Calling a kmalloc_node on a possible node which is not yet onlined can
lead to panic. Currently node_present_pages() doesn't verify the node is
online before accessing the pgdat for the node. However pgdat struct may
not be available resulting in a crash.
NIP [c0000000003d55f4] ___slab_alloc+0x1f4/0x760
LR [c0000000003d5b94] __slab_alloc+0x34/0x60
Call Trace:
[c0000008b3783960] [c0000000003d5734] ___slab_alloc+0x334/0x760 (unreliable)
[c0000008b3783a40] [c0000000003d5b94] __slab_alloc+0x34/0x60
[c0000008b3783a70] [c0000000003d6fa0] __kmalloc_node+0x110/0x490
[c0000008b3783af0] [c0000000003443d8] kvmalloc_node+0x58/0x110
[c0000008b3783b30] [c0000000003fee38] mem_cgroup_css_online+0x108/0x270
[c0000008b3783b90] [c000000000235aa8] online_css+0x48/0xd0
[c0000008b3783bc0] [c00000000023eaec] cgroup_apply_control_enable+0x2ec/0x4d0
[c0000008b3783ca0] [c000000000242318] cgroup_mkdir+0x228/0x5f0
[c0000008b3783d10] [c00000000051e170] kernfs_iop_mkdir+0x90/0xf0
[c0000008b3783d50] [c00000000043dc00] vfs_mkdir+0x110/0x230
[c0000008b3783da0] [c000000000441c90] do_mkdirat+0xb0/0x1a0
[c0000008b3783e20] [c00000000000b278] system_call+0x5c/0x68
Fix this by verifying the node is online before accessing the pgdat
structure. Fix the same for node_spanned_pages() too.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
include/linux/mmzone.h | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
Currently while allocating a slab for a offline node, we use its
associated node_numa_mem to search for a partial slab. If we don't find
a partial slab, we try allocating a slab from the offline node using
__alloc_pages_node. However this is bound to fail.
NIP [c00000000039a300] __alloc_pages_nodemask+0x130/0x3b0
LR [c00000000039a3c4] __alloc_pages_nodemask+0x1f4/0x3b0
Call Trace:
[c0000008b36837f0] [c00000000039a3b4] __alloc_pages_nodemask+0x1e4/0x3b0 (unreliable)
[c0000008b3683870] [c0000000003d1ff8] new_slab+0x128/0xcf0
[c0000008b3683950] [c0000000003d6060] ___slab_alloc+0x410/0x820
[c0000008b3683a40] [c0000000003d64a4] __slab_alloc+0x34/0x60
[c0000008b3683a70] [c0000000003d78b0] __kmalloc_node+0x110/0x490
[c0000008b3683af0] [c000000000343a08] kvmalloc_node+0x58/0x110
[c0000008b3683b30] [c0000000003ffd44] mem_cgroup_css_online+0x104/0x270
[c0000008b3683b90] [c000000000234e08] online_css+0x48/0xd0
[c0000008b3683bc0] [c00000000023dedc] cgroup_apply_control_enable+0x2ec/0x4d0
[c0000008b3683ca0] [c0000000002416f8] cgroup_mkdir+0x228/0x5f0
[c0000008b3683d10] [c000000000520360] kernfs_iop_mkdir+0x90/0xf0
[c0000008b3683d50] [c00000000043e400] vfs_mkdir+0x110/0x230
[c0000008b3683da0] [c000000000441ee0] do_mkdirat+0xb0/0x1a0
[c0000008b3683e20] [c00000000000b278] system_call+0x5c/0x68
Mitigate this by allocating the new slab from the node_numa_mem.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
Changelog v1 -> v2:
- Handled comments from Vlastimil Babka
- Now node gets set to node_numa_mem in new_slab_objects.
mm/slub.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
For a memoryless or offline nodes, node_numa_mem refers to a N_MEMORY
fallback node. Currently kernel has an API set_numa_mem that sets
node_numa_mem for memoryless node. However this API cannot be used for
offline nodes. Hence all offline nodes will have their node_numa_mem set
to 0. However systems can themselves have node 0 as offline i.e
memoryless and cpuless at this time. In such cases,
node_to_mem_node() fails to provide a N_MEMORY fallback node.
