search for: mgeneral

Steve Capper dixit: >> Do we need the fpregs saved even though klibc doesn't do fp? >For gcc targetting Aarch64, We can only guarantee that d8-d15 are >left alone when -mgeneral-regs-only is supplied for building klibc >and any software linked against klibc. I would much prefer to We can enforce this in klcc, just like -mregparm=3 is used by the i386 target. bye, //mirabilos -- ?Cool, /usr/share/doc/mksh/examples/uhr.gz ist ja ein Grund, mksh auf jedem System zu inst...

On 11/08/2013 09:12 AM, Steve Capper wrote: > + > +/* > + * x19-x28 are callee saved, also save fp, lr, sp. > + * d8-d15 are also callee saved. > + */ > + > +struct __jmp_buf { > + uint64_t __gregs[13]; > + uint64_t __fpregs[8]; > +}; > + Since the index of these arrays have no connection with what is stored in them, they should be named fields in the structure,

...00000..edc3312 --- /dev/null +++ b/usr/include/arch/arm64/klibc/archsetjmp.h @@ -0,0 +1,22 @@ +/* + * arch/arm64/include/klibc/archsetjmp.h + */ + +#ifndef _KLIBC_ARCHSETJMP_H +#define _KLIBC_ARCHSETJMP_H + +/* + * x19-x28 are callee saved, also save fp, lr, sp. + * d8-d15 are unused as we specify -mgeneral-regs-only as a build flag. + */ + +struct __jmp_buf { + uint64_t __x19, __x20, __x21, __x22; + uint64_t __x23, __x24, __x25, __x26; + uint64_t __x27, __x28, __x29, __x30; + uint64_t __sp; +}; + +typedef struct __jmp_buf jmp_buf[1]; + +#endif /* _SETJMP_H */ diff --git a/usr/include/arch/m32r/kli...

...uapi -Iarch/arm64/include/generated/uapi -I./include/uapi -Iinclude/generated/uapi -include ./include/linux/kconfig.h -D__KERNEL__ -mlittle-endian -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -std=gnu89 -mgeneral-regs-only -fno-asynchronous-unwind-tables -fno-delete-null-pointer-checks -O2 --param=allow-store-data-races=0 -Wframe-larger-than=2048 -fno-stack-protector -Wno-unused-but-set-variable -fno-omit-frame-pointer -fno-optimize-sibling-calls -fno-var-tracking-assignments -g -Wdeclaration-after-statemen...

...MCONFIG | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/usr/klibc/arch/arm64/MCONFIG b/usr/klibc/arch/arm64/MCONFIG index e31ffea79b0b9..f8741ff295d5c 100644 --- a/usr/klibc/arch/arm64/MCONFIG +++ b/usr/klibc/arch/arm64/MCONFIG @@ -20,7 +20,7 @@ KLIBCREQFLAGS += -fno-exceptions -mgeneral-regs-only # On arm64, binaries are normally loaded at 4MB. Place klibc.so # a little before that at 2MB to prevent overlap. -KLIBCSHAREDFLAGS = -Ttext-segment 0x0200000 +KLIBCSHAREDFLAGS = $(LD_IMAGE_BASE_OPT) 0x0200000 # Kernel has never used stack trampolines KLIBCEXECSTACK := n -- 2.28....

...; in them, they should be named fields in the structure, not an array. > > Do we need the fpregs saved even though klibc doesn't do fp? > I agree about the named struct, it looks a lot nicer, thanks. For gcc targetting Aarch64, We can only guarantee that d8-d15 are left alone when -mgeneral-regs-only is supplied for building klibc and any software linked against klibc. I would much prefer to save/restore d8-d15 to avoid potential future headaches. > > diff --git a/usr/klibc/arch/arm64/pipe.c b/usr/klibc/arch/arm64/pipe.c > > new file mode 100644 > > index 0000000..f...

