The e300c3 cores has an fpu and supports floating point instructions. I've confirmed that -mhard-float compiles and boots core-image-minimal with no floating point related issue. I've also compiled a program generating a floating point instruction and run it on the core I have in front of me (mpc8314e) The e300c2 does not have an fpu. Maybe it is just a copy and past error.
When you say copy and paste error - where? Which document/file? What are you looking for as a resolution here? For a new machine definition with a tune file using -mhard-float? My ppc hardware is broken currently. Perhaps you would care to provide a patch and test results?
Created attachment 1324 [details] proposed patch
(In reply to comment #1) I apologize that my report wasn't detailed enough. > When you say copy and paste error - where? Which document/file? meta/conf/machine/include/tune-ppce300c2.inc . As I explained, the e300c2 core does not have an fpu, the e300c3 does. http://www.freescale.com/files/32bit/doc/ref_manual/e300coreRM.pdf > What are you looking for as a resolution here? For a new machine definition > with a tune file using -mhard-float? To help ;-). If it has an fpu, it would be nice to use it. > My ppc hardware is broken currently. Perhaps you would care to provide a > patch and test results? I do not have MPC8315rdb, but a board based on the MPC8314e Soc, which also has an e300c3 core. Patch proposed.
This seems to have slipped through the cracks. Kevin is this something you could check into?
This issue has been fixed in 1.7 by the following commit: commit 3e528d3e4bc542ca2f654c54363c56af0fb9f61b Author: Mats Kärrman <Mats.Karrman@tritech.se> Date: Tue Apr 1 16:20:34 2014 +0000 Make ppce300c3 tune hard-float by default The tuning file for PowerPC e300c3 is soft-float. In OE-classic it was hard- float and it should be as the c3 has an fpu. I have modified the tuning file to include both a hard-float version (using the existing ppce300c3 name) and an optional soft-float version (called ppce300c3-nf). The following patch also passes a "--with-cpu=e300c3" argument to GLIBC. For this to have any effect the sqrt/sqrtf implementations added by the "glibc.fix_sqrt2.patch" are required and also an additional "Implies" file (added to the mentioned patch as a separate patch for eglibc_2.19). Tested with eglibc 2.19 on PowerPC MPC5125. (From OE-Core rev: 9d502ca8551fd461f869395b1b7e62d6dcf59a84) Signed-off-by: Mats Karrman <mats.karrman@tritech.se> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
http://git.yoctoproject.org/cgit.cgi/poky/commit/?id=3e528d3e4bc542ca2f654c54363c56af0fb9f61b
Verified on master: ccf5a4d2342c4fd40e4e0b332ea7e34e0bf175d4