summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--arch/riscv/include/asm/atomic.h103
-rw-r--r--arch/riscv/include/asm/barrier.h23
-rw-r--r--arch/riscv/include/asm/bitops.h2
-rw-r--r--arch/riscv/include/asm/cacheflush.h30
-rw-r--r--arch/riscv/include/asm/io.h2
-rw-r--r--arch/riscv/include/asm/mmu.h4
-rw-r--r--arch/riscv/include/asm/mmu_context.h44
-rw-r--r--arch/riscv/include/asm/pgtable.h58
-rw-r--r--arch/riscv/include/asm/spinlock.h11
-rw-r--r--arch/riscv/include/asm/tlbflush.h7
-rw-r--r--arch/riscv/include/asm/vdso-syscalls.h28
-rw-r--r--arch/riscv/include/asm/vdso.h4
-rw-r--r--arch/riscv/kernel/smp.c48
-rw-r--r--arch/riscv/kernel/sys_riscv.c33
-rw-r--r--arch/riscv/kernel/syscall_table.c2
-rw-r--r--arch/riscv/kernel/vdso/Makefile7
-rw-r--r--arch/riscv/kernel/vdso/clock_getres.S26
-rw-r--r--arch/riscv/kernel/vdso/clock_gettime.S26
-rw-r--r--arch/riscv/kernel/vdso/flush_icache.S31
-rw-r--r--arch/riscv/kernel/vdso/getcpu.S26
-rw-r--r--arch/riscv/kernel/vdso/gettimeofday.S26
-rw-r--r--arch/riscv/kernel/vdso/vdso.lds.S7
-rw-r--r--arch/riscv/mm/Makefile1
-rw-r--r--arch/riscv/mm/cacheflush.c23
-rw-r--r--include/linux/libgcc.h (renamed from include/lib/libgcc.h)0
-rw-r--r--lib/ashldi3.c2
-rw-r--r--lib/ashrdi3.c2
-rw-r--r--lib/cmpdi2.c2
-rw-r--r--lib/lshrdi3.c2
-rw-r--r--lib/muldi3.c2
-rw-r--r--lib/ucmpdi2.c2
31 files changed, 459 insertions, 125 deletions
diff --git a/arch/riscv/include/asm/atomic.h b/arch/riscv/include/asm/atomic.h
index e2e37c57cbeb..e65d1cd89e28 100644
--- a/arch/riscv/include/asm/atomic.h
+++ b/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) \
diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
index 183534b7c39b..773c4e039cd7 100644
--- a/arch/riscv/include/asm/barrier.h
+++ b/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__ */
diff --git a/arch/riscv/include/asm/bitops.h b/arch/riscv/include/asm/bitops.h
index 7c281ef1d583..f30daf26f08f 100644
--- a/arch/riscv/include/asm/bitops.h
+++ b/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, )
diff --git a/arch/riscv/include/asm/cacheflush.h b/arch/riscv/include/asm/cacheflush.h
index 0595585013b0..efd89a88d2d0 100644
--- a/arch/riscv/include/asm/cacheflush.h
+++ b/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 */
diff --git a/arch/riscv/include/asm/io.h b/arch/riscv/include/asm/io.h
index 2a5c7c371e2d..a82ce599b639 100644
--- a/arch/riscv/include/asm/io.h
+++ b/arch/riscv/include/asm/io.h
@@ -252,7 +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; \
diff --git a/arch/riscv/include/asm/mmu.h b/arch/riscv/include/asm/mmu.h
index 66805cba9a27..5df2dccdba12 100644
--- a/arch/riscv/include/asm/mmu.h
+++ b/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__ */
diff --git a/arch/riscv/include/asm/mmu_context.h b/arch/riscv/include/asm/mmu_context.h
index 1cd5172882c2..97424834dce2 100644
--- a/arch/riscv/include/asm/mmu_context.h
+++ b/arch/riscv/include/asm/mmu_context.h
@@ -1,5 +1,6 @@
/*
* 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
@@ -20,6 +21,7 @@
#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)
@@ -47,12 +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);
}
}
diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
index 3399257780b2..2cbd92ed1629 100644
--- a/arch/riscv/include/asm/pgtable.h
+++ b/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,
diff --git a/arch/riscv/include/asm/spinlock.h b/arch/riscv/include/asm/spinlock.h
index 04c71d938afd..2fd27e8ef1fd 100644
--- a/arch/riscv/include/asm/spinlock.h
+++ b/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)
diff --git a/arch/riscv/include/asm/tlbflush.h b/arch/riscv/include/asm/tlbflush.h
index 5ee4ae370b5e..715b0f10af58 100644
--- a/arch/riscv/include/asm/tlbflush.h
+++ b/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");
diff --git a/arch/riscv/include/asm/vdso-syscalls.h b/arch/riscv/include/asm/vdso-syscalls.h
new file mode 100644
index 000000000000..a2ccf1894929
--- /dev/null
+++ b/arch/riscv/include/asm/vdso-syscalls.h
@@ -0,0 +1,28 @@
+/*
+ * 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 */
diff --git a/arch/riscv/include/asm/vdso.h b/arch/riscv/include/asm/vdso.h
index 602f61257553..541544d64c33 100644
--- a/arch/riscv/include/asm/vdso.h
+++ b/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 */
diff --git a/arch/riscv/kernel/smp.c b/arch/riscv/kernel/smp.c
index 17014c68e9fe..6d3962435720 100644
--- a/arch/riscv/kernel/smp.c
+++ b/arch/riscv/kernel/smp.c
@@ -115,3 +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();
+}
diff --git a/arch/riscv/kernel/sys_riscv.c b/arch/riscv/kernel/sys_riscv.c
index 4351be7d0533..a2ae936a093e 100644
--- a/arch/riscv/kernel/sys_riscv.c
+++ b/arch/riscv/kernel/sys_riscv.c
@@ -14,8 +14,8 @@
*/
#include <linux/syscalls.h>
-#include <asm/cmpxchg.h>
#include <asm/unistd.h>
+#include <asm/cacheflush.h>
static long riscv_sys_mmap(unsigned long addr, unsigned long len,
unsigned long prot, unsigned long flags,
@@ -47,3 +47,34 @@ SYSCALL_DEFINE6(mmap2, unsigned long, addr, unsigned long, len,
return riscv_sys_mmap(addr, len, prot, flags, fd, offset, 12);
}
#endif /* !CONFIG_64BIT */
+
+#ifdef CONFIG_SMP
+/*
+ * Allows the instruction cache to be flushed from userspace. Despite RISC-V
+ * having a direct 'fence.i' instruction available to userspace (which we
+ * can't trap!), that's not actually viable when running on Linux because the
+ * kernel might schedule a process on another hart. There is no way for
+ * userspace to handle this without invoking the kernel (as it doesn't know the
+ * thread->hart mappings), so we've defined a RISC-V specific system call to
+ * flush the instruction cache.
+ *
+ * sys_riscv_flush_icache() is defined to flush the instruction cache over an
+ * address range, with the flush applying to either all threads or just the
+ * caller. We don't currently do anything with the address range, that's just
+ * in there for forwards compatibility.
+ */
+SYSCALL_DEFINE3(riscv_flush_icache, uintptr_t, start, uintptr_t, end,
+ uintptr_t, flags)
+{
+ struct mm_struct *mm = current->mm;
+ bool local = (flags & SYS_RISCV_FLUSH_ICACHE_LOCAL) != 0;
+
+ /* Check the reserved flags. */
+ if (unlikely(flags & !SYS_RISCV_FLUSH_ICACHE_ALL))
+ return -EINVAL;
+
+ flush_icache_mm(mm, local);
+
+ return 0;
+}
+#endif
diff --git a/arch/riscv/kernel/syscall_table.c b/arch/riscv/kernel/syscall_table.c
index 4e30dc5fb593..a5bd6401f95e 100644
--- a/arch/riscv/kernel/syscall_table.c
+++ b/arch/riscv/kernel/syscall_table.c
@@ -15,6 +15,7 @@
#include <linux/linkage.h>
#include <linux/syscalls.h>
#include <asm-generic/syscalls.h>
+#include <asm/vdso.h>
#undef __SYSCALL
#define __SYSCALL(nr, call) [nr] = (call),
@@ -22,4 +23,5 @@
void *sys_call_table[__NR_syscalls] = {
[0 ... __NR_syscalls - 1] = sys_ni_syscall,
#include <asm/unistd.h>
+#include <asm/vdso-syscalls.h>
};
diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makefile
index 523d0a8ac8db..324568d33921 100644
--- a/arch/riscv/kernel/vdso/Makefile
+++ b/arch/riscv/kernel/vdso/Makefile
@@ -1,7 +1,12 @@
# Copied from arch/tile/kernel/vdso/Makefile
# Symbols present in the vdso
-vdso-syms = rt_sigreturn
+vdso-syms = rt_sigreturn
+vdso-syms += gettimeofday
+vdso-syms += clock_gettime
+vdso-syms += clock_getres
+vdso-syms += getcpu
+vdso-syms += flush_icache
# Files to link into the vdso
obj-vdso = $(patsubst %, %.o, $(vdso-syms))
diff --git a/arch/riscv/kernel/vdso/clock_getres.S b/arch/riscv/kernel/vdso/clock_getres.S
new file mode 100644
index 000000000000..edf7e2339648
--- /dev/null
+++ b/arch/riscv/kernel/vdso/clock_getres.S
@@ -0,0 +1,26 @@
+/*
+ * 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
+ * as published by the Free Software Foundation, version 2.
