op, I, asm_type, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
-{												\
-	__asm__ __volatile__ (									\
-		"amo" #asm_op "." #asm_type " zero, %1, %0"					\
-		: "+A" (v->counter)								\
-		: "r" (I)									\
-		: "memory");									\
+#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)				\
+static __always_inline void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)	\
+{											\
+	__asm__ __volatile__ (								\
+		"amo" #asm_op "." #asm_type " zero, %1, %0"				\
+		: "+A" (v->counter)							\
+		: "r" (I)								\
+		: "memory");								\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I)			\
-        ATOMIC_OP (op, asm_op, c_op, I, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, I)			\
+        ATOMIC_OP (op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I)			\
-        ATOMIC_OP (op, asm_op, c_op, I, w,  int,   )	\
-        ATOMIC_OP (op, asm_op, c_op, I, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, I)			\
+        ATOMIC_OP (op, asm_op, I, w,  int,   )	\
+        ATOMIC_OP (op, asm_op, I, d, long, 64)
 #endif
 
-ATOMIC_OPS(add, add, +,  i)
-ATOMIC_OPS(sub, add, +, -i)
-ATOMIC_OPS(and, and, &,  i)
-ATOMIC_OPS( or,  or, |,  i)
-ATOMIC_OPS(xor, xor, ^,  i)
+ATOMIC_OPS(add, add,  i)
+ATOMIC_OPS(sub, add, -i)
+ATOMIC_OPS(and, and,  i)
+ATOMIC_OPS( or,  or,  i)
+ATOMIC_OPS(xor, xor,  i)
 
 #undef ATOMIC_OP
 #undef ATOMIC_OPS
@@ -83,7 +83,7 @@ ATOMIC_OPS(xor, xor, ^,  i)
  * There's two flavors of these: the arithmatic ops have both fetch and return
  * versions, while the logical ops only have fetch versions.
  */
-#define ATOMIC_FETCH_OP(op, asm_op, c_op, I, asm_or, c_or, asm_type, c_type, prefix)			\
+#define ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, asm_type, c_type, prefix)				\
 static __always_inline c_type atomic##prefix##_fetch_##op##c_or(c_type i, atomic##prefix##_t *v)	\
 {													\
 	register c_type ret;										\
@@ -103,13 +103,13 @@ static __always_inline c_type atomic##prefix##_##op##_return##c_or(c_type i, ato
 
 #ifdef CONFIG_GENERIC_ATOMIC64
 #define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op, c_op, I, asm_or, c_or, w,  int,   )	\
+        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )
 #else
 #define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op, c_op, I, asm_or, c_or, w,  int,   )	\
+        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP (op, asm_op, c_op, I, asm_or, c_or, d, long, 64)	\
+        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, d, long, 64)	\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, d, long, 64)
 #endif
 
@@ -126,28 +126,28 @@ ATOMIC_OPS(sub, add, +, -i, .aqrl,         )
 #undef ATOMIC_OPS
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
+        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )		\
-        ATOMIC_FETCH_OP(op, asm_op, c_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
+        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )	\
+        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, d, long, 64)
 #endif
 
-ATOMIC_OPS(and, and, &,  i,      , _relaxed)
-ATOMIC_OPS(and, and, &,  i, .aq  , _acquire)
-ATOMIC_OPS(and, and, &,  i, .rl  , _release)
-ATOMIC_OPS(and, and, &,  i, .aqrl,         )
+ATOMIC_OPS(and, and, i,      , _relaxed)
+ATOMIC_OPS(and, and, i, .aq  , _acquire)
+ATOMIC_OPS(and, and, i, .rl  , _release)
+ATOMIC_OPS(and, and, i, .aqrl,         )
 
-ATOMIC_OPS( or,  or, |,  i,      , _relaxed)
-ATOMIC_OPS( or,  or, |,  i, .aq  , _acquire)
-ATOMIC_OPS( or,  or, |,  i, .rl  , _release)
-ATOMIC_OPS( or,  or, |,  i, .aqrl,         )
+ATOMIC_OPS( or,  or, i,      , _relaxed)
+ATOMIC_OPS( or,  or, i, .aq  , _acquire)
+ATOMIC_OPS( or,  or, i, .rl  , _release)
+ATOMIC_OPS( or,  or, i, .aqrl,         )
 
