target/arm: Split out aa64_va_parameter_tbi, aa64_va_parameter_tbid

For the purpose of rebuild_hflags_a64, we do not need to compute
all of the va parameters, only tbi. Moreover, we can compute them
in a form that is more useful to storing in hflags.

This eliminates the need for aa64_va_parameter_both, so fold that
in to aa64_va_parameter. The remaining calls to aa64_va_parameter
are in get_phys_addr_lpae and in pauth_helper.c.

This reduces the total cpu consumption of aa64_va_parameter in a
kernel boot plus a kvm guest kernel boot from 3% to 0.5%.

Backports commit b830a5ee82e66f54697dcc6450fe9239b7412d13 from qemu

qemu/target/arm/helper.c

@@ -10089,12 +10089,34 @@ static uint8_t convert_stage2_attrs(CPUARMState *env, uint8_t s2attrs)
 }
 #endif /* !CONFIG_USER_ONLY */
 
-ARMVAParameters aa64_va_parameters_both(CPUARMState *env, uint64_t va,
+static int aa64_va_parameter_tbi(uint64_t tcr, ARMMMUIdx mmu_idx)
-                                        ARMMMUIdx mmu_idx)
+{
+    if (regime_has_2_ranges(mmu_idx)) {
+        return extract64(tcr, 37, 2);
+    } else if (mmu_idx == ARMMMUIdx_Stage2) {
+        return 0; /* VTCR_EL2 */
+    } else {
+        return extract32(tcr, 20, 1);
+    }
+}
+
+static int aa64_va_parameter_tbid(uint64_t tcr, ARMMMUIdx mmu_idx)
+{
+    if (regime_has_2_ranges(mmu_idx)) {
+        return extract64(tcr, 51, 2);
+    } else if (mmu_idx == ARMMMUIdx_Stage2) {
+        return 0; /* VTCR_EL2 */
+    } else {
+        return extract32(tcr, 29, 1);
+    }
+}
+
+ARMVAParameters aa64_va_parameters(CPUARMState *env, uint64_t va,
+                                   ARMMMUIdx mmu_idx, bool data)
 {
     uint64_t tcr = regime_tcr(env, mmu_idx)->raw_tcr;
-    bool tbi, tbid, epd, hpd, using16k, using64k;
+    bool epd, hpd, using16k, using64k;
-    int select, tsz;
+    int select, tsz, tbi;
 
     if (!regime_has_2_ranges(mmu_idx)) {
         select = 0;
@@ -10103,11 +10125,9 @@ ARMVAParameters aa64_va_parameters_both(CPUARMState *env, uint64_t va,
         using16k = extract32(tcr, 15, 1);
         if (mmu_idx == ARMMMUIdx_Stage2) {
             /* VTCR_EL2 */
-            tbi = tbid = hpd = false;
+            hpd = false;
         } else {
-            tbi = extract32(tcr, 20, 1);
             hpd = extract32(tcr, 24, 1);
-            tbid = extract32(tcr, 29, 1);
         }
         epd = false;
     } else {
@@ -10121,28 +10141,30 @@ ARMVAParameters aa64_va_parameters_both(CPUARMState *env, uint64_t va,
             epd = extract32(tcr, 7, 1);
             using64k = extract32(tcr, 14, 1);
             using16k = extract32(tcr, 15, 1);
-            tbi = extract64(tcr, 37, 1);
             hpd = extract64(tcr, 41, 1);
-            tbid = extract64(tcr, 51, 1);
         } else {
             int tg = extract32(tcr, 30, 2);
             using16k = tg == 1;
             using64k = tg == 3;
             tsz = extract32(tcr, 16, 6);
             epd = extract32(tcr, 23, 1);
-            tbi = extract64(tcr, 38, 1);
             hpd = extract64(tcr, 42, 1);
-            tbid = extract64(tcr, 52, 1);
         }
     }
     tsz = MIN(tsz, 39);  /* TODO: ARMv8.4-TTST */
     tsz = MAX(tsz, 16);  /* TODO: ARMv8.2-LVA  */
 
+    /* Present TBI as a composite with TBID.  */
+    tbi = aa64_va_parameter_tbi(tcr, mmu_idx);
+    if (!data) {
+        tbi &= ~aa64_va_parameter_tbid(tcr, mmu_idx);
+    }
+    tbi = (tbi >> select) & 1;
+
     return (ARMVAParameters) {
         .tsz = tsz,
         .select = select,
         .tbi = tbi,
-        .tbid = tbid,
         .epd = epd,
         .hpd = hpd,
         .using16k = using16k,
@@ -10150,16 +10172,6 @@ ARMVAParameters aa64_va_parameters_both(CPUARMState *env, uint64_t va,
     };
 }
 
-ARMVAParameters aa64_va_parameters(CPUARMState *env, uint64_t va,
-                                   ARMMMUIdx mmu_idx, bool data)
-{
-    ARMVAParameters ret = aa64_va_parameters_both(env, va, mmu_idx);
-
-    /* Present TBI as a composite with TBID.  */
-    ret.tbi &= (data || !ret.tbid);
-    return ret;
-}
-
 #ifndef CONFIG_USER_ONLY
 static ARMVAParameters aa32_va_parameters(CPUARMState *env, uint32_t va,
                                           ARMMMUIdx mmu_idx)
@@ -11940,18 +11952,12 @@ void cpu_get_tb_cpu_state(CPUARMState *env, target_ulong *pc,
         ARMMMUIdx stage1 = stage_1_mmu_idx(mmu_idx);
 
         {
-            ARMVAParameters p0 = aa64_va_parameters_both(env, 0, stage1);
+            uint64_t tcr = regime_tcr(env, mmu_idx)->raw_tcr;
             int tbii, tbid;
 
             /* Get control bits for tagged addresses.  */
-            if (regime_has_2_ranges(mmu_idx)) {
+            tbid = aa64_va_parameter_tbi(tcr, mmu_idx);
-                ARMVAParameters p1 = aa64_va_parameters_both(env, -1, stage1);
+            tbii = tbid & ~aa64_va_parameter_tbid(tcr, mmu_idx);
-                tbid = (p1.tbi << 1) | p0.tbi;
-                tbii = tbid & ~((p1.tbid << 1) | p0.tbid);
-            } else {
-                tbid = p0.tbi;
-                tbii = tbid & !p0.tbid;
-            }
 
             flags = FIELD_DP32(flags, TBFLAG_A64, TBII, tbii);
             flags = FIELD_DP32(flags, TBFLAG_A64, TBID, tbid);

qemu/target/arm/internals.h

@@ -1129,15 +1129,12 @@ typedef struct ARMVAParameters {
     unsigned tsz    : 8;
     unsigned select : 1;
     bool tbi        : 1;
-    bool tbid       : 1;
     bool epd        : 1;
     bool hpd        : 1;
     bool using16k   : 1;
     bool using64k   : 1;
 } ARMVAParameters;
 
-ARMVAParameters aa64_va_parameters_both(CPUARMState *env, uint64_t va,
-                                        ARMMMUIdx mmu_idx);
 ARMVAParameters aa64_va_parameters(CPUARMState *env, uint64_t va,
                                    ARMMMUIdx mmu_idx, bool data);