qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: Daniel Henrique Barboza <danielhb413@gmail.com>
Cc: qemu-ppc@nongnu.org, qemu-devel@nongnu.org, groug@kaod.org
Subject: Re: [PATCH v5 4/4] spapr: move FORM1 verifications to do_client_architecture_support()
Date: Tue, 7 Sep 2021 11:04:09 +1000	[thread overview]
Message-ID: <YTa6if+0ww3H437j@yekko> (raw)
In-Reply-To: <20210907002527.412013-5-danielhb413@gmail.com>

[-- Attachment #1: Type: text/plain, Size: 6887 bytes --]

On Mon, Sep 06, 2021 at 09:25:27PM -0300, Daniel Henrique Barboza wrote:
> FORM2 NUMA affinity is prepared to deal with empty (memory/cpu less)
> NUMA nodes. This is used by the DAX KMEM driver to locate a PAPR SCM
> device that has a different latency than the original NUMA node from the
> regular memory. FORM2 is also enable to deal with asymmetric NUMA
> distances gracefully, something that our FORM1 implementation doesn't
> do.
> 
> Move these FORM1 verifications to a new function and wait until after
> CAS, when we're sure that we're sticking with FORM1, to enforce them.
> 
> Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
> ---
>  hw/ppc/spapr.c              | 33 -------------------------
>  hw/ppc/spapr_hcall.c        |  6 +++++
>  hw/ppc/spapr_numa.c         | 49 ++++++++++++++++++++++++++++++++-----
>  include/hw/ppc/spapr_numa.h |  1 +
>  4 files changed, 50 insertions(+), 39 deletions(-)
> 
> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
> index 8d98e3b08a..c974c07fb8 100644
> --- a/hw/ppc/spapr.c
> +++ b/hw/ppc/spapr.c
> @@ -2797,39 +2797,6 @@ static void spapr_machine_init(MachineState *machine)
>      /* init CPUs */
>      spapr_init_cpus(spapr);
>  
> -    /*
> -     * check we don't have a memory-less/cpu-less NUMA node
> -     * Firmware relies on the existing memory/cpu topology to provide the
> -     * NUMA topology to the kernel.
> -     * And the linux kernel needs to know the NUMA topology at start
> -     * to be able to hotplug CPUs later.
> -     */
> -    if (machine->numa_state->num_nodes) {
> -        for (i = 0; i < machine->numa_state->num_nodes; ++i) {
> -            /* check for memory-less node */
> -            if (machine->numa_state->nodes[i].node_mem == 0) {
> -                CPUState *cs;
> -                int found = 0;
> -                /* check for cpu-less node */
> -                CPU_FOREACH(cs) {
> -                    PowerPCCPU *cpu = POWERPC_CPU(cs);
> -                    if (cpu->node_id == i) {
> -                        found = 1;
> -                        break;
> -                    }
> -                }
> -                /* memory-less and cpu-less node */
> -                if (!found) {
> -                    error_report(
> -                       "Memory-less/cpu-less nodes are not supported (node %d)",
> -                                 i);
> -                    exit(1);
> -                }
> -            }
> -        }
> -
> -    }
> -
>      spapr->gpu_numa_id = spapr_numa_initial_nvgpu_numa_id(machine);
>  
>      if ((!kvm_enabled() || kvmppc_has_cap_mmu_radix()) &&
> diff --git a/hw/ppc/spapr_hcall.c b/hw/ppc/spapr_hcall.c
> index 7efbe93f4b..27ee713600 100644
> --- a/hw/ppc/spapr_hcall.c
> +++ b/hw/ppc/spapr_hcall.c
> @@ -1202,9 +1202,15 @@ target_ulong do_client_architecture_support(PowerPCCPU *cpu,
>       * If the guest chooses FORM2 we need to reset the associativity
>       * information - it is being defaulted to FORM1 during
>       * spapr_machine_reset().
> +     *
> +     * If we're sure that we'll be using FORM1, verify now if we have
> +     * a configuration or condition that is not available for FORM1
> +     * (namely asymmetric NUMA topologies and empty NUMA nodes).
>       */
>      if (spapr_ovec_test(spapr->ov5_cas, OV5_FORM2_AFFINITY)) {
>          spapr_numa_associativity_reset(spapr);
> +    } else {
> +        spapr_numa_check_FORM1_constraints(MACHINE(spapr));

Couldn't you put this call into one of the existing FORM1 functions?