-ATOMIC_OPS(xor, xor, ^,  i,      , _relaxed)
-ATOMIC_OPS(xor, xor, ^,  i, .aq  , _acquire)
-ATOMIC_OPS(xor, xor, ^,  i, .rl  , _release)
-ATOMIC_OPS(xor, xor, ^,  i, .aqrl,         )
+ATOMIC_OPS(xor, xor, i,      , _relaxed)
+ATOMIC_OPS(xor, xor, i, .aq  , _acquire)
+ATOMIC_OPS(xor, xor, i, .rl  , _release)
+ATOMIC_OPS(xor, xor, i, .aqrl,         )
 
 #undef ATOMIC_OPS
 
@@ -182,13 +182,13 @@ ATOMIC_OPS(add_negative, add,  <, 0)
 #undef ATOMIC_OP
 #undef ATOMIC_OPS
 
-#define ATOMIC_OP(op, func_op, c_op, I, c_type, prefix)				\
+#define ATOMIC_OP(op, func_op, I, c_type, prefix)				\
 static __always_inline void atomic##prefix##_##op(atomic##prefix##_t *v)	\
 {										\
 	atomic##prefix##_##func_op(I, v);					\
 }
 
-#define ATOMIC_FETCH_OP(op, func_op, c_op, I, c_type, prefix)				\
+#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)					\
 static __always_inline c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)	\
 {											\
 	return atomic##prefix##_fetch_##func_op(I, v);					\
@@ -202,16 +202,16 @@ static __always_inline c_type atomic##prefix##_##op##_return(atomic##prefix##_t
 
 #ifdef CONFIG_GENERIC_ATOMIC64
 #define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op, c_op, I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op, c_op, I,  int,   )				\
+        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
+        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )
 #else
 #define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op, c_op, I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op, c_op, I,  int,   )				\
+        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
+        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )				\
-        ATOMIC_OP       (op, asm_op, c_op, I, long, 64)				\
-        ATOMIC_FETCH_OP (op, asm_op, c_op, I, long, 64)				\
+        ATOMIC_OP       (op, asm_op,       I, long, 64)				\
+        ATOMIC_FETCH_OP (op, asm_op,       I, long, 64)				\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I, long, 64)
 #endif
 
@@ -300,8 +300,13 @@ static __always_inline long atomic64_inc_not_zero(atomic64_t *v)
 
 /*
  * atomic_{cmp,}xchg is required to have exactly the same ordering semantics as
- * {cmp,}xchg and the operations that return, so they need a barrier.  We just
- * use the other implementations directly.
+ * {cmp,}xchg and the operations that return, so they need a barrier.
+ */
+/*
+ * FIXME: atomic_cmpxchg_{acquire,release,relaxed} are all implemented by
+ * assigning the same barrier to both the LR and SC operations, but that might
+ * not make any sense.  We're waiting on a memory model specification to
+ * determine exactly what the right thing to do is here.
  */
 #define ATOMIC_OP(c_t, prefix, c_or, size, asm_or)						\
 static __always_inline c_t atomic##prefix##_cmpxchg##c_or(atomic##prefix##_t *v, c_t o, c_t n) 	\

arch/riscv/include/asm/barrier.h

@@ -38,29 +38,6 @@
 #define smp_rmb()	RISCV_FENCE(r,r)
 #define smp_wmb()	RISCV_FENCE(w,w)
 
-/*
- * These fences exist to enforce ordering around the relaxed AMOs.  The
- * documentation defines that
- * "
- *     atomic_fetch_add();
- *   is equivalent to:
- *     smp_mb__before_atomic();
- *     atomic_fetch_add_relaxed();
- *     smp_mb__after_atomic();
- * "
- * So we emit full fences on both sides.
- */
-#define __smb_mb__before_atomic()	smp_mb()
-#define __smb_mb__after_atomic()	smp_mb()
-
-/*
- * These barriers prevent accesses performed outside a spinlock from being moved
- * inside a spinlock.  Since RISC-V sets the aq/rl bits on our spinlock only
- * enforce release consistency, we need full fences here.
- */
-#define smb_mb__before_spinlock()	smp_mb()
-#define smb_mb__after_spinlock()	smp_mb()
-
 #include <asm-generic/barrier.h>
 
