From: Tejun Heo <tj@kernel.org>
To: linux-kernel@vger.kernel.org, x86@kernel.org, yinghai@kernel.org,
brgerst@gmail.com, gorcunov@gmail.com, shaohui.zheng@intel.com,
rientjes@google.com, mingo@elte.hu, hpa@linux.intel.com,
ankita@in.ibm.com
Cc: Tejun Heo <tj@kernel.org>
Subject: [PATCH 30/33] x86-64, NUMA: Wrap node ID during emulation
Date: Wed, 16 Feb 2011 13:21:04 +0100 [thread overview]
Message-ID: <1297858867-25981-31-git-send-email-tj@kernel.org> (raw)
In-Reply-To: <1297858867-25981-1-git-send-email-tj@kernel.org>
Both emulation layout functions - split_nodes[_size]_interleave() -
didn't wrap emulated nid while laying out the fake nodes and tried to
avoid interating over the specified number of nodes, which is fragile.
Now that the emulation code generates numa_meminfo, the node memblks
don't need to be consecutive and emulated node IDs can simply wrap.
This makes the code more robust and is necessary for updates to better
handle the cases where the physical nodes are interleaved.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Shaohui Zheng <shaohui.zheng@intel.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: H. Peter Anvin <hpa@linux.intel.com>
---
arch/x86/mm/numa_64.c | 12 ++----------
1 files changed, 2 insertions(+), 10 deletions(-)
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 30c087d..040d0ff 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -768,15 +768,7 @@ static int __init split_nodes_interleave(struct numa_meminfo *ei,
memblock_x86_hole_size(end, physnodes[i].end) < size)
end = physnodes[i].end;
- /*
- * Avoid allocating more nodes than requested, which can
- * happen as a result of rounding down each node's size
- * to FAKE_NODE_MIN_SIZE.
- */
- if (nodes_weight(physnode_mask) + nid >= nr_nodes)
- end = physnodes[i].end;
-
- ret = emu_setup_memblk(ei, nid++, i,
+ ret = emu_setup_memblk(ei, nid++ % nr_nodes, i,
physnodes[i].start,
min(end, physnodes[i].end));
if (ret < 0)
@@ -873,7 +865,7 @@ static int __init split_nodes_size_interleave(struct numa_meminfo *ei,
memblock_x86_hole_size(end, physnodes[i].end) < size)
end = physnodes[i].end;
- ret = emu_setup_memblk(ei, nid++, i,
+ ret = emu_setup_memblk(ei, nid++ % MAX_NUMNODES, i,
physnodes[i].start,
min(end, physnodes[i].end));
if (ret < 0)
--
1.7.1
next prev parent reply other threads:[~2011-02-16 12:23 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-02-16 12:20 [PATCHSET x86/numa] x86-64, NUMA: bring sanity to NUMA config/emulation Tejun Heo
2011-02-16 12:20 ` [PATCH 01/33] x86-64, NUMA: Make dummy node initialization path similar to non-dummy ones Tejun Heo
2011-02-16 12:20 ` [PATCH 02/33] x86-64, NUMA: Simplify hotplug node handling in acpi_numa_memory_affinity_init() Tejun Heo
2011-02-16 12:20 ` [PATCH 03/33] x86, NUMA: Drop @start/last_pfn from initmem_init() Tejun Heo
2011-02-16 12:20 ` [PATCH 04/33] x86-64, NUMA: Unify {acpi|amd}_{numa_init|scan_nodes}() arguments and return values Tejun Heo
2011-02-16 12:20 ` [PATCH 05/33] x86-64, NUMA: Wrap acpi_numa_init() so that failure can be indicated by return value Tejun Heo
2011-02-16 12:20 ` [PATCH 06/33] x86, NUMA: Move *_numa_init() invocations into initmem_init() Tejun Heo
2011-02-16 12:20 ` [PATCH 07/33] x86-64, NUMA: Restructure initmem_init() Tejun Heo
2011-02-16 12:20 ` [PATCH 08/33] x86-64, NUMA: Use common {cpu|mem}_nodes_parsed Tejun Heo
