Introduce TCGOpcode for memory barrier

This commit introduces the TCGOpcode for memory barrier instruction.

This opcode takes an argument which is the type of memory barrier
which should be generated.

Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Message-Id: <20160714202026.9727-2-bobby.prani@gmail.com>
Signed-off-by: Richard Henderson <rth@twiddle.net>

5 files changed, 55 insertions, 0 deletions

diff --git a/tcg/README b/tcg/README
index ce8bebab37..1d48aa963f 100644
--- a/tcg/README
+++ b/tcg/README
@@ -402,6 +402,23 @@ double-word product T0.  The later is returned in two single-word outputs.
 
 Similar to mulu2, except the two inputs T1 and T2 are signed.
 
+********* Memory Barrier support
+
+* mb <$arg>
+
+Generate a target memory barrier instruction to ensure memory ordering as being
+enforced by a corresponding guest memory barrier instruction. The ordering
+enforced by the backend may be stricter than the ordering required by the guest.
+It cannot be weaker. This opcode takes a constant argument which is required to
+generate the appropriate barrier instruction. The backend should take care to
+emit the target barrier instruction only when necessary i.e., for SMP guests and
+when MTTCG is enabled.
+
+The guest translators should generate this opcode for all guest instructions
+which have ordering side effects.
+
+Please see docs/atomics.txt for more information on memory barriers.
+
 ********* 64-bit guest on 32-bit host support
 
 The following opcodes are internal to TCG.  Thus they are to be implemented by
diff --git a/tcg/tcg-op.c b/tcg/tcg-op.c
index 0243c99094..291d50bb7d 100644
--- a/tcg/tcg-op.c
+++ b/tcg/tcg-op.c
@@ -148,6 +148,23 @@ void tcg_gen_op6(TCGContext *ctx, TCGOpcode opc, TCGArg a1, TCGArg a2,
     tcg_emit_op(ctx, opc, pi);
 }
 
+void tcg_gen_mb(TCGBar mb_type)
+{
+    bool emit_barriers = true;
+
+#ifndef CONFIG_USER_ONLY
+    /* TODO: When MTTCG is available for system mode, we will check
+     * the following condition and enable emit_barriers
+     * (qemu_tcg_mttcg_enabled() && smp_cpus > 1)
+     */
+    emit_barriers = false;
+#endif
+
+    if (emit_barriers) {
+        tcg_gen_op1(&tcg_ctx, INDEX_op_mb, mb_type);
+    }
+}
+
 /* 32 bit ops */
 
 void tcg_gen_addi_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2)
diff --git a/tcg/tcg-op.h b/tcg/tcg-op.h
index f217e80747..02cb376681 100644
--- a/tcg/tcg-op.h
+++ b/tcg/tcg-op.h
@@ -261,6 +261,8 @@ static inline void tcg_gen_br(TCGLabel *l)
     tcg_gen_op1(&tcg_ctx, INDEX_op_br, label_arg(l));
 }
 
+void tcg_gen_mb(TCGBar);
+
 /* Helper calls. */
 
 /* 32 bit ops */
diff --git a/tcg/tcg-opc.h b/tcg/tcg-opc.h
index 6d0410c4b9..45528d2192 100644
--- a/tcg/tcg-opc.h
+++ b/tcg/tcg-opc.h
@@ -42,6 +42,8 @@ DEF(br, 0, 0, 1, TCG_OPF_BB_END)
 # define IMPL64  TCG_OPF_64BIT
 #endif
 
+DEF(mb, 0, 0, 1, 0)
+
 DEF(mov_i32, 1, 1, 0, TCG_OPF_NOT_PRESENT)
 DEF(movi_i32, 1, 0, 1, TCG_OPF_NOT_PRESENT)
 DEF(setcond_i32, 1, 2, 1, 0)
diff --git a/tcg/tcg.h b/tcg/tcg.h
index 0384d7adf0..c9949aa79b 100644
--- a/tcg/tcg.h
+++ b/tcg/tcg.h
@@ -465,6 +465,23 @@ static inline intptr_t QEMU_ARTIFICIAL GET_TCGV_PTR(TCGv_ptr t)
 #define TCG_CALL_DUMMY_TCGV     MAKE_TCGV_I32(-1)
 #define TCG_CALL_DUMMY_ARG      ((TCGArg)(-1))
 
+typedef enum {
+    /* Used to indicate the type of accesses on which ordering
+       is to be ensured.  Modeled after SPARC barriers.  */
+    TCG_MO_LD_LD  = 0x01,
+    TCG_MO_ST_LD  = 0x02,
+    TCG_MO_LD_ST  = 0x04,
+    TCG_MO_ST_ST  = 0x08,
+    TCG_MO_ALL    = 0x0F,  /* OR of the above */
+
+    /* Used to indicate the kind of ordering which is to be ensured by the
+       instruction.  These types are derived from x86/aarch64 instructions.
+       It should be noted that these are different from C11 semantics.  */
+    TCG_BAR_LDAQ  = 0x10,  /* Following ops will not come forward */
+    TCG_BAR_STRL  = 0x20,  /* Previous ops will not be delayed */
+    TCG_BAR_SC    = 0x30,  /* No ops cross barrier; OR of the above */
+} TCGBar;
+
 /* Conditions.  Note that these are laid out for easy manipulation by
    the functions below:
      bit 0 is used for inverting;