From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: elena.ufimtseva@oracle.com, keir@xen.org,
ian.campbell@citrix.com, stefano.stabellini@eu.citrix.com,
andrew.cooper3@citrix.com, dario.faggioli@citrix.com,
ian.jackson@eu.citrix.com, xen-devel@lists.xen.org,
jbeulich@suse.com, wei.liu2@citrix.com
Subject: Re: [PATCH v5 1/8] numa: __node_distance() should return u8
Date: Fri, 20 Mar 2015 10:25:30 -0400 [thread overview]
Message-ID: <20150320142530.GI29188@l.oracle.com> (raw)
In-Reply-To: <1426802044-19444-2-git-send-email-boris.ostrovsky@oracle.com>
On Thu, Mar 19, 2015 at 05:53:57PM -0400, Boris Ostrovsky wrote:
> SLIT values are byte-sized and some of them (0-9 and 255) have
> special meaning. Adjust __node_distance() to reflect this and
> modify scrub_heap_pages() and do_sysctl() to deal with
> __node_distance() returning an invalid SLIT entry.
>
> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> ---
>
> Changes in v5:
> * XEN_SYSCTL_numainfo knows about NUMA_NO_DISTANCE
> * Cleaner changes in __node_distance()
>
> xen/arch/x86/srat.c | 13 ++++++++++---
> xen/common/page_alloc.c | 4 ++--
> xen/common/sysctl.c | 6 +++++-
> xen/include/asm-x86/numa.h | 2 +-
> xen/include/xen/numa.h | 3 ++-
> 5 files changed, 20 insertions(+), 8 deletions(-)
>
> diff --git a/xen/arch/x86/srat.c b/xen/arch/x86/srat.c
> index dfabba3..92c89a5 100644
> --- a/xen/arch/x86/srat.c
> +++ b/xen/arch/x86/srat.c
> @@ -496,14 +496,21 @@ static unsigned node_to_pxm(nodeid_t n)
> return 0;
> }
>
> -int __node_distance(nodeid_t a, nodeid_t b)
> +u8 __node_distance(nodeid_t a, nodeid_t b)
> {
> - int index;
> + unsigned index;
> + u8 slit_val;
>
> if (!acpi_slit)
> return a == b ? 10 : 20;
> index = acpi_slit->locality_count * node_to_pxm(a);
> - return acpi_slit->entry[index + node_to_pxm(b)];
> + slit_val = acpi_slit->entry[index + node_to_pxm(b)];
> +
> + /* ACPI defines 0xff as an unreachable node and 0-9 are undefined */
> + if ((slit_val == 0xff) || (slit_val <= 9))
> + return NUMA_NO_DISTANCE;
> + else
> + return slit_val;
> }
>
> EXPORT_SYMBOL(__node_distance);
> diff --git a/xen/common/page_alloc.c b/xen/common/page_alloc.c
> index d999296..bfb356e 100644
> --- a/xen/common/page_alloc.c
> +++ b/xen/common/page_alloc.c
> @@ -1434,13 +1434,13 @@ void __init scrub_heap_pages(void)
> /* Figure out which NODE CPUs are close. */
> for_each_online_node ( j )
> {
> - int distance;
> + u8 distance;
>
> if ( cpumask_empty(&node_to_cpumask(j)) )
> continue;
>
> distance = __node_distance(i, j);
> - if ( distance < last_distance )
> + if ( (distance < last_distance) && (distance != NUMA_NO_DISTANCE) )
> {
> last_distance = distance;
> best_node = j;
> diff --git a/xen/common/sysctl.c b/xen/common/sysctl.c
> index 0cb6ee1..6fdd029 100644
> --- a/xen/common/sysctl.c
> +++ b/xen/common/sysctl.c
> @@ -304,7 +304,11 @@ long do_sysctl(XEN_GUEST_HANDLE_PARAM(xen_sysctl_t) u_sysctl)
> {
> uint32_t distance = ~0u;
> if ( node_online(i) && node_online(j) )
> - distance = __node_distance(i, j);
> + {
> + u8 d = __node_distance(i, j);
> + if ( d != NUMA_NO_DISTANCE )
> + distance = d;
> + }
> if ( copy_to_guest_offset(
> ni->node_to_node_distance,
> i*(max_node_index+1) + j, &distance, 1) )
> diff --git a/xen/include/asm-x86/numa.h b/xen/include/asm-x86/numa.h
> index cc5b5d1..7a489d3 100644
> --- a/xen/include/asm-x86/numa.h
> +++ b/xen/include/asm-x86/numa.h
> @@ -85,6 +85,6 @@ extern int valid_numa_range(u64 start, u64 end, nodeid_t node);
