Merge tag 'pull-tcg-20230701' of https://gitlab.com/rth7680/qemu into…

… staging

dbus: Two hot fixes, per request of Marc-André Lureau
accel/tcg: Fix tb_invalidate_phys_range iteration
fpu: Add float64_to_int{32,64}_modulo
tcg: Reduce scope of tcg_assert_listed_vecop
target/nios2: Explicitly ask for target-endian loads
linux-user: Avoid mmap of the last byte of the reserved_va

# -----BEGIN PGP SIGNATURE-----
#
# iQFRBAABCgA7FiEEekgeeIaLTbaoWgXAZN846K9+IV8FAmSfzXwdHHJpY2hhcmQu
# aGVuZGVyc29uQGxpbmFyby5vcmcACgkQZN846K9+IV+GMAgAicMA7dZEUNiKT1co
# pwQNF/aQehs3a+UYcHFZRQWjwNsXzDrPRTAyBkDFrzR2ILxKlpPw2JBRiqrr9pqj
# YWit0pHVv/OAYfSEzcqUaIeWyAh2xlAT4IbSz+sLcPBdPgUwm3z0Y7mTz3kUAkB2
# gXO/iuoD8ORwgSnFvH+FSws16kr1x/8cAaObY7BupUhS7hK8M9zsCehhk6ssxv7+
# EpR0kDIeoC2kjJLvQAoGW4DPzfmAvVmI/OiJKpqrAlTJIeAkngalSuaxj/t9Dte6
# zy4h8JW5VbHw3qLxTvg42/Pk4AiweBh38hpUfLQ2cprO7dy+T9qS2v8CGnMzrmeB
# kzlIMg==
# =a7vA
# -----END PGP SIGNATURE-----
# gpg: Signature made Sat 01 Jul 2023 08:53:48 AM CEST
# gpg:                using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F
# gpg:                issuer "richard.henderson@linaro.org"
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [ultimate]

* tag 'pull-tcg-20230701' of https://gitlab.com/rth7680/qemu:
  linux-user: Avoid mmap of the last byte of the reserved_va
  target/nios2 : Explicitly ask for target-endian loads and stores
  tcg: Reduce tcg_assert_listed_vecop() scope
  target/arm: Use float64_to_int32_modulo for FJCVTZS
  target/alpha: Use float64_to_int64_modulo for CVTTQ
  tests/tcg/alpha: Add test for cvttq
  fpu: Add float64_to_int{32,64}_modulo
  accel/tcg: Assert one page in tb_invalidate_phys_page_range__locked
  accel/tcg: Fix start page passed to tb_invalidate_phys_page_range__locked
  audio: dbus requires pixman
  ui/dbus: fix build errors in dbus_update_gl_cb and dbus_call_update_gl

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

accel/tcg/tb-maint.c

@@ -1092,6 +1092,9 @@ tb_invalidate_phys_page_range__locked(struct page_collection *pages,
     TranslationBlock *current_tb = retaddr ? tcg_tb_lookup(retaddr) : NULL;
 #endif /* TARGET_HAS_PRECISE_SMC */
 
+    /* Range may not cross a page. */
+    tcg_debug_assert(((start ^ last) & TARGET_PAGE_MASK) == 0);
+
     /*
      * We remove all the TBs in the range [start, last].
      * XXX: see if in some cases it could be faster to invalidate all the code
@@ -1182,15 +1185,17 @@ void tb_invalidate_phys_range(tb_page_addr_t start, tb_page_addr_t last)
     index_last = last >> TARGET_PAGE_BITS;
     for (index = start >> TARGET_PAGE_BITS; index <= index_last; index++) {
         PageDesc *pd = page_find(index);
-        tb_page_addr_t bound;
+        tb_page_addr_t page_start, page_last;
 
