All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: Reza Arbab <arbab@linux.ibm.com>
Cc: Leonardo Augusto Guimaraes Garcia <lagarcia@linux.ibm.com>,
	Alexey Kardashevskiy <aik@ozlabs.ru>,
	Daniel Henrique Barboza <danielhb@linux.ibm.com>,
	Daniel Henrique Barboza <danielhb413@gmail.com>,
	qemu-devel@nongnu.org, Greg Kurz <groug@kaod.org>,
	qemu-ppc@nongnu.org
Subject: Re: [PATCH v4] spapr: Add a new level of NUMA for GPUs
Date: Fri, 17 Jul 2020 10:07:59 +1000	[thread overview]
Message-ID: <20200717000759.GE5607@umbus.fritz.box> (raw)
In-Reply-To: <20200716225655.24289-1-arbab@linux.ibm.com>

[-- Attachment #1: Type: text/plain, Size: 7282 bytes --]

On Thu, Jul 16, 2020 at 05:56:55PM -0500, Reza Arbab wrote:
> NUMA nodes corresponding to GPU memory currently have the same
> affinity/distance as normal memory nodes. Add a third NUMA associativity
> reference point enabling us to give GPU nodes more distance.
> 
> This is guest visible information, which shouldn't change under a
> running guest across migration between different qemu versions, so make
> the change effective only in new (pseries > 5.0) machine types.
> 
> Before, `numactl -H` output in a guest with 4 GPUs (nodes 2-5):
> 
> node distances:
> node   0   1   2   3   4   5
>   0:  10  40  40  40  40  40
>   1:  40  10  40  40  40  40
>   2:  40  40  10  40  40  40
>   3:  40  40  40  10  40  40
>   4:  40  40  40  40  10  40
>   5:  40  40  40  40  40  10
> 
> After:
> 
> node distances:
> node   0   1   2   3   4   5
>   0:  10  40  80  80  80  80
>   1:  40  10  80  80  80  80
>   2:  80  80  10  80  80  80
>   3:  80  80  80  10  80  80
>   4:  80  80  80  80  10  80
>   5:  80  80  80  80  80  10
> 
> These are the same distances as on the host, mirroring the change made
> to host firmware in skiboot commit f845a648b8cb ("numa/associativity:
> Add a new level of NUMA for GPU's").

Applied to ppc-for-5.1.

