Re: [PATCH 2/4] linux-user: Fix accept4(SOCK_NONBLOCK) syscall

2023-07-07 Thread Helge Deller
On 7/7/23 22:15, Richard Henderson wrote: On 7/7/23 14:19, Helge Deller wrote: The accept4() syscall takes two flags only: SOCK_NONBLOCK and SOCK_CLOEXEC. Even the real Linux kernel returns -EINVAL if any other bits have been set. Change the implementation of accept4() to recognize those two va

Re: [PATCH 2/4] linux-user: Fix accept4(SOCK_NONBLOCK) syscall

2023-07-07 Thread Richard Henderson
On 7/7/23 14:19, Helge Deller wrote: The accept4() syscall takes two flags only: SOCK_NONBLOCK and SOCK_CLOEXEC. Even the real Linux kernel returns -EINVAL if any other bits have been set. Change the implementation of accept4() to recognize those two values only, instead of using the fcntl_flags

[PATCH 2/4] linux-user: Fix accept4(SOCK_NONBLOCK) syscall

2023-07-07 Thread Helge Deller
The accept4() syscall takes two flags only: SOCK_NONBLOCK and SOCK_CLOEXEC. Even the real Linux kernel returns -EINVAL if any other bits have been set. Change the implementation of accept4() to recognize those two values only, instead of using the fcntl_flags_tbl[] bitmask translation. Beside thi