powerpc/mm: Trace tlbie(l) instructions

Add a trace point for tlbie(l) (Translation Lookaside Buffer Invalidate
Entry (Local)) instructions.

The tlbie instruction has changed over the years, so not all versions
accept the same operands. Use the ISA v3 field operands because they are
the most verbose, we may change them in future.

Example output:

  qemu-system-ppc-5371  [016]  1412.369519: tlbie:
  	tlbie with lpid 0, local 1, rb=67bd8900174c11c1, rs=0, ric=0 prs=0 r=0
Signed-off-by: Balbir Singh <bsingharora@gmail.com>
[mpe: Add some missing trace_tlbie()s, reword change log]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

arch/powerpc/include/asm/trace.h

@@ -168,6 +168,39 @@ TRACE_EVENT(hash_fault,
 		      __entry->addr, __entry->access, __entry->trap)
 );
 
+
+TRACE_EVENT(tlbie,
+
+	TP_PROTO(unsigned long lpid, unsigned long local, unsigned long rb,
+		unsigned long rs, unsigned long ric, unsigned long prs,
+		unsigned long r),
+	TP_ARGS(lpid, local, rb, rs, ric, prs, r),
+	TP_STRUCT__entry(
+		__field(unsigned long, lpid)
+		__field(unsigned long, local)
+		__field(unsigned long, rb)
+		__field(unsigned long, rs)
+		__field(unsigned long, ric)
+		__field(unsigned long, prs)
+		__field(unsigned long, r)
+		),
+
+	TP_fast_assign(
+		__entry->lpid = lpid;
+		__entry->local = local;
+		__entry->rb = rb;
+		__entry->rs = rs;
+		__entry->ric = ric;
+		__entry->prs = prs;
+		__entry->r = r;
+		),
+
+	TP_printk("lpid=%ld, local=%ld, rb=0x%lx, rs=0x%lx, ric=0x%lx, "
+		"prs=0x%lx, r=0x%lx", __entry->lpid, __entry->local,
+		__entry->rb, __entry->rs, __entry->ric, __entry->prs,
+		__entry->r)
+);
+
 #endif /* _TRACE_POWERPC_H */
 
 #undef TRACE_INCLUDE_PATH

arch/powerpc/kvm/book3s_hv_rm_mmu.c

@@ -15,6 +15,7 @@
 #include <linux/log2.h>
 
 #include <asm/tlbflush.h>
+#include <asm/trace.h>
 #include <asm/kvm_ppc.h>
 #include <asm/kvm_book3s.h>
 #include <asm/book3s/64/mmu-hash.h>
@@ -443,17 +444,23 @@ static void do_tlbies(struct kvm *kvm, unsigned long *rbvalues,
 			cpu_relax();
 		if (need_sync)
 			asm volatile("ptesync" : : : "memory");
-		for (i = 0; i < npages; ++i)
+		for (i = 0; i < npages; ++i) {
 			asm volatile(PPC_TLBIE_5(%0,%1,0,0,0) : :
 				     "r" (rbvalues[i]), "r" (kvm->arch.lpid));
+			trace_tlbie(kvm->arch.lpid, 0, rbvalues[i],
+				kvm->arch.lpid, 0, 0, 0);
+		}
 		asm volatile("eieio; tlbsync; ptesync" : : : "memory");
 		kvm->arch.tlbie_lock = 0;
 	} else {
 		if (need_sync)
 			asm volatile("ptesync" : : : "memory");
-		for (i = 0; i < npages; ++i)
+		for (i = 0; i < npages; ++i) {
 			asm volatile(PPC_TLBIEL(%0,%1,0,0,0) : :
 				     "r" (rbvalues[i]), "r" (0));
+			trace_tlbie(kvm->arch.lpid, 1, rbvalues[i],
+				0, 0, 0, 0);
+		}
 		asm volatile("ptesync" : : : "memory");
 	}
 }

arch/powerpc/mm/hash_native_64.c

@@ -23,6 +23,7 @@
 #include <asm/mmu_context.h>
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
+#include <asm/trace.h>
 #include <asm/tlb.h>
 #include <asm/cputable.h>
 #include <asm/udbg.h>
@@ -98,6 +99,7 @@ static inline void __tlbie(unsigned long vpn, int psize, int apsize, int ssize)
 			     : "memory");
 		break;
 	}
+	trace_tlbie(0, 0, va, 0, 0, 0, 0);
 }
 
 static inline void __tlbiel(unsigned long vpn, int psize, int apsize, int ssize)
@@ -147,6 +149,7 @@ static inline void __tlbiel(unsigned long vpn, int psize, int apsize, int ssize)
 			     : "memory");
 		break;
 	}
+	trace_tlbie(0, 1, va, 0, 0, 0, 0);
 
 }
 

arch/powerpc/mm/hash_utils_64.c

@@ -810,6 +810,8 @@ static void update_hid_for_hash(void)
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(0), "i"(0), "i"(2), "r"(0) : "memory");
 	asm volatile("eieio; tlbsync; ptesync; isync; slbia": : :"memory");
+	trace_tlbie(0, 0, rb, 0, 2, 0, 0);
+
 	/*
 	 * now switch the HID
 	 */

arch/powerpc/mm/pgtable-radix.c

@@ -20,6 +20,7 @@
 #include <asm/firmware.h>
 #include <asm/powernv.h>
 #include <asm/sections.h>
+#include <asm/trace.h>
 
 #include <trace/events/thp.h>
 
@@ -198,6 +199,7 @@ static void __init radix_init_pgtable(void)
 	asm volatile(PPC_TLBIE_5(%0,%1,2,1,1) : :
 		     "r" (TLBIEL_INVAL_SET_LPID), "r" (0));
 	asm volatile("eieio; tlbsync; ptesync" : : : "memory");
+	trace_tlbie(0, 0, TLBIEL_INVAL_SET_LPID, 0, 2, 1, 1);
 }
 
 static void __init radix_init_partition_table(void)
@@ -324,6 +326,9 @@ static void update_hid_for_radix(void)
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(1), "i"(1), "i"(2), "r"(0) : "memory");
 	asm volatile("eieio; tlbsync; ptesync; isync; slbia": : :"memory");
+	trace_tlbie(0, 0, rb, 0, 2, 0, 1);
+	trace_tlbie(0, 0, rb, 0, 2, 1, 1);
+
 	/*
 	 * now switch the HID
 	 */

arch/powerpc/mm/pgtable_64.c

@@ -47,6 +47,7 @@
 #include <asm/smp.h>
 #include <asm/machdep.h>
 #include <asm/tlb.h>
+#include <asm/trace.h>
 #include <asm/processor.h>
 #include <asm/cputable.h>
 #include <asm/sections.h>
@@ -477,12 +478,15 @@ void mmu_partition_table_set_entry(unsigned int lpid, unsigned long dw0,
 	 * use of this partition ID was, not the new use.
 	 */
 	asm volatile("ptesync" : : : "memory");
-	if (old & PATB_HR)
+	if (old & PATB_HR) {
 		asm volatile(PPC_TLBIE_5(%0,%1,2,0,1) : :
 			     "r" (TLBIEL_INVAL_SET_LPID), "r" (lpid));
-	else
+		trace_tlbie(lpid, 0, TLBIEL_INVAL_SET_LPID, lpid, 2, 0, 1);
+	} else {
 		asm volatile(PPC_TLBIE_5(%0,%1,2,0,0) : :
 			     "r" (TLBIEL_INVAL_SET_LPID), "r" (lpid));
+		trace_tlbie(lpid, 0, TLBIEL_INVAL_SET_LPID, lpid, 2, 0, 0);
+	}
 	asm volatile("eieio; tlbsync; ptesync" : : : "memory");
 }
 EXPORT_SYMBOL_GPL(mmu_partition_table_set_entry);

arch/powerpc/mm/tlb-radix.c

@@ -16,6 +16,7 @@
 
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
+#include <asm/trace.h>
 
 
 #define RIC_FLUSH_TLB 0
@@ -35,6 +36,7 @@ static inline void __tlbiel_pid(unsigned long pid, int set,
 
 	asm volatile(PPC_TLBIEL(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
+	trace_tlbie(0, 1, rb, rs, ric, prs, r);
 }
 
 /*
@@ -87,6 +89,7 @@ static inline void _tlbie_pid(unsigned long pid, unsigned long ric)
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
+	trace_tlbie(0, 0, rb, rs, ric, prs, r);
 }
 
 static inline void _tlbiel_va(unsigned long va, unsigned long pid,
@@ -104,6 +107,7 @@ static inline void _tlbiel_va(unsigned long va, unsigned long pid,
 	asm volatile(PPC_TLBIEL(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
 	asm volatile("ptesync": : :"memory");
+	trace_tlbie(0, 1, rb, rs, ric, prs, r);
 }
 
 static inline void _tlbie_va(unsigned long va, unsigned long pid,
@@ -121,6 +125,7 @@ static inline void _tlbie_va(unsigned long va, unsigned long pid,
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
+	trace_tlbie(0, 0, rb, rs, ric, prs, r);
 }
 
 /*
@@ -377,6 +382,7 @@ void radix__flush_tlb_lpid_va(unsigned long lpid, unsigned long gpa,
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
+	trace_tlbie(lpid, 0, rb, rs, ric, prs, r);
 }
 EXPORT_SYMBOL(radix__flush_tlb_lpid_va);
 
@@ -394,6 +400,7 @@ void radix__flush_tlb_lpid(unsigned long lpid)
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
+	trace_tlbie(lpid, 0, rb, rs, ric, prs, r);
 }
 EXPORT_SYMBOL(radix__flush_tlb_lpid);
 
@@ -420,12 +427,14 @@ void radix__flush_tlb_all(void)
 	 */
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(1), "i"(ric), "r"(rs) : "memory");
+	trace_tlbie(0, 0, rb, rs, ric, prs, r);
 	/*
 	 * now flush host entires by passing PRS = 0 and LPID == 0
 	 */
 	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
 		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(0) : "memory");
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
+	trace_tlbie(0, 0, rb, 0, ric, prs, r);
 }
 
 void radix__flush_tlb_pte_p9_dd1(unsigned long old_pte, struct mm_struct *mm,