powerpc/book3s: Flush SLB/TLBs if we get SLB/TLB machine check errors on power8.

This patch handles the memory errors on power8. If we get a machine check
exception due to SLB or TLB errors, then flush SLBs/TLBs and reload SLBs to
recover.

Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

3 files changed, 41 insertions, 0 deletions

diff --git a/arch/powerpc/include/asm/mce.h b/arch/powerpc/include/asm/mce.h
index 8157d4eaead6..e3ffa825b970 100644
--- a/arch/powerpc/include/asm/mce.h
+++ b/arch/powerpc/include/asm/mce.h
@@ -64,4 +64,7 @@
                                          P7_DSISR_MC_SLB_MULTIHIT | \
                                          P7_DSISR_MC_SLB_MULTIHIT_PARITY)
 
+#define P8_DSISR_MC_SLB_ERRORS          (P7_DSISR_MC_SLB_ERRORS | \
+                                         P8_DSISR_MC_ERAT_MULTIHIT_SEC)
+
 #endif /* __ASM_PPC64_MCE_H__ */
diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index c54188bcdd9e..6c8dd5da4de5 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -74,6 +74,7 @@ extern void __restore_cpu_a2(void);
 extern void __flush_tlb_power7(unsigned long inval_selector);
 extern void __flush_tlb_power8(unsigned long inval_selector);
 extern long __machine_check_early_realmode_p7(struct pt_regs *regs);
+extern long __machine_check_early_realmode_p8(struct pt_regs *regs);
 #endif /* CONFIG_PPC64 */
 #if defined(CONFIG_E500)
 extern void __setup_cpu_e5500(unsigned long offset, struct cpu_spec* spec);
@@ -462,6 +463,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
                 .cpu_setup              = __setup_cpu_power8,
                 .cpu_restore            = __restore_cpu_power8,
                 .flush_tlb              = __flush_tlb_power8,
+                .machine_check_early    = __machine_check_early_realmode_p8,
                 .platform               = "power8",
         },
         {       /* Power7 */
@@ -521,6 +523,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
                 .cpu_setup              = __setup_cpu_power8,
                 .cpu_restore            = __restore_cpu_power8,
                 .flush_tlb              = __flush_tlb_power8,
+                .machine_check_early    = __machine_check_early_realmode_p8,
                 .platform               = "power8",
         },
         {       /* Power8 */
@@ -540,6 +543,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
                 .cpu_setup              = __setup_cpu_power8,
                 .cpu_restore            = __restore_cpu_power8,
                 .flush_tlb              = __flush_tlb_power8,
+                .machine_check_early    = __machine_check_early_realmode_p8,
                 .platform               = "power8",
         },
         {       /* Cell Broadband Engine */
diff --git a/arch/powerpc/kernel/mce_power.c b/arch/powerpc/kernel/mce_power.c
index 690547319b03..60a217f11275 100644
--- a/arch/powerpc/kernel/mce_power.c
+++ b/arch/powerpc/kernel/mce_power.c
@@ -148,3 +148,37 @@ long __machine_check_early_realmode_p7(struct pt_regs *regs)
         /* TODO: Decode machine check reason. */
         return handled;
 }
+
+static long mce_handle_ierror_p8(uint64_t srr1)
+{
+        long handled = 0;
+
+        handled = mce_handle_common_ierror(srr1);
+
+        if (P7_SRR1_MC_IFETCH(srr1) == P8_SRR1_MC_IFETCH_ERAT_MULTIHIT) {
+                flush_and_reload_slb();
+                handled = 1;
+        }
+        return handled;
+}
+
+static long mce_handle_derror_p8(uint64_t dsisr)
+{
+        return mce_handle_derror(dsisr, P8_DSISR_MC_SLB_ERRORS);
+}
+
+long __machine_check_early_realmode_p8(struct pt_regs *regs)
+{
+        uint64_t srr1;
+        long handled = 1;
+
+        srr1 = regs->msr;
+
+        if (P7_SRR1_MC_LOADSTORE(srr1))
+                handled = mce_handle_derror_p8(regs->dsisr);
+        else
+                handled = mce_handle_ierror_p8(srr1);
+
+        /* TODO: Decode machine check reason. */
+        return handled;
+}