...nd that there is no fp support in klibc. Thorsten Glaser <tg at mirbsd.de> wrote: >Steve Capper dixit: > >>> Do we need the fpregs saved even though klibc doesn't do fp? > >>For gcc targetting Aarch64, We can only guarantee that d8-d15 are >>left alone when -mgeneral-regs-only is supplied for building klibc >>and any software linked against klibc. I would much prefer to > >We can enforce this in klcc, just like -mregparm=3 is used by >the i386 target. > >bye, >//mirabilos -- Sent from my mobile phone. Please pardon brevity and lack of...

Specifying -no-implicit-float prevents LLVM from using non-GPR registers for purely integer operations. This is useful for operating systems (such as Wind River's VxWorks) that support tasks that do not save all registers on context switch. This presents an interesting problem for variadic functions that may optionally take non-integer arguments (e.g. printf style functions). Should non-GPR

...". Whether or not simd (neon) is enabled by default is an orthogonal issue. We plan on implementing this interface for AArch64 Clang in future, and completely dropping the current support for -mfpu. This means that -march will become the preferred way to specify the target CPU/architecture. -mgeneral-regs-only: This is an option currently supported by AArch64 GCC, which we would also like to implement. Passing this option to the compiler should ensure that it generates code which only uses the general purpose registers. Otherwise, the compiler should throw an error. Is this proposal reasonable...

...KLIBC_ASMMACROS_H */ diff --git a/usr/klibc/arch/arm64/setjmp.S b/usr/klibc/arch/arm64/setjmp.S index 13ab99d..284e328 100644 --- a/usr/klibc/arch/arm64/setjmp.S +++ b/usr/klibc/arch/arm64/setjmp.S @@ -4,8 +4,6 @@ # setjmp/longjmp for arm64 # -#include <klibc/asmmacros.h> - # we specify -mgeneral-regs-only as a build flag thus do not need to # save d8-d15 diff --git a/usr/klibc/arch/arm64/sysstub.ph b/usr/klibc/arch/arm64/sysstub.ph index 47cbfd9..095b1e8 100644 --- a/usr/klibc/arch/arm64/sysstub.ph +++ b/usr/klibc/arch/arm64/sysstub.ph @@ -10,7 +10,6 @@ sub make_sysstub($$$$$@) {...

.../generated/uapi -I./include/uapi > -Iinclude/generated/uapi -include ./include/linux/kconfig.h > -D__KERNEL__ -mlittle-endian -Wall -Wundef -Wstrict-prototypes > -Wno-trigraphs -fno-strict-aliasing -fno-common > -Werror-implicit-function-declaration -Wno-format-security -std=gnu89 > -mgeneral-regs-only -fno-asynchronous-unwind-tables > -fno-delete-null-pointer-checks -O2 --param=allow-store-data-races=0 > -Wframe-larger-than=2048 -fno-stack-protector > -Wno-unused-but-set-variable -fno-omit-frame-pointer > -fno-optimize-sibling-calls -fno-var-tracking-assignments -g > -Wd...

Hello, Here is V2 of the arm64 support for klibc patch set. Notable changes since the original series: * fp regs dropped from setjmp/longjmp * chmod, lstat and stat re-implemented with *at functions. * open64 merged into open. As with the original, this series is to be applied against the latest klibc, just after 25a66fa README.klibc: update build information V2 has been tested on x86_64

...G_AEABI),y) KLIBCREQFLAGS += -mabi=aapcs-linux -mno-thumb-interwork else diff --git a/usr/klibc/arch/arm64/MCONFIG b/usr/klibc/arch/arm64/MCONFIG index 82664a7..6d22847 100644 --- a/usr/klibc/arch/arm64/MCONFIG +++ b/usr/klibc/arch/arm64/MCONFIG @@ -20,4 +20,4 @@ KLIBCREQFLAGS += -fno-exceptions -mgeneral-regs-only # On arm64, binaries are normally loaded at 4MB. Place klibc.so # a little before that at 2MB to prevent overlap. -KLIBCSHAREDFLAGS = -Ttext 0x0200000 +KLIBCSHAREDFLAGS = -Ttext-segment 0x0200000 diff --git a/usr/klibc/arch/m68k/MCONFIG b/usr/klibc/arch/m68k/MCONFIG index 7d4615d..3f4...

