linux-mips
[Top] [All Lists]

Re: [PATCH v2 1/3] MIPS: add a common mips_cyc2ns()

To: Wu Zhangin <wuzhangjin@gmail.com>
Subject: Re: [PATCH v2 1/3] MIPS: add a common mips_cyc2ns()
From: David Daney <ddaney@caviumnetworks.com>
Date: Wed, 07 Apr 2010 09:48:16 -0700
Cc: Ralf Baechle <ralf@linux-mips.org>, linux-mips@linux-mips.org, Ralf Rösch <roesch.ralf@web.de>
In-reply-to: <9e1889ed5fa23dfaa1ad432ebb4b8f837f6668b4.1270655886.git.wuzhangjin@gmail.com>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <cover.1270653461.git.wuzhangjin@gmail.com> <9e1889ed5fa23dfaa1ad432ebb4b8f837f6668b4.1270655886.git.wuzhangjin@gmail.com>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.9) Gecko/20100330 Fedora/3.0.4-1.fc12 Thunderbird/3.0.4
On 04/07/2010 09:05 AM, Wu Zhangin wrote:
From: Wu Zhangjin<wuzhangjin@gmail.com>

Changes:

v1 ->  v2:

   o change the old mips_sched_clock() to mips_cyc2ns() and modify the
   arguments to support 32bit.
   o add 32bit support: use a smaller shift to avoid the quick overflow
   of 64bit arithmatic and balance the overhead of the 128bit arithmatic
   and the precision lost with the smaller shift.

----------------------

Because the high resolution sched_clock() for r4k has the same overflow
problem and solution mentioned in "MIPS: Octeon: Use non-overflowing
arithmetic in sched_clock".

     "With typical mult and shift values, the calculation for Octeon's
     sched_clock overflows when using 64-bit arithmetic.  Use 128-bit
     calculations instead."

To reduce the duplication, This patch abstracts the solution into an
inline funciton mips_cyc2ns() into arch/mips/include/asm/time.h from
arch/mips/cavium-octeon/csrc-octeon.c.

Two patches for Cavium and R4K will be sent out respectively to use this
common function.

Signed-off-by: Wu Zhangjin<wuzhangjin@gmail.com>
---
  arch/mips/include/asm/time.h |   38 ++++++++++++++++++++++++++++++++++++++
  1 files changed, 38 insertions(+), 0 deletions(-)

diff --git a/arch/mips/include/asm/time.h b/arch/mips/include/asm/time.h
index c7f1bfe..898f0e0 100644
--- a/arch/mips/include/asm/time.h
+++ b/arch/mips/include/asm/time.h
@@ -96,4 +96,42 @@ static inline void clockevent_set_clock(struct 
clock_event_device *cd,
        clockevents_calc_mult_shift(cd, clock, 4);
  }

+static inline unsigned long long mips_cyc2ns(u64 cyc, u64 mult, u64 shift)
+{
+#ifdef CONFIG_32BIT
+       /*
+        * To balance the overhead of 128bit-arithematic and the precision
+        * lost, we choose a smaller shift to avoid the quick overflow as the
+        * X86&  ARM does. please refer to arch/x86/kernel/tsc.c and
+        * arch/arm/plat-orion/time.c
+        */
+       return (cyc * mult)>>  shift;
+#else /* CONFIG_64BIT */
+       /* 64-bit arithmatic can overflow, so use 128-bit.  */
+#if (__GNUC__<  4) || ((__GNUC__ == 4)&&  (__GNUC_MINOR__<= 3))
+       u64 t1, t2, t3;
+       unsigned long long rv;
+
+       asm (
+               "dmultu\t%[cyc],%[mult]\n\t"
+               "nor\t%[t1],$0,%[shift]\n\t"
+               "mfhi\t%[t2]\n\t"
+               "mflo\t%[t3]\n\t"
+               "dsll\t%[t2],%[t2],1\n\t"
+               "dsrlv\t%[rv],%[t3],%[shift]\n\t"
+               "dsllv\t%[t1],%[t2],%[t1]\n\t"
+               "or\t%[rv],%[t1],%[rv]\n\t"
+               : [rv] "=&r" (rv), [t1] "=&r" (t1), [t2] "=&r" (t2), [t3] "=&r" 
(t3)
+               : [cyc] "r" (cyc), [mult] "r" (mult), [shift] "r" (shift)
+               : "hi", "lo");
+       return rv;
+#else  /* GCC>  4.3 do it the easy way.  */
+       unsigned int __attribute__((mode(TI))) t = cyc;
+
+       t = (t * mult)>>  shift;
+       return (unsigned long long)t;
+#endif
+#endif /* CONFIG_64BIT */
+}
+
  #endif /* _ASM_TIME_H */

It turns out that all GCC versions can handle the inline asm way. It has also been noted that the default Debian compiler somehow has problems with the 'easy way'.

Therefore, I would recommend gitting rid of the GCC version conditionals and just leave the inline asm.

David Daney

<Prev in Thread] Current Thread [Next in Thread>