(added Roman) On Tue, Jun 11, 2019 at 11:03:49AM +0100, Mark Rutland wrote: > On Mon, Jun 10, 2019 at 01:26:15PM -0400, Qian Cai wrote: > > On Mon, 2019-06-10 at 12:43 +0100, Will Deacon wrote: > > > On Tue, Jun 04, 2019 at 03:23:38PM +0100, Mark Rutland wrote: > > > > On Tue, Jun 04, 2019 at 10:00:36AM -0400, Qian Cai wrote: > > > > > The commit "arm64: switch to generic version of pte allocation" > > > > > introduced endless failures during boot like, > > > > > > > > > > kobject_add_internal failed for pgd_cache(285:chronyd.service) (error: > > > > > -2 parent: cgroup) > > > > > > > > > > It turns out __GFP_ACCOUNT is passed to kernel page table allocations > > > > > and then later memcg finds out those don't belong to any cgroup. > > > > > > > > Mike, I understood from [1] that this wasn't expected to be a problem, > > > > as the accounting should bypass kernel threads. > > > > > > > > Was that assumption wrong, or is something different happening here? > > > > > > > > > > > > > > backtrace: > > > > > kobject_add_internal > > > > > kobject_init_and_add > > > > > sysfs_slab_add+0x1a8 > > > > > __kmem_cache_create > > > > > create_cache > > > > > memcg_create_kmem_cache > > > > > memcg_kmem_cache_create_func > > > > > process_one_work > > > > > worker_thread > > > > > kthread > > > > > > > > > > Signed-off-by: Qian Cai <cai@xxxxxx> > > > > > --- > > > > > arch/arm64/mm/pgd.c | 2 +- > > > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > > > > > > > diff --git a/arch/arm64/mm/pgd.c b/arch/arm64/mm/pgd.c > > > > > index 769516cb6677..53c48f5c8765 100644 > > > > > --- a/arch/arm64/mm/pgd.c > > > > > +++ b/arch/arm64/mm/pgd.c > > > > > @@ -38,7 +38,7 @@ pgd_t *pgd_alloc(struct mm_struct *mm) > > > > > if (PGD_SIZE == PAGE_SIZE) > > > > > return (pgd_t *)__get_free_page(gfp); > > > > > else > > > > > - return kmem_cache_alloc(pgd_cache, gfp); > > > > > + return kmem_cache_alloc(pgd_cache, GFP_PGTABLE_KERNEL); > > > > > > > > This is used to allocate PGDs for both user and kernel pagetables (e.g. > > > > for the efi runtime services), so while this may fix the regression, I'm > > > > not sure it's the right fix. > > > > > > > > Do we need a separate pgd_alloc_kernel()? > > > > > > So can I take the above for -rc5, or is somebody else working on a different > > > fix to implement pgd_alloc_kernel()? > > > > The offensive commit "arm64: switch to generic version of pte allocation" is not > > yet in the mainline, but only in the Andrew's tree and linux-next, and I doubt > > Andrew will push this out any time sooner given it is broken. > > I'd assumed that Mike would respin these patches to implement and use > pgd_alloc_kernel() (or take gfp flags) and the updated patches would > replace these in akpm's tree. > > Mike, could you confirm what your plan is? I'm happy to review/test > updated patches for arm64. The log Qian Cai posted at [1] and partially cited below confirms that the failure happens when *user* PGDs are allocated and the addition of __GFP_ACCOUNT to gfp flags used by pgd_alloc() only uncovered another issue. I'm still failing to reproduce it with qemu and I'm not really familiar with slub/memcg code to say anything smart about it. Will keep looking. Note, that as failures start way after efi_virtmap_init() that allocates a PGD for efi_mm, there are no real fixes required for the original series, except that the check for mm == &init_mm I copied for some reason from powerpc is bogus and can be removed. I surely can add pgd_alloc_kernel() to be used by the EFI code to make sure we won't run into issues with memcg in the future. [ 82.125966] Freeing unused kernel memory: 28672K [ 87.940365] Checked W+X mappings: passed, no W+X pages found [ 87.946769] Run /init as init process [ 88.040040] systemd[1]: System time before build time, advancing clock. [ 88.054593] systemd[1]: Failed to insert module 'autofs4': No such file or directory [ 88.374129] modprobe (1726) used greatest stack depth: 28464 bytes left [ 88.470108] systemd[1]: systemd 239 running in system mode. (+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD +IDN2 -IDN +PCRE2 default-hierarchy=legacy) [ 88.498398] systemd[1]: Detected architecture arm64. [ 88.506517] systemd[1]: Running in initial RAM disk. [ 89.621995] mkdir (1730) used greatest stack depth: 27872 bytes left [ 90.222658] random: systemd: uninitialized urandom read (16 bytes read) [ 90.230072] systemd[1]: Reached target Swap. [ 90.240205] random: systemd: uninitialized urandom read (16 bytes read) [ 90.251088] systemd[1]: Reached target Timers. [ 90.261303] random: systemd: uninitialized urandom read (16 bytes read) [ 90.271209] systemd[1]: Listening on udev Control Socket. [ 90.283238] systemd[1]: Reached target Local File Systems. [ 90.296232] systemd[1]: Reached target Slices. [ 90.307239] systemd[1]: Listening on udev Kernel Socket. [ 90.608597] kobject_add_internal failed for pgd_cache(13:init.scope) (error: -2 parent: cgroup) [ 90.678007] kobject_add_internal failed for pgd_cache(13:init.scope)(error: -2 parent: cgroup) [ 90.713260] kobject_add_internal failed for pgd_cache(21:systemd-tmpfiles-setup.service) (error: -2 parent: cgroup) [ 90.820012] systemd-tmpfile (1759) used greatest stack depth: 27184 bytes left [ 90.861942] kobject_add_internal failed for pgd_cache(13:init.scope) error: -2 parent: cgroup) > Thanks, > Mark. > [1] https://cailca.github.io/files/dmesg.txt -- Sincerely yours, Mike.
