All of lore.kernel.org
 help / color / mirror / Atom feed
* [Qemu-devel] muldiv64
@ 2010-10-29 19:30 malc
  2010-10-29 20:45 ` [Qemu-devel] muldiv64 Blue Swirl
  0 siblings, 1 reply; 3+ messages in thread
From: malc @ 2010-10-29 19:30 UTC (permalink / raw)
  To: qemu-devel; +Cc: Blue Swirl, Stefan Hajnoczi, Eduardo Cruz


c57c846a80f9306aa2c6cf7efdef45ed42723fac broke gus, Blue Swirl said that
he put muldiv64 into qemu-timer.h since putting it elsewhere broke
something and he can't remember what, anyways gus has nothing to do with
timers (just like muldiv64) so I've tried following and built all targets
on linux with -tracing-backend=nop (-tracing-backend=simple doesn't even
build for the reasons I can not understand), so unless someone will point
me to a scenario where following doesn't work it will be applied, thanks.

diff --git a/hw/omap_clk.c b/hw/omap_clk.c
index 10c9c43..6bcabef 100644
--- a/hw/omap_clk.c
+++ b/hw/omap_clk.c
@@ -20,7 +20,6 @@
  */
 #include "hw.h"
 #include "omap.h"
-#include "qemu-timer.h" /* for muldiv64() */
 
 struct clk {
     const char *name;
diff --git a/qemu-common.h b/qemu-common.h
index 2fbc27f..d31f366 100644
--- a/qemu-common.h
+++ b/qemu-common.h
@@ -309,6 +309,30 @@ static inline uint8_t from_bcd(uint8_t val)
     return ((val >> 4) * 10) + (val & 0x0f);
 }
 
+/* compute with 96 bit intermediate result: (a*b)/c */
+static inline uint64_t muldiv64(uint64_t a, uint32_t b, uint32_t c)
+{
+    union {
+        uint64_t ll;
+        struct {
+#ifdef HOST_WORDS_BIGENDIAN
+            uint32_t high, low;
+#else
+            uint32_t low, high;
+#endif
+        } l;
+    } u, res;
+    uint64_t rl, rh;
+
+    u.ll = a;
+    rl = (uint64_t)u.l.low * (uint64_t)b;
+    rh = (uint64_t)u.l.high * (uint64_t)b;
+    rh += (rl >> 32);
+    res.l.high = rh / c;
+    res.l.low = (((rh % c) << 32) + (rl & 0xffffffff)) / c;
+    return res.ll;
+}
+
 #include "module.h"
 
 #endif
diff --git a/qemu-timer.h b/qemu-timer.h
index 299e387..8cd8f83 100644
--- a/qemu-timer.h
+++ b/qemu-timer.h
@@ -59,30 +59,6 @@ static inline int64_t get_ticks_per_sec(void)
     return 1000000000LL;
 }
 
-/* compute with 96 bit intermediate result: (a*b)/c */
-static inline uint64_t muldiv64(uint64_t a, uint32_t b, uint32_t c)
-{
-    union {
-        uint64_t ll;
-        struct {
-#ifdef HOST_WORDS_BIGENDIAN
-            uint32_t high, low;
-#else
-            uint32_t low, high;
-#endif
-        } l;
-    } u, res;
-    uint64_t rl, rh;
-
-    u.ll = a;
-    rl = (uint64_t)u.l.low * (uint64_t)b;
-    rh = (uint64_t)u.l.high * (uint64_t)b;
-    rh += (rl >> 32);
-    res.l.high = rh / c;
-    res.l.low = (((rh % c) << 32) + (rl & 0xffffffff)) / c;
-    return res.ll;
-}
-
 /* real time host monotonic timer */
 static inline int64_t get_clock_realtime(void)
 {

-- 
mailto:av1474@comtv.ru

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* [Qemu-devel] Re: muldiv64
  2010-10-29 19:30 [Qemu-devel] muldiv64 malc
@ 2010-10-29 20:45 ` Blue Swirl
  2010-10-29 21:44   ` malc
  0 siblings, 1 reply; 3+ messages in thread
From: Blue Swirl @ 2010-10-29 20:45 UTC (permalink / raw)
  To: malc; +Cc: Stefan Hajnoczi, qemu-devel, Eduardo Cruz

On Fri, Oct 29, 2010 at 7:30 PM, malc <av1474@comtv.ru> wrote:
>
> c57c846a80f9306aa2c6cf7efdef45ed42723fac broke gus, Blue Swirl said that
> he put muldiv64 into qemu-timer.h since putting it elsewhere broke
> something and he can't remember what, anyways gus has nothing to do with
> timers (just like muldiv64) so I've tried following and built all targets
> on linux with -tracing-backend=nop (-tracing-backend=simple doesn't even
> build for the reasons I can not understand), so unless someone will point
> me to a scenario where following doesn't work it will be applied, thanks.

Maybe other changes I made later fixed the breakage, now with your
patch applied there seem to be no problems.

Acked, except for the description.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [Qemu-devel] Re: muldiv64
  2010-10-29 20:45 ` [Qemu-devel] muldiv64 Blue Swirl
@ 2010-10-29 21:44   ` malc
  0 siblings, 0 replies; 3+ messages in thread
From: malc @ 2010-10-29 21:44 UTC (permalink / raw)
  To: Blue Swirl; +Cc: Stefan Hajnoczi, qemu-devel, Eduardo Cruz

On Fri, 29 Oct 2010, Blue Swirl wrote:

> On Fri, Oct 29, 2010 at 7:30 PM, malc <av1474@comtv.ru> wrote:
> >
> > c57c846a80f9306aa2c6cf7efdef45ed42723fac broke gus, Blue Swirl said that
> > he put muldiv64 into qemu-timer.h since putting it elsewhere broke
> > something and he can't remember what, anyways gus has nothing to do with
> > timers (just like muldiv64) so I've tried following and built all targets
> > on linux with -tracing-backend=nop (-tracing-backend=simple doesn't even
> > build for the reasons I can not understand), so unless someone will point
> > me to a scenario where following doesn't work it will be applied, thanks.
> 
> Maybe other changes I made later fixed the breakage, now with your
> patch applied there seem to be no problems.
> 
> Acked, except for the description.
> 

Applied.

-- 
mailto:av1474@comtv.ru

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2010-10-29 21:45 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-10-29 19:30 [Qemu-devel] muldiv64 malc
2010-10-29 20:45 ` [Qemu-devel] muldiv64 Blue Swirl
2010-10-29 21:44   ` malc

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.