search for: __udivsi3

...s! __divsi3 (18 instructions) perform a DIV after 2 calls of abs(), plus a final negation, instead of just a single IDIV __modsi3 (14 instructions) calls __divsi3 (18 instructions) __divmodsi4 (17 instructions) calls __divsi3 (18 instructions) __udivsi3 (52 instructions) does NOT use DIV, but performs BITWISE division using shifts and additions! __umodsi3 (14 instructions) calls __udivsi3 (52 instructions) __udivmodsi4 (17 instructions) calls __udivsi3 (52 instructions) __muldi3 (41 instructions) performs a "long&...

...calls of abs(), >> plus a final negation, instead of just >> a single IDIV >> __modsi3 (14 instructions) calls __divsi3 (18 instructions) >> __divmodsi4 (17 instructions) calls __divsi3 (18 instructions) >> >> __udivsi3 (52 instructions) does NOT use DIV, but performs BITWISE >> division using shifts and additions! >> __umodsi3 (14 instructions) calls __udivsi3 (52 instructions) >> __udivmodsi4 (17 instructions) calls __udivsi3 (52 instructions) >> >> __mul...

.../klibc/libgcc/__udivmodsi4.c 2002-09-16 09:45:47.000000000 -0700 +++ klibc-0.114/klibc/libgcc/__udivmodsi4.c 2004-02-22 04:39:00.000000000 -0800 @@ -1,3 +1,7 @@ +/* + * libgcc/__udivmodsi4.c + */ + #include <klibc/diverr.h> #include <stdint.h> diff -ruN klibc-0.114_orig/klibc/libgcc/__udivsi3.c klibc-0.114/klibc/libgcc/__udivsi3.c --- klibc-0.114_orig/klibc/libgcc/__udivsi3.c 2002-09-16 09:45:47.000000000 -0700 +++ klibc-0.114/klibc/libgcc/__udivsi3.c 2004-02-22 04:39:00.000000000 -0800 @@ -1,5 +1,5 @@ /* - * libgcc/__divsi3.c + * libgcc/__udivsi3.c */ #include <stdint.h> di...

...tor.com|ChangeSet|20020723064121|00000|f8a756c36aa22cad C c Add a bunch of libgcc objects. K 41984 O -rw-rw-r-- P klibc/arch/ia64/Makefile.inc ------------------------------------------------ D13 1 I13 9 arch/$(ARCH)/pipe.o \ libgcc/__divdi3.o \ libgcc/__divsi3.o \ libgcc/__udivdi3.o \ libgcc/__udivsi3.o \ libgcc/__umodsi3.o \ libgcc/__umoddi3.o \ libgcc/__udivmodsi4.o \ libgcc/__udivmoddi4.o # Patch checksum=c61569d9 -------------- next part -------------- A non-text attachment was scrubbed... Name: klibc-libgcc-fixes.diff Type: text/x-patch Size: 1114 bytes Desc: not available Url : http:/...

Hi everyone, I have profiled Opus on AArch64. I just run opus_demo with some pcm files. Following is time proportion of FFT with different bitrate. Bitrate | Time cost by FFT/iFFT 24kb/s | 15% 48kb/s | 15% 96kb/s | 13% Any comment? I want some data close to real application, any suggestion? Thanks, Phil Wang -------------- next part -------------- An HTML attachment was scrubbed... URL:

> Looks as though whomever implemented the call to __aeabi_idiv0 wanted > to be conservative for non EABI targets. How could it prevent him from providing default implementation of __aeabi_idiv0() for EABI targets? > AFAIK, gnueabi targets recognize all EABI functions, so that should > work well. Not sure I understand you, nothing in compiler-rt defines these functions, they are

...ld +++ b/usr/klibc/arch/ia64/Kbuild @@ -2,7 +2,7 @@ # klibc files for ia64 # -klib-y := vfork.o setjmp.o pipe.o syscall.o +klib-y := vfork.o setjmp.o sigaction.o pipe.o syscall.o klib-y += ../../libgcc/__divdi3.o ../../libgcc/__divsi3.o klib-y += ../../libgcc/__udivdi3.o ../../libgcc/__udivsi3.o diff --git a/usr/klibc/arch/ia64/sigaction.c b/usr/klibc/arch/ia64/sigaction.c new file mode 100644 index 00000000..c1ea5fdd --- /dev/null +++ b/usr/klibc/arch/ia64/sigaction.c @@ -0,0 +1,50 @@ +/* + * sigaction.c + */ + +#include <signal.h> +#include <sys/syscall.h> + +/* We use -mno...

...ld +++ b/usr/klibc/arch/ia64/Kbuild @@ -2,7 +2,7 @@ # klibc files for ia64 # -klib-y := vfork.o setjmp.o pipe.o syscall.o +klib-y := vfork.o setjmp.o sigaction.o pipe.o syscall.o klib-y += ../../libgcc/__divdi3.o ../../libgcc/__divsi3.o klib-y += ../../libgcc/__udivdi3.o ../../libgcc/__udivsi3.o diff --git a/usr/klibc/arch/ia64/sigaction.c b/usr/klibc/arch/ia64/sigaction.c new file mode 100644 index 00000000..e512bdcd --- /dev/null +++ b/usr/klibc/arch/ia64/sigaction.c @@ -0,0 +1,50 @@ +/* + * sigaction.c + */ + +#include <signal.h> +#include <sys/syscall.h> + +/* We use -mno...

...t (long a, long b) > { > int neg = 0; > long res; > > if (a < 0) > { > a = -a; > neg = 1; > } > > if (b < 0) > { > b = -b; > neg = !neg; > } > > res = a*b; //(unsigned long)a / (unsigned long)b; // will call __udivsi3 > > if (neg) > res = -res; > > return res; > } > > > This gets compiled into > > ; Function Attrs: norecurse nounwind readnone > define dso_local i32 @test(i32 %a, i32 %b) local_unnamed_addr #0 { > entry: > %cmp = icmp slt i32 %a, 0 > %sub =...

Changing the order of the checks in CodeGenPrepare::optimizeSelectInst() sounds good to me. But you may need to go further for optimum performance. For example, we may be canonicalizing math/logic IR patterns into 'select' such as in the recent: https://reviews.llvm.org/D67799 So if you want those to become ALU ops again rather than branches, then you need to do the transform later in

This fixes a long standing issue where it was not possible to do "make usr/klibc/arch/x86_64/longjmp.o" in the kernel. The principle is that all .o files to be part of klibc are listed with klib-y. For each directory a klib.list file is made that specify all .o file and the final AR then adds all .o files to create libc.a. This patch introduce the infrastructure and converts x86_64 to

...> long test (long a, long b) > { > int neg = 0; > long res; > > if (a < 0) > { > a = -a; > neg = 1; > } > > if (b < 0) > { > b = -b; > neg = !neg; > } > > res = a*b; //(unsigned long)a / (unsigned long)b; // will call __udivsi3 > > if (neg) > res = -res; > > return res; > } > > > This gets compiled into > > ; Function Attrs: norecurse nounwind readnone > define dso_local i32 @test(i32 %a, i32 %b) local_unnamed_addr #0 { > entry: > %cmp = icmp slt i32 %a, 0 > %sub = sub n...

...not clang. > > long test (long a, long b) > { > int neg = 0; > long res; > > if (a < 0) > { > a = -a; > neg = 1; > } > > if (b < 0) > { > b = -b; > neg = !neg; > } > > res = a*b; //(unsigned long)a / (unsigned long)b; // will call __udivsi3 > > if (neg) > res = -res; > > return res; > } > > > This gets compiled into > > ; Function Attrs: norecurse nounwind readnone > define dso_local i32 @test(i32 %a, i32 %b) local_unnamed_addr #0 { > entry: > %cmp = icmp slt i32 %a, 0 > %sub = sub nsw i32...

...included from the main Makefile, and that pathnames should be +# accordingly. +# + +KLIBCARCHOBJS = \ + arch/$(KLIBCARCH)/vfork.o \ + arch/$(KLIBCARCH)/setjmp.o \ + arch/$(KLIBCARCH)/pipe.o \ + arch/$(KLIBCARCH)/syscall.o \ + libgcc/__divdi3.o \ + libgcc/__divsi3.o \ + libgcc/__udivdi3.o \ + libgcc/__udivsi3.o \ + libgcc/__umodsi3.o \ + libgcc/__umoddi3.o \ + libgcc/__udivmodsi4.o \ + libgcc/__udivmoddi4.o + +KLIBCARCHSOOBJS = $(patsubst %o,%.lo,%(KLIBCARCHOBJS)) + +archclean: diff --git a/usr/klibc/arch/ia64/crt0.S b/usr/klibc/arch/ia64/crt0.S new file mode 100644 index 0000000..722276e --- /dev/null...

...if (a < 0) >>> { >>> a = -a; >>> neg = 1; >>> } >>> >>> if (b < 0) >>> { >>> b = -b; >>> neg = !neg; >>> } >>> >>> res = a*b; //(unsigned long)a / (unsigned long)b; // will call __udivsi3 >>> >>> if (neg) >>> res = -res; >>> >>> return res; >>> } >>> >>> >>> This gets compiled into >>> >>> ; Function Attrs: norecurse nounwind readnone >>> define dso_local i32 @test(i32 %...

...t; a = -a; >>>> neg = 1; >>>> } >>>> >>>> if (b < 0) >>>> { >>>> b = -b; >>>> neg = !neg; >>>> } >>>> >>>> res = a*b; //(unsigned long)a / (unsigned long)b; // will call __udivsi3 >>>> >>>> if (neg) >>>> res = -res; >>>> >>>> return res; >>>> } >>>> >>>> >>>> This gets compiled into >>>> >>>> ; Function Attrs: norecurse nounwind readnone &...

Hello everyone, this is the fourth version of the patch series that introduces ARMv7 with virtualization extensions support in Xen. The series allows Xen and Dom0 to boot on a Cortex-A15 based Versatile Express simulator. See the following announce email for more informations about what we are trying to achieve, as well as the original git history: See

Hello everyone, this is the very first version of the patch series that introduces ARMv7 with virtualization extensions support in Xen. The series allows Xen and Dom0 to boot on a Cortex-A15 based Versatile Express simulator. See the following announce email for more informations about what we are trying to achieve, as well as the original git history: See

I have updated the klibc patchset based on feedback received. In particular, the patchset has been reorganized so as not to break git-bisect. Additionally, this updates the patch base to 2.6.17-git12 (d38b69689c349f35502b92e20dafb30c62d49d63) and klibc 1.4.8; the main difference on the klibc side is removal of obsolete code. This is also available as a git tree at: