RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-23 18:25:08 +00:00
Code Issues Activity Actions

Kernel: More signal handling improvements.

Browse source 
Finally fixed the weird flaky crashing when resizing Terminal windows.
It was because we were dispatching a signal to "current" from the scheduler.
Yet another thing I dislike about even having a "current" process while
we're in the scheduler. Not sure yet how to fix this.

Let the signal handler's kernel stack be a kmalloc() allocation for now.
Once we can do allocation of consecutive physical pages in the supervisor
memory region, we can use that for all types of kernel stacks.
This commit is contained in:
Andreas Kling 2019-03-05 12:50:55 +01:00
parent 086a0fc969
commit 91031346e5
6 changed files with 93 additions and 55 deletions
Download patch file Download diff file Expand all files Collapse all files

7
Kernel/Scheduler.cpp
View file

@ -162,6 +162,11 @@ bool Scheduler::pick_next()
Process::for_each_living([] (auto& process) {
if (!process.has_unmasked_pending_signals())
return true;
// FIXME: It would be nice if the Scheduler didn't have to worry about who is "current"
// For now, avoid dispatching signals to "current" and do it in a scheduling pass
// while some other process is interrupted. Otherwise a mess will be made.
if (&process == current)
return true;
// We know how to interrupt blocked processes, but if they are just executing
// at some random point in the kernel, let them continue. They'll be in userspace
// sooner or later and we can deliver the signal then.
@ -199,7 +204,7 @@ bool Scheduler::pick_next()
if (process->state() == Process::Runnable || process->state() == Process::Running) {
#ifdef SCHEDULER_DEBUG
dbgprintf("switch to %s(%u) @ %w:%x\n", process->name().characters(), process->pid(), process->tss().cs, process->tss().eip);
kprintf("switch to %s(%u) @ %w:%x\n", process->name().characters(), process->pid(), process->tss().cs, process->tss().eip);
#endif
return context_switch(*process);
}