Re: Gutenprint 5.3.4: only on 32-bit failures with "Image is too long/wide for the page" for certain PPDs

On 9/4/21 12:50 PM, Solomon Peachy wrote:
> On Sat, Sep 04, 2021 at 10:03:33AM -0400, Robert Krawitz wrote:
>>> Only on 32-bit (i586) the cupsfilter command fails for the PPDs
>>> listed below with output excerpts from the full test log file.
> 
>> All of these are dyesub printers, which might or might not be the issue.
> 
> What these printers share in common is non-printable margins that must 
> be present in the data stream (eg the imageable area is 1844 pixels wide 
> but the printer requires 1920 pixels to be sent per row) and that is 
> represented by page margins.
> 
> In this case it sounds like we're getting hit with behavioral 
> (rounding?) differences between x87 and SSE2 floating point.

Sigh...have to figure out how to fix this.  Maybe we have to allow something like a millipoint of
slop.  Interesting that it seems to have only hit the dyesub printers; is the issue that those are
the only ones with non-integer widths (in points)?