Fix-up coefficients and smallest denormal value for ieeeDouble in exp()#3388
Fix-up coefficients and smallest denormal value for ieeeDouble in exp()#3388ibuclaw wants to merge 1 commit intodlang:masterfrom
Conversation
|
By the way, how do you guys get the represented hexadecimal float? I've used |
There was a problem hiding this comment.
FYI, C2 was changed from the (128bit number) 1.428606820309417232121458176568075500134E-6L;, if we are going to have a branch for each float format, might as well have the number in the correct precision.
|
Ping. @MartinNowak - this is for 2.068 release |
|
Also need to sort out https://issues.dlang.org/show_bug.cgi?id=14732 |
|
Ping? |
|
I'd opportunistically merge but have literally no iidea if this pull does the right thing. You might want to target the stable branch for 2.068 though. |
There was a problem hiding this comment.
Those are the same coefficients as above.
There was a problem hiding this comment.
Yes, the coefficients should be identical regardless of what platform / float format we are using.
|
Rebased and repopend against stable, see #3514. |
3 participants
Addresses complaints in #3285
Apart from the failing unittest....