...9;re having explicit checks for no SSE2 cases here and there, so -mno-sse / -mkernel / -msoft-float is unaffected 4. The use of floating point arguments should be considered as "explicit use FP" and therefore should be allowed by the spirit of -mno-implicit-float flag 5. We're having -mgeneral-regs-only these days, so more strict cases are properly recognized as well. Any thoughts? PS: This fixes PR36507 and the fix is in https://reviews.llvm.org/D62639 On Mon, Nov 27, 2017 at 6:01 PM Nasser, Salim via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > Specifying -no-implici...

...dicated page or vDSO for signal return since 2.6.13 +KLIBCEXECSTACK := n diff --git a/usr/klibc/arch/arm64/MCONFIG b/usr/klibc/arch/arm64/MCONFIG index 6d22847e..e31ffea7 100644 --- a/usr/klibc/arch/arm64/MCONFIG +++ b/usr/klibc/arch/arm64/MCONFIG @@ -21,3 +21,6 @@ KLIBCREQFLAGS += -fno-exceptions -mgeneral-regs-only # On arm64, binaries are normally loaded at 4MB. Place klibc.so # a little before that at 2MB to prevent overlap. KLIBCSHAREDFLAGS = -Ttext-segment 0x0200000 + +# Kernel has never used stack trampolines +KLIBCEXECSTACK := n diff --git a/usr/klibc/arch/i386/MCONFIG b/usr/klibc/arch/i386...

On 9 July 2016 at 08:30, Markus Mayer <markus.mayer at broadcom.com> wrote: > On 9 July 2016 at 05:04, Luis de Bethencourt <luisbg at osg.samsung.com> wrote: >> On 08/07/16 23:43, Markus Mayer wrote: >>> Add a collection of generic functions to convert strings to lowercase >>> or uppercase. >>> >>> Changing the case of a string (with or

This reverts commit 9d8d648e604026b32cad00a84ed6c29cbd157641, which broke signal handing on some architectures. On m68k and parisc, signal return depends on a trampoline that the kernel writes on the stack. On alpha, s390, and sparc (32-bit), we can avoid this by providing our own function as sa_restorer, but we currently don't. Signed-off-by: Ben Hutchings <ben at decadent.org.uk>

...) KLIBCREQFLAGS += -mabi=aapcs-linux -mno-thumb-interwork else diff --git a/usr/klibc/arch/arm64/MCONFIG b/usr/klibc/arch/arm64/MCONFIG index 6d22847e670a..7ab22282ebbe 100644 --- a/usr/klibc/arch/arm64/MCONFIG +++ b/usr/klibc/arch/arm64/MCONFIG @@ -20,4 +20,4 @@ KLIBCREQFLAGS += -fno-exceptions -mgeneral-regs-only # On arm64, binaries are normally loaded at 4MB. Place klibc.so # a little before that at 2MB to prevent overlap. -KLIBCSHAREDFLAGS = -Ttext-segment 0x0200000 +KLIBCSHAREDFLAGS = $(IMAGE_BASE) 0x0200000 diff --git a/usr/klibc/arch/i386/MCONFIG b/usr/klibc/arch/i386/MCONFIG index 07046...

...linkflags when building the shared version of the library # This address needs to be reachable using normal inter-module # calls, and work on the memory models for this architecture -#KLIBCSHAREDFLAGS = -Ttext 0x01800200 +KLIBCSHAREDFLAGS = -Ttext 0x01800200 #KLIBCREQFLAGS += #KLIBCOPTFLAGS += -mgeneral-regs-only diff --git a/usr/klibc/arch/aarch64/crt0.S b/usr/klibc/arch/aarch64/crt0.S index b0de74b..21406c2 100644 --- a/usr/klibc/arch/aarch64/crt0.S +++ b/usr/klibc/arch/aarch64/crt0.S @@ -13,7 +13,7 @@ .globl _start _start: - stp x29, x30, [sp, -16]! - mov x29, sp + mov...