Mitigate this by having a new API that sets the default node_numa_mem for
offline nodes to be first_memory_node.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
include/asm-generic/topology.h | 3 +++
include/linux/topology.h | 7 +++++++
2 files changed, 10 insertions(+)
@@ -147,6 +147,13 @@ static inline int node_to_mem_node(int node)}#endif+#ifndef reset_numa_mem+staticinlinevoidreset_numa_mem(intnode)+{+_node_numa_mem_[node]=first_memory_node;+}+#endif+#ifndef numa_mem_id/* Returns the number of the nearest Node with memory */staticinlineintnuma_mem_id(void)
Currently fallback nodes for offline nodes aren't set. Hence by default
node 0 ends up being the default node. However node 0 might be offline.
Fix this by explicitly setting fallback node. Ensure first_memory_node
is set before kernel does explicit setting of fallback node.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
Changelog v1 -> v2:
- Handled comments from Bharata B Rao
- Dont use dump_numa_cpu_topology to set fallback nodes
arch/powerpc/include/asm/topology.h | 16 ++++++++++++++++
arch/powerpc/kernel/smp.c | 1 +
2 files changed, 17 insertions(+)
From: Michal Hocko <mhocko@suse.com> Date: 2020-03-18 10:03:05
On Wed 18-03-20 12:58:07, Srikar Dronamraju wrote:
quoted hunk
Calling a kmalloc_node on a possible node which is not yet onlined can
lead to panic. Currently node_present_pages() doesn't verify the node is
online before accessing the pgdat for the node. However pgdat struct may
not be available resulting in a crash.
NIP [c0000000003d55f4] ___slab_alloc+0x1f4/0x760
LR [c0000000003d5b94] __slab_alloc+0x34/0x60
Call Trace:
[c0000008b3783960] [c0000000003d5734] ___slab_alloc+0x334/0x760 (unreliable)
[c0000008b3783a40] [c0000000003d5b94] __slab_alloc+0x34/0x60
[c0000008b3783a70] [c0000000003d6fa0] __kmalloc_node+0x110/0x490
[c0000008b3783af0] [c0000000003443d8] kvmalloc_node+0x58/0x110
[c0000008b3783b30] [c0000000003fee38] mem_cgroup_css_online+0x108/0x270
[c0000008b3783b90] [c000000000235aa8] online_css+0x48/0xd0
[c0000008b3783bc0] [c00000000023eaec] cgroup_apply_control_enable+0x2ec/0x4d0
[c0000008b3783ca0] [c000000000242318] cgroup_mkdir+0x228/0x5f0
[c0000008b3783d10] [c00000000051e170] kernfs_iop_mkdir+0x90/0xf0
[c0000008b3783d50] [c00000000043dc00] vfs_mkdir+0x110/0x230
[c0000008b3783da0] [c000000000441c90] do_mkdirat+0xb0/0x1a0
[c0000008b3783e20] [c00000000000b278] system_call+0x5c/0x68
Fix this by verifying the node is online before accessing the pgdat
structure. Fix the same for node_spanned_pages() too.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
include/linux/mmzone.h | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
I believe this is a wrong approach. We really do not want to special
case all the places which require NODE_DATA. Can we please go and
allocate pgdat for all possible nodes?
The current state of memory less hacks subtle bugs poping up here and
there just prove that we should have done that from the very begining
IMHO.
* Michal Hocko [off-list ref] [2020-03-18 11:02:56]:
On Wed 18-03-20 12:58:07, Srikar Dronamraju wrote:
quoted
Calling a kmalloc_node on a possible node which is not yet onlined can
lead to panic. Currently node_present_pages() doesn't verify the node is
online before accessing the pgdat for the node. However pgdat struct may
not be available resulting in a crash.
NIP [c0000000003d55f4] ___slab_alloc+0x1f4/0x760
LR [c0000000003d5b94] __slab_alloc+0x34/0x60
Call Trace:
[c0000008b3783960] [c0000000003d5734] ___slab_alloc+0x334/0x760 (unreliable)
[c0000008b3783a40] [c0000000003d5b94] __slab_alloc+0x34/0x60
[c0000008b3783a70] [c0000000003d6fa0] __kmalloc_node+0x110/0x490
[c0000008b3783af0] [c0000000003443d8] kvmalloc_node+0x58/0x110
[c0000008b3783b30] [c0000000003fee38] mem_cgroup_css_online+0x108/0x270
[c0000008b3783b90] [c000000000235aa8] online_css+0x48/0xd0
[c0000008b3783bc0] [c00000000023eaec] cgroup_apply_control_enable+0x2ec/0x4d0
[c0000008b3783ca0] [c000000000242318] cgroup_mkdir+0x228/0x5f0
[c0000008b3783d10] [c00000000051e170] kernfs_iop_mkdir+0x90/0xf0
[c0000008b3783d50] [c00000000043dc00] vfs_mkdir+0x110/0x230
[c0000008b3783da0] [c000000000441c90] do_mkdirat+0xb0/0x1a0
[c0000008b3783e20] [c00000000000b278] system_call+0x5c/0x68
Fix this by verifying the node is online before accessing the pgdat
structure. Fix the same for node_spanned_pages() too.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
include/linux/mmzone.h | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
I believe this is a wrong approach. We really do not want to special
case all the places which require NODE_DATA. Can we please go and
allocate pgdat for all possible nodes?
I can do that but the question I had was should we make this change just for
Powerpc or should the change be for other archs.
NODE_DATA initialization always seems to be in arch specific code.
The other archs that are affected seem to be mips, sh and sparc
These archs seem to have making an assumption that NODE_DATA has to be local
only,
For example on sparc / arch/sparc/mm/init_64.c in allocate_node_data function.
NODE_DATA(nid) = memblock_alloc_node(sizeof(struct pglist_data),
SMP_CACHE_BYTES, nid);
if (!NODE_DATA(nid)) {
prom_printf("Cannot allocate pglist_data for nid[%d]\n", nid);
prom_halt();
}
NODE_DATA(nid)->node_id = nid;
So even if I make changes to allocate NODE_DATA from fallback node, I may not
be able to test them.
So please let me know your thoughts around the same.
The current state of memory less hacks subtle bugs poping up here and
there just prove that we should have done that from the very begining
IMHO.
I believe this is a wrong approach. We really do not want to special
case all the places which require NODE_DATA. Can we please go and
allocate pgdat for all possible nodes?
I can do that but the question I had was should we make this change just for
Powerpc or should the change be for other archs.
No, we shouldn't, really. If NODE_DATA is non-null for all possible
nodes then this shouldn't be really necessary and arch specific.
NODE_DATA initialization always seems to be in arch specific code.
The other archs that are affected seem to be mips, sh and sparc
These archs seem to have making an assumption that NODE_DATA has to be local
only,
Which is all good and fine for nodes that hold some memory. If those
architectures support memory less nodes at all then I do not see any
problem to have remote pgdata.
For example on sparc / arch/sparc/mm/init_64.c in allocate_node_data function.
NODE_DATA(nid) = memblock_alloc_node(sizeof(struct pglist_data),
SMP_CACHE_BYTES, nid);
if (!NODE_DATA(nid)) {
prom_printf("Cannot allocate pglist_data for nid[%d]\n", nid);
prom_halt();
}
NODE_DATA(nid)->node_id = nid;
This code is not about memroy less nodes, is it? It looks more like a
allocation failure panic-like handling because there is not enough
memory to hold pgdat. This also strongly suggests that this platform
doesn't really expect memory less nodes in the early init path.
So even if I make changes to allocate NODE_DATA from fallback node, I may not
be able to test them.
Please try to focus on the architecture you can test for. From the
existing reports I have seen this looks mostly to be a problem for x86
and ppc
--
Michal Hocko
SUSE Labs
On Wed 18-03-20 12:58:07, Srikar Dronamraju wrote:
quoted
Calling a kmalloc_node on a possible node which is not yet onlined can
lead to panic. Currently node_present_pages() doesn't verify the node is
online before accessing the pgdat for the node. However pgdat struct may
not be available resulting in a crash.