         if (pd == NULL) {
             continue;
         }
         assert_page_locked(pd);
-        bound = (index << TARGET_PAGE_BITS) | ~TARGET_PAGE_MASK;
-        bound = MIN(bound, last);
-        tb_invalidate_phys_page_range__locked(pages, pd, start, bound, 0);
+        page_start = index << TARGET_PAGE_BITS;
+        page_last = page_start | ~TARGET_PAGE_MASK;
+        page_last = MIN(page_last, last);
+        tb_invalidate_phys_page_range__locked(pages, pd,
+                                              page_start, page_last, 0);
     }
     page_collection_unlock(pages);
 }

audio/meson.build

@@ -31,7 +31,7 @@ endforeach
 
 if dbus_display
     module_ss = ss.source_set()
-    module_ss.add(when: gio, if_true: files('dbusaudio.c'))
+    module_ss.add(when: [gio, pixman], if_true: files('dbusaudio.c'))
     audio_modules += {'dbus': module_ss}
 endif
 

fpu/softfloat-parts.c.inc

@@ -1181,6 +1181,84 @@ static uint64_t partsN(float_to_uint)(FloatPartsN *p, FloatRoundMode rmode,
     return r;
 }
 
+/*
+ * Like partsN(float_to_sint), except do not saturate the result.
+ * Instead, return the rounded unbounded precision two's compliment result,
+ * modulo 2**(bitsm1 + 1).
+ */
+static int64_t partsN(float_to_sint_modulo)(FloatPartsN *p,
+                                            FloatRoundMode rmode,
+                                            int bitsm1, float_status *s)
+{
+    int flags = 0;
+    uint64_t r;
+    bool overflow = false;
+
+    switch (p->cls) {
+    case float_class_snan:
+        flags |= float_flag_invalid_snan;
+        /* fall through */
+    case float_class_qnan:
+        flags |= float_flag_invalid;
+        r = 0;
+        break;
+
+    case float_class_inf:
+        overflow = true;
+        r = 0;
+        break;
+
+    case float_class_zero:
+        return 0;
+
+    case float_class_normal:
+        /* TODO: N - 2 is frac_size for rounding; could use input fmt. */
+        if (parts_round_to_int_normal(p, rmode, 0, N - 2)) {
+            flags = float_flag_inexact;
+        }
+
+        if (p->exp <= DECOMPOSED_BINARY_POINT) {
+            /*
+             * Because we rounded to integral, and exp < 64,
+             * we know frac_low is zero.
+             */
+            r = p->frac_hi >> (DECOMPOSED_BINARY_POINT - p->exp);
+            if (p->exp < bitsm1) {
+                /* Result in range. */
+            } else if (p->exp == bitsm1) {
+                /* The only in-range value is INT_MIN. */
+                overflow = !p->sign || p->frac_hi != DECOMPOSED_IMPLICIT_BIT;
+            } else {
+                overflow = true;
+            }
+        } else {
+            /* Overflow, but there might still be bits to return. */
+            int shl = p->exp - DECOMPOSED_BINARY_POINT;
+            if (shl < N) {
+                frac_shl(p, shl);
+                r = p->frac_hi;
+            } else {
+                r = 0;
+            }
+            overflow = true;
+        }
+
+        if (p->sign) {
+            r = -r;
+        }
+        break;
+
+    default:
+        g_assert_not_reached();
+    }
+
+    if (overflow) {
+        flags = float_flag_invalid | float_flag_invalid_cvti;
+    }
+    float_raise(flags, s);
+    return r;
+}
+
 /*
  * Integer to float conversions
  *

fpu/softfloat.c

@@ -852,11 +852,24 @@ static uint64_t parts128_float_to_uint(FloatParts128 *p, FloatRoundMode rmode,
 #define parts_float_to_uint(P, R, Z, M, S) \
     PARTS_GENERIC_64_128(float_to_uint, P)(P, R, Z, M, S)
 
