hw/registerfields: Use 64-bit bitfield for FIELD_DP64

If we have a field that's wider than 32-bits, we need a data type wide enough to
be able to create the bitfield used to deposit the value.

Signed-off-by: Joe Komlodi <joe.komlodi@xilinx.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1626805903-162860-3-git-send-email-joe.komlodi@xilinx.com
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

diff --git a/include/hw/registerfields.h b/include/hw/registerfields.h
index 9a03ac55e41682bed5cf67749adf6177df4b810b..f2a3c9c41f7b8cfff27b8fea8e5473f657bbb25a 100644 (file)
--- a/include/hw/registerfields.h
+++ b/include/hw/registerfields.h
@@ -95,7 +95,7 @@
     _d; })
 #define FIELD_DP64(storage, reg, field, val) ({                           \
     struct {                                                              \
-        unsigned int v:R_ ## reg ## _ ## field ## _LENGTH;                \
+        uint64_t v:R_ ## reg ## _ ## field ## _LENGTH;                \
     } _v = { .v = val };                                                  \
     uint64_t _d;                                                          \
     _d = deposit64((storage), R_ ## reg ## _ ## field ## _SHIFT,          \