qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [Qemu-devel] [PATCH v2] memory: inline and optimize devend_memop
@ 2019-09-12 14:18 Paolo Bonzini
  2019-09-12 17:54 ` Richard Henderson
  0 siblings, 1 reply; 2+ messages in thread
From: Paolo Bonzini @ 2019-09-12 14:18 UTC (permalink / raw)
  To: qemu-devel; +Cc: Tony Nguyen, richard.henderson

devend_memop can rely on the fact that the result is always either
0 or MO_BSWAP, corresponding respectively to host endianness and
the opposite.  Native (target) endianness in turn can be either
the host endianness, in which case MO_BSWAP is only returned for
host-opposite endianness, or the opposite, in which case 0 is only
returned for host endianness.

With this in mind, devend_memop can be compiled as a setcond+shift
for every target.  Do this and, while at it, move it to
include/exec/memory.h since !NEED_CPU_H files do not (and should not)
need it.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 include/exec/memory.h | 19 ++++++++++++++++++-
 memory.c              | 18 ------------------
 2 files changed, 18 insertions(+), 19 deletions(-)

diff --git a/include/exec/memory.h b/include/exec/memory.h
index 2dd810259d..2e83c96580 100644
--- a/include/exec/memory.h
+++ b/include/exec/memory.h
@@ -2201,8 +2201,25 @@ address_space_write_cached(MemoryRegionCache *cache, hwaddr addr,
     }
 }
 
+#ifdef NEED_CPU_H
 /* enum device_endian to MemOp.  */
-MemOp devend_memop(enum device_endian end);
+static inline MemOp devend_memop(enum device_endian end)
+{
+    QEMU_BUILD_BUG_ON(DEVICE_HOST_ENDIAN != DEVICE_LITTLE_ENDIAN &&
+                      DEVICE_HOST_ENDIAN != DEVICE_BIG_ENDIAN);
+
+#ifdef BSWAP_NEEDED
+    /* Swap if non-host endianness or native (target) endianness */
+    return (end == DEVICE_HOST_ENDIAN) ? 0 : MO_BSWAP;
+#else
+    const int non_host_endianness =
+        DEVICE_LITTLE_ENDIAN ^ DEVICE_BIG_ENDIAN ^ DEVICE_HOST_ENDIAN;
+
+    /* In this case, native (target) endianness needs no swap.  */
+    return (end == non_host_endianness) ? MO_BSWAP : 0;
+#endif
+}
+#endif
 
 #endif
 
diff --git a/memory.c b/memory.c
index 61a254c3f9..b9dd6b94ca 100644
--- a/memory.c
+++ b/memory.c
@@ -3267,21 +3267,3 @@ static void memory_register_types(void)
 }
 
 type_init(memory_register_types)
-
-MemOp devend_memop(enum device_endian end)
-{
-    static MemOp conv[] = {
-        [DEVICE_LITTLE_ENDIAN] = MO_LE,
-        [DEVICE_BIG_ENDIAN] = MO_BE,
-        [DEVICE_NATIVE_ENDIAN] = MO_TE,
-        [DEVICE_HOST_ENDIAN] = 0,
-    };
-    switch (end) {
-    case DEVICE_LITTLE_ENDIAN:
-    case DEVICE_BIG_ENDIAN:
-    case DEVICE_NATIVE_ENDIAN:
-        return conv[end];
-    default:
-        g_assert_not_reached();
-    }
-}
-- 
2.21.0



^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [Qemu-devel] [PATCH v2] memory: inline and optimize devend_memop
  2019-09-12 14:18 [Qemu-devel] [PATCH v2] memory: inline and optimize devend_memop Paolo Bonzini
@ 2019-09-12 17:54 ` Richard Henderson
  0 siblings, 0 replies; 2+ messages in thread
From: Richard Henderson @ 2019-09-12 17:54 UTC (permalink / raw)
  To: Paolo Bonzini, qemu-devel; +Cc: Tony Nguyen

On 9/12/19 10:18 AM, Paolo Bonzini wrote:
> -MemOp devend_memop(enum device_endian end);
> +static inline MemOp devend_memop(enum device_endian end)

Ah ha, yes of course, the static inline.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~




^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2019-09-12 17:55 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-09-12 14:18 [Qemu-devel] [PATCH v2] memory: inline and optimize devend_memop Paolo Bonzini
2019-09-12 17:54 ` Richard Henderson

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).