NIP [c0000000003d55f4] ___slab_alloc+0x1f4/0x760
LR [c0000000003d5b94] __slab_alloc+0x34/0x60
Call Trace:
[c0000008b3783960] [c0000000003d5734] ___slab_alloc+0x334/0x760 (unreliable)
[c0000008b3783a40] [c0000000003d5b94] __slab_alloc+0x34/0x60
[c0000008b3783a70] [c0000000003d6fa0] __kmalloc_node+0x110/0x490
[c0000008b3783af0] [c0000000003443d8] kvmalloc_node+0x58/0x110
[c0000008b3783b30] [c0000000003fee38] mem_cgroup_css_online+0x108/0x270
[c0000008b3783b90] [c000000000235aa8] online_css+0x48/0xd0
[c0000008b3783bc0] [c00000000023eaec] cgroup_apply_control_enable+0x2ec/0x4d0
[c0000008b3783ca0] [c000000000242318] cgroup_mkdir+0x228/0x5f0
[c0000008b3783d10] [c00000000051e170] kernfs_iop_mkdir+0x90/0xf0
[c0000008b3783d50] [c00000000043dc00] vfs_mkdir+0x110/0x230
[c0000008b3783da0] [c000000000441c90] do_mkdirat+0xb0/0x1a0
[c0000008b3783e20] [c00000000000b278] system_call+0x5c/0x68
Fix this by verifying the node is online before accessing the pgdat
structure. Fix the same for node_spanned_pages() too.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
include/linux/mmzone.h | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
I believe this is a wrong approach. We really do not want to special
case all the places which require NODE_DATA. Can we please go and
allocate pgdat for all possible nodes?
The current state of memory less hacks subtle bugs poping up here and
there just prove that we should have done that from the very begining
IMHO.
Yes. So here's an alternative proposal for fixing the current situation in SLUB,
before the long-term solution of having all possible nodes provide valid pgdat
with zonelists:
- fix SLUB with the hunk at the end of this mail - the point is to use NUMA_NO_NODE
as fallback instead of node_to_mem_node()
- this removes all uses of node_to_mem_node (luckily it's just SLUB),
kill it completely instead of trying to fix it up
- patch 1/4 is not needed with the fix
- perhaps many of your other patches are alss not needed
- once we get the long-term solution, some of the !node_online() checks can be removed
----8<----
From: Michal Hocko <mhocko@suse.com> Date: 2020-03-18 12:52:25
On Wed 18-03-20 12:53:32, Vlastimil Babka wrote:
[...]
Yes. So here's an alternative proposal for fixing the current situation in SLUB,
before the long-term solution of having all possible nodes provide valid pgdat
with zonelists:
- fix SLUB with the hunk at the end of this mail - the point is to use NUMA_NO_NODE
as fallback instead of node_to_mem_node()
I am not familiar with SLUB to review.
- this removes all uses of node_to_mem_node (luckily it's just SLUB),
kill it completely instead of trying to fix it up
Sounds like a good plan to me. The code shouldn't really care.
--
Michal Hocko
SUSE Labs
From: Christopher Lameter <hidden> Date: 2020-03-18 19:20:43
On Wed, 18 Mar 2020, Srikar Dronamraju wrote:
For a memoryless or offline nodes, node_numa_mem refers to a N_MEMORY
fallback node. Currently kernel has an API set_numa_mem that sets
node_numa_mem for memoryless node. However this API cannot be used for
offline nodes. Hence all offline nodes will have their node_numa_mem set
to 0. However systems can themselves have node 0 as offline i.e
That is a significant change to the basic assumptions for memory less
nodes. Node 0 needed to have memory and processors. Not sure what else
may break.
From: Michael Ellerman <mpe@ellerman.id.au> Date: 2020-03-19 00:32:15
Vlastimil Babka [off-list ref] writes:
On 3/18/20 11:02 AM, Michal Hocko wrote:
quoted
On Wed 18-03-20 12:58:07, Srikar Dronamraju wrote:
quoted
Calling a kmalloc_node on a possible node which is not yet onlined can
lead to panic. Currently node_present_pages() doesn't verify the node is
online before accessing the pgdat for the node. However pgdat struct may
not be available resulting in a crash.