> 
> Signed-off-by: Reza Arbab <arbab@linux.ibm.com>
> ---
> v4:
> * Use nvslot->numa_id for distinction at all levels of ibm,associativity
> * Use ARRAY_SIZE(refpoints)
> * Rebase
> 
> v3:
> * Squash into one patch
> * Add PHB compat property
> ---
>  hw/ppc/spapr.c              | 21 +++++++++++++++++++--
>  hw/ppc/spapr_pci.c          |  2 ++
>  hw/ppc/spapr_pci_nvlink2.c  | 13 ++++++++++---
>  include/hw/pci-host/spapr.h |  1 +
>  include/hw/ppc/spapr.h      |  1 +
>  5 files changed, 33 insertions(+), 5 deletions(-)
> 
> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
> index 299908cc7396..0ae293ec9431 100644
> --- a/hw/ppc/spapr.c
> +++ b/hw/ppc/spapr.c
> @@ -890,10 +890,16 @@ static int spapr_dt_rng(void *fdt)
>  static void spapr_dt_rtas(SpaprMachineState *spapr, void *fdt)
>  {
>      MachineState *ms = MACHINE(spapr);
> +    SpaprMachineClass *smc = SPAPR_MACHINE_GET_CLASS(ms);
>      int rtas;
>      GString *hypertas = g_string_sized_new(256);
>      GString *qemu_hypertas = g_string_sized_new(256);
> -    uint32_t refpoints[] = { cpu_to_be32(0x4), cpu_to_be32(0x4) };
> +    uint32_t refpoints[] = {
> +        cpu_to_be32(0x4),
> +        cpu_to_be32(0x4),
> +        cpu_to_be32(0x2),
> +    };
> +    uint32_t nr_refpoints = ARRAY_SIZE(refpoints);
>      uint64_t max_device_addr = MACHINE(spapr)->device_memory->base +
>          memory_region_size(&MACHINE(spapr)->device_memory->mr);
>      uint32_t lrdr_capacity[] = {
> @@ -945,8 +951,12 @@ static void spapr_dt_rtas(SpaprMachineState *spapr, void *fdt)
>                       qemu_hypertas->str, qemu_hypertas->len));
>      g_string_free(qemu_hypertas, TRUE);
>  
> +    if (smc->pre_5_1_assoc_refpoints) {
> +        nr_refpoints = 2;
> +    }
> +
>      _FDT(fdt_setprop(fdt, rtas, "ibm,associativity-reference-points",
> -                     refpoints, sizeof(refpoints)));
> +                     refpoints, nr_refpoints * sizeof(refpoints[0])));
>  
>      _FDT(fdt_setprop(fdt, rtas, "ibm,max-associativity-domains",
>                       maxdomains, sizeof(maxdomains)));
> @@ -4584,9 +4594,16 @@ DEFINE_SPAPR_MACHINE(5_1, "5.1", true);
>   */
>  static void spapr_machine_5_0_class_options(MachineClass *mc)
>  {
> +    SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
> +    static GlobalProperty compat[] = {
> +        { TYPE_SPAPR_PCI_HOST_BRIDGE, "pre-5.1-associativity", "on" },
> +    };
> +
>      spapr_machine_5_1_class_options(mc);
>      compat_props_add(mc->compat_props, hw_compat_5_0, hw_compat_5_0_len);
> +    compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
>      mc->numa_mem_supported = true;
> +    smc->pre_5_1_assoc_refpoints = true;
>  }
>  
>  DEFINE_SPAPR_MACHINE(5_0, "5.0", false);
> diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
> index 2a6a48744aaa..16739334e35f 100644
> --- a/hw/ppc/spapr_pci.c
> +++ b/hw/ppc/spapr_pci.c
> @@ -2035,6 +2035,8 @@ static Property spapr_phb_properties[] = {
>                       pcie_ecs, true),
>      DEFINE_PROP_UINT64("gpa", SpaprPhbState, nv2_gpa_win_addr, 0),
>      DEFINE_PROP_UINT64("atsd", SpaprPhbState, nv2_atsd_win_addr, 0),
> +    DEFINE_PROP_BOOL("pre-5.1-associativity", SpaprPhbState,
> +                     pre_5_1_assoc, false),
>      DEFINE_PROP_END_OF_LIST(),
>  };
>  
> diff --git a/hw/ppc/spapr_pci_nvlink2.c b/hw/ppc/spapr_pci_nvlink2.c
> index dd8cd6db9654..76ae77ebc851 100644
> --- a/hw/ppc/spapr_pci_nvlink2.c
> +++ b/hw/ppc/spapr_pci_nvlink2.c
> @@ -362,9 +362,9 @@ void spapr_phb_nvgpu_ram_populate_dt(SpaprPhbState *sphb, void *fdt)
>                                                      &error_abort);
>          uint32_t associativity[] = {
>              cpu_to_be32(0x4),
> -            SPAPR_GPU_NUMA_ID,
> -            SPAPR_GPU_NUMA_ID,
> -            SPAPR_GPU_NUMA_ID,
> +            cpu_to_be32(nvslot->numa_id),
> +            cpu_to_be32(nvslot->numa_id),
> +            cpu_to_be32(nvslot->numa_id),
>              cpu_to_be32(nvslot->numa_id)
>          };
>          uint64_t size = object_property_get_uint(nv_mrobj, "size", NULL);
> @@ -375,6 +375,13 @@ void spapr_phb_nvgpu_ram_populate_dt(SpaprPhbState *sphb, void *fdt)
>          _FDT(off);
>          _FDT((fdt_setprop_string(fdt, off, "device_type", "memory")));
>          _FDT((fdt_setprop(fdt, off, "reg", mem_reg, sizeof(mem_reg))));
> +
> +        if (sphb->pre_5_1_assoc) {
> +            associativity[1] = SPAPR_GPU_NUMA_ID;
> +            associativity[2] = SPAPR_GPU_NUMA_ID;
> +            associativity[3] = SPAPR_GPU_NUMA_ID;
> +        }
> +
>          _FDT((fdt_setprop(fdt, off, "ibm,associativity", associativity,
>                            sizeof(associativity))));
>  
> diff --git a/include/hw/pci-host/spapr.h b/include/hw/pci-host/spapr.h
> index 8877ff51fbf7..600eb55c3488 100644
> --- a/include/hw/pci-host/spapr.h
> +++ b/include/hw/pci-host/spapr.h
> @@ -94,6 +94,7 @@ struct SpaprPhbState {
>      hwaddr nv2_gpa_win_addr;
>      hwaddr nv2_atsd_win_addr;
>      SpaprPhbPciNvGpuConfig *nvgpus;
> +    bool pre_5_1_assoc;
>  };
>  
>  #define SPAPR_PCI_MEM_WIN_BUS_OFFSET 0x80000000ULL
> diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h
> index c421410e3fb8..3134d339e8fe 100644
> --- a/include/hw/ppc/spapr.h
> +++ b/include/hw/ppc/spapr.h
> @@ -129,6 +129,7 @@ struct SpaprMachineClass {
>      bool linux_pci_probe;
>      bool smp_threads_vsmt; /* set VSMT to smp_threads by default */
>      hwaddr rma_limit;          /* clamp the RMA to this size */
> +    bool pre_5_1_assoc_refpoints;
>  
>      void (*phb_placement)(SpaprMachineState *spapr, uint32_t index,
>                            uint64_t *buid, hwaddr *pio, 

-- 
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:[~2020-07-17  0:09 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-16 22:56 [PATCH v4] spapr: Add a new level of NUMA for GPUs Reza Arbab
2020-07-17  0:07 ` 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=20200717000759.GE5607@umbus.fritz.box \
    --to=david@gibson.dropbear.id.au \
    --cc=aik@ozlabs.ru \
    --cc=arbab@linux.ibm.com \
    --cc=danielhb413@gmail.com \
    --cc=danielhb@linux.ibm.com \
    --cc=groug@kaod.org \
    --cc=lagarcia@linux.ibm.com \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.