RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-31 09:38:11 +00:00
Code Issues Activity Actions

Kernel+LibPthread+LibC: Add a naive futex and use it for pthread_cond_t

Browse source 
This patch implements a simple version of the futex (fast userspace
mutex) API in the kernel and uses it to make the pthread_cond_t API's
block instead of busily sched_yield().

An arbitrary userspace address is passed to the kernel as a "token"
that identifies the futex and you can then FUTEX_WAIT and FUTEX_WAKE
that specific userspace address.

FUTEX_WAIT corresponds to pthread_cond_wait() and FUTEX_WAKE is used
for pthread_cond_signal() and pthread_cond_broadcast().

I'm pretty sure I'm missing something in this implementation, but it's
hopefully okay for a start. :^)
This commit is contained in:
Andreas Kling 2019-12-22 21:29:47 +01:00
parent 4b8b100b83
commit 4a8683ea68
9 changed files with 99 additions and 55 deletions
Download patch file Download diff file Expand all files Collapse all files

10
Kernel/Syscall.h
View file

@ -150,7 +150,8 @@ typedef u32 socklen_t;
__ENUMERATE_SYSCALL(set_shared_buffer_volatile) \
__ENUMERATE_SYSCALL(profiling_enable) \
__ENUMERATE_SYSCALL(profiling_disable) \
__ENUMERATE_SYSCALL(get_kernel_info_page)
__ENUMERATE_SYSCALL(get_kernel_info_page) \
__ENUMERATE_SYSCALL(futex)
namespace Syscall {
@ -260,6 +261,13 @@ struct SC_setsockopt_params {
socklen_t value_size;
};
struct SC_futex_params {
i32* userspace_address;
int futex_op;
i32 val;
const timespec* timeout;
};
struct SC_create_thread_params {
unsigned int m_detach_state = 0; // JOINABLE or DETACHED
int m_schedule_priority = 2; // ThreadPriority::Normal