Messages in this thread | | | Date | Wed, 18 Oct 2023 20:26:31 +0800 | Subject | Re: [PATCH v2] x86/mm: Drop 4MB restriction on minimal NUMA node memory size | From | Qi Zheng <> |
| |
Hi all,
On 2023/10/18 18:42, Ingo Molnar wrote: > > * Mike Rapoport <rppt@kernel.org> wrote: > >> From: "Mike Rapoport (IBM)" <rppt@kernel.org> >> >> Qi Zheng reports crashes in a production environment and provides a >> simplified example as a reproducer: >> >> For example, if we use qemu to start a two NUMA node kernel, >> one of the nodes has 2M memory (less than NODE_MIN_SIZE), >> and the other node has 2G, then we will encounter the >> following panic: >> >> [ 0.149844] BUG: kernel NULL pointer dereference, address: 0000000000000000 >> [ 0.150783] #PF: supervisor write access in kernel mode >> [ 0.151488] #PF: error_code(0x0002) - not-present page >> <...> >> [ 0.156056] RIP: 0010:_raw_spin_lock_irqsave+0x22/0x40 >> <...> >> [ 0.169781] Call Trace: >> [ 0.170159] <TASK> >> [ 0.170448] deactivate_slab+0x187/0x3c0 >> [ 0.171031] ? bootstrap+0x1b/0x10e >> [ 0.171559] ? preempt_count_sub+0x9/0xa0 >> [ 0.172145] ? kmem_cache_alloc+0x12c/0x440 >> [ 0.172735] ? bootstrap+0x1b/0x10e >> [ 0.173236] bootstrap+0x6b/0x10e >> [ 0.173720] kmem_cache_init+0x10a/0x188 >> [ 0.174240] start_kernel+0x415/0x6ac >> [ 0.174738] secondary_startup_64_no_verify+0xe0/0xeb >> [ 0.175417] </TASK> >> [ 0.175713] Modules linked in: >> [ 0.176117] CR2: 0000000000000000 >> >> The crashes happen because of inconsistency between nodemask that has >> nodes with less than 4MB as memoryless and the actual memory fed into >> core mm. > > Presumably the core MM got fixed too to not just crash, but provide some > sort of warning? > >> The commit 9391a3f9c7f1 ("[PATCH] x86_64: Clear more state when ignoring >> empty node in SRAT parsing") that introduced minimal size of a NUMA node >> does not explain why a node size cannot be less than 4MB and what boot >> failures this restriction might fix. >> >> Since then a lot has changed and core mm won't confuse badly about small >> node sizes. > > Core MM won't get confused ... other than by the above weird Qemu topology, > to which it responds with a ... NULL pointer dereference? > > Seems quite close to the literal definition of 'get confused badly' to me, > and doesn't give me the warm fuzzy feeling that giving the core MM even > *more* weird topologies is super safe ... :-/ > >> Drop the limitation for the minimal node size. > > While I agree with dropping the limitation, and I agree that 9391a3f9c7f1 > should have provided more of a justification, I believe a core MM fix is in > order as well, for it to not crash. [ If it's fixed upstream already, > please reference the relevant commit ID. ]
Agree. I posted a fixed patchset[1] before, maybe we can reconsider it. :)
[1]. https://lore.kernel.org/lkml/20230215152412.13368-1-zhengqi.arch@bytedance.com/
For memoryless node, this patchset skip it and fallback to other nodes when build its zonelists.
Say we have node0 and node1, and node0 is memoryless, then:
[ 0.102400] Fallback order for Node 0: 1 [ 0.102931] Fallback order for Node 1: 1
In this way, we will not allocate pages from memoryless node0. Then the crash problem under the weird Qemu topology will be fixed.
Thanks, Qi
> > Also, the changelog spelling & general presentation were quite low quality > - I've fixed it up a bit below, please carry this version going forward. > Please spell-check your patches before sending out Nth versions of it, > maybe maintainers are skipping them for a reason! > > Thanks, > > Ingo > > =================> > From: "Mike Rapoport (IBM)" <rppt@kernel.org> > Date: Tue, 17 Oct 2023 09:22:15 +0300 > Subject: [PATCH] x86/mm: Drop 4MB restriction on minimal NUMA node memory size > > Qi Zheng reported crashes in a production environment and provided a > simplified example as a reproducer: > > | For example, if we use qemu to start a two NUMA node kernel, > | one of the nodes has 2M memory (less than NODE_MIN_SIZE), > | and the other node has 2G, then we will encounter the > | following panic: > | > | BUG: kernel NULL pointer dereference, address: 0000000000000000 > | <...> > | RIP: 0010:_raw_spin_lock_irqsave+0x22/0x40 > | <...> > | Call Trace: > | <TASK> > | deactivate_slab() > | bootstrap() > | kmem_cache_init() > | start_kernel() > | secondary_startup_64_no_verify() > > The crashes happen because of inconsistency between the nodemask that > has nodes with less than 4MB as memoryless, and the actual memory fed > into the core mm. > > The commit: > > 9391a3f9c7f1 ("[PATCH] x86_64: Clear more state when ignoring empty node in SRAT parsing") > > ... that introduced minimal size of a NUMA node does not explain why > a node size cannot be less than 4MB and what boot failures this > restriction might fix. > > In the 17 years since then a lot has changed and core mm won't get > confused about small node sizes. > > Drop the limitation for the minimal node size. > > [ mingo: Improved changelog clarity. ] > > Reported-by: Qi Zheng <zhengqi.arch@bytedance.com> > Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org> > Not-Yet-Signed-off-by: Ingo Molnar <mingo@kernel.org> > Acked-by: David Hildenbrand <david@redhat.com> > Acked-by: Michal Hocko <mhocko@suse.com> > Link: https://lore.kernel.org/all/20230212110305.93670-1-zhengqi.arch@bytedance.com/ > --- > arch/x86/include/asm/numa.h | 7 ------- > arch/x86/mm/numa.c | 7 ------- > 2 files changed, 14 deletions(-) > > diff --git a/arch/x86/include/asm/numa.h b/arch/x86/include/asm/numa.h > index e3bae2b60a0d..ef2844d69173 100644 > --- a/arch/x86/include/asm/numa.h > +++ b/arch/x86/include/asm/numa.h > @@ -12,13 +12,6 @@ > > #define NR_NODE_MEMBLKS (MAX_NUMNODES*2) > > -/* > - * Too small node sizes may confuse the VM badly. Usually they > - * result from BIOS bugs. So dont recognize nodes as standalone > - * NUMA entities that have less than this amount of RAM listed: > - */ > -#define NODE_MIN_SIZE (4*1024*1024) > - > extern int numa_off; > > /* > diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c > index c01c5506fd4a..aa39d678fe81 100644 > --- a/arch/x86/mm/numa.c > +++ b/arch/x86/mm/numa.c > @@ -602,13 +602,6 @@ static int __init numa_register_memblks(struct numa_meminfo *mi) > if (start >= end) > continue; > > - /* > - * Don't confuse VM with a node that doesn't have the > - * minimum amount of memory: > - */ > - if (end && (end - start) < NODE_MIN_SIZE) > - continue; > - > alloc_node_data(nid); > } >
| |