For discussion of Alpine Linux development and developer support

[alpine-devel] [RFC] mipsel*hf support

Message ID
Sender timestamp
DKIM signature
Download raw message
Patch: +2 -0
This hints at enabling two additional MIPS variants in Alpine:

1) mipseln8hf, MIPS32R2 with 64-bit IEEE 754-2008 (nan2008) FPU unit

The primary (and probably the only) target is BE-T1 SoC (with P5600
core) as found on BFK3 board. M51xx-based systems (if any) might be
compatible too.

2) mipselhf, MIPS32R2 with 64-bit IEEE 754-1985 (legacy) FPU unit

This is mostly for MIPS64R2 systems using O32 ABI. Loongson 3A/B,
probably Octeon III. Older MIPS32R2 cores (24K*f, 74K*f) might fit
too however impractical due to usually low amount of RAM.

What is the Alpine policy on source-only ports (since there is no
intention to provide builder for these targets)?

Is there any interest in mipsel*hf at all?

Should I send mipsel*hf (versus generic mips*) patches?

(not in the list so please CC me).

The patch below just illustrates the difference from existing mips*.

 main/gcc/APKBUILD | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/main/gcc/APKBUILD b/main/gcc/APKBUILD
index 0cf9b1eddc..35ce17ac1d 100644
--- a/main/gcc/APKBUILD
+++ b/main/gcc/APKBUILD
@@ -226,6 +226,8 @@ build() {
 	mips-*-*-*)		_arch_configure="--with-arch=mips32 --with-mips-plt --with-float=soft --with-abi=32";;
 	mips64-*-*-*)		_arch_configure="--with-arch=mips3 --with-tune=mips64 --with-mips-plt --with-float=soft --with-abi=64";;
 	mips64el-*-*-*)		_arch_configure="--with-arch=mips3 --with-tune=mips64 --with-mips-plt --with-float=soft --with-abi=64";;
+	mipsel-*-*-*n8hf)	_arch_configure="--with-arch=mips32r2 --with-mips-plt --with-float=hard --with-fp_32=64 --with-nan=2008 --with-tune=p5600";;
+	mipsel-*-*-*hf)		_arch_configure="--with-arch=mips32r2 --with-mips-plt --with-float=hard --with-fp_32=64 --with-tune=loongson3a";;
 	mipsel-*-*-*)		_arch_configure="--with-arch=mips32 --with-mips-plt --with-float=soft --with-abi=32";;
 	powerpc-*-*-*)		_arch_configure="--enable-secureplt --enable-decimal-float=no";;
 	powerpc64*-*-*-*)	_arch_configure="--with-abi=elfv2 --enable-secureplt --enable-decimal-float=no --enable-targets=powerpcle-linux";;