+ *
+ * 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.
+ */
+
+#include <linux/linkage.h>
+#include <asm/unistd.h>
+
+ .text
+/* int __vdso_clock_getres(clockid_t clock_id, struct timespec *res); */
+ENTRY(__vdso_clock_getres)
+ .cfi_startproc
+ /* For now, just do the syscall. */
+ li a7, __NR_clock_getres
+ ecall
+ ret
+ .cfi_endproc
+ENDPROC(__vdso_clock_getres)
diff --git a/arch/riscv/kernel/vdso/clock_gettime.S b/arch/riscv/kernel/vdso/clock_gettime.S
new file mode 100644
index 000000000000..aac65676c6d5
--- /dev/null
+++ b/arch/riscv/kernel/vdso/clock_gettime.S
@@ -0,0 +1,26 @@
+/*
+ * 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
+ * as published by the Free Software Foundation, version 2.
+ *
+ * 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.
+ */
+
+#include <linux/linkage.h>
+#include <asm/unistd.h>
+
+ .text
+/* int __vdso_clock_gettime(clockid_t clock_id, struct timespec *tp); */
+ENTRY(__vdso_clock_gettime)
+ .cfi_startproc
+ /* For now, just do the syscall. */
+ li a7, __NR_clock_gettime
+ ecall
+ ret
+ .cfi_endproc
+ENDPROC(__vdso_clock_gettime)
diff --git a/arch/riscv/kernel/vdso/flush_icache.S b/arch/riscv/kernel/vdso/flush_icache.S
new file mode 100644
index 000000000000..b0fbad74e873
--- /dev/null
+++ b/arch/riscv/kernel/vdso/flush_icache.S
@@ -0,0 +1,31 @@
+/*
+ * 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
+ * as published by the Free Software Foundation, version 2.
+ *
+ * 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.
+ */
+
+#include <linux/linkage.h>
+#include <asm/unistd.h>
+#include <asm/vdso-syscalls.h>
+
+ .text
+/* int __vdso_flush_icache(void *start, void *end, unsigned long flags); */
+ENTRY(__vdso_flush_icache)
+ .cfi_startproc
+#ifdef CONFIG_SMP
+ li a7, __NR_riscv_flush_icache
+ ecall
+#else
+ fence.i
+ li a0, 0
+#endif
+ ret
+ .cfi_endproc
+ENDPROC(__vdso_flush_icache)
diff --git a/arch/riscv/kernel/vdso/getcpu.S b/arch/riscv/kernel/vdso/getcpu.S
new file mode 100644
index 000000000000..cc7e98924484
--- /dev/null
+++ b/arch/riscv/kernel/vdso/getcpu.S
@@ -0,0 +1,26 @@
+/*
+ * 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
+ * as published by the Free Software Foundation, version 2.
+ *
+ * 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.
+ */
+
+#include <linux/linkage.h>
+#include <asm/unistd.h>
+
+ .text
+/* int __vdso_getcpu(unsigned *cpu, unsigned *node, void *unused); */
+ENTRY(__vdso_getcpu)
+ .cfi_startproc
+ /* For now, just do the syscall. */
+ li a7, __NR_getcpu
+ ecall
+ ret
+ .cfi_endproc
+ENDPROC(__vdso_getcpu)
diff --git a/arch/riscv/kernel/vdso/gettimeofday.S b/arch/riscv/kernel/vdso/gettimeofday.S
new file mode 100644
index 000000000000..da85d33e8990
--- /dev/null
+++ b/arch/riscv/kernel/vdso/gettimeofday.S
@@ -0,0 +1,26 @@
+/*
+ * 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
+ * as published by the Free Software Foundation, version 2.
+ *
+ * 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.
+ */
+
+#include <linux/linkage.h>
+#include <asm/unistd.h>
+
+ .text
+/* int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz); */
+ENTRY(__vdso_gettimeofday)
+ .cfi_startproc
+ /* For now, just do the syscall. */
+ li a7, __NR_gettimeofday
+ ecall
+ ret
+ .cfi_endproc
+ENDPROC(__vdso_gettimeofday)
diff --git a/arch/riscv/kernel/vdso/vdso.lds.S b/arch/riscv/kernel/vdso/vdso.lds.S
index 8c9dce95c11d..cd1d47e0724b 100644
--- a/arch/riscv/kernel/vdso/vdso.lds.S
+++ b/arch/riscv/kernel/vdso/vdso.lds.S
@@ -70,8 +70,11 @@ VERSION
LINUX_4.15 {
global:
__vdso_rt_sigreturn;
- __vdso_cmpxchg32;
- __vdso_cmpxchg64;
+ __vdso_gettimeofday;
+ __vdso_clock_gettime;
+ __vdso_clock_getres;
+ __vdso_getcpu;
+ __vdso_flush_icache;
local: *;
};
}
diff --git a/arch/riscv/mm/Makefile b/arch/riscv/mm/Makefile
index 81f7d9ce6d88..eb22ab49b3e0 100644
--- a/arch/riscv/mm/Makefile
+++ b/arch/riscv/mm/Makefile
@@ -2,3 +2,4 @@ obj-y += init.o
obj-y += fault.o
obj-y += extable.o
obj-y += ioremap.o
+obj-y += cacheflush.o
diff --git a/arch/riscv/mm/cacheflush.c b/arch/riscv/mm/cacheflush.c
new file mode 100644
index 000000000000..498c0a0814fe
--- /dev/null
+++ b/arch/riscv/mm/cacheflush.c
@@ -0,0 +1,23 @@
+/*
+ * 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
+ * as published by the Free Software Foundation, version 2.
+ *
+ * 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.
+ */
+
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+
+void flush_icache_pte(pte_t pte)
+{
+ struct page *page = pte_page(pte);
+
+ if (!test_and_set_bit(PG_dcache_clean, &page->flags))
+ flush_icache_all();
+}
diff --git a/include/lib/libgcc.h b/include/linux/libgcc.h
index 32e1e0f4b2d0..32e1e0f4b2d0 100644
--- a/include/lib/libgcc.h
+++ b/include/linux/libgcc.h
diff --git a/lib/ashldi3.c b/lib/ashldi3.c
index 1b6087db95a5..3ffc46e3bb6c 100644
--- a/lib/ashldi3.c
+++ b/lib/ashldi3.c
@@ -16,7 +16,7 @@
#include <linux/export.h>
-#include <lib/libgcc.h>
+#include <linux/libgcc.h>
long long notrace __ashldi3(long long u, word_type b)
{
diff --git a/lib/ashrdi3.c b/lib/ashrdi3.c
index 2e67c97ac65a..ea054550f0e8 100644
--- a/lib/ashrdi3.c
+++ b/lib/ashrdi3.c
@@ -16,7 +16,7 @@
#include <linux/export.h>
-#include <lib/libgcc.h>
+#include <linux/libgcc.h>
long long notrace __ashrdi3(long long u, word_type b)
{
diff --git a/lib/cmpdi2.c b/lib/cmpdi2.c
index 6d7ebf6c2b86..2250da7e503e 100644
--- a/lib/cmpdi2.c
+++ b/lib/cmpdi2.c
@@ -16,7 +16,7 @@
#include <linux/export.h>
-#include <lib/libgcc.h>
+#include <linux/libgcc.h>
word_type notrace __cmpdi2(long long a, long long b)
{
diff --git a/lib/lshrdi3.c b/lib/lshrdi3.c
index 8e845f4bb65f..99cfa5721f2d 100644
--- a/lib/lshrdi3.c
+++ b/lib/lshrdi3.c
@@ -17,7 +17,7 @@
*/
#include <linux/module.h>
-#include <lib/libgcc.h>
+#include <linux/libgcc.h>
long long notrace __lshrdi3(long long u, word_type b)
{
diff --git a/lib/muldi3.c b/lib/muldi3.c
index 88938543e10a..54c8b3123376 100644
--- a/lib/muldi3.c
+++ b/lib/muldi3.c
@@ -15,7 +15,7 @@
*/
#include <linux/export.h>
-#include <lib/libgcc.h>
+#include <linux/libgcc.h>
#define W_TYPE_SIZE 32
diff --git a/lib/ucmpdi2.c b/lib/ucmpdi2.c
index 49a53505c8e3..25ca2d4c1e19 100644
--- a/lib/ucmpdi2.c
+++ b/lib/ucmpdi2.c
@@ -15,7 +15,7 @@
*/
#include <linux/module.h>
-#include <lib/libgcc.h>
+#include <linux/libgcc.h>
word_type __ucmpdi2(unsigned long long a, unsigned long long b)
{