Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-10 Thread Nils Becker
2016-02-09 18:02 GMT+01:00 Gregor Thalhammer : >> It is not suitable as a standard for numpy. > > Why should numpy not provide fast transcendental math functions? For linear algebra it supports fast implementations, even non-free (MKL). Wouldn’t it be nice if numpy outperforms C? Floating point op

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-09 Thread Matthew Brett
On Tue, Feb 9, 2016 at 7:06 AM, Daπid wrote: > On 8 February 2016 at 18:36, Nathaniel Smith wrote: >> I would be highly suspicious that this speed comes at the expense of >> accuracy... My impression is that there's a lot of room to make >> speed/accuracy tradeoffs in these functions, and modern

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-09 Thread Gregor Thalhammer
> Am 09.02.2016 um 11:21 schrieb Nils Becker : > > 2016-02-08 18:54 GMT+01:00 Julian Taylor >: > > which version of glibm was used here? There are significant difference > > in performance between versions. > > Also the input ranges are very important for th

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-09 Thread Daπid
On 8 February 2016 at 18:36, Nathaniel Smith wrote: > I would be highly suspicious that this speed comes at the expense of > accuracy... My impression is that there's a lot of room to make > speed/accuracy tradeoffs in these functions, and modern glibc's libm has > seen a fair amount of scrutiny b

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-09 Thread Nils Becker
2016-02-08 18:54 GMT+01:00 Julian Taylor : > which version of glibm was used here? There are significant difference > in performance between versions. > Also the input ranges are very important for these functions, depending > on input the speed of these functions can vary by factors of 1000. > > g

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-08 Thread Gregor Thalhammer
> Am 08.02.2016 um 18:36 schrieb Nathaniel Smith : > > On Feb 8, 2016 3:04 AM, "Nils Becker" > wrote: > > > [...] > > Very superficial benchmarks (see below) seem devastating for gnu libm. It > > seems that openlibm (compiled with gcc -mtune=native -O3) performs r

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-08 Thread Julian Taylor
On 02/08/2016 06:36 PM, Nathaniel Smith wrote: > On Feb 8, 2016 3:04 AM, "Nils Becker" > wrote: >> > [...] >> Very superficial benchmarks (see below) seem devastating for gnu libm. > It seems that openlibm (compiled with gcc -mtune=native -O3) performs > really well

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-08 Thread Nathaniel Smith
On Feb 8, 2016 3:04 AM, "Nils Becker" wrote: > [...] > Very superficial benchmarks (see below) seem devastating for gnu libm. It seems that openlibm (compiled with gcc -mtune=native -O3) performs really well and intels libm implementation is the best (on my intel CPU). I did not check the accuracy

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-08 Thread Nils Becker
> The npy_math functions are used if otherwise unavailable OR if someone > has at some point noticed that say glibc 2.4-2.10 has a bad quality > tan (or whatever) and added a special case hack that checks for those > particular library versions and uses our built-in version instead. > It's not the

Re: [Numpy-discussion] Linking other libm-Implementation

2016-02-07 Thread Nathaniel Smith
On Sun, Feb 7, 2016 at 4:39 PM, Nils Becker wrote: > Hi all, > > I wanted to know if there is any sane way to build numpy while linking to a > different implementation of libm? > A drop-in replacement for libm (e.g. openlibm) should in principle work, I > guess, but I did not manage to actually ma