NIP [c0000000003d55f4] ___slab_alloc+0x1f4/0x760
LR [c0000000003d5b94] __slab_alloc+0x34/0x60
Call Trace:
[c0000008b3783960] [c0000000003d5734] ___slab_alloc+0x334/0x760 (unreliable)
[c0000008b3783a40] [c0000000003d5b94] __slab_alloc+0x34/0x60
[c0000008b3783a70] [c0000000003d6fa0] __kmalloc_node+0x110/0x490
[c0000008b3783af0] [c0000000003443d8] kvmalloc_node+0x58/0x110
[c0000008b3783b30] [c0000000003fee38] mem_cgroup_css_online+0x108/0x270
[c0000008b3783b90] [c000000000235aa8] online_css+0x48/0xd0
[c0000008b3783bc0] [c00000000023eaec] cgroup_apply_control_enable+0x2ec/0x4d0
[c0000008b3783ca0] [c000000000242318] cgroup_mkdir+0x228/0x5f0
[c0000008b3783d10] [c00000000051e170] kernfs_iop_mkdir+0x90/0xf0
[c0000008b3783d50] [c00000000043dc00] vfs_mkdir+0x110/0x230
[c0000008b3783da0] [c000000000441c90] do_mkdirat+0xb0/0x1a0
[c0000008b3783e20] [c00000000000b278] system_call+0x5c/0x68
Fix this by verifying the node is online before accessing the pgdat
structure. Fix the same for node_spanned_pages() too.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
include/linux/mmzone.h | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
I believe this is a wrong approach. We really do not want to special
case all the places which require NODE_DATA. Can we please go and
allocate pgdat for all possible nodes?
The current state of memory less hacks subtle bugs poping up here and
there just prove that we should have done that from the very begining
IMHO.
Yes. So here's an alternative proposal for fixing the current situation in SLUB,
before the long-term solution of having all possible nodes provide valid pgdat
with zonelists:
- fix SLUB with the hunk at the end of this mail - the point is to use NUMA_NO_NODE
as fallback instead of node_to_mem_node()
- this removes all uses of node_to_mem_node (luckily it's just SLUB),
kill it completely instead of trying to fix it up
- patch 1/4 is not needed with the fix
- perhaps many of your other patches are alss not needed
- once we get the long-term solution, some of the !node_online() checks can be removed
@@ -1973,8 +1973,6 @@ static void *get_partial(struct kmem_cache *s, gfp_t flags, int node, if (node == NUMA_NO_NODE) searchnode = numa_mem_id();- else if (!node_present_pages(node))- searchnode = node_to_mem_node(node); object = get_partial_node(s, get_node(s, searchnode), c, flags); if (object || node != NUMA_NO_NODE)
@@ -2568,12 +2566,15 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, redo: if (unlikely(!node_match(page, node))) {- int searchnode = node;-- if (node != NUMA_NO_NODE && !node_present_pages(node))- searchnode = node_to_mem_node(node);-- if (unlikely(!node_match(page, searchnode))) {+ /*+ * node_match() false implies node != NUMA_NO_NODE+ * but if the node is not online and has no pages, just
From: Michael Ellerman <mpe@ellerman.id.au> Date: 2020-03-19 01:11:58
Michael Ellerman [off-list ref] writes:
Vlastimil Babka [off-list ref] writes:
quoted
On 3/18/20 11:02 AM, Michal Hocko wrote:
quoted
On Wed 18-03-20 12:58:07, Srikar Dronamraju wrote:
quoted
Calling a kmalloc_node on a possible node which is not yet onlined can
lead to panic. Currently node_present_pages() doesn't verify the node is
online before accessing the pgdat for the node. However pgdat struct may
not be available resulting in a crash.
