similar to: Fix segfault in new exit code on parisc, klibc-1.1.3

The PA-RISC ELF ABI puts the function pointer to be registered with atexit in %r23. Use this instead of passing in NULL. While it's unlikely to be used, doing the right thing is good. Signed-off-by: Kyle McMartin <kyle@parisc-linux.org> diff --git a/klibc/arch/parisc/crt0.S b/klibc/arch/parisc/crt0.S index 7428443..a9b6e96 100644 --- a/klibc/arch/parisc/crt0.S +++

Implement "pid_t vfork(void)" for parisc. Signed-off-by: Kyle McMartin <kyle at parisc-linux.org> --- Ugh. vfork() me harder. Kbuild | 2 +- vfork.S | 31 +++++++++++++++++++++++++++++++ 2 files changed, 32 insertions(+), 1 deletion(-) diff --git a/usr/klibc/arch/parisc/Kbuild b/usr/klibc/arch/parisc/Kbuild index d57a873..57ca5c2 100644 --- a/usr/klibc/arch/parisc/Kbuild

Reimplement __common_syscall in assembler. The 32-bit ABI says that the fifth and sixth arguments to the function are passed on the stack, but our syscall ABI says they are passed in %arg4 and %arg5 like the 64-bit ABI. Also, tried to optimize the code slightly by making use of the cmpb delay slot to load the errno return of -1. Signed-off-by: Kyle McMartin <kyle@parisc-linux.org> diff

The parts of klibc specific to the parisc architecture. Signed-off-by: H. Peter Anvin <hpa at zytor.com> --- commit 078d6614054391efe17093f8d70340e2c0644ffb tree 63a4bf899e5ca2ef3c0a8e9ef3098273012f7a33 parent ebd2860ad3dc19cb11fd5b9cc235cab54e9165f4 author H. Peter Anvin <hpa at zytor.com> Sun, 25 Jun 2006 16:58:36 -0700 committer H. Peter Anvin <hpa at zytor.com> Sun, 25 Jun

Hello Kyle, May be out of context but before I forget: > > + > > + /* store %r22 to errno... */ > > + ldil L%errno,%r1 > > + ldo R%errno(%r1),%r1 Could it not be 'load32 errno, %r1' (the assembly.h macro)? Cheers, Joel ---------- Club Scarlet : Tout le monde gagne! Si vous devenez aujourd'hui Scarlet One grace a un client existant de Scarlet, vous

The iitlbp instruction needs the space register explicitly specified with newer toolchains. Specify sr0, but it doesn't really matter as the insn is just used to crash the program. Signed-off-by: Kyle McMartin <kyle@parisc-linux.org> diff --git a/klibc/arch/parisc/crt0.S b/klibc/arch/parisc/crt0.S index fb0bd37..7adc6c4 100644 --- a/klibc/arch/parisc/crt0.S +++

I’m not sure if it’s truly deterministic. It always gives the same results (so far) on my machine but I’m not sure that’s enough. My guess is it’s probably going to be deterministic on one machine but might well not be deterministic across environments. Like it might give varying results if cross compiled on different hosts, macOS vs intel Linux vs arm vs s390. (Obviously AVR is always a cross

We have a test that looks like this… define void @array16_store() { ; CHECK-LABEL: array16_store: ; CHECK: ldi [[REG1:r[0-9]+]], 204 ; CHECK: ldi [[REG2:r[0-9]+]], 170 ; CHECK: sts int.array+3, [[REG2]] ; CHECK: sts int.array+2, [[REG1]] ; CHECK: ldi [[REG1:r[0-9]+]], 187 ; CHECK: ldi [[REG2:r[0-9]+]], 170 ; CHECK: sts int.array+1, [[REG2]] ; CHECK: sts int.array, [[REG1]] ; CHECK: ldi

On PA-RISC machines running in 64-bit mode, uname -m returns `parisc64' instead of `parisc.' Thus, we must sed this to match like a few other platforms. From: Kyle McMartin <kyle@parisc-linux.org> Signed-off-by: Kyle McMartin <kyle@parisc-linux.org> diff --git a/Makefile b/Makefile index 2b057b8..16a026d 100644 --- a/Makefile +++ b/Makefile @@ -14,7 +14,7 @@ include