 #endif /* __ASSEMBLY__ */

arch/riscv/include/asm/bitops.h

@@ -67,7 +67,7 @@
 		: "memory");
 
 #define __test_and_op_bit(op, mod, nr, addr) 			\
-	__test_and_op_bit_ord(op, mod, nr, addr, )
+	__test_and_op_bit_ord(op, mod, nr, addr, .aqrl)
 #define __op_bit(op, mod, nr, addr)				\
 	__op_bit_ord(op, mod, nr, addr, )
 

arch/riscv/include/asm/bug.h

@@ -27,8 +27,8 @@
 typedef u32 bug_insn_t;
 
 #ifdef CONFIG_GENERIC_BUG_RELATIVE_POINTERS
-#define __BUG_ENTRY_ADDR	INT " 1b - 2b"
-#define __BUG_ENTRY_FILE	INT " %0 - 2b"
+#define __BUG_ENTRY_ADDR	RISCV_INT " 1b - 2b"
+#define __BUG_ENTRY_FILE	RISCV_INT " %0 - 2b"
 #else
 #define __BUG_ENTRY_ADDR	RISCV_PTR " 1b"
 #define __BUG_ENTRY_FILE	RISCV_PTR " %0"
@@ -38,7 +38,7 @@ typedef u32 bug_insn_t;
 #define __BUG_ENTRY			\
 	__BUG_ENTRY_ADDR "\n\t"		\
 	__BUG_ENTRY_FILE "\n\t"		\
-	SHORT " %1"
+	RISCV_SHORT " %1"
 #else
 #define __BUG_ENTRY			\
 	__BUG_ENTRY_ADDR

arch/riscv/include/asm/cacheflush.h

@@ -18,22 +18,44 @@
 
 #undef flush_icache_range
 #undef flush_icache_user_range
+#undef flush_dcache_page
 
 static inline void local_flush_icache_all(void)
 {
 	asm volatile ("fence.i" ::: "memory");
 }
 
+#define PG_dcache_clean PG_arch_1
+
+static inline void flush_dcache_page(struct page *page)
+{
+	if (test_bit(PG_dcache_clean, &page->flags))
+		clear_bit(PG_dcache_clean, &page->flags);
+}
+
+/*
+ * RISC-V doesn't have an instruction to flush parts of the instruction cache,
+ * so instead we just flush the whole thing.
+ */
+#define flush_icache_range(start, end) flush_icache_all()
+#define flush_icache_user_range(vma, pg, addr, len) flush_icache_all()
+
 #ifndef CONFIG_SMP
 
-#define flush_icache_range(start, end) local_flush_icache_all()
-#define flush_icache_user_range(vma, pg, addr, len) local_flush_icache_all()
+#define flush_icache_all() local_flush_icache_all()
+#define flush_icache_mm(mm, local) flush_icache_all()
 
 #else /* CONFIG_SMP */
 
-#define flush_icache_range(start, end) sbi_remote_fence_i(0)
-#define flush_icache_user_range(vma, pg, addr, len) sbi_remote_fence_i(0)
+#define flush_icache_all() sbi_remote_fence_i(0)
+void flush_icache_mm(struct mm_struct *mm, bool local);
 
 #endif /* CONFIG_SMP */
 
+/*
+ * Bits in sys_riscv_flush_icache()'s flags argument.
+ */
+#define SYS_RISCV_FLUSH_ICACHE_LOCAL 1UL
+#define SYS_RISCV_FLUSH_ICACHE_ALL   (SYS_RISCV_FLUSH_ICACHE_LOCAL)
+
 #endif /* _ASM_RISCV_CACHEFLUSH_H */

arch/riscv/include/asm/io.h

@@ -19,6 +19,8 @@
 #ifndef _ASM_RISCV_IO_H
 #define _ASM_RISCV_IO_H
 
+#include <linux/types.h>
+
 #ifdef CONFIG_MMU
 
 extern void __iomem *ioremap(phys_addr_t offset, unsigned long size);
@@ -32,7 +34,7 @@ extern void __iomem *ioremap(phys_addr_t offset, unsigned long size);
 #define ioremap_wc(addr, size) ioremap((addr), (size))
 #define ioremap_wt(addr, size) ioremap((addr), (size))
 