NIP [c0000000003d55f4] ___slab_alloc+0x1f4/0x760
LR [c0000000003d5b94] __slab_alloc+0x34/0x60
Call Trace:
[c0000008b3783960] [c0000000003d5734] ___slab_alloc+0x334/0x760 (unreliable)
[c0000008b3783a40] [c0000000003d5b94] __slab_alloc+0x34/0x60
[c0000008b3783a70] [c0000000003d6fa0] __kmalloc_node+0x110/0x490
[c0000008b3783af0] [c0000000003443d8] kvmalloc_node+0x58/0x110
[c0000008b3783b30] [c0000000003fee38] mem_cgroup_css_online+0x108/0x270
[c0000008b3783b90] [c000000000235aa8] online_css+0x48/0xd0
[c0000008b3783bc0] [c00000000023eaec] cgroup_apply_control_enable+0x2ec/0x4d0
[c0000008b3783ca0] [c000000000242318] cgroup_mkdir+0x228/0x5f0
[c0000008b3783d10] [c00000000051e170] kernfs_iop_mkdir+0x90/0xf0
[c0000008b3783d50] [c00000000043dc00] vfs_mkdir+0x110/0x230
[c0000008b3783da0] [c000000000441c90] do_mkdirat+0xb0/0x1a0
[c0000008b3783e20] [c00000000000b278] system_call+0x5c/0x68
Fix this by verifying the node is online before accessing the pgdat
structure. Fix the same for node_spanned_pages() too.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Sachin Sant <redacted>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <redacted>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <redacted>
Cc: Kirill Tkhai <redacted>
Cc: Vlastimil Babka <redacted>
Cc: Srikar Dronamraju <redacted>
Cc: Bharata B Rao <redacted>
Cc: Nathan Lynch <redacted>
Reported-by: Sachin Sant <redacted>
Tested-by: Sachin Sant <redacted>
Signed-off-by: Srikar Dronamraju <redacted>
---
include/linux/mmzone.h | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
I believe this is a wrong approach. We really do not want to special
case all the places which require NODE_DATA. Can we please go and
allocate pgdat for all possible nodes?
The current state of memory less hacks subtle bugs poping up here and
there just prove that we should have done that from the very begining
IMHO.
Yes. So here's an alternative proposal for fixing the current situation in SLUB,
before the long-term solution of having all possible nodes provide valid pgdat
with zonelists:
- fix SLUB with the hunk at the end of this mail - the point is to use NUMA_NO_NODE
as fallback instead of node_to_mem_node()
- this removes all uses of node_to_mem_node (luckily it's just SLUB),
kill it completely instead of trying to fix it up
- patch 1/4 is not needed with the fix
- perhaps many of your other patches are alss not needed
- once we get the long-term solution, some of the !node_online() checks can be removed
@@ -1973,8 +1973,6 @@ static void *get_partial(struct kmem_cache *s, gfp_t flags, int node, if (node == NUMA_NO_NODE) searchnode = numa_mem_id();- else if (!node_present_pages(node))- searchnode = node_to_mem_node(node); object = get_partial_node(s, get_node(s, searchnode), c, flags); if (object || node != NUMA_NO_NODE)
@@ -2568,12 +2566,15 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, redo: if (unlikely(!node_match(page, node))) {- int searchnode = node;-- if (node != NUMA_NO_NODE && !node_present_pages(node))- searchnode = node_to_mem_node(node);-- if (unlikely(!node_match(page, searchnode))) {+ /*+ * node_match() false implies node != NUMA_NO_NODE+ * but if the node is not online and has no pages, just
^
this should be 'or' ?
Sorry I see you've already fixed this in the version you posted.
cheers
From: Michal Hocko <mhocko@suse.com> Date: 2020-03-19 07:44:43
On Wed 18-03-20 19:20:41, Cristopher Lameter wrote:
On Wed, 18 Mar 2020, Srikar Dronamraju wrote:
quoted
For a memoryless or offline nodes, node_numa_mem refers to a N_MEMORY
fallback node. Currently kernel has an API set_numa_mem that sets
node_numa_mem for memoryless node. However this API cannot be used for
offline nodes. Hence all offline nodes will have their node_numa_mem set
to 0. However systems can themselves have node 0 as offline i.e
That is a significant change to the basic assumptions for memory less
nodes. Node 0 needed to have memory and processors. Not sure what else
may break.
This assumption is simply incorrect. There many examples but just
one from top of my head 3e8589963773 ("memcg: make it work on sparse
non-0-node systems"). We simply have to forget that some nodes are
special.
--
Michal Hocko
SUSE Labs
Why don't we need the node_present_pages() check here?
Page allocator is fine with a node without present pages, as long as there's a
zonelist, which online nodes must have (ideally all possible nodes should have,
and then we can remove this).
SLUB on the other hand doesn't allocate cache per-cpu structures for nodes
without present pages (understandably) that's why the other place includes the
node_present_pages() check.
Thanks