From: George Dunlap <george.dunlap@eu.citrix.com>
To: Dario Faggioli <dario.faggioli@citrix.com>
Cc: Marcus Granado <Marcus.Granado@eu.citrix.com>,
Dan Magenheimer <dan.magenheimer@oracle.com>,
Ian Campbell <Ian.Campbell@citrix.com>,
Anil Madhavapeddy <anil@recoil.org>,
Andrew Cooper <Andrew.Cooper3@citrix.com>,
Juergen Gross <juergen.gross@ts.fujitsu.com>,
Ian Jackson <Ian.Jackson@eu.citrix.com>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
Jan Beulich <JBeulich@suse.com>,
Daniel De Graaf <dgdegra@tycho.nsa.gov>,
Matt Wilson <msw@amazon.com>
Subject: Re: [PATCH 06 of 10 v2] libxl: allow for explicitly specifying node-affinity
Date: Fri, 21 Dec 2012 15:30:25 +0000 [thread overview]
Message-ID: <50D48091.5070409@eu.citrix.com> (raw)
In-Reply-To: <3c196445edb57baadf4f.1355944042@Solace>
On 19/12/12 19:07, Dario Faggioli wrote:
> By introducing a nodemap in libxl_domain_build_info and
> providing the get/set methods to deal with it.
>
> Signed-off-by: Dario Faggioli <dario.faggioli@citrix.com>
> Acked-by: Juergen Gross <juergen.gross@ts.fujitsu.com>
I think you'll probably need to add a line like the following:
#define LIBXL_HAVE_NODEAFFINITY 1
So that people wanting to build against different versions of the
library can behave appropriately. But IanC or IanJ would be the final
word on that, I think.
-George
>
> diff --git a/tools/libxl/libxl.c b/tools/libxl/libxl.c
> --- a/tools/libxl/libxl.c
> +++ b/tools/libxl/libxl.c
> @@ -4142,6 +4142,26 @@ int libxl_set_vcpuaffinity_all(libxl_ctx
> return rc;
> }
>
> +int libxl_domain_set_nodeaffinity(libxl_ctx *ctx, uint32_t domid,
> + libxl_bitmap *nodemap)
> +{
> + if (xc_domain_node_setaffinity(ctx->xch, domid, nodemap->map)) {
> + LIBXL__LOG_ERRNO(ctx, LIBXL__LOG_ERROR, "setting node affinity");
> + return ERROR_FAIL;
> + }
> + return 0;
> +}
> +
> +int libxl_domain_get_nodeaffinity(libxl_ctx *ctx, uint32_t domid,
> + libxl_bitmap *nodemap)
> +{
> + if (xc_domain_node_getaffinity(ctx->xch, domid, nodemap->map)) {
> + LIBXL__LOG_ERRNO(ctx, LIBXL__LOG_ERROR, "getting node affinity");
> + return ERROR_FAIL;
> + }
> + return 0;
> +}
> +
> int libxl_set_vcpuonline(libxl_ctx *ctx, uint32_t domid, libxl_bitmap *cpumap)
> {
> GC_INIT(ctx);
> diff --git a/tools/libxl/libxl.h b/tools/libxl/libxl.h
> --- a/tools/libxl/libxl.h
> +++ b/tools/libxl/libxl.h
> @@ -861,6 +861,10 @@ int libxl_set_vcpuaffinity(libxl_ctx *ct
> libxl_bitmap *cpumap);
> int libxl_set_vcpuaffinity_all(libxl_ctx *ctx, uint32_t domid,
> unsigned int max_vcpus, libxl_bitmap *cpumap);
> +int libxl_domain_set_nodeaffinity(libxl_ctx *ctx, uint32_t domid,
> + libxl_bitmap *nodemap);
> +int libxl_domain_get_nodeaffinity(libxl_ctx *ctx, uint32_t domid,
> + libxl_bitmap *nodemap);
> int libxl_set_vcpuonline(libxl_ctx *ctx, uint32_t domid, libxl_bitmap *cpumap);
>
> libxl_scheduler libxl_get_scheduler(libxl_ctx *ctx);
> diff --git a/tools/libxl/libxl_create.c b/tools/libxl/libxl_create.c
> --- a/tools/libxl/libxl_create.c
> +++ b/tools/libxl/libxl_create.c
> @@ -184,6 +184,12 @@ int libxl__domain_build_info_setdefault(
>
> libxl_defbool_setdefault(&b_info->numa_placement, true);
>
> + if (!b_info->nodemap.size) {
> + if (libxl_node_bitmap_alloc(CTX, &b_info->nodemap, 0))
> + return ERROR_FAIL;
> + libxl_bitmap_set_any(&b_info->nodemap);
> + }
> +
> if (b_info->max_memkb == LIBXL_MEMKB_DEFAULT)
> b_info->max_memkb = 32 * 1024;
> if (b_info->target_memkb == LIBXL_MEMKB_DEFAULT)
> diff --git a/tools/libxl/libxl_dom.c b/tools/libxl/libxl_dom.c
> --- a/tools/libxl/libxl_dom.c
> +++ b/tools/libxl/libxl_dom.c
> @@ -230,6 +230,7 @@ int libxl__build_pre(libxl__gc *gc, uint
> if (rc)
> return rc;
> }
> + libxl_domain_set_nodeaffinity(ctx, domid, &info->nodemap);
> libxl_set_vcpuaffinity_all(ctx, domid, info->max_vcpus, &info->cpumap);
>
