From: Szabolcs Nagy <szabolcs.nagy@arm.com>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Cc: libc-alpha@sourceware.org, Christian Brauner <brauner@kernel.org>
Subject: Re: [PATCH v2 4/9] aarch64: Add the clone3 wrapper
Date: Wed, 2 Nov 2022 12:12:51 +0000 [thread overview]
Message-ID: <Y2Jew7cPZqOg6dzL@arm.com> (raw)
In-Reply-To: <20220930192613.3491147-5-adhemerval.zanella@linaro.org>
The 09/30/2022 16:26, Adhemerval Zanella via Libc-alpha wrote:
> It follow the internal signature:
>
> extern int clone3 (struct clone_args *__cl_args, size_t __size,
> int (*__func) (void *__arg), void *__arg);
>
> And x86_64 semantics to return EINVAL if either cl_args or func
> is NULL. The stack is 16-byte aligned prior executing func.
"x86_64 semantics" sounds wrong: maybe this should be documented?
i'd expect 0 cl_args/func to be UB like in most posix apis.
and aligning sp in the child fails if signals are allowed there
(pthreads does not allow signals now, direct callers might).
i dont know if that's a concert (or if unaligned stack is
something we should fix up in clone3).
>
> Checked on aarch64-linux-gnu.
> ---
> sysdeps/unix/sysv/linux/aarch64/clone3.S | 90 ++++++++++++++++++++++++
> sysdeps/unix/sysv/linux/aarch64/sysdep.h | 2 +
> 2 files changed, 92 insertions(+)
> create mode 100644 sysdeps/unix/sysv/linux/aarch64/clone3.S
>
> diff --git a/sysdeps/unix/sysv/linux/aarch64/clone3.S b/sysdeps/unix/sysv/linux/aarch64/clone3.S
> new file mode 100644
> index 0000000000..dba93430eb
> --- /dev/null
> +++ b/sysdeps/unix/sysv/linux/aarch64/clone3.S
> @@ -0,0 +1,90 @@
> +/* The clone3 syscall wrapper. Linux/aarch64 version.
> + Copyright (C) 2022 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/>. */
> +
> +#include <sysdep.h>
> +#define _ERRNO_H 1
> +#include <bits/errno.h>
> +
> +/* The userland implementation is:
> + int clone3 (struct clone_args *cl_args, size_t size,
> + 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:
> + x0: cl_args
> + x1: size
> + x2: func
> + x3: arg */
> +
> + .text
> +ENTRY(__clone3)
> + PTR_ARG (0)
> + PTR_ARG (1)
> + PTR_ARG (3)
> + PTR_ARG (4)
> + /* Save args for the child. */
> + mov x10, x0 /* cl_args */
> + mov x11, x2 /* func */
> + mov x12, x3 /* args */
> +
> + /* Sanity check args. */
> + mov x0, #-EINVAL
> + cbz x10, .Lsyscall_error /* No NULL cl_args pointer. */
> + cbz x2, .Lsyscall_error /* No NULL function pointer. */
> +
> + /* Do the system call, the kernel expects:
> + x8: system call number
> + x0: cl_args
> + x1: size */
> + mov x0, x10
> + mov x8, #SYS_ify(clone3)
> + svc 0x0
> +
> + cmp x0, #0
> + beq thread_start
> + blt .Lsyscall_error
> + RET
> +PSEUDO_END (__clone3)
> +
> + .align 4
> + .type thread_start, %function
> +thread_start:
> + cfi_startproc
> + cfi_undefined (x30)
> + mov x29, 0
> +
> + /* Align sp. */
> + mov x0, sp
> + and x0, x0, -16
> + mov sp, x0
> +
> + /* Pick the function arg and execute. */
> + mov x0, x12
> + blr x11
> +
> + /* We are done, pass the return value through x0. */
> + mov x8, #SYS_ify(exit)
> + svc 0x0
> + cfi_endproc
> + .size thread_start, .-thread_start
> +
> +libc_hidden_def (__clone3)
> +weak_alias (__clone3, clone3)
> diff --git a/sysdeps/unix/sysv/linux/aarch64/sysdep.h b/sysdeps/unix/sysv/linux/aarch64/sysdep.h
> index f1853e012f..42bb22f5e6 100644
> --- a/sysdeps/unix/sysv/linux/aarch64/sysdep.h
> +++ b/sysdeps/unix/sysv/linux/aarch64/sysdep.h
> @@ -164,6 +164,8 @@
> # define HAVE_CLOCK_GETTIME64_VSYSCALL "__kernel_clock_gettime"
> # define HAVE_GETTIMEOFDAY_VSYSCALL "__kernel_gettimeofday"
>
> +# define HAVE_CLONE3_WRAPPER 1
> +
> # undef INTERNAL_SYSCALL_RAW
> # define INTERNAL_SYSCALL_RAW(name, nr, args...) \
> ({ long _sys_result; \
> --
> 2.34.1
>
next prev parent reply other threads:[~2022-11-02 12:13 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-30 19:26 [PATCH v2 0/9] Optimize posix_spawn signal setup with clone3 Adhemerval Zanella
2022-09-30 19:26 ` [PATCH v2 1/9] linux: Do not reset signal handler in posix_spawn if it is already SIG_DFL Adhemerval Zanella
2023-01-11 21:06 ` Carlos O'Donell
2022-09-30 19:26 ` [PATCH v2 2/9] linux: Add clone3 CLONE_CLEAR_SIGHAND optimization to posix_spawn Adhemerval Zanella
2023-01-11 21:06 ` Carlos O'Donell
2022-09-30 19:26 ` [PATCH v2 3/9] powerpc64le: Add the clone3 wrapper Adhemerval Zanella
2022-09-30 19:26 ` [PATCH v2 4/9] aarch64: " Adhemerval Zanella
2022-11-02 12:12 ` Szabolcs Nagy [this message]
2022-11-03 13:15 ` Adhemerval Zanella Netto
2022-11-03 14:01 ` Szabolcs Nagy
2022-11-03 16:22 ` Adhemerval Zanella Netto
2022-11-03 16:31 ` Szabolcs Nagy
2022-11-03 16:39 ` Adhemerval Zanella Netto
2022-11-03 16:52 ` Szabolcs Nagy
2022-11-03 16:55 ` Adhemerval Zanella Netto
2022-11-03 20:55 ` H.J. Lu
2022-11-03 21:28 ` Adhemerval Zanella Netto
2022-11-03 21:22 ` Adhemerval Zanella Netto
2022-11-03 21:58 ` H.J. Lu
2022-11-04 12:32 ` Adhemerval Zanella Netto
2022-09-30 19:26 ` [PATCH v2 5/9] s390x: " Adhemerval Zanella
2022-09-30 19:26 ` [PATCH v2 6/9] riscv: " Adhemerval Zanella
2022-09-30 19:26 ` [PATCH v2 7/9] arm: " Adhemerval Zanella
2022-09-30 19:26 ` [PATCH v2 8/9] mips: " Adhemerval Zanella
2022-09-30 19:26 ` [PATCH v2 9/9] Linux: optimize clone3 internal usage Adhemerval Zanella
2023-01-11 21:12 ` Carlos O'Donell
2022-10-27 16:48 ` [PATCH v2 0/9] Optimize posix_spawn signal setup with clone3 Adhemerval Zanella Netto
2023-01-11 21:11 ` Carlos O'Donell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Y2Jew7cPZqOg6dzL@arm.com \
--to=szabolcs.nagy@arm.com \
--cc=adhemerval.zanella@linaro.org \
--cc=brauner@kernel.org \
--cc=libc-alpha@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).