1 |
Hi |
2 |
I too have been trying to set hard maverick crunch FPU code |
3 |
generation as the default for a compiler here, and I see some people |
4 |
here made the same mistake of configuring GCC using |
5 |
--with-float-abi=softfp |
6 |
This option does nothing and is silently ignored (well ok, "configure" |
7 |
sets $with_float_abi but nothing ever inspects that) - you still have |
8 |
to supply --with-float-abi=softfp at runtime to get hard FP |
9 |
instructions, obtaining soft float otherwise. |
10 |
|
11 |
There is currently no way to select softfp Maverick code generation by default. |
12 |
for GCC 4.3.2, the closest configure options are, as well as |
13 |
--with-float=softfp and --with-fpu=maverick: |
14 |
|
15 |
--with-arch=armv4t the build fails when it tries to compile hard FPA |
16 |
instructions into libgcc, |
17 |
--with-cpu=ep9312 it compiles armv5t instructions into libgcc |
18 |
--with-arch=ep9312 the assembler barfs saying that ep9312 is not a |
19 |
recognised architecture. |
20 |
|
21 |
There is even one check that looks for --target=ep9312-* and selects |
22 |
default fpu=Maverick but I very much doubt that works. |
23 |
|
24 |
This whole area is bit-rotten. I think the 'ep9312' should be |
25 |
eliminated,or mapped into arm920t+maverick early on, and everything |
26 |
should work off -m[cf]pu= the same as every other fpu. Unfortunately, |
27 |
checks for ep9312 are spread through the config scripts and source |
28 |
like a fungus. |
29 |
|
30 |
M |