Message ID | alpine.DEB.2.21.2002070031280.818@digraph.polyomino.org.uk |
---|---|
State | New |
Headers | show |
Series | Update syscall lists for Linux 5.5 | expand |
On 07/02/20 06:01, Joseph Myers wrote: > Linux 5.5 has no new syscalls to add to syscall-names.list, but it > does newly enable the clone3 syscall for AArch64. This patch updates > the kernel version listed in syscall-names.list and regenerates the > AArch64 arch-syscall.h. > > Tested with build-many-glibcs.py. > LGTM. Siddhesh
* Joseph Myers: > Linux 5.5 has no new syscalls to add to syscall-names.list, but it > does newly enable the clone3 syscall for AArch64. This patch updates > the kernel version listed in syscall-names.list and regenerates the > AArch64 arch-syscall.h. Have you been able to use build-many-glibcs.py for the update? Does the update-syscalls subcommand fit your needs?
On Fri, 7 Feb 2020, Florian Weimer wrote: > * Joseph Myers: > > > Linux 5.5 has no new syscalls to add to syscall-names.list, but it > > does newly enable the clone3 syscall for AArch64. This patch updates > > the kernel version listed in syscall-names.list and regenerates the > > AArch64 arch-syscall.h. > > Have you been able to use build-many-glibcs.py for the update? Does > the update-syscalls subcommand fit your needs? I used update-syscalls successfully for the regeneration (I only named aarch64-linux-gnu on the update-syscalls run since there weren't any new all-architectures syscalls, so haven't tested the case of an all-architectures update).
On 2020-02-07 00:31, Joseph Myers wrote: > Linux 5.5 has no new syscalls to add to syscall-names.list, but it > does newly enable the clone3 syscall for AArch64. This patch updates In practice clone3 has also been enabled in Linux 5.4.12. This cause the misc/tst-glibcsyscalls test to fail when building glibc 2.31 with Linux kernel headers >= 5.4.12: original exit status 1 error: kernel syscall 'clone3' (435) not known to glibc info: glibc tables are based on kernel version 5.4 info: installed kernel headers are version 5.4 Therefore should we backport this commit into the 2.31 stable branch?
* Aurelien Jarno: > On 2020-02-07 00:31, Joseph Myers wrote: >> Linux 5.5 has no new syscalls to add to syscall-names.list, but it >> does newly enable the clone3 syscall for AArch64. This patch updates > > In practice clone3 has also been enabled in Linux 5.4.12. This cause the > misc/tst-glibcsyscalls test to fail when building glibc 2.31 with Linux > kernel headers >= 5.4.12: > > original exit status 1 > error: kernel syscall 'clone3' (435) not known to glibc > info: glibc tables are based on kernel version 5.4 > info: installed kernel headers are version 5.4 > > Therefore should we backport this commit into the 2.31 stable branch? Yes, we should. Thanks.
diff --git a/sysdeps/unix/sysv/linux/aarch64/arch-syscall.h b/sysdeps/unix/sysv/linux/aarch64/arch-syscall.h index 9378387..c847194 100644 --- a/sysdeps/unix/sysv/linux/aarch64/arch-syscall.h +++ b/sysdeps/unix/sysv/linux/aarch64/arch-syscall.h @@ -17,6 +17,7 @@ #define __NR_clock_nanosleep 115 #define __NR_clock_settime 112 #define __NR_clone 220 +#define __NR_clone3 435 #define __NR_close 57 #define __NR_connect 203 #define __NR_copy_file_range 285 diff --git a/sysdeps/unix/sysv/linux/syscall-names.list b/sysdeps/unix/sysv/linux/syscall-names.list index 36e087d..3d89814 100644 --- a/sysdeps/unix/sysv/linux/syscall-names.list +++ b/sysdeps/unix/sysv/linux/syscall-names.list @@ -21,8 +21,8 @@ # This file can list all potential system calls. The names are only # used if the installed kernel headers also provide them. -# The list of system calls is current as of Linux 5.4. -kernel 5.4 +# The list of system calls is current as of Linux 5.5. +kernel 5.5 FAST_atomic_update FAST_cmpxchg