Re: [PATCH] SMP initialization: detection and enumeration

[Almudena Garcia]

I attach a new version of my patch, fixing some errors and following the latest comments about this.
This time, I've generated the files manually from "git diff", instead using "git format-patch", so the patches could contain little format mistakes. Excuse me about this issue.

El jue., 30 jul. 2020 a las 23:46, Almudena Garcia (<liberamenso10000@gmail.com>) escribió:

> You don't. Read about GDT and segments register, and segmentation in
> general. In the GDT you'd only store a pointer to the per-cpu data, and
> processors will load it.

> > Is there any way to avoid using an array for that?

> Yes, you'd just store the ID in the per-cpu data.

Thanks, I'll take notes about this

El jue., 30 jul. 2020 a las 23:35, Samuel Thibault (<samuel.thibault@gnu.org>) escribió:

Almudena Garcia, le jeu. 30 juil. 2020 23:30:55 +0200, a ecrit:
> > As Richard said, we just want to have different GDTs on the different
> > processors, so that we wan use the fs segment register to implement TLS
> > in the kernel and have per-cpu data cost essentially the same as global
> > data.
>
> Yes, but how can I store the relation APIC ID - Kernel ID in the GDT?

You don't. Read about GDT and segments register, and segmentation in
general. In the GDT you'd only store a pointer to the per-cpu data, and
processors will load it.

> Is there any way to avoid using an array for that?

Yes, you'd just store the ID in the per-cpu data.

Samuel

0005-smp-Add-smp-pseudoclass.patch
Description: Text Data

0003-vm_kern-Add-function-to-map-table.patch
Description: Text Data

0001-configfrag_redefine_NCPUS.patch
Description: Text Data

0004-smp-Add-APIC-finder-ACPI.patch
Description: Text Data

0002-Add-APIC-pseudoclass.patch
Description: Text Data

0006-model_dep.c-Add-new-smp_init-call.patch
Description: Text Data