[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ft-devel] FT_MulDiv optimization
From: |
Behdad Esfahbod |
Subject: |
Re: [ft-devel] FT_MulDiv optimization |
Date: |
Fri, 04 Jul 2014 14:06:40 -0400 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.6.0 |
On 14-07-04 01:57 PM, Werner LEMBERG wrote:
>
>>>>> Well, at the same time it reduces the available range for `c' by
>>>>> more than 50%.
>>>>
>>>> I was greedy with the space for a and b. With a + b < 2*46340
>>>> instead, c can use the same old limit. We can expand c a lot further
>>>> by making a+b only a bit smaller.
>>>
>>> OK. Another reason for a statistical analysis :-)
>>
>> Not worth it. If Alexei keeps a + b < 2*46340 then this is a clear
>> improvement and needs no further analysis.
>
> Hmm. We have three variables (a, b, c) in the function, and Alexei's
> patch increases the number of possible (a,b) pairs, and at the same
> time it reduces the number of possible c values for the shortcut. You
> say that this is a good thing, but why?
See attached patch which is a clear improvement in space and speed. This is
what he means by not being greedy.
--
behdad
http://behdad.org/
ft-muldiv-opt.patch
Description: Text Data
- [ft-devel] FT_MulDiv optimization, Alexei Podtelezhnikov, 2014/07/03
- Re: [ft-devel] FT_MulDiv optimization, Alexei Podtelezhnikov, 2014/07/04
- Re: [ft-devel] FT_MulDiv optimization, Behdad Esfahbod, 2014/07/04
- Re: [ft-devel] FT_MulDiv optimization, Alexei Podtelezhnikov, 2014/07/04
- Re: [ft-devel] FT_MulDiv optimization, Behdad Esfahbod, 2014/07/04
- Re: [ft-devel] FT_MulDiv optimization, Werner LEMBERG, 2014/07/04
- Re: [ft-devel] FT_MulDiv optimization, Alexei Podtelezhnikov, 2014/07/04
- Re: [ft-devel] FT_MulDiv optimization, Werner LEMBERG, 2014/07/05
- Re: [ft-devel] FT_MulDiv optimization, Alexei Podtelezhnikov, 2014/07/05
- Re: [ft-devel] FT_MulDiv optimization, Werner LEMBERG, 2014/07/08