+static int64_t parts64_float_to_sint_modulo(FloatParts64 *p,
+                                            FloatRoundMode rmode,
+                                            int bitsm1, float_status *s);
+static int64_t parts128_float_to_sint_modulo(FloatParts128 *p,
+                                             FloatRoundMode rmode,
+                                             int bitsm1, float_status *s);
+
+#define parts_float_to_sint_modulo(P, R, M, S) \
+    PARTS_GENERIC_64_128(float_to_sint_modulo, P)(P, R, M, S)
+
 static void parts64_sint_to_float(FloatParts64 *p, int64_t a,
                                   int scale, float_status *s);
 static void parts128_sint_to_float(FloatParts128 *p, int64_t a,
                                    int scale, float_status *s);
 
+#define parts_float_to_sint(P, R, Z, MN, MX, S) \
+    PARTS_GENERIC_64_128(float_to_sint, P)(P, R, Z, MN, MX, S)
+
 #define parts_sint_to_float(P, I, Z, S) \
     PARTS_GENERIC_64_128(sint_to_float, P)(P, I, Z, S)
 
@@ -3409,6 +3422,24 @@ int64_t bfloat16_to_int64_round_to_zero(bfloat16 a, float_status *s)
     return bfloat16_to_int64_scalbn(a, float_round_to_zero, 0, s);
 }
 
+int32_t float64_to_int32_modulo(float64 a, FloatRoundMode rmode,
+                                float_status *s)
+{
+    FloatParts64 p;
+
+    float64_unpack_canonical(&p, a, s);
+    return parts_float_to_sint_modulo(&p, rmode, 31, s);
+}
+
+int64_t float64_to_int64_modulo(float64 a, FloatRoundMode rmode,
+                                float_status *s)
+{
+    FloatParts64 p;
+
+    float64_unpack_canonical(&p, a, s);
+    return parts_float_to_sint_modulo(&p, rmode, 63, s);
+}
+
 /*
  * Floating-point to unsigned integer conversions
  */

include/fpu/softfloat.h

@@ -751,6 +751,9 @@ int16_t float64_to_int16_round_to_zero(float64, float_status *status);
 int32_t float64_to_int32_round_to_zero(float64, float_status *status);
 int64_t float64_to_int64_round_to_zero(float64, float_status *status);
 
+int32_t float64_to_int32_modulo(float64, FloatRoundMode, float_status *status);
+int64_t float64_to_int64_modulo(float64, FloatRoundMode, float_status *status);
+
 uint16_t float64_to_uint16_scalbn(float64, FloatRoundMode, int, float_status *);
 uint32_t float64_to_uint32_scalbn(float64, FloatRoundMode, int, float_status *);
 uint64_t float64_to_uint64_scalbn(float64, FloatRoundMode, int, float_status *);

include/tcg/tcg.h

@@ -1135,12 +1135,6 @@ uint64_t dup_const(unsigned vece, uint64_t c);
         : (qemu_build_not_reached_always(), 0))                    \
      : dup_const(VECE, C))
 
-#ifdef CONFIG_DEBUG_TCG
-void tcg_assert_listed_vecop(TCGOpcode);
-#else
-static inline void tcg_assert_listed_vecop(TCGOpcode op) { }
-#endif
-
 static inline const TCGOpcode *tcg_swap_vecop_list(const TCGOpcode *n)
 {
 #ifdef CONFIG_DEBUG_TCG

linux-user/mmap.c

@@ -281,9 +281,15 @@ static abi_ulong mmap_find_vma_reserved(abi_ulong start, abi_ulong size,
     /* Note that start and size have already been aligned by mmap_find_vma. */
 
     end_addr = start + size;
+    /*
+     * Start at the top of the address space, ignoring the last page.
+     * If reserved_va == UINT32_MAX, then end_addr wraps to 0,
+     * throwing the rest of the calculations off.
+     * TODO: rewrite using last_addr instead.
+     * TODO: use the interval tree instead of probing every page.
+     */
     if (start > reserved_va - size) {
-        /* Start at the top of the address space.  */
-        end_addr = ((reserved_va + 1 - size) & -align) + size;
+        end_addr = ((reserved_va - size) & -align) + size;
         looped = true;
     }
 
