Re: [PATCH -next] arm64/mm: fix a bogus GFP flag in pgd_alloc()

[Mike Rapoport]

(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.