2011-02-16 12:20 ` [PATCH 09/33] x86-64, NUMA: Remove local variable found from amd_numa_init() Tejun Heo
2011-02-16 12:20 ` [PATCH 10/33] x86-64, NUMA: Move apicid to numa mapping initialization from amd_scan_nodes() to amd_numa_init() Tejun Heo
2011-02-16 12:20 ` [PATCH 11/33] x86-64, NUMA: Use common numa_nodes[] Tejun Heo
2011-02-16 12:20 ` [PATCH 12/33] x86-64, NUMA: Kill {acpi|amd}_get_nodes() Tejun Heo
2011-02-16 12:20 ` [PATCH 13/33] x86-64, NUMA: Factor out memblk handling into numa_{add|register}_memblk() Tejun Heo
2011-02-16 16:15 ` [PATCH UPDATED " Tejun Heo
2011-02-16 12:20 ` [PATCH 14/33] x86-64, NUMA: Unify use of memblk in all init methods Tejun Heo
2011-02-16 12:20 ` [PATCH 15/33] x86-64, NUMA: Unify the rest of memblk registration Tejun Heo
2011-02-16 12:20 ` [PATCH 16/33] x86-64, NUMA: Kill {acpi|amd|dummy}_scan_nodes() Tejun Heo
2011-02-16 12:20 ` [PATCH 17/33] x86-64, NUMA: Remove %NULL @nodeids handling from compute_hash_shift() Tejun Heo
2011-02-16 12:20 ` [PATCH 18/33] x86-64, NUMA: Introduce struct numa_meminfo Tejun Heo
2011-02-16 12:20 ` [PATCH 19/33] x86-64, NUMA: Separate out numa_cleanup_meminfo() Tejun Heo
2011-02-16 12:20 ` [PATCH 20/33] x86-64, NUMA: make numa_cleanup_meminfo() prettier Tejun Heo
2011-02-16 12:20 ` [PATCH 21/33] x86-64, NUMA: consolidate and improve memblk sanity checks Tejun Heo
2011-02-16 12:20 ` [PATCH 22/33] x86-64, NUMA: Add common find_node_by_addr() Tejun Heo
2011-02-16 12:20 ` [PATCH 23/33] x86-64, NUMA: Kill numa_nodes[] Tejun Heo
2011-02-16 12:20 ` [PATCH 24/33] x86-64, NUMA: Rename cpu_nodes_parsed to numa_nodes_parsed Tejun Heo
2011-02-16 12:20 ` [PATCH 25/33] x86-64, NUMA: Kill mem_nodes_parsed Tejun Heo
2011-02-16 12:21 ` [PATCH 26/33] x86-64, NUMA: Implement generic node distance handling Tejun Heo
2011-02-16 12:21 ` [PATCH 27/33] x86-64, NUMA: Trivial changes to prepare for emulation updates Tejun Heo
2011-02-16 12:21 ` [PATCH 28/33] x86-64, NUMA: Build and use direct emulated nid -> phys nid mapping Tejun Heo
2011-02-16 14:14 ` [PATCH UPDATED " Tejun Heo
2011-02-16 12:21 ` [PATCH 29/33] x86-64, NUMA: Make emulation code build numa_meminfo and share the registration path Tejun Heo
2011-02-16 12:21 ` Tejun Heo [this message]
2011-02-16 12:21 ` [PATCH 31/33] x86-64, NUMA: Emulate directly from numa_meminfo Tejun Heo
2011-02-16 12:21 ` [PATCH 32/33] x86-64, NUMA: Unify emulated apicid -> node mapping transformation Tejun Heo
2011-02-16 12:21 ` [PATCH 33/33] x86-64, NUMA: Unify emulated distance mapping Tejun Heo
2011-02-16 12:52 ` [PATCHSET x86/numa] x86-64, NUMA: bring sanity to NUMA config/emulation Ingo Molnar
2011-02-16 14:17 ` Tejun Heo
2011-02-16 15:53 ` Ingo Molnar
2011-02-16 16:23 ` Tejun Heo
2011-02-16 17:29 ` Ingo Molnar
2011-02-16 17:33 ` Tejun Heo
2011-02-17 12:35 ` [boot crash] " Ingo Molnar
2011-02-17 12:48 ` Tejun Heo
2011-02-17 16:10 ` Ingo Molnar
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=1297858867-25981-31-git-send-email-tj@kernel.org \
--to=tj@kernel.org \
--cc=ankita@in.ibm.com \
--cc=brgerst@gmail.com \
--cc=gorcunov@gmail.com \
--cc=hpa@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=rientjes@google.com \
--cc=shaohui.zheng@intel.com \
--cc=x86@kernel.org \
--cc=yinghai@kernel.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