-extern void iounmap(void __iomem *addr);
+extern void iounmap(volatile void __iomem *addr);
 
 #endif /* CONFIG_MMU */
 
@@ -250,7 +252,7 @@ static inline u64 __raw_readq(const volatile void __iomem *addr)
 			const ctype *buf = buffer;				\
 										\
 			do {							\
-				__raw_writeq(*buf++, addr);			\
+				__raw_write ## len(*buf++, addr);		\
 			} while (--count);					\
 		}								\
 		afence;								\
@@ -266,9 +268,9 @@ __io_reads_ins(reads, u32, l, __io_br(), __io_ar())
 __io_reads_ins(ins,  u8, b, __io_pbr(), __io_par())
 __io_reads_ins(ins, u16, w, __io_pbr(), __io_par())
 __io_reads_ins(ins, u32, l, __io_pbr(), __io_par())
-#define insb(addr, buffer, count) __insb((void __iomem *)addr, buffer, count)
-#define insw(addr, buffer, count) __insw((void __iomem *)addr, buffer, count)
-#define insl(addr, buffer, count) __insl((void __iomem *)addr, buffer, count)
+#define insb(addr, buffer, count) __insb((void __iomem *)(long)addr, buffer, count)
+#define insw(addr, buffer, count) __insw((void __iomem *)(long)addr, buffer, count)
+#define insl(addr, buffer, count) __insl((void __iomem *)(long)addr, buffer, count)
 
 __io_writes_outs(writes,  u8, b, __io_bw(), __io_aw())
 __io_writes_outs(writes, u16, w, __io_bw(), __io_aw())
@@ -280,9 +282,9 @@ __io_writes_outs(writes, u32, l, __io_bw(), __io_aw())
 __io_writes_outs(outs,  u8, b, __io_pbw(), __io_paw())
 __io_writes_outs(outs, u16, w, __io_pbw(), __io_paw())
 __io_writes_outs(outs, u32, l, __io_pbw(), __io_paw())
-#define outsb(addr, buffer, count) __outsb((void __iomem *)addr, buffer, count)
-#define outsw(addr, buffer, count) __outsw((void __iomem *)addr, buffer, count)
-#define outsl(addr, buffer, count) __outsl((void __iomem *)addr, buffer, count)
+#define outsb(addr, buffer, count) __outsb((void __iomem *)(long)addr, buffer, count)
+#define outsw(addr, buffer, count) __outsw((void __iomem *)(long)addr, buffer, count)
+#define outsl(addr, buffer, count) __outsl((void __iomem *)(long)addr, buffer, count)
 
 #ifdef CONFIG_64BIT
 __io_reads_ins(reads, u64, q, __io_br(), __io_ar())

arch/riscv/include/asm/mmu.h

@@ -19,6 +19,10 @@
 
 typedef struct {
 	void *vdso;
+#ifdef CONFIG_SMP
+	/* A local icache flush is needed before user execution can resume. */
+	cpumask_t icache_stale_mask;
+#endif
 } mm_context_t;
 
 #endif /* __ASSEMBLY__ */

arch/riscv/include/asm/mmu_context.h

 /*
  * Copyright (C) 2012 Regents of the University of California
+ * Copyright (C) 2017 SiFive
  *
  *   This program is free software; you can redistribute it and/or
  *   modify it under the terms of the GNU General Public License
@@ -14,11 +15,13 @@
 #ifndef _ASM_RISCV_MMU_CONTEXT_H
 #define _ASM_RISCV_MMU_CONTEXT_H
 
