Hello,
Paul Burton wrote:
> This series allows the kernel to be built without any support for
> floating point. There are 2 main motivations for this:
>
> - The specification for nanoMIPS floating point support has not yet
> been finalized, so building the kernel without floating point will
> be useful for initial nanoMIPS support. Current nanoMIPS hardware,
> the MIPS I7200 CPU, has no FPU.
>
> - If userland is known not to use floating point instructions then
> omiting the kernel's support for it can produce respectable code
> size savings - around 112KB for 64r6el_defconfig.
>
> Applies cleanly atop v4.20-rc1.
>
> Thanks,
> Paul
>
> Paul Burton (20):
> MIPS: Hide CONFIG_MIPS_O32_FP64_SUPPORT prompt for >= MIPSr6
> MIPS: BCM5xxx: Remove dead init_fpu code
> MIPS: Simplify FP context initialization
> MIPS: Ensure emulated FP sets PF_USED_MATH
> MIPS: Drop forward declarations of sigcontext in asm/fpu.h
> MIPS: Better abstract R2300 FPU usage in Kconfig
> MIPS: Introduce CONFIG_MIPS_FP_SUPPORT
> MIPS: Hardcode cpu_has_fpu=0 when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: Stub asm/fpu.h functions
> MIPS: cpu-probe: Avoid probing FPU when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: traps: Never enable FPU when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: branch: Remove FP branch handling when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: unaligned: Remove FP & MSA code when unsupported
> MIPS: ptrace: Remove FP support when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: signal: Remove FP context support when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: Avoid FP ELF checks when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: Avoid FCSR sanitization when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: Don't compile math-emu when CONFIG_MIPS_FP_SUPPORT=n
> MIPS: Remove struct task_struct fpu state when
> CONFIG_MIPS_FP_SUPPORT=n
> MIPS: Allow FP support to be disabled
>
> arch/mips/Kconfig | 29 ++-
> arch/mips/Makefile | 2 +-
> arch/mips/include/asm/cpu-features.h | 11 +-
> arch/mips/include/asm/dsemul.h | 29 ++-
> arch/mips/include/asm/elf.h | 26 ++-
> arch/mips/include/asm/fpu.h | 145 +++++++++---
> arch/mips/include/asm/fpu_emulator.h | 11 -
> arch/mips/include/asm/processor.h | 19 +-
> arch/mips/include/asm/switch_to.h | 6 +-
> arch/mips/kernel/Makefile | 3 +-
> arch/mips/kernel/asm-offsets.c | 7 +-
> arch/mips/kernel/bmips_5xxx_init.S | 6 -
> arch/mips/kernel/branch.c | 34 +--
> arch/mips/kernel/cpu-probe.c | 54 +++--
> arch/mips/kernel/elf.c | 4 +
> arch/mips/kernel/genex.S | 2 +
> arch/mips/kernel/mips-r2-to-r6-emul.c | 7 -
> arch/mips/kernel/ptrace.c | 319 +++++++++++++-------------
> arch/mips/kernel/ptrace32.c | 33 +--
> arch/mips/kernel/r2300_fpu.S | 58 -----
> arch/mips/kernel/r4k_fpu.S | 144 ------------
> arch/mips/kernel/signal.c | 39 +++-
> arch/mips/kernel/traps.c | 117 ++++++----
> arch/mips/kernel/unaligned.c | 38 +--
> arch/mips/kvm/Kconfig | 1 +
> arch/mips/math-emu/cp1emu.c | 7 +
> 26 files changed, 600 insertions(+), 551 deletions(-)
Series applied to mips-next.
Thanks,
Paul
[ This message was auto-generated; if you believe anything is incorrect
then please email paul.burton@xxxxxxxx to report it. ]
