From: Peter Zijlstra <peterz@infradead.org>
To: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Ray Lee <ray-lk@madrabbit.org>,
adobriyan@gmail.com, Ingo Molnar <mingo@elte.hu>,
"Zhang, Yanmin" <yanmin_zhang@linux.intel.com>,
Dhaval Giani <dhaval@linux.vnet.ibm.com>,
LKML <linux-kernel@vger.kernel.org>,
Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>,
Aneesh Kumar KV <aneesh.kumar@linux.vnet.ibm.com>,
Balbir Singh <balbir@in.ibm.com>,
Chris Friesen <cfriesen@nortel.com>
Subject: Re: VolanoMark regression with 2.6.27-rc1
Date: Thu, 21 Aug 2008 10:17:35 +0200 [thread overview]
Message-ID: <1219306655.8651.90.camel@twins> (raw)
In-Reply-To: <200808211611.17889.nickpiggin@yahoo.com.au>
On Thu, 2008-08-21 at 16:11 +1000, Nick Piggin wrote:
> On Thursday 21 August 2008 06:56, Peter Zijlstra wrote:
> > On Wed, 2008-08-20 at 22:30 +0200, Peter Zijlstra wrote:
>
> > > works for the above example, but when I make it long long, so as to
> > > match the longest supported type, it goes boom again - for as of yet
> > > unknown reasons.
> >
> > Ok, people pointed out I got my promotion rules mixed up, I casted the
> > result of the division to signed, instead of ending up with a signed
> > division.
> >
> > #define avg(x, y) ({ \
> > typeof(x) _avg1 = (x); \
> > typeof(y) _avg2 = (y); \
> > (void) (&_avg1 == &_avg2); \
> > (typeof(x))(_avg1 + ((long long)_avg2 - _avg1)/2); })
> >
> > seems to work.
>
> Right, I guess that will work, but unfortunately the code gen on 32-bit
> is a monstrosity. If you're going to cast to 64-bit anyway, we might as
> well then just do the normal add rather than playing the game to avoid
> overflow.
>
> Secondly, this is operating on the fixed point scaled load numbers, so in
> the case of the scheduler I wouldn't worry too much about rounding... also
> in most integer operations, rounding down is less surprising than rounding
> up like the last code did.
>
> I still don't know whether it is appropriate to put it into kernel.h
> (because of rounding, and variability when it comes to what type size will
> hold the sum of parameters), but for the scheduler, I would use this:
>
> ((unsigned long long)a + b) / 2;
Right - anyway the point is moot - as Yanmin says it still sucks rocks.
But since I couldn't let it rest :-)
---
Index: linux-2.6/include/linux/kernel.h
===================================================================
--- linux-2.6.orig/include/linux/kernel.h
+++ linux-2.6/include/linux/kernel.h
@@ -367,6 +367,45 @@ static inline char *pack_hex_byte(char *
(void) (&_max1 == &_max2); \
_max1 > _max2 ? _max1 : _max2; })
+#define __avg_t(type, x, y) ({ \
+ typeof(x) __avg1 = (x); \
+ typeof(y) __avg2 = (y); \
+ __avg1 + ((type)(__avg2 - __avg1))/2; })
+
+extern void avg_unknown_size(void);
+
+#define __avg(x, y) ({ \
+ typeof(x) ret; \
+ switch (sizeof(ret)) { \
+ case 1: \
+ ret = __avg_t(s8, x, y); \
+ break; \
+ case 2: \
+ ret = __avg_t(s16, x, y); \
+ break; \
