From: Carlos O'Donell via Libc-alpha <libc-alpha@sourceware.org>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>,
libc-alpha@sourceware.org, Szabolcs Nagy <szabolcs.nagy@arm.com>,
"H . J . Lu" <hjl.tools@gmail.com>
Cc: Christian Brauner <brauner@kernel.org>
Subject: Re: [PATCH v2 02/11] linux: Extend internal clone3 documentation
Date: Wed, 11 Jan 2023 16:35:09 -0500 [thread overview]
Message-ID: <f5852ad3-1a4d-8e74-af9d-7cac69f86288@redhat.com> (raw)
In-Reply-To: <20221104190112.2566409-3-adhemerval.zanella@linaro.org>
On 11/4/22 15:01, Adhemerval Zanella via Libc-alpha wrote:
> Different than kernel, clone3 returns EINVAL for NULL struct
> clone_args or function pointer. This is similar to clone
> interface that return EINVAL for NULL function argument.
Please post v4.
Suggest:
Document that the internal __clone3 returns EINVAL for a NULL struct
clone_args or function pointer and that this differs from the kernel.
This behavior is similar to the clone interface that returns EINVAL
for a NULL function argument.
> It also clean up the Linux clone3.h interface, since it not
> currently exported.
Suggest:
Additionally cleanup the Linux clone3.h interface (not currently exported).
>
> Checked on x86_64-linux-gnu.
> ---
> include/clone_internal.h | 24 +++++++++++++++++++-----
> sysdeps/unix/sysv/linux/clone3.h | 10 +---------
> 2 files changed, 20 insertions(+), 14 deletions(-)
>
> diff --git a/include/clone_internal.h b/include/clone_internal.h
> index 4b23ef33ce..73b8114df4 100644
> --- a/include/clone_internal.h
> +++ b/include/clone_internal.h
> @@ -1,10 +1,24 @@
> -#ifndef _CLONE3_H
> -#include_next <clone3.h>
> +#ifndef _CLONE_INTERNAL_H
> +#define _CLONE_INTERNAL_H
>
> -extern __typeof (clone3) __clone3;
> +#include <clone3.h>
>
> -/* The internal wrapper of clone/clone2 and clone3. If __clone3 returns
> - -1 with ENOSYS, fall back to clone or clone2. */
> +/* The clone3 syscall provides a superset of the functionality of the clone
> + interface. The kernel might extend __CL_ARGS struct in the future, with
> + each version with a diffent __SIZE. If the child is created, it will
> + start __FUNC function with __ARG arguments.
OK.
> +
> + Different than kernel, the implementation also returns EINVAL for an
> + invalid NULL __CL_ARGS or __FUNC (similar to __clone).
Suggest:
Unlike the kernel the implementation returns EINVAL for an invalid NULL __CL_ARGS
or __FUNC (similar to __clone).
> +
This has nothing to do with the ABI in glibc so I'd avoid "ABI" here.
Suggest:
This function is only implemented if the machine defines HAVE_CLONE3_WRAPPER.
> + This function is only implemented if the ABI defines HAVE_CLONE3_WRAPPER.
> +*/
> +extern int __clone3 (struct clone_args *__cl_args, size_t __size,
> + int (*__func) (void *__arg), void *__arg);
> +
> +/* The internal wrapper of clone/clone2 and clone3. Different than __clone3,
> + it will align the stack if required. If __clone3 returns -1 with ENOSYS,
> + fall back to clone or clone2. */
> extern int __clone_internal (struct clone_args *__cl_args,
> int (*__func) (void *__arg), void *__arg);
>
> diff --git a/sysdeps/unix/sysv/linux/clone3.h b/sysdeps/unix/sysv/linux/clone3.h
> index 889014a6a9..7c7e2cdcb0 100644
> --- a/sysdeps/unix/sysv/linux/clone3.h
> +++ b/sysdeps/unix/sysv/linux/clone3.h
> @@ -1,4 +1,4 @@
> -/* The wrapper of clone3.
> +/* The clone3 kernel interface definitions.
> Copyright (C) 2021-2022 Free Software Foundation, Inc.
> This file is part of the GNU C Library.
>
> @@ -23,8 +23,6 @@
> #include <stddef.h>
> #include <bits/types.h>
>
> -__BEGIN_DECLS
> -
> /* The unsigned 64-bit and 8-byte aligned integer type. */
> typedef __U64_TYPE __aligned_uint64_t __attribute__ ((__aligned__ (8)));
>
> @@ -58,10 +56,4 @@ struct clone_args
> __aligned_uint64_t cgroup;
> };
>
> -/* The wrapper of clone3. */
> -extern int clone3 (struct clone_args *__cl_args, size_t __size,
> - int (*__func) (void *__arg), void *__arg);
> -
> -__END_DECLS
> -
OK.
> #endif /* clone3.h */
--
Cheers,
Carlos.
next prev parent reply other threads:[~2023-01-11 21:35 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-04 19:01 [PATCH v2 00/11] Optimize posix_spawn signal setup with clone3 Adhemerval Zanella via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 01/11] linux: Do not reset signal handler in posix_spawn if it is already SIG_DFL Adhemerval Zanella via Libc-alpha
2023-01-11 21:27 ` Carlos O'Donell via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 02/11] linux: Extend internal clone3 documentation Adhemerval Zanella via Libc-alpha
2022-11-07 16:40 ` Szabolcs Nagy via Libc-alpha
2023-01-11 21:35 ` Carlos O'Donell via Libc-alpha [this message]
2022-11-04 19:01 ` [PATCH v2 03/11] Linux: Do not align the stack for __clone3 Adhemerval Zanella via Libc-alpha
2023-01-11 21:43 ` Carlos O'Donell via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 04/11] linux: Add clone3 CLONE_CLEAR_SIGHAND optimization to posix_spawn Adhemerval Zanella via Libc-alpha
2023-01-11 21:44 ` Carlos O'Donell via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 05/11] powerpc64: Add the clone3 wrapper Adhemerval Zanella via Libc-alpha
2022-11-28 16:38 ` Paul E Murphy via Libc-alpha
2022-11-29 20:43 ` Adhemerval Zanella Netto via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 06/11] aarch64: " Adhemerval Zanella via Libc-alpha
2022-11-07 16:14 ` Szabolcs Nagy via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 07/11] s390x: " Adhemerval Zanella via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 08/11] riscv: " Adhemerval Zanella via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 09/11] arm: " Adhemerval Zanella via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 10/11] mips: " Adhemerval Zanella via Libc-alpha
2022-11-04 19:01 ` [PATCH v2 11/11] Linux: optimize clone3 internal usage Adhemerval Zanella via Libc-alpha
2023-01-11 21:45 ` Carlos O'Donell via Libc-alpha
2023-01-11 21:24 ` [PATCH v2 00/11] Optimize posix_spawn signal setup with clone3 Carlos O'Donell via Libc-alpha
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
List information: https://www.gnu.org/software/libc/involved.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=f5852ad3-1a4d-8e74-af9d-7cac69f86288@redhat.com \
--to=libc-alpha@sourceware.org \
--cc=adhemerval.zanella@linaro.org \
--cc=brauner@kernel.org \
--cc=carlos@redhat.com \
--cc=hjl.tools@gmail.com \
--cc=szabolcs.nagy@arm.com \
/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).