>      }
>  
>      /*
> diff --git a/hw/ppc/spapr_numa.c b/hw/ppc/spapr_numa.c
> index ca276e16cb..0c57d03184 100644
> --- a/hw/ppc/spapr_numa.c
> +++ b/hw/ppc/spapr_numa.c
> @@ -155,6 +155,49 @@ static void spapr_numa_define_associativity_domains(SpaprMachineState *spapr)
>  
>  }
>  
> +void spapr_numa_check_FORM1_constraints(MachineState *machine)
> +{
> +    int i;
> +
> +    if (!spapr_numa_is_symmetrical(machine)) {
> +        error_report("Asymmetrical NUMA topologies aren't supported "
> +                     "in the pSeries machine");

Error message needs an update since they are now possible with FORM2.

> +        exit(EXIT_FAILURE);
> +    }
> +
> +    /*
> +     * check we don't have a memory-less/cpu-less NUMA node
> +     * Firmware relies on the existing memory/cpu topology to provide the
> +     * NUMA topology to the kernel.
> +     * And the linux kernel needs to know the NUMA topology at start
> +     * to be able to hotplug CPUs later.
> +     */
> +    if (machine->numa_state->num_nodes) {
> +        for (i = 0; i < machine->numa_state->num_nodes; ++i) {
> +            /* check for memory-less node */
> +            if (machine->numa_state->nodes[i].node_mem == 0) {
> +                CPUState *cs;
> +                int found = 0;
> +                /* check for cpu-less node */
> +                CPU_FOREACH(cs) {
> +                    PowerPCCPU *cpu = POWERPC_CPU(cs);
> +                    if (cpu->node_id == i) {
> +                        found = 1;
> +                        break;
> +                    }
> +                }
> +                /* memory-less and cpu-less node */
> +                if (!found) {
> +                    error_report(
> +                       "Memory-less/cpu-less nodes are not supported (node %d)",
> +                                 i);
> +                    exit(EXIT_FAILURE);
> +                }
> +            }
> +        }
> +    }
> +}
> +
>  /*
>   * Set NUMA machine state data based on FORM1 affinity semantics.
>   */
> @@ -172,12 +215,6 @@ static void spapr_numa_FORM1_affinity_init(SpaprMachineState *spapr,
>          return;
>      }
>  
> -    if (!spapr_numa_is_symmetrical(machine)) {
> -        error_report("Asymmetrical NUMA topologies aren't supported "
> -                     "in the pSeries machine");
> -        exit(EXIT_FAILURE);
> -    }
> -
>      spapr_numa_define_associativity_domains(spapr);
>  }
>  
> diff --git a/include/hw/ppc/spapr_numa.h b/include/hw/ppc/spapr_numa.h
> index 0e457bba57..b5a19cb3f1 100644
> --- a/include/hw/ppc/spapr_numa.h
> +++ b/include/hw/ppc/spapr_numa.h
> @@ -25,5 +25,6 @@ int spapr_numa_fixup_cpu_dt(SpaprMachineState *spapr, void *fdt,
>  int spapr_numa_write_assoc_lookup_arrays(SpaprMachineState *spapr, void *fdt,
>                                           int offset);
>  unsigned int spapr_numa_initial_nvgpu_numa_id(MachineState *machine);
> +void spapr_numa_check_FORM1_constraints(MachineState *machine);
>  
>  #endif /* HW_SPAPR_NUMA_H */

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

      reply	other threads:[~2021-09-07  1:08 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-07  0:25 [PATCH v5 0/4] pSeries FORM2 affinity support Daniel Henrique Barboza
2021-09-07  0:25 ` [PATCH v5 1/4] spapr: move NUMA associativity init to machine reset Daniel Henrique Barboza
2021-09-07  0:37   ` David Gibson
2021-09-07  7:10     ` Greg Kurz
2021-09-07  9:23       ` David Gibson
2021-09-10 19:57         ` Daniel Henrique Barboza
2021-09-11  3:53           ` David Gibson
2021-09-07  0:25 ` [PATCH v5 2/4] spapr_numa.c: split FORM1 code into helpers Daniel Henrique Barboza
2021-09-07  0:39   ` David Gibson
2021-09-07  0:25 ` [PATCH v5 3/4] spapr_numa.c: base FORM2 NUMA affinity support Daniel Henrique Barboza
2021-09-07  1:02   ` David Gibson
2021-09-07 10:07     ` Daniel Henrique Barboza
2021-09-08  1:54       ` David Gibson
2021-09-07  7:50   ` Greg Kurz
2021-09-07  0:25 ` [PATCH v5 4/4] spapr: move FORM1 verifications to do_client_architecture_support() Daniel Henrique Barboza
2021-09-07  1:04   ` David Gibson [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YTa6if+0ww3H437j@yekko \
    --to=david@gibson.dropbear.id.au \
    --cc=danielhb413@gmail.com \
    --cc=groug@kaod.org \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).