[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GLPSOL in webassemby faster than native ?

From: Domingo Alvarez Duarte
Subject: Re: GLPSOL in webassemby faster than native ?
Date: Fri, 25 Sep 2020 10:04:02 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.10.0

Hello Michael !

Thank you for reply !

I'll take into account the use of possible wider float format for intermediary values using something like your suggestion of redefinable type like "glp_double_t" (actually in gcc 9 in linux x86 "double_t" and "double" are equal).

But also leave the possibility of have it working (somehow) with float32, float64, float80, float128, ... everywhere.

Again help/suggestions/comments are welcome !

Cheers !

On 24/9/20 21:18, Michael Hennebry wrote:
On Thu, 24 Sep 2020, Domingo Alvarez Duarte wrote:

I just got glpsol with "long double" working and add binaries for anyone that want to test then here

As noted there it'll benefit from tuning the constants in src/glpk_real.h

Any help/suggestion/comment is welcome !

Note that using long doubles everywhere
would slow down a memory bound computation.
Fetching more data would be required.

One might introduce glp_double_t.
C99 uses double_t for the type in which unnamed
intermediate results of double calculations are stored.

Use glp_double_t for locals that are not arrays,
especially running totals.
Do the casts to ensure that desired calculations
are actually done in glp_double_t.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]