[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1260850127-9766-17-git-send-email-zamsden@redhat.com>
Date: Mon, 14 Dec 2009 18:08:43 -1000
From: Zachary Amsden <zamsden@...hat.com>
To: kvm@...r.kernel.org
Cc: Zachary Amsden <zamsden@...hat.com>, Avi Kivity <avi@...hat.com>,
Marcelo Tosatti <mtosatti@...hat.com>,
Joerg Roedel <joerg.roedel@....com>,
linux-kernel@...r.kernel.org, Dor Laor <dlaor@...hat.com>
Subject: [PATCH RFC: kvm tsc virtualization 16/20] Fix 32-bit mult_precise
Turns out it wasn't very precise; it needs to work on 64-bit values.
Signed-off-by: Zachary Amsden <zamsden@...hat.com>
---
arch/x86/kvm/x86.c | 18 +++++++++++++++---
1 files changed, 15 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 7e2ba3e..792c895 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -803,15 +803,27 @@ static void compute_best_multiplier(unsigned long a, unsigned long b,
*s = shift;
}
-static inline unsigned long mult_precise(unsigned long val, unsigned long mult,
- int shift)
+static inline u64 mult_precise(u64 val, unsigned long mult, int shift)
{
+#if (BITS_PER_LONG == 64)
unsigned long top, bot;
- __asm__ ( "mul %3; shrd %1, %0" :
+ __asm__ ( "mulq %3; shrdq %1, %0" :
"=&a" (bot), "=&d" (top) :
"0" (mult), "rm" (val), "c" (shift));
return bot;
+#else
+ unsigned long ltop, lbot;
+ unsigned long htop, hbot;
+
+ __asm__ ( "mull %3; shrd %1, %0" :
+ "=&a" (lbot), "=&d" (ltop) :
+ "0" (mult), "rm" (long)(val), "c" (shift));
+ __asm__ ( "mull %3; shrd %1, %0" :
+ "=&a" (hbot), "=&d" (htop) :
+ "0" (mult), "rm" (long)(val >> 32), "c" (shift));
+ return (u64)lbot + ((u64)ltop + (u64)hbot) << 32;
+#endif
}
static inline u64 compute_ref_tsc(void)
--
1.6.5.2
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists