linux-kernel - Re: [tip:x86/urgent] arch/x86/mm/srat: Skip NUMA_NO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Sun, 26 Jan 2014 00:43:21 -0800
From:	Yinghai Lu <yinghai@...nel.org>
To:	Ingo Molnar <mingo@...nel.org>, "H. Peter Anvin" <hpa@...or.com>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Yinghai Lu <yinghai@...nel.org>,
	Toshi Kani <toshi.kani@...com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	David Rientjes <rientjes@...gle.com>
Cc:	"linux-tip-commits@...r.kernel.org" 
	<linux-tip-commits@...r.kernel.org>
Subject: Re: [tip:x86/urgent] arch/x86/mm/srat: Skip NUMA_NO_NODE while
 parsing SLIT

On Sat, Jan 25, 2014 at 6:25 AM, tip-bot for Toshi Kani
<tipbot@...or.com> wrote:
> Commit-ID:  a85eba8814631d0d48361c8b9a7ee0984e80c03c
> Gitweb:     http://git.kernel.org/tip/a85eba8814631d0d48361c8b9a7ee0984e80c03c
> Author:     Toshi Kani <toshi.kani@...com>
> AuthorDate: Tue, 21 Jan 2014 14:33:15 -0800
> Committer:  Ingo Molnar <mingo@...nel.org>
> CommitDate: Sat, 25 Jan 2014 09:13:35 +0100
>
> arch/x86/mm/srat: Skip NUMA_NO_NODE while parsing SLIT
>
> When ACPI SLIT table has an I/O locality (i.e. a locality
> unique to an I/O device), numa_set_distance() emits this warning
> message:
>
>  NUMA: Warning: node ids are out of bound, from=-1 to=-1 distance=10
>
> acpi_numa_slit_init() calls numa_set_distance() with
> pxm_to_node(), which assumes that all localities have been
> parsed with SRAT previously.  SRAT does not list I/O localities,
> where as SLIT lists all localities including I/Os.  Hence,
> pxm_to_node() returns NUMA_NO_NODE (-1) for an I/O locality.
>
> I/O localities are not supported and are ignored today, but emitting
> such warning message leads to unnecessary confusion.
>
> Change acpi_numa_slit_init() to avoid calling
> numa_set_distance() with NUMA_NO_NODE.
>
> Signed-off-by: Toshi Kani <toshi.kani@...com>
> Acked-by: David Rientjes <rientjes@...gle.com>
> Signed-off-by: Andrew Morton <akpm@...ux-foundation.org>
> Cc: Yinghai Lu <yinghai@...nel.org>
> Link: http://lkml.kernel.org/n/tip-dSvpjjvp8aMzs1ybkftxohlh@git.kernel.org
> Signed-off-by: Ingo Molnar <mingo@...nel.org>
> ---
>  arch/x86/mm/srat.c | 16 +++++++++++++---
>  1 file changed, 13 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/mm/srat.c b/arch/x86/mm/srat.c
> index 266ca91..5ecf651 100644
> --- a/arch/x86/mm/srat.c
> +++ b/arch/x86/mm/srat.c
> @@ -42,15 +42,25 @@ static __init inline int srat_disabled(void)
>         return acpi_numa < 0;
>  }
>
> -/* Callback for SLIT parsing */
> +/*
> + * Callback for SLIT parsing.  pxm_to_node() returns NUMA_NO_NODE for
> + * I/O localities since SRAT does not list them.  I/O localities are
> + * not supported at this point.
> + */
>  void __init acpi_numa_slit_init(struct acpi_table_slit *slit)
>  {
>         int i, j;
>
> -       for (i = 0; i < slit->locality_count; i++)
> -               for (j = 0; j < slit->locality_count; j++)
> +       for (i = 0; i < slit->locality_count; i++) {
> +               if (pxm_to_node(i) == NUMA_NO_NODE)
> +                       continue;
> +               for (j = 0; j < slit->locality_count; j++) {
> +                       if (pxm_to_node(j) == NUMA_NO_NODE)
> +                               continue;
>                         numa_set_distance(pxm_to_node(i), pxm_to_node(j),
>                                 slit->entry[slit->locality_count * i + j]);
> +               }
> +       }
>  }
>
>  /* Callback for Proximity Domain -> x2APIC mapping */

wonder if the patch that i sent one year ago is better.

https://lkml.org/lkml/2013/1/21/559

as it avoid calling extra calling of pxm_to_node(i).

From: Yinghai Lu <yinghai@...nel.org>
Subject: [PATCH] x86, mm: Skip unknown PXM early in SLIT parsing

Some systems one bios could support 2 sockets and 4 sockets,
and SLIT is the same, aka 4x4.

For 2 sockets configuration, SRAT will only have PXM0 and PXM2.

So we will get warning:
NUMA: Warning: node ids are out of bound, from=0 to=-1 distance=15

Need to skip PXM1 and PXM2 as there is no responding node,
To avoid uncorrect warning.

Signed-off-by: Yinghai Lu <yinghai@...nel.org>

---
 arch/x86/mm/srat.c |   17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

Index: linux-2.6/arch/x86/mm/srat.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/srat.c
+++ linux-2.6/arch/x86/mm/srat.c
@@ -46,11 +46,22 @@ static __init inline int srat_disabled(v
 void __init acpi_numa_slit_init(struct acpi_table_slit *slit)
 {
     int i, j;
+    int from_node, to_node;

-    for (i = 0; i < slit->locality_count; i++)
-        for (j = 0; j < slit->locality_count; j++)
-            numa_set_distance(pxm_to_node(i), pxm_to_node(j),
+    for (i = 0; i < slit->locality_count; i++) {
+        from_node = pxm_to_node(i);
+        if (from_node < 0)
+            continue; /* skip unknown PXM */
+
+        for (j = 0; j < slit->locality_count; j++) {
+            to_node = pxm_to_node(j);
+            if (to_node < 0)
+                continue; /* skip unknown PXM */
+
+            numa_set_distance(from_node, to_node,
                 slit->entry[slit->locality_count * i + j]);
+        }
+    }
 }

 /* Callback for Proximity Domain -> x2APIC mapping */
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/