Re: #215067 mozilla FTBFS

2003-10-23 Thread Adam C Powell IV
tags 215067 patch thanks On Thu, 2003-10-23 at 14:07, Philip Blundell wrote: > On Thu, 2003-10-23 at 13:40, Adam C Powell IV wrote: > > Is it secretly big-endian for > > its float/double emulation, and little-endian for ints? > > More or less, yeah. It's actually only doubles that are affected.

Re: #215067 mozilla FTBFS

2003-10-23 Thread Philip Blundell
On Thu, 2003-10-23 at 13:40, Adam C Powell IV wrote: > Is it secretly big-endian for > its float/double emulation, and little-endian for ints? More or less, yeah. It's actually only doubles that are affected. I suggest you try changing #if defined(IS_LITTLE_ENDIAN) into #if defined(IS_LITT

Re: #215067 mozilla FTBFS

2003-10-23 Thread Peter Naulls
In message <[EMAIL PROTECTED]> Nicholas Clark <[EMAIL PROTECTED]> wrote: > On Thu, Oct 23, 2003 at 08:40:57AM -0400, Adam C Powell IV wrote: > > > Does ARM store doubles in a non-IEEE way? Is it secretly big-endian for > > its float/double emulation, and little-endian for ints? What e

Re: #215067 mozilla FTBFS

2003-10-23 Thread Nicholas Clark
On Thu, Oct 23, 2003 at 08:40:57AM -0400, Adam C Powell IV wrote: > Does ARM store doubles in a non-IEEE way? Is it secretly big-endian for > its float/double emulation, and little-endian for ints? What else could > be wrong with mozilla's assumptions about double format? I suspect you've found

Re: #215067 mozilla FTBFS

2003-10-23 Thread Ralph Siemsen
On Thu, Oct 23, 2003 at 09:57:52AM -0400, Whaley, Robert wrote: > I think for now, only the old FPA memory layout is used in Debian > binaries. I think this is the format you describe in your message > that's causing the problem. This is true on the Netwinder redhat-like distribution as well.

Re: #215067 mozilla FTBFS

2003-10-23 Thread Ralph Siemsen
On Thu, Oct 23, 2003 at 08:40:57AM -0400, Adam C Powell IV wrote: > Does ARM store doubles in a non-IEEE way? Is it secretly big-endian for > its float/double emulation, and little-endian for ints? What else could > be wrong with mozilla's assumptions about double format? If you're manipulatin

RE: #215067 mozilla FTBFS

2003-10-23 Thread Whaley, Robert
ARM has 2 different floating point formats. The newer one is called VFP and is ANSI/IEEE 754-195 compliant. I'm not sure which processors (if any yet) have this hardware. The older floating point format is called FPA. Most ARM processors don't have floating point hardware. These are typically

Re: #215067 mozilla FTBFS

2003-10-23 Thread Adam C Powell IV
Okay ARM hackers, as a user/neophyte, I need your help. As discussed below (the good stuff is at the end), I've traced the mozilla segfault to their PR_dtoa function, which converts doubles to strings. Because it directly manipulates the bits of doubles, it is making some gross errors, like appar