+ case 4: \
+ ret = __avg_t(s32, x, y); \
+ break; \
+ case 8: \
+ ret = __avg_t(s64, x, y); \
+ break; \
+ default: \
+ avg_unknown_size(); \
+ break; \
+ } \
+ ret; })
+
+#define avg(x, y) ({ \
+ typeof(x) _avg1 = (x); \
+ typeof(y) _avg2 = (y); \
+ (void) (&_avg1 == &_avg2); \
+ __avg(_avg1, _avg2); })
+
+#define avg_t(type, x, y) ({ \
+ type _avg1 = (x); \
+ type _avg2 = (y); \
+ __avg(_avg1, _avg2); })
+
/**
* clamp - return a value clamped to a given range with strict typechecking
* @val: current value
next prev parent reply other threads:[~2008-08-21 8:18 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-07-31 3:20 VolanoMark regression with 2.6.27-rc1 Zhang, Yanmin
2008-07-31 7:31 ` Zhang, Yanmin
2008-07-31 7:39 ` Peter Zijlstra
2008-07-31 7:49 ` Zhang, Yanmin
2008-08-01 0:39 ` Zhang, Yanmin
2008-08-01 2:35 ` Miao Xie
2008-08-01 3:08 ` Zhang, Yanmin
2008-08-01 5:14 ` Dhaval Giani
2008-08-04 5:04 ` Zhang, Yanmin
2008-08-04 5:22 ` Dhaval Giani
2008-08-04 5:37 ` Zhang, Yanmin
2008-08-04 5:53 ` Dhaval Giani
2008-08-04 6:26 ` Peter Zijlstra
2008-08-04 6:26 ` Peter Zijlstra
2008-08-04 7:05 ` Dhaval Giani
2008-08-04 7:12 ` Peter Zijlstra
2030-08-06 3:26 ` Zhang, Yanmin
2008-08-08 7:30 ` Peter Zijlstra
[not found] ` <20080811185008.GA29291@linux.vnet.ibm.com>
[not found] ` <1912726331.25608.235.camel@ymzhang>
[not found] ` <20080817115035.GA32223@linux.vnet.ibm.com>
[not found] ` <20080818052155.GA5063@linux.vnet.ibm.com>
2008-08-20 7:24 ` Zhang, Yanmin
2008-08-20 7:41 ` Peter Zijlstra
2008-08-20 10:51 ` Ingo Molnar
2008-08-20 13:32 ` Peter Zijlstra
2008-08-20 13:47 ` Ingo Molnar
2008-08-21 2:25 ` Zhang, Yanmin
2008-08-21 6:16 ` Ingo Molnar
2008-08-21 6:48 ` Zhang, Yanmin
2008-08-29 3:35 ` Zhang, Yanmin
2008-08-29 3:38 ` Zhang, Yanmin
2008-08-20 14:32 ` adobriyan
2008-08-20 14:33 ` Peter Zijlstra
2008-08-20 15:10 ` Nick Piggin
2008-08-20 15:15 ` Peter Zijlstra
2008-08-20 16:29 ` Ray Lee
2008-08-20 16:51 ` Peter Zijlstra
2008-08-20 17:21 ` Peter Zijlstra
2008-08-20 17:55 ` Nick Piggin
2008-08-20 18:15 ` Ray Lee
2008-08-20 20:30 ` Peter Zijlstra
2008-08-20 20:56 ` Peter Zijlstra
2008-08-21 6:11 ` Nick Piggin
2008-08-21 8:17 ` Peter Zijlstra [this message]
2008-08-21 6:15 ` Ingo Molnar
2008-08-20 20:58 ` Ray Lee
2008-08-20 21:04 ` Peter Zijlstra
2008-08-21 6:12 ` Ingo Molnar
2030-08-13 8:50 ` Zhang, Yanmin
2008-08-04 6:54 ` Peter Zijlstra
2008-08-15 15:37 ` Ingo Molnar
2008-08-01 12:25 ` Hugh Dickins
2008-08-04 0:54 ` Zhang, Yanmin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1219306655.8651.90.camel@twins \
--to=peterz@infradead.org \
--cc=adobriyan@gmail.com \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=balbir@in.ibm.com \
--cc=cfriesen@nortel.com \
--cc=dhaval@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=nickpiggin@yahoo.com.au \
--cc=ray-lk@madrabbit.org \
--cc=vatsa@linux.vnet.ibm.com \
--cc=yanmin_zhang@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox