* [PATCH] mempolicy: check home_node is in the nodes of policy @ 2024-01-26 13:22 Chunsheng Luo 2024-01-28 8:29 ` Andrew Morton 2024-01-29 15:12 ` Gregory Price 0 siblings, 2 replies; 6+ messages in thread From: Chunsheng Luo @ 2024-01-26 13:22 UTC (permalink / raw) To: akpm; +Cc: linux-mm, linux-kernel, Chunsheng Luo set_mempolicy_home_node should be used after setting the memory policy. If the home_node isn't in the nodes of policy, we should return failure to avoid misunderstanding. Signed-off-by: Chunsheng Luo <luochunsheng@ustc.edu> --- mm/mempolicy.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 10a590ee1c89..9282be2ae18e 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1536,6 +1536,12 @@ SYSCALL_DEFINE4(set_mempolicy_home_node, unsigned long, start, unsigned long, le err = -EOPNOTSUPP; break; } + + if (!node_isset(home_node, old->nodes)) { + err = -EINVAL; + break; + } + new = mpol_dup(old); if (IS_ERR(new)) { err = PTR_ERR(new); -- 2.43.0 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] mempolicy: check home_node is in the nodes of policy 2024-01-26 13:22 [PATCH] mempolicy: check home_node is in the nodes of policy Chunsheng Luo @ 2024-01-28 8:29 ` Andrew Morton 2024-01-29 6:15 ` ustc 2024-01-29 15:12 ` Gregory Price 1 sibling, 1 reply; 6+ messages in thread From: Andrew Morton @ 2024-01-28 8:29 UTC (permalink / raw) To: Chunsheng Luo; +Cc: linux-mm, linux-kernel On Fri, 26 Jan 2024 08:22:40 -0500 Chunsheng Luo <luochunsheng@ustc.edu> wrote: > set_mempolicy_home_node should be used after setting the memory > policy. If the home_node isn't in the nodes of policy, we should > return failure to avoid misunderstanding. Thanks. Under what circumstances does userspace trigger this issue? ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] mempolicy: check home_node is in the nodes of policy 2024-01-28 8:29 ` Andrew Morton @ 2024-01-29 6:15 ` ustc 0 siblings, 0 replies; 6+ messages in thread From: ustc @ 2024-01-29 6:15 UTC (permalink / raw) To: Andrew Morton; +Cc: linux-mm, linux-kernel For example, In a system with NUMA nodes 0,1,2,3, i mbind process to node 0-2 and set home_node to node 3, it will not be allocated from node 3, then from node closer to node 3. But i think home_node should be set directly from node 0-2, which makes more sense. So i think it needs to return failure to prompt user. On 2024/1/28 16:29, Andrew Morton wrote: > On Fri, 26 Jan 2024 08:22:40 -0500 Chunsheng Luo <luochunsheng@ustc.edu> wrote: > >> set_mempolicy_home_node should be used after setting the memory >> policy. If the home_node isn't in the nodes of policy, we should >> return failure to avoid misunderstanding. > Thanks. Under what circumstances does userspace trigger this issue? > ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] mempolicy: check home_node is in the nodes of policy 2024-01-26 13:22 [PATCH] mempolicy: check home_node is in the nodes of policy Chunsheng Luo 2024-01-28 8:29 ` Andrew Morton @ 2024-01-29 15:12 ` Gregory Price 2024-01-30 2:29 ` Chunsheng Luo 1 sibling, 1 reply; 6+ messages in thread From: Gregory Price @ 2024-01-29 15:12 UTC (permalink / raw) To: Chunsheng Luo; +Cc: akpm, linux-mm, linux-kernel On Fri, Jan 26, 2024 at 08:22:40AM -0500, Chunsheng Luo wrote: > set_mempolicy_home_node should be used after setting the memory > policy. If the home_node isn't in the nodes of policy, we should > return failure to avoid misunderstanding. > > Signed-off-by: Chunsheng Luo <luochunsheng@ustc.edu> > --- > mm/mempolicy.c | 6 ++++++ > 1 file changed, 6 insertions(+) > Since it's not possible to add/remove a node to a mask without also erasing the home node, this seems reasonable. e.g. this is what happens presently mbind(0-2) : mask(0,1,2), home_node(NUMA_NO_NODE) home_node(3) : mask(0,1,2), home_node(3) mbind(0-3) : mask(0,1,2,3), home_node(NUMA_NO_NODE) However, it is possible for a cgroup migration or a change to cpusets.mems_allowed to change a nodemask without somping the home_node. e.g.: mbind(2-3) : mask(2-3), home_node(NUMA_NO_NODE) home_node(3) : mask(2-3), home_node(3) cpusets(0-1) : mask(0-1), home_node(3) Should the rebind code also shift the home-node or un-set it accordingly to keep the mask/home_node behavior consistent with the syscalls? (see mpol_rebind_nodemask) > diff --git a/mm/mempolicy.c b/mm/mempolicy.c > index 10a590ee1c89..9282be2ae18e 100644 > --- a/mm/mempolicy.c > +++ b/mm/mempolicy.c > @@ -1536,6 +1536,12 @@ SYSCALL_DEFINE4(set_mempolicy_home_node, unsigned long, start, unsigned long, le > err = -EOPNOTSUPP; > break; > } > + > + if (!node_isset(home_node, old->nodes)) { > + err = -EINVAL; > + break; > + } > + > new = mpol_dup(old); > if (IS_ERR(new)) { > err = PTR_ERR(new); > -- > 2.43.0 > > ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] mempolicy: check home_node is in the nodes of policy 2024-01-29 15:12 ` Gregory Price @ 2024-01-30 2:29 ` Chunsheng Luo 2024-01-30 4:02 ` Chunsheng Luo 0 siblings, 1 reply; 6+ messages in thread From: Chunsheng Luo @ 2024-01-30 2:29 UTC (permalink / raw) To: Gregory Price; +Cc: akpm, linux-mm, linux-kernel [-- Attachment #1: Type: text/plain, Size: 2132 bytes --] On 2024/1/29 23:12, Gregory Price wrote: > On Fri, Jan 26, 2024 at 08:22:40AM -0500, Chunsheng Luo wrote: >> set_mempolicy_home_node should be used after setting the memory >> policy. If the home_node isn't in the nodes of policy, we should >> return failure to avoid misunderstanding. >> >> Signed-off-by: Chunsheng Luo<luochunsheng@ustc.edu> >> --- >> mm/mempolicy.c | 6 ++++++ >> 1 file changed, 6 insertions(+) >> > Since it's not possible to add/remove a node to a mask without also > erasing the home node, this seems reasonable. > > e.g. this is what happens presently > mbind(0-2) : mask(0,1,2), home_node(NUMA_NO_NODE) > home_node(3) : mask(0,1,2), home_node(3) > mbind(0-3) : mask(0,1,2,3), home_node(NUMA_NO_NODE) > > However, it is possible for a cgroup migration or a change to > cpusets.mems_allowed to change a nodemask without somping the home_node. > > e.g.: > mbind(2-3) : mask(2-3), home_node(NUMA_NO_NODE) > home_node(3) : mask(2-3), home_node(3) > cpusets(0-1) : mask(0-1), home_node(3) > > Should the rebind code also shift the home-node or un-set it accordingly > to keep the mask/home_node behavior consistent with the syscalls? > > (see mpol_rebind_nodemask) Thank you for your reply. First, home_node can only be set for VMA policy. second, the result of cgroup migration is consistent with the result of the first e.g. when process mem_allowed updates, vma policy will alse be updated. Function call: update_tasks_nodemask ->mpol_rebind_mm -> for_each_vma: mpol_rebind_policy -> mpol_rebind_nodemask > >> diff --git a/mm/mempolicy.c b/mm/mempolicy.c >> index 10a590ee1c89..9282be2ae18e 100644 >> --- a/mm/mempolicy.c >> +++ b/mm/mempolicy.c >> @@ -1536,6 +1536,12 @@ SYSCALL_DEFINE4(set_mempolicy_home_node, unsigned long, start, unsigned long, le >> err = -EOPNOTSUPP; >> break; >> } >> + >> + if (!node_isset(home_node, old->nodes)) { >> + err = -EINVAL; >> + break; >> + } >> + >> new = mpol_dup(old); >> if (IS_ERR(new)) { >> err = PTR_ERR(new); >> -- >> 2.43.0 >> >> [-- Attachment #2: Type: text/html, Size: 3004 bytes --] ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] mempolicy: check home_node is in the nodes of policy 2024-01-30 2:29 ` Chunsheng Luo @ 2024-01-30 4:02 ` Chunsheng Luo 0 siblings, 0 replies; 6+ messages in thread From: Chunsheng Luo @ 2024-01-30 4:02 UTC (permalink / raw) To: Gregory Price; +Cc: akpm, linux-mm, linux-kernel [-- Attachment #1: Type: text/plain, Size: 2434 bytes --] On 2024/1/30 10:29, Chunsheng Luo wrote: > > > On 2024/1/29 23:12, Gregory Price wrote: >> On Fri, Jan 26, 2024 at 08:22:40AM -0500, Chunsheng Luo wrote: >>> set_mempolicy_home_node should be used after setting the memory >>> policy. If the home_node isn't in the nodes of policy, we should >>> return failure to avoid misunderstanding. >>> >>> Signed-off-by: Chunsheng Luo<luochunsheng@ustc.edu> >>> --- >>> mm/mempolicy.c | 6 ++++++ >>> 1 file changed, 6 insertions(+) >>> >> Since it's not possible to add/remove a node to a mask without also >> erasing the home node, this seems reasonable. >> >> e.g. this is what happens presently >> mbind(0-2) : mask(0,1,2), home_node(NUMA_NO_NODE) >> home_node(3) : mask(0,1,2), home_node(3) >> mbind(0-3) : mask(0,1,2,3), home_node(NUMA_NO_NODE) >> >> However, it is possible for a cgroup migration or a change to >> cpusets.mems_allowed to change a nodemask without somping the home_node. >> >> e.g.: >> mbind(2-3) : mask(2-3), home_node(NUMA_NO_NODE) >> home_node(3) : mask(2-3), home_node(3) >> cpusets(0-1) : mask(0-1), home_node(3) >> >> Should the rebind code also shift the home-node or un-set it accordingly >> to keep the mask/home_node behavior consistent with the syscalls? >> >> (see mpol_rebind_nodemask) > Thank you for your reply. > First, home_node can only be set for VMA policy. > second, the result of cgroup migration is consistent with the result of the first e.g. > when process mem_allowed updates, vma policy will alse be updated. > Function call: > update_tasks_nodemask > ->mpol_rebind_mm > -> for_each_vma: mpol_rebind_policy > -> mpol_rebind_nodemask Sorry, I make a mistake. The mpol_rebind_nodemask() only modifies the nodemask of the vma policy, but doesn't modify the home_node, so the behavior is different from do_mbind(), so you are right. >>> diff --git a/mm/mempolicy.c b/mm/mempolicy.c >>> index 10a590ee1c89..9282be2ae18e 100644 >>> --- a/mm/mempolicy.c >>> +++ b/mm/mempolicy.c >>> @@ -1536,6 +1536,12 @@ SYSCALL_DEFINE4(set_mempolicy_home_node, unsigned long, start, unsigned long, le >>> err = -EOPNOTSUPP; >>> break; >>> } >>> + >>> + if (!node_isset(home_node, old->nodes)) { >>> + err = -EINVAL; >>> + break; >>> + } >>> + >>> new = mpol_dup(old); >>> if (IS_ERR(new)) { >>> err = PTR_ERR(new); >>> -- >>> 2.43.0 >>> >>> [-- Attachment #2: Type: text/html, Size: 3661 bytes --] ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2024-01-30 4:02 UTC | newest] Thread overview: 6+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2024-01-26 13:22 [PATCH] mempolicy: check home_node is in the nodes of policy Chunsheng Luo 2024-01-28 8:29 ` Andrew Morton 2024-01-29 6:15 ` ustc 2024-01-29 15:12 ` Gregory Price 2024-01-30 2:29 ` Chunsheng Luo 2024-01-30 4:02 ` Chunsheng Luo
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).