+#include <linux/mm_types.h>
 #include <asm-generic/mm_hooks.h>
 
 #include <linux/mm.h>
 #include <linux/sched.h>
 #include <asm/tlbflush.h>
+#include <asm/cacheflush.h>
 
 static inline void enter_lazy_tlb(struct mm_struct *mm,
 	struct task_struct *task)
@@ -46,12 +49,54 @@ static inline void set_pgdir(pgd_t *pgd)
 	csr_write(sptbr, virt_to_pfn(pgd) | SPTBR_MODE);
 }
 
+/*
+ * When necessary, performs a deferred icache flush for the given MM context,
+ * on the local CPU.  RISC-V has no direct mechanism for instruction cache
+ * shoot downs, so instead we send an IPI that informs the remote harts they
+ * need to flush their local instruction caches.  To avoid pathologically slow
+ * behavior in a common case (a bunch of single-hart processes on a many-hart
+ * machine, ie 'make -j') we avoid the IPIs for harts that are not currently
+ * executing a MM context and instead schedule a deferred local instruction
+ * cache flush to be performed before execution resumes on each hart.  This
+ * actually performs that local instruction cache flush, which implicitly only
+ * refers to the current hart.
+ */
+static inline void flush_icache_deferred(struct mm_struct *mm)
+{
+#ifdef CONFIG_SMP
+	unsigned int cpu = smp_processor_id();
+	cpumask_t *mask = &mm->context.icache_stale_mask;
+
+	if (cpumask_test_cpu(cpu, mask)) {
+		cpumask_clear_cpu(cpu, mask);
+		/*
+		 * Ensure the remote hart's writes are visible to this hart.
+		 * This pairs with a barrier in flush_icache_mm.
+		 */
+		smp_mb();
+		local_flush_icache_all();
+	}
+#endif
+}
+
 static inline void switch_mm(struct mm_struct *prev,
 	struct mm_struct *next, struct task_struct *task)
 {
 	if (likely(prev != next)) {
+		/*
+		 * Mark the current MM context as inactive, and the next as
+		 * active.  This is at least used by the icache flushing
+		 * routines in order to determine who should
+		 */
+		unsigned int cpu = smp_processor_id();
+
+		cpumask_clear_cpu(cpu, mm_cpumask(prev));
+		cpumask_set_cpu(cpu, mm_cpumask(next));
+
 		set_pgdir(next->pgd);
 		local_flush_tlb_all();
+
+		flush_icache_deferred(next);
 	}
 }
 

arch/riscv/include/asm/pgtable.h

@@ -178,28 +178,6 @@ static inline pte_t *pte_offset_kernel(pmd_t *pmd, unsigned long addr)
 #define pte_offset_map(dir, addr)	pte_offset_kernel((dir), (addr))
 #define pte_unmap(pte)			((void)(pte))
 
-/*
- * Certain architectures need to do special things when PTEs within
- * a page table are directly modified.  Thus, the following hook is
- * made available.
- */
-static inline void set_pte(pte_t *ptep, pte_t pteval)
-{
-	*ptep = pteval;
-}
-
-static inline void set_pte_at(struct mm_struct *mm,
-	unsigned long addr, pte_t *ptep, pte_t pteval)
-{
-	set_pte(ptep, pteval);
-}
-
-static inline void pte_clear(struct mm_struct *mm,
-	unsigned long addr, pte_t *ptep)
-{
-	set_pte_at(mm, addr, ptep, __pte(0));
-}
-
 static inline int pte_present(pte_t pte)
 {
 	return (pte_val(pte) & _PAGE_PRESENT);
@@ -210,21 +188,22 @@ static inline int pte_none(pte_t pte)
 	return (pte_val(pte) == 0);
 }
 
-/* static inline int pte_read(pte_t pte) */
-
 static inline int pte_write(pte_t pte)
 {
 	return pte_val(pte) & _PAGE_WRITE;
 }
 
+static inline int pte_exec(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_EXEC;
+}
+
 static inline int pte_huge(pte_t pte)
 {
 	return pte_present(pte)
 		&& (pte_val(pte) & (_PAGE_READ | _PAGE_WRITE | _PAGE_EXEC));
 }
 
