| Message ID | 1453135700-2404-1-git-send-email-Vincent.Riera@imgtec.com |
|---|---|
| State | Accepted |
| Commit | 190aa2b98f8f88563beb48b4b03269e7acbcfe77 |
| Headers | show |
Vicente, On Mon, 18 Jan 2016 16:48:20 +0000, Vicente Olivert Riera wrote: > Recent MIPS compilers have an stricter agreement between the compiler > and the assembler on the ABI. The compiler now passes -msoft-float to > the assembler where it previously did not. > > libffi uses MIPS hard-float instructions so when we try to build it for > soft-float using a recent compiler it fails because of the explained > above. This package should be fixed in order to add support for > soft-float. > > In the meantime we can add a little hack in order to be able to build it > for soft-float. Basically we add ".set hardfloat" so the assembler will > not disable the hard-float instructions. The build will of course > include some hard-float bits which shouldn't exist in a soft-float > build, and that may cause runtime problems. But that's what we already > had before and we had lived with it. > > We choose to apply this temporary fix because it only affects MIPS > soft-float builds. Otherwise we would need to disable libffi and > recursively all the packages depending on it (python, libglib2...) for > MIPS soft-float, which would be a bit overkill. > > Fixes: > > http://autobuild.buildroot.net/results/6b0607b7cb0ac5525c2e47e819301f38bd2d6b30/ > > Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com> > --- > Changes v1 -> v2: > - Add autobuild URL Tested-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> (building with our Sourcery MIPS configuration fails without the patch, works with this patch applied.) Thanks! Thomas
>>>>> "Vicente" == Vicente Olivert Riera <Vincent.Riera@imgtec.com> writes: > Recent MIPS compilers have an stricter agreement between the compiler > and the assembler on the ABI. The compiler now passes -msoft-float to > the assembler where it previously did not. > libffi uses MIPS hard-float instructions so when we try to build it for > soft-float using a recent compiler it fails because of the explained > above. This package should be fixed in order to add support for > soft-float. > In the meantime we can add a little hack in order to be able to build it > for soft-float. Basically we add ".set hardfloat" so the assembler will > not disable the hard-float instructions. The build will of course > include some hard-float bits which shouldn't exist in a soft-float > build, and that may cause runtime problems. But that's what we already > had before and we had lived with it. > We choose to apply this temporary fix because it only affects MIPS > soft-float builds. Otherwise we would need to disable libffi and > recursively all the packages depending on it (python, libglib2...) for > MIPS soft-float, which would be a bit overkill. > Fixes: > http://autobuild.buildroot.net/results/6b0607b7cb0ac5525c2e47e819301f38bd2d6b30/ > Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com> > --- > Changes v1 -> v2: > - Add autobuild URL Committed, thanks.
diff --git a/package/libffi/0003-libffi-enable-hardfloat-in-the-MIPS-assembly-code.patch b/package/libffi/0003-libffi-enable-hardfloat-in-the-MIPS-assembly-code.patch new file mode 100644 index 0000000..776990d --- /dev/null +++ b/package/libffi/0003-libffi-enable-hardfloat-in-the-MIPS-assembly-code.patch @@ -0,0 +1,41 @@ +From 48bc37fabbc685b1e3293055bd33ca66c619305e Mon Sep 17 00:00:00 2001 +From: Vicente Olivert Riera <Vincent.Riera@imgtec.com> +Date: Wed, 13 Jan 2016 14:49:59 +0000 +Subject: [PATCH] libffi: enable hardfloat in the MIPS assembly code + +This way it will be possible to build it for soft-float. This is only a +temporary fix. The package needs to be fixed properly. + +Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com> +--- + src/mips/n32.S | 1 + + src/mips/o32.S | 1 + + 2 files changed, 2 insertions(+) + +diff --git a/src/mips/n32.S b/src/mips/n32.S +index c6985d3..dc842d5 100644 +--- a/src/mips/n32.S ++++ b/src/mips/n32.S +@@ -44,6 +44,7 @@ + .abicalls + #endif + .set mips4 ++ .set hardfloat + .text + .align 2 + .globl ffi_call_N32 +diff --git a/src/mips/o32.S b/src/mips/o32.S +index eb27981..b653daf 100644 +--- a/src/mips/o32.S ++++ b/src/mips/o32.S +@@ -42,6 +42,7 @@ + #define RA_OFF (SIZEOF_FRAME - 1 * FFI_SIZEOF_ARG) + + .abicalls ++ .set hardfloat + .text + .align 2 + .globl ffi_call_O32 +-- +2.4.10 +
Recent MIPS compilers have an stricter agreement between the compiler and the assembler on the ABI. The compiler now passes -msoft-float to the assembler where it previously did not. libffi uses MIPS hard-float instructions so when we try to build it for soft-float using a recent compiler it fails because of the explained above. This package should be fixed in order to add support for soft-float. In the meantime we can add a little hack in order to be able to build it for soft-float. Basically we add ".set hardfloat" so the assembler will not disable the hard-float instructions. The build will of course include some hard-float bits which shouldn't exist in a soft-float build, and that may cause runtime problems. But that's what we already had before and we had lived with it. We choose to apply this temporary fix because it only affects MIPS soft-float builds. Otherwise we would need to disable libffi and recursively all the packages depending on it (python, libglib2...) for MIPS soft-float, which would be a bit overkill. Fixes: http://autobuild.buildroot.net/results/6b0607b7cb0ac5525c2e47e819301f38bd2d6b30/ Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com> --- Changes v1 -> v2: - Add autobuild URL ...nable-hardfloat-in-the-MIPS-assembly-code.patch | 41 ++++++++++++++++++++++ 1 file changed, 41 insertions(+) create mode 100644 package/libffi/0003-libffi-enable-hardfloat-in-the-MIPS-assembly-code.patch