Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

Samuel Thibault, le dim. 11 févr. 2024 16:50:51 +0100, a ecrit: > Almudena Garcia, le dim. 11 févr. 2024 16:23:17 +0100, a ecrit: > > But, how can I add the processors to the pset from userspace? > > Using processor set RPCs > > Probably we are missing an RPC to get the pset containing APs, but

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

Almudena Garcia, le dim. 11 févr. 2024 16:23:17 +0100, a ecrit: > But, how can I add the processors to the pset from userspace? Using processor set RPCs Probably we are missing an RPC to get the pset containing APs, but that's not hard to add. Samuel

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

But, how can I add the processors to the pset from userspace? I want to test the parallel environment from Hurd, although gnumach boots with 1 processor El dom, 11 feb 2024 a las 15:36, Samuel Thibault () escribió: > Samuel Thibault, le dim. 11 févr. 2024 15:13:51 +0100, a ecrit: > > By being

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

Samuel Thibault, le dim. 11 févr. 2024 15:13:51 +0100, a ecrit: > By being able to select which block gets exposed to parallelism, you can > selectively check which blocks work and which don't. (while also being already useful to be able to run make in parallel ;) Samuel

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

Almudena Garcia, le dim. 11 févr. 2024 11:17:01 +, a ecrit: > With this old patch I got to boot and executive a bunch of threads splitting > in multiple cpus > > https://git.zammit.org/gnumach-sv.git/commit/?h=fixes=0fe92b6b52726bcd2976863d344117dad8d19694 > > With the patch I quote and

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

With this old patch I got to boot and executive a bunch of threads splitting in multiple cpus https://git.zammit.org/gnumach-sv.git/commit/?h=fixes=0fe92b6b52726bcd2976863d344117dad8d19694 So I prefer to check which is the block of this that really cause the problem. With the patch I quote

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

Almudena Garcia, le dim. 11 févr. 2024 10:48:23 +, a ecrit: > I don't understand which is the objective of this. In an SMP system, all > processor are equal in the scheduler. Why do you add this difference? That's what we had already discussed at the time you introduced the initial SMP

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

Damien Zammit, le dim. 11 févr. 2024 07:10:24 +, a ecrit: > This has the effect of running with one cpu only with smp, > but has the ability to enable APs in userspace with the right > processor set RPCs. > --- > ddb/db_print.c | 4 +++- > kern/machine.c | 7 ++- > kern/processor.c

Re: [PATCH gnumach] smp: Create AP processor set and put all APs inside it

I don't understand which is the objective of this. In an SMP system, all processor are equal in the scheduler. Why do you add this difference? El domingo 11 de febrero de 2024, Damien Zammit escribió: > This has the effect of running with one cpu only with smp, > but has the ability to enable

[PATCH gnumach] smp: Create AP processor set and put all APs inside it

This has the effect of running with one cpu only with smp, but has the ability to enable APs in userspace with the right processor set RPCs. --- ddb/db_print.c | 4 +++- kern/machine.c | 7 ++- kern/processor.c | 30 +++--- kern/processor.h | 1 + 4 files