> #endif
>
> void srat_parse_regions(u64 addr);
> -extern int __node_distance(nodeid_t a, nodeid_t b);
> +extern u8 __node_distance(nodeid_t a, nodeid_t b);
>
> #endif
> diff --git a/xen/include/xen/numa.h b/xen/include/xen/numa.h
> index ac4b391..7aef1a8 100644
> --- a/xen/include/xen/numa.h
> +++ b/xen/include/xen/numa.h
> @@ -7,7 +7,8 @@
> #define NODES_SHIFT 0
> #endif
>
> -#define NUMA_NO_NODE 0xFF
> +#define NUMA_NO_NODE 0xFF
> +#define NUMA_NO_DISTANCE 0xFF
>
> #define MAX_NUMNODES (1 << NODES_SHIFT)
>
> --
> 1.7.1
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2015-03-20 14:25 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-19 21:53 [PATCH v5 0/8] Display IO topology when PXM data is available (plus some cleanup) Boris Ostrovsky
2015-03-19 21:53 ` [PATCH v5 1/8] numa: __node_distance() should return u8 Boris Ostrovsky
2015-03-20 14:25 ` Konrad Rzeszutek Wilk [this message]
2015-03-19 21:53 ` [PATCH v5 2/8] pci: Stash device's PXM information in struct pci_dev Boris Ostrovsky
2015-03-20 14:01 ` Konrad Rzeszutek Wilk
2015-03-19 21:53 ` [PATCH v5 3/8] sysctl: Make XEN_SYSCTL_topologyinfo sysctl a little more efficient Boris Ostrovsky
2015-03-20 13:25 ` Ian Campbell
2015-03-20 14:00 ` Boris Ostrovsky
2015-03-20 14:10 ` Ian Campbell
2015-03-25 16:13 ` Andrew Cooper
2015-03-19 21:54 ` [PATCH v5 4/8] sysctl: Make XEN_SYSCTL_numainfo " Boris Ostrovsky
2015-03-20 13:26 ` Ian Campbell
2015-03-20 16:15 ` Jan Beulich
2015-03-19 21:54 ` [PATCH v5 5/8] sysctl: Add sysctl interface for querying PCI topology Boris Ostrovsky
2015-03-20 14:21 ` Konrad Rzeszutek Wilk
2015-03-20 16:26 ` Jan Beulich
2015-03-20 20:01 ` Boris Ostrovsky
2015-03-23 8:15 ` Jan Beulich
2015-03-23 13:58 ` Boris Ostrovsky
2015-03-19 21:54 ` [PATCH v5 6/8] libxl/libxc: Move libxl_get_cpu_topology()'s hypercall buffer management to libxc Boris Ostrovsky
2015-03-20 13:54 ` Ian Campbell
2015-03-20 14:24 ` Boris Ostrovsky
2015-03-20 14:30 ` Ian Campbell
2015-03-19 21:54 ` [PATCH v5 7/8] libxl/libxc: Move libxl_get_numainfo()'s " Boris Ostrovsky
2015-03-20 13:49 ` Konrad Rzeszutek Wilk
2015-03-20 14:10 ` Boris Ostrovsky
2015-03-20 14:21 ` Ian Campbell
2015-03-20 13:56 ` Ian Campbell
2015-03-20 14:31 ` Boris Ostrovsky
2015-03-20 14:46 ` Ian Campbell
2015-03-20 14:59 ` Boris Ostrovsky
2015-03-19 21:54 ` [PATCH v5 8/8] libxl: Add interface for querying hypervisor about PCI topology Boris Ostrovsky
2015-03-23 12:42 ` [PATCH v5 0/8] Display IO topology when PXM data is available (plus some cleanup) Julien Grall
2015-03-23 13:47 ` Boris Ostrovsky
2015-03-23 14:30 ` George Dunlap
2015-03-23 14:33 ` Boris Ostrovsky
2015-03-23 14:42 ` Ian Campbell
2015-03-23 15:26 ` Julien Grall
2015-03-23 15:27 ` Jan Beulich
2015-03-23 15:33 ` Boris Ostrovsky
2015-03-23 15:46 ` Jan Beulich
2015-03-23 14:30 ` Jan Beulich
2015-03-23 14:38 ` George Dunlap
2015-03-23 15:17 ` Ian Campbell
2015-03-23 15:21 ` Julien Grall
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=20150320142530.GI29188@l.oracle.com \
--to=konrad.wilk@oracle.com \
--cc=andrew.cooper3@citrix.com \
--cc=boris.ostrovsky@oracle.com \
--cc=dario.faggioli@citrix.com \
--cc=elena.ufimtseva@oracle.com \
--cc=ian.campbell@citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=jbeulich@suse.com \
--cc=keir@xen.org \
--cc=stefano.stabellini@eu.citrix.com \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xen.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.