> xc_domain_setmaxmem(ctx->xch, domid, info->target_memkb + LIBXL_MAXMEM_CONSTANT);
> diff --git a/tools/libxl/libxl_types.idl b/tools/libxl/libxl_types.idl
> --- a/tools/libxl/libxl_types.idl
> +++ b/tools/libxl/libxl_types.idl
> @@ -261,6 +261,7 @@ libxl_domain_build_info = Struct("domain
> ("max_vcpus", integer),
> ("avail_vcpus", libxl_bitmap),
> ("cpumap", libxl_bitmap),
> + ("nodemap", libxl_bitmap),
> ("numa_placement", libxl_defbool),
> ("tsc_mode", libxl_tsc_mode),
> ("max_memkb", MemKB),
next prev parent reply other threads:[~2012-12-21 15:30 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-19 19:07 [PATCH 00 of 10 v2] NUMA aware credit scheduling Dario Faggioli
2012-12-19 19:07 ` [PATCH 01 of 10 v2] xen, libxc: rename xenctl_cpumap to xenctl_bitmap Dario Faggioli
2012-12-20 9:17 ` Jan Beulich
2012-12-20 9:35 ` Dario Faggioli
2012-12-19 19:07 ` [PATCH 02 of 10 v2] xen, libxc: introduce node maps and masks Dario Faggioli
2012-12-20 9:18 ` Jan Beulich
2012-12-20 9:55 ` Dario Faggioli
2012-12-20 14:33 ` George Dunlap
2012-12-20 14:52 ` Jan Beulich
2012-12-20 15:13 ` Dario Faggioli
2012-12-19 19:07 ` [PATCH 03 of 10 v2] xen: sched_credit: let the scheduler know about node-affinity Dario Faggioli
2012-12-20 6:44 ` Juergen Gross
2012-12-20 8:16 ` Dario Faggioli
2012-12-20 8:25 ` Juergen Gross
2012-12-20 8:33 ` Dario Faggioli
2012-12-20 8:39 ` Juergen Gross
2012-12-20 8:58 ` Dario Faggioli
2012-12-20 15:28 ` George Dunlap
2012-12-20 16:00 ` Dario Faggioli
2012-12-20 9:22 ` Jan Beulich
2012-12-20 15:56 ` George Dunlap
2012-12-20 17:12 ` Dario Faggioli
2012-12-20 16:48 ` George Dunlap
2012-12-20 18:18 ` Dario Faggioli
2012-12-21 14:29 ` George Dunlap
2012-12-21 16:07 ` Dario Faggioli
2012-12-20 20:21 ` George Dunlap
2012-12-21 0:18 ` Dario Faggioli
2012-12-21 14:56 ` George Dunlap
2012-12-21 16:13 ` Dario Faggioli
2012-12-19 19:07 ` [PATCH 04 of 10 v2] xen: allow for explicitly specifying node-affinity Dario Faggioli
2012-12-21 15:17 ` George Dunlap
2012-12-21 16:17 ` Dario Faggioli
2013-01-03 16:05 ` Daniel De Graaf
2012-12-19 19:07 ` [PATCH 05 of 10 v2] libxc: " Dario Faggioli
2012-12-21 15:19 ` George Dunlap
2012-12-21 16:27 ` Dario Faggioli
2012-12-19 19:07 ` [PATCH 06 of 10 v2] libxl: " Dario Faggioli
2012-12-21 15:30 ` George Dunlap [this message]
2012-12-21 16:18 ` Dario Faggioli
2012-12-21 17:02 ` Ian Jackson
2012-12-21 17:09 ` Dario Faggioli
2012-12-19 19:07 ` [PATCH 07 of 10 v2] libxl: optimize the calculation of how many VCPUs can run on a candidate Dario Faggioli
2012-12-20 8:41 ` Ian Campbell
2012-12-20 9:24 ` Dario Faggioli
2012-12-21 16:00 ` George Dunlap
2012-12-21 16:23 ` Dario Faggioli
2012-12-19 19:07 ` [PATCH 08 of 10 v2] libxl: automatic placement deals with node-affinity Dario Faggioli
2012-12-21 16:22 ` George Dunlap
2012-12-19 19:07 ` [PATCH 09 of 10 v2] xl: add node-affinity to the output of `xl list` Dario Faggioli
2012-12-21 16:34 ` George Dunlap
2012-12-21 16:54 ` Dario Faggioli
2012-12-19 19:07 ` [PATCH 10 of 10 v2] docs: rearrange and update NUMA placement documentation Dario Faggioli
2012-12-19 23:16 ` [PATCH 00 of 10 v2] NUMA aware credit scheduling Dario Faggioli
2013-01-11 12:19 ` Ian Campbell
2013-01-11 13:57 ` Dario Faggioli
2013-01-11 14:09 ` Ian Campbell
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=50D48091.5070409@eu.citrix.com \
--to=george.dunlap@eu.citrix.com \
--cc=Andrew.Cooper3@citrix.com \
--cc=Ian.Campbell@citrix.com \
--cc=Ian.Jackson@eu.citrix.com \
--cc=JBeulich@suse.com \
--cc=Marcus.Granado@eu.citrix.com \
--cc=anil@recoil.org \
--cc=dan.magenheimer@oracle.com \
--cc=dario.faggioli@citrix.com \
--cc=dgdegra@tycho.nsa.gov \
--cc=juergen.gross@ts.fujitsu.com \
--cc=msw@amazon.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 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).