On 16 April 2011 at 18:26, Aurelien Jarno wrote:
| Package: sprng
| Version: 2.0a-6
| Severity: serious
| Tags: patch
|
| Since GCC 4.4 it's not possible anymore to use the 'h' constraints for
| MIPS inline assembly code when doing a multiplication. That's why sprng
| fails to build from source on mips and mipsel.
|
| That said GCC supports 32x32 => 64 multiplication on 32-bit architecture
| for a lot of time, so there is no need to use assembly code for that.
| The patch below fixes the problem by using standard multiplication
| instead of assembly code.
Wow, thanks.
I'll fold that in -- thanks a bunch for looking into this and the patch.
Dirk
|
| diff -u sprng-2.0a/SRC/pmlcg/longlong.h sprng-2.0a/SRC/pmlcg/longlong.h
| --- sprng-2.0a/SRC/pmlcg/longlong.h
| +++ sprng-2.0a/SRC/pmlcg/longlong.h
| @@ -701,23 +701,12 @@
| #endif /* __m88000__ */
|
| #if defined (__mips__) && W_TYPE_SIZE == 32
| -#if __GNUC__ > 2 || __GNUC_MINOR__ >= 7
| -#define umul_ppmm(w1, w0, u, v) \
| - __asm__ ("multu %2,%3" \
| - : "=l" ((USItype)(w0)), \
| - "=h" ((USItype)(w1)) \
| - : "d" ((USItype)(u)), \
| - "d" ((USItype)(v)))
| -#else
| -#define umul_ppmm(w1, w0, u, v) \
| - __asm__ ("multu %2,%3"
| - "\n mflo %0"
| - "\n mfhi %1" \
| - : "=d" ((USItype)(w0)), \
| - "=d" ((USItype)(w1)) \
| - : "d" ((USItype)(u)), \
| - "d" ((USItype)(v)))
| -#endif
| +#define umul_ppmm(w1, w0, u, v) \
| + do { \
| + UDItype __x = (UDItype) (USItype) (u) * (USItype) (v); \
| + (w1) = (USItype) (__x >> 32); \
| + (w0) = (USItype) (__x); \
| + } while (0)
| #define UMUL_TIME 10
| #define UDIV_TIME 100
| #endif /* __mips__ */
|
|
| -- System Information:
| Debian Release: wheezy/sid
| APT prefers unstable
| APT policy: (500, 'unstable'), (1, 'experimental')
| Architecture: mips (mips64)
|
| Kernel: Linux 2.6.32-5-5kc-malta
| Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
| Shell: /bin/sh linked to /bin/bash
|
|
--
Dirk Eddelbuettel | [email protected] | http://dirk.eddelbuettel.com
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