I'm a new LLVM developer contributing patches for the AVR platform and I'm trying to understand which parts of the code base are malfunctioning in my case. This LLVM IR... define hidden i32 @setServoAngle3(i32) local_unnamed_addr { entry: %1 = call i32 @myExternalFunction1(i32 %0, i32 119) ret i32 %1 } declare i32 @myExternalFunction1(i32, i32) Is being lowered to this assembly

Commit-ID: b71dd57f6a784962681ac05aa686b28db8668609 Gitweb: http://git.kernel.org/?p=libs/klibc/klibc.git;a=commit;h=b71dd57f6a784962681ac05aa686b28db8668609 Author: Ben Hutchings <ben at decadent.org.uk> AuthorDate: Mon, 21 Jan 2019 03:55:35 +0000 Committer: Ben Hutchings <ben at decadent.org.uk> CommitDate: Mon, 21 Jan 2019 03:55:35 +0000 [klibc] parisc: Fix vfork() The

Got this bug report for Ubuntu's klibc: https://launchpad.net/distros/ubuntu/+source/klibc/+bug/37528 > In the fstype tool there is a problem if the lower 16 bit of number of > free inodes in a ext3 filesystem happens to match minix filesystem > magic. > > This will cause the system not to boot since ubuntu will attempt to > mount the filesystem as a minix filesystem. I

Hi Dear All, This is a function which contains Covariance Ratio and Likelihood Distance values (CVRi, LDi). i want to compute the all row's values, that is run this function for nrow(X) times. The X and Y matrices are;

The SVR4 ABI PPC supplement isn't considered canonical on Linux. Instead this arch uses the LSB ABI documents. r7 is not guaranteed to be initialised to any particular value, and the termination function pointer is not provided at all. I still have a couple more archs to test, don't rush out 1.1.5 yet. ;) --- klibc-1.1.3.orig/klibc/arch/ppc/crt0.S +++ klibc-1.1.3/klibc/arch/ppc/crt0.S

+Lang Hames <lhames at gmail.com>, Master Regent of the Three <No, Two sir> JITs On Thu, Jul 28, 2016 at 12:31 PM Sean Ogden via llvm-dev < llvm-dev at lists.llvm.org> wrote: > I needed to be able to update stub pointers for hot functions that get > recompiled in a lazy JIT that uses CompileOnDemandLayer. In order to do > this I added a method that allows pointers to

I noticed the following this week. No idea how long its been happening because I have not tested deleting folders hardly at all. Steps to reproduce: 1. Create a folder "NewFolder", click on it in thunderbird so an index dir is created: ls -ldi Maildir/.NewFolder/ 1504062 drwx------ 5 mcdouga9 egrstaff 4096 Nov 16 19:44 Maildir/.NewFolder/ ls -ldi

I needed to be able to update stub pointers for hot functions that get recompiled in a lazy JIT that uses CompileOnDemandLayer. In order to do this I added a method that allows pointers to be updated but does not expose any of the other internals of the COD layer. Does anyone have a cleaner way to do this? Has something to facilitate this already been added? Would it be possible to merge this

Hi all, I am working on an llvm backend for a processor with a relative simple instruction set. For small loops, the code that is produced depends heavily on how the loop is specified: The less information we provide to clang, the better the loop code becomes... Any idea how I can learn llvm that we don't have load/store instructions with register index, so that it is more efficient to

Hi, there I am writing an backend, and I met a problem. We don't have load/store instructions for vector predicate registers(vpr for short). The hardware has 64 vector registers(vr for short) and 8 vector predicate registers. And there is no move instructions between vr and vpr. vr supports many operations, and vpr supports vpror, vprxor, vprand and vprinv operations. A vr has 512 bits, and

It does work. I just tested it on my JIT. Thanks! As for the part that couples them too tightly, would you recommend I just keep my own specialized version of CompileOnDemandLayer.h that includes this functionality, or do you have any ideas for a cleaner way to do this? I've noticed a couple of people asking for support for updating stub pointers for functions that are optimized at runtime,