arm64: mm: Remove arch_apei_flush_tlb_one()
authorJames Morse <james.morse@arm.com>
Mon, 6 Nov 2017 18:44:26 +0000 (18:44 +0000)
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>
Tue, 7 Nov 2017 11:13:33 +0000 (12:13 +0100)
Nothing calls arch_apei_flush_tlb_one() anymore, instead relying on
__set_fixmap() to do the invalidation. Remove it.

Move the IPI-considered-harmful comment to __set_fixmap().

Signed-off-by: James Morse <james.morse@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Tested-by: Tyler Baicar <tbaicar@codeaurora.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: All applicable <stable@vger.kernel.org>
arch/arm64/include/asm/acpi.h
arch/arm64/mm/mmu.c

index 59cca1d6ec547270adbd56a4e2265b9f9fc34375..32f465a80e4e86d5b13fd2d92925b16cceb0bf07 100644 (file)
@@ -126,18 +126,6 @@ static inline const char *acpi_get_enable_method(int cpu)
  */
 #define acpi_disable_cmcff 1
 pgprot_t arch_apei_get_mem_attribute(phys_addr_t addr);
-
-/*
- * Despite its name, this function must still broadcast the TLB
- * invalidation in order to ensure other CPUs don't end up with junk
- * entries as a result of speculation. Unusually, its also called in
- * IRQ context (ghes_iounmap_irq) so if we ever need to use IPIs for
- * TLB broadcasting, then we're in trouble here.
- */
-static inline void arch_apei_flush_tlb_one(unsigned long addr)
-{
-       flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
-}
 #endif /* CONFIG_ACPI_APEI */
 
 #ifdef CONFIG_ACPI_NUMA
index f1eb15e0e8642d2a74c7d19d18e8e8d3ae5a7062..267d2b79d52d6e3918a18a2c590d797bdec6d3cb 100644 (file)
@@ -778,6 +778,10 @@ void __init early_fixmap_init(void)
        }
 }
 
+/*
+ * Unusually, this is also called in IRQ context (ghes_iounmap_irq) so if we
+ * ever need to use IPIs for TLB broadcasting, then we're in trouble here.
+ */
 void __set_fixmap(enum fixed_addresses idx,
                               phys_addr_t phys, pgprot_t flags)
 {