Page 2 of 2
Re: Bochs "physical adress not available" after enabeling paging
Posted: Tue Jan 21, 2025 12:44 pm
by Nezmar
Octocontrabass wrote: ↑Tue Jan 21, 2025 12:37 pm
It does throw a page fault. When it tries to handle the page fault, it encounters another page fault. When it tries to handle
that page fault, it encounters a third page fault and shuts down. The chipset detects the shutdown and resets the CPU.
How comes it encounter a page fault when handling a page fault? Should paging not be disabled when calling the interrupt?
Re: Bochs "physical adress not available" after enabeling paging
Posted: Tue Jan 21, 2025 12:54 pm
by Octocontrabass
Nezmar wrote: ↑Tue Jan 21, 2025 12:42 pmHow am i supposed to do it then?
Normally you reserve part of the virtual address space for your kernel and keep your entire kernel mapped the same way all the time.
Nezmar wrote: ↑Tue Jan 21, 2025 12:44 pmHow comes it encounter a page fault when handling a page fault? Should paging not be disabled when calling the interrupt?
Paging stays enabled. Where did you hear otherwise?
Re: Bochs "physical adress not available" after enabeling paging
Posted: Tue Jan 21, 2025 1:30 pm
by Nezmar
So basically the kernel still needs to be mapped in the processes page directory so that it can call interrupts and all?
Re: Bochs "physical adress not available" after enabeling paging
Posted: Tue Jan 21, 2025 2:22 pm
by Nezmar
Well after reading the wiki again I realized that I have made some incorrect assumptions about how paging works.
Anyway thanks a lot for the help. I think I understand it now.