[PATCH v5 4/5] x86-64: Add the clone3 wrapper

H.J. Lu hjl.tools@gmail.com
Sat May 22 01:52:36 GMT 2021


On Thu, May 20, 2021 at 11:39 AM Noah Goldstein <goldstein.w.n@gmail.com> wrote:
>
> On Thu, May 20, 2021 at 2:35 PM Noah Goldstein <goldstein.w.n@gmail.com> wrote:
> >
> > On Sat, May 15, 2021 at 9:23 AM H.J. Lu via Libc-alpha
> > <libc-alpha@sourceware.org> wrote:
> > >
> > > extern int clone3 (struct clone_args *__cl_args,
> > >                    int (*__func) (void *__arg), void *__arg);
> > > ---
> > >  sysdeps/unix/sysv/linux/x86_64/clone3.S | 92 +++++++++++++++++++++++++
> > >  sysdeps/unix/sysv/linux/x86_64/sysdep.h |  2 +
> > >  2 files changed, 94 insertions(+)
> > >  create mode 100644 sysdeps/unix/sysv/linux/x86_64/clone3.S
> > >
> > > diff --git a/sysdeps/unix/sysv/linux/x86_64/clone3.S b/sysdeps/unix/sysv/linux/x86_64/clone3.S
> > > new file mode 100644
> > > index 0000000000..f7d4036a6a
> > > --- /dev/null
> > > +++ b/sysdeps/unix/sysv/linux/x86_64/clone3.S
> > > @@ -0,0 +1,92 @@
> > > +/* The clone3 syscall wrapper.  Linux/x86-64 version.
> > > +   Copyright (C) 2021 Free Software Foundation, Inc.
> > > +   This file is part of the GNU C Library.
> > > +
> > > +   The GNU C Library is free software; you can redistribute it and/or
> > > +   modify it under the terms of the GNU Lesser General Public
> > > +   License as published by the Free Software Foundation; either
> > > +   version 2.1 of the License, or (at your option) any later version.
> > > +
> > > +   The GNU C Library is distributed in the hope that it will be useful,
> > > +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> > > +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> > > +   Lesser General Public License for more details.
> > > +
> > > +   You should have received a copy of the GNU Lesser General Public
> > > +   License along with the GNU C Library; if not, see
> > > +   <https://www.gnu.org/licenses/>.  */
> > > +
> > > +/* clone3() is even more special than fork() as it mucks with stacks
> > > +   and invokes a function in the right context after its all over.  */
> > > +
> > > +#include <sysdep.h>
> > > +#include <clone-offsets.h>
> > > +
> > > +/* The userland implementation is:
> > > +   int clone3 (struct clone_args *cl_args, int (*func)(void *arg),
> > > +              void *arg);
> > > +   the kernel entry is:
> > > +   int clone3 (struct clone_args *cl_args, size_t size);
> > > +
> > > +   The parameters are passed in registers from userland:
> > > +   rdi: cl_args
> > > +   rsi: func
> > > +   rdx: arg
> > > +
> > > +   The kernel expects:
> > > +   rax: system call number
> > > +   rdi: cl_args
> > > +   rsi: size  */
> > > +
> > > +        .text
> > > +ENTRY (__clone3)
> > > +       /* Sanity check arguments.  */
> > > +       movq    $-EINVAL, %rax
> >
> > Can this be movl?

Yes.  Fixed.

> > > +       testq   %rdi, %rdi              /* No NULL cl_args pointer.  */
> > > +       jz      SYSCALL_ERROR_LABEL
> > > +       testq   %rsi, %rsi              /* No NULL function pointer.  */
> > > +       jz      SYSCALL_ERROR_LABEL
> > > +
> > > +       /* Save the function pointer in R8 which is preserved by the
> > > +          syscall.  */
> > > +       movq    %rsi, %r8
> > > +
> > > +       /* Put sizeof (struct clone_args) in ESI.  */
> > > +       movl    $CLONE_ARGS_SIZE , %esi
> > > +
> > > +       /* Do the system call.  */
> > > +       movl    $SYS_ify(clone3), %eax
> > > +
> > > +       /* End FDE now, because in the child the unwind info will be
> > > +          wrong.  */
> > > +       cfi_endproc
> > > +       syscall
> > > +
> > > +       test    %RAX_LP, %RAX_LP
> > > +       jl      SYSCALL_ERROR_LABEL
> > > +       jz      L(thread_start)
> > > +
> >
> > Is expectation to go to L(thread_start)? If so
> > think jnz L(ret) and fallthrough is probably
> > better.
>
> Or better take the error check branch off
> the critical path with jnz L(error_or_ret) then jl
> in L(error_or_ret)

I don't think the clone wrapper is on the critical path.
Since the same code is executed by both child and parent.
I check the error return first.

> >
> > > +       ret
> > > +
> > > +L(thread_start):
> > > +       cfi_startproc
> > > +       /* Clearing frame pointer is insufficient, use CFI.  */
> > > +       cfi_undefined (rip)
> > > +       /* Clear the frame pointer.  The ABI suggests this be done, to mark
> > > +          the outermost frame obviously.  */
> > > +       xorl    %ebp, %ebp
> > > +
> > > +       /* Set up arguments for the function call.  */
> > > +       movq    %rdx, %rdi      /* Argument.  */
> > > +       call    *%r8            /* Call function.  */
> > > +       /* Call exit with return value from function call. */
> > > +       movq    %rax, %rdi
> > > +       movl    $SYS_ify(exit), %eax
> > > +       syscall
> > > +       cfi_endproc
> > > +
> > > +       cfi_startproc
> > > +PSEUDO_END (__clone3)
> > > +
> > > +libc_hidden_def (__clone3)
> > > +weak_alias (__clone3, clone3)
> > > diff --git a/sysdeps/unix/sysv/linux/x86_64/sysdep.h b/sysdeps/unix/sysv/linux/x86_64/sysdep.h
> > > index dbad2c788a..f26ffc68ae 100644
> > > --- a/sysdeps/unix/sysv/linux/x86_64/sysdep.h
> > > +++ b/sysdeps/unix/sysv/linux/x86_64/sysdep.h
> > > @@ -377,6 +377,8 @@
> > >  # define HAVE_GETCPU_VSYSCALL          "__vdso_getcpu"
> > >  # define HAVE_CLOCK_GETRES64_VSYSCALL   "__vdso_clock_getres"
> > >
> > > +# define HAVE_CLONE3_WAPPER                    1
> > > +
> > >  # define SINGLE_THREAD_BY_GLOBAL               1
> > >
> > >  #endif /* __ASSEMBLER__ */
> > > --
> > > 2.31.1
> > >

Thanks.

-- 
H.J.


More information about the Libc-alpha mailing list