-/* static inline int pte_exec(pte_t pte) */
-
 static inline int pte_dirty(pte_t pte)
 {
 	return pte_val(pte) & _PAGE_DIRTY;
@@ -311,6 +290,33 @@ static inline int pte_same(pte_t pte_a, pte_t pte_b)
 	return pte_val(pte_a) == pte_val(pte_b);
 }
 
+/*
+ * Certain architectures need to do special things when PTEs within
+ * a page table are directly modified.  Thus, the following hook is
+ * made available.
+ */
+static inline void set_pte(pte_t *ptep, pte_t pteval)
+{
+	*ptep = pteval;
+}
+
+void flush_icache_pte(pte_t pte);
+
+static inline void set_pte_at(struct mm_struct *mm,
+	unsigned long addr, pte_t *ptep, pte_t pteval)
+{
+	if (pte_present(pteval) && pte_exec(pteval))
+		flush_icache_pte(pteval);
+
+	set_pte(ptep, pteval);
+}
+
+static inline void pte_clear(struct mm_struct *mm,
+	unsigned long addr, pte_t *ptep)
+{
+	set_pte_at(mm, addr, ptep, __pte(0));
+}
+
 #define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
 static inline int ptep_set_access_flags(struct vm_area_struct *vma,
 					unsigned long address, pte_t *ptep,

arch/riscv/include/asm/spinlock.h

@@ -24,7 +24,7 @@
 
 /* FIXME: Replace this with a ticket lock, like MIPS. */
 
-#define arch_spin_is_locked(x)	((x)->lock != 0)
+#define arch_spin_is_locked(x)	(READ_ONCE((x)->lock) != 0)
 
 static inline void arch_spin_unlock(arch_spinlock_t *lock)
 {
@@ -58,15 +58,6 @@ static inline void arch_spin_lock(arch_spinlock_t *lock)
 	}
 }
 
-static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
-{
-	smp_rmb();
-	do {
-		cpu_relax();
-	} while (arch_spin_is_locked(lock));
-	smp_acquire__after_ctrl_dep();
-}
-
 /***********************************************************/
 
 static inline void arch_read_lock(arch_rwlock_t *lock)

arch/riscv/include/asm/timex.h

@@ -18,7 +18,7 @@
 
 typedef unsigned long cycles_t;
 
-static inline cycles_t get_cycles(void)
+static inline cycles_t get_cycles_inline(void)
 {
 	cycles_t n;
 
@@ -27,6 +27,7 @@ static inline cycles_t get_cycles(void)
 		: "=r" (n));
 	return n;
 }
+#define get_cycles get_cycles_inline
 
 #ifdef CONFIG_64BIT
 static inline uint64_t get_cycles64(void)

arch/riscv/include/asm/tlbflush.h

@@ -17,7 +17,12 @@
 
 #ifdef CONFIG_MMU
 
-/* Flush entire local TLB */
+#include <linux/mm_types.h>
+
+/*
+ * Flush entire local TLB.  'sfence.vma' implicitly fences with the instruction
+ * cache as well, so a 'fence.i' is not necessary.
+ */
 static inline void local_flush_tlb_all(void)
 {
 	__asm__ __volatile__ ("sfence.vma" : : : "memory");

arch/riscv/include/asm/vdso-syscalls.h

+/*
+ * Copyright (C) 2017 SiFive
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef _ASM_RISCV_VDSO_SYSCALLS_H
+#define _ASM_RISCV_VDSO_SYSCALLS_H
+
+#ifdef CONFIG_SMP
+
+/* These syscalls are only used by the vDSO and are not in the uapi. */
+#define __NR_riscv_flush_icache (__NR_arch_specific_syscall + 15)
+__SYSCALL(__NR_riscv_flush_icache, sys_riscv_flush_icache)
+
+#endif
+
+#endif /* _ASM_RISCV_VDSO_H */

arch/riscv/include/asm/vdso.h

@@ -38,4 +38,8 @@ struct vdso_data {
 	(void __user *)((unsigned long)(base) + __vdso_##name);			\
 })
 
+#ifdef CONFIG_SMP
+asmlinkage long sys_riscv_flush_icache(uintptr_t, uintptr_t, uintptr_t);
+#endif
+
 #endif /* _ASM_RISCV_VDSO_H */

arch/riscv/kernel/head.S

@@ -152,6 +152,3 @@ END(_start)
 __PAGE_ALIGNED_BSS
 	/* Empty zero page */
 	.balign PAGE_SIZE
-ENTRY(empty_zero_page)
-	.fill (empty_zero_page + PAGE_SIZE) - ., 1, 0x00
-END(empty_zero_page)

arch/riscv/kernel/riscv_ksyms.c

@@ -12,4 +12,7 @@
 /*
  * Assembly functions that may be used (directly or indirectly) by modules
  */
+EXPORT_SYMBOL(__clear_user);
 EXPORT_SYMBOL(__copy_user);
+EXPORT_SYMBOL(memset);
+EXPORT_SYMBOL(memcpy);

arch/riscv/kernel/setup.c

@@ -58,7 +58,12 @@ static char __initdata builtin_cmdline[COMMAND_LINE_SIZE] = CONFIG_CMDLINE;
 #endif /* CONFIG_CMDLINE_BOOL */
 
 unsigned long va_pa_offset;
+EXPORT_SYMBOL(va_pa_offset);
 unsigned long pfn_base;
+EXPORT_SYMBOL(pfn_base);
+
+unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)] __page_aligned_bss;
+EXPORT_SYMBOL(empty_zero_page);
 
 /* The lucky hart to first increment this variable will boot the other cores */
 atomic_t hart_lottery;

arch/riscv/kernel/smp.c

@@ -38,6 +38,13 @@ enum ipi_message_type {
 	IPI_MAX
 };
 
+
+/* Unsupported */
+int setup_profiling_timer(unsigned int multiplier)
+{
+	return -EINVAL;
+}
+
 irqreturn_t handle_ipi(void)
 {
 	unsigned long *pending_ipis = &ipi_data[smp_processor_id()].bits;
@@ -108,3 +115,51 @@ void smp_send_reschedule(int cpu)
 {
 	send_ipi_message(cpumask_of(cpu), IPI_RESCHEDULE);
 }
+
+/*
+ * Performs an icache flush for the given MM context.  RISC-V has no direct
+ * mechanism for instruction cache shoot downs, so instead we send an IPI that
+ * informs the remote harts they need to flush their local instruction caches.
+ * To avoid pathologically slow behavior in a common case (a bunch of
+ * single-hart processes on a many-hart machine, ie 'make -j') we avoid the
+ * IPIs for harts that are not currently executing a MM context and instead
+ * schedule a deferred local instruction cache flush to be performed before
+ * execution resumes on each hart.
+ */
+void flush_icache_mm(struct mm_struct *mm, bool local)
+{
+	unsigned int cpu;
+	cpumask_t others, *mask;
+
+	preempt_disable();
+
+	/* Mark every hart's icache as needing a flush for this MM. */
+	mask = &mm->context.icache_stale_mask;
+	cpumask_setall(mask);
+	/* Flush this hart's I$ now, and mark it as flushed. */
+	cpu = smp_processor_id();
+	cpumask_clear_cpu(cpu, mask);
+	local_flush_icache_all();
+
+	/*
+	 * Flush the I$ of other harts concurrently executing, and mark them as
+	 * flushed.
+	 */
+	cpumask_andnot(&others, mm_cpumask(mm), cpumask_of(cpu));
+	local |= cpumask_empty(&others);
+	if (mm != current->active_mm || !local)
+		sbi_remote_fence_i(others.bits);
+	else {
+		/*
+		 * It's assumed that at least one strongly ordered operation is
+		 * performed on this hart between setting a hart's cpumask bit
+		 * and scheduling this MM context on that hart.  Sending an SBI
+		 * remote message will do this, but in the case where no
+		 * messages are sent we still need to order this hart's writes
+		 * with flush_icache_deferred().
+		 */
+		smp_mb();
+	}
+
+	preempt_enable();
+}