@@ -296,8 +302,8 @@ static abi_ulong mmap_find_vma_reserved(abi_ulong start, abi_ulong size,
                 /* Failure.  The entire address space has been searched.  */
                 return (abi_ulong)-1;
             }
-            /* Re-start at the top of the address space.  */
-            addr = end_addr = ((reserved_va + 1 - size) & -align) + size;
+            /* Re-start at the top of the address space (see above). */
+            addr = end_addr = ((reserved_va - size) & -align) + size;
             looped = true;
         } else {
             prot = page_get_flags(addr);

target/alpha/fpu_helper.c

@@ -453,78 +453,29 @@ uint64_t helper_cvtqs(CPUAlphaState *env, uint64_t a)
 
 static uint64_t do_cvttq(CPUAlphaState *env, uint64_t a, int roundmode)
 {
-    uint64_t frac, ret = 0;
-    uint32_t exp, sign, exc = 0;
-    int shift;
+    float64 fa;
+    int64_t ret;
+    uint32_t exc;
 
-    sign = (a >> 63);
-    exp = (uint32_t)(a >> 52) & 0x7ff;
-    frac = a & 0xfffffffffffffull;
+    fa = t_to_float64(a);
+    ret = float64_to_int64_modulo(fa, roundmode, &FP_STATUS);
 
-    if (exp == 0) {
-        if (unlikely(frac != 0) && !env->fp_status.flush_inputs_to_zero) {
-            goto do_underflow;
-        }
-    } else if (exp == 0x7ff) {
-        exc = FPCR_INV;
-    } else {
-        /* Restore implicit bit.  */
-        frac |= 0x10000000000000ull;
-
-        shift = exp - 1023 - 52;
-        if (shift >= 0) {
-            /* In this case the number is so large that we must shift
-               the fraction left.  There is no rounding to do.  */
-            if (shift < 64) {
-                ret = frac << shift;
-            }
-            /* Check for overflow.  Note the special case of -0x1p63.  */
-            if (shift >= 11 && a != 0xC3E0000000000000ull) {
-                exc = FPCR_IOV | FPCR_INE;
-            }
-        } else {
-            uint64_t round;
-
-            /* In this case the number is smaller than the fraction as
-               represented by the 52 bit number.  Here we must think
-               about rounding the result.  Handle this by shifting the
-               fractional part of the number into the high bits of ROUND.
-               This will let us efficiently handle round-to-nearest.  */
-            shift = -shift;
-            if (shift < 63) {
-                ret = frac >> shift;
-                round = frac << (64 - shift);
-            } else {
-                /* The exponent is so small we shift out everything.
-                   Leave a sticky bit for proper rounding below.  */
-            do_underflow:
-                round = 1;
-            }
+    exc = get_float_exception_flags(&FP_STATUS);
+    if (unlikely(exc)) {
+        set_float_exception_flags(0, &FP_STATUS);
 
-            if (round) {
-                exc = FPCR_INE;
-                switch (roundmode) {
-                case float_round_nearest_even:
-                    if (round == (1ull << 63)) {
-                        /* Fraction is exactly 0.5; round to even.  */
-                        ret += (ret & 1);
-                    } else if (round > (1ull << 63)) {
-                        ret += 1;
-                    }
-                    break;
-                case float_round_to_zero:
-                    break;
-                case float_round_up:
-                    ret += 1 - sign;
-                    break;
-                case float_round_down:
-                    ret += sign;
-                    break;
-                }
+        /* We need to massage the resulting exceptions. */
+        if (exc & float_flag_invalid_cvti) {
+            /* Overflow, either normal or infinity. */
+            if (float64_is_infinity(fa)) {
+                exc = FPCR_INV;
+            } else {
+                exc = FPCR_IOV | FPCR_INE;
             }
-        }
-        if (sign) {
-            ret = -ret;
+        } else if (exc & float_flag_invalid) {
+            exc = FPCR_INV;
+        } else if (exc & float_flag_inexact) {
+            exc = FPCR_INE;
         }
     }
     env->error_code = exc;