Re: mm/pgtable: add stubs for {pmd/pub}_{set/clear}_huge
From: Geert Uytterhoeven
Date: Tue Jul 13 2021 - 10:39:39 EST
Hi Christophe,
On Fri, Jul 2, 2021 at 10:16 PM Linux Kernel Mailing List
<linux-kernel@xxxxxxxxxxxxxxx> wrote:
> Commit: c742199a014de23ee92055c2473d91fe5561ffdf
> Parent: 79c1c594f49a88fba9744cb5c85978c6b1b365ec
> Refname: refs/heads/master
> Web: https://git.kernel.org/torvalds/c/c742199a014de23ee92055c2473d91fe5561ffdf
> Author: Christophe Leroy <christophe.leroy@xxxxxxxxxx>
> AuthorDate: Wed Jun 30 18:48:03 2021 -0700
> Committer: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
> CommitDate: Wed Jun 30 20:47:26 2021 -0700
>
> mm/pgtable: add stubs for {pmd/pub}_{set/clear}_huge
>
> For architectures with no PMD and/or no PUD, add stubs similar to what we
> have for architectures without P4D.
>
> [christophe.leroy@xxxxxxxxxx: arm64: define only {pud/pmd}_{set/clear}_huge when useful]
> Link: https://lkml.kernel.org/r/73ec95f40cafbbb69bdfb43a7f53876fd845b0ce.1620990479.git.christophe.leroy@xxxxxxxxxx
> [christophe.leroy@xxxxxxxxxx: x86: define only {pud/pmd}_{set/clear}_huge when useful]
> Link: https://lkml.kernel.org/r/7fbf1b6bc3e15c07c24fa45278d57064f14c896b.1620930415.git.christophe.leroy@xxxxxxxxxx
>
> Link: https://lkml.kernel.org/r/5ac5976419350e8e048d463a64cae449eb3ba4b0.1620795204.git.christophe.leroy@xxxxxxxxxx
> Signed-off-by: Christophe Leroy <christophe.leroy@xxxxxxxxxx>
> Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
> Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx>
> Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
> Cc: Mike Rapoport <rppt@xxxxxxxxxx>
> Cc: Nicholas Piggin <npiggin@xxxxxxxxx>
> Cc: Paul Mackerras <paulus@xxxxxxxxx>
> Cc: Uladzislau Rezki <uladzislau.rezki@xxxxxxxx>
> Cc: Naresh Kamboju <naresh.kamboju@xxxxxxxxxx>
> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Since this commit, both WARN_ON() tests in pud_huge_tests() are
triggering on arm64 if CONFIG_ARM64_VA_BITS_39=y and
CONFIG_PGTABLE_LEVELS=3:
------------[ cut here ]------------
WARNING: CPU: 0 PID: 1 at mm/debug_vm_pgtable.c:438
debug_vm_pgtable+0x6d0/0x7c4
Modules linked in:
CPU: 0 PID: 1 Comm: swapper/0 Not tainted
5.13.0-salvator-x-00209-gc742199a014d #1068
Hardware name: Renesas Salvator-X 2nd version board based on r8a77951 (DT)
pstate: a0400005 (NzCv daif +PAN -UAO -TCO BTYPE=--)
pc : debug_vm_pgtable+0x6d0/0x7c4
lr : debug_vm_pgtable+0x6cc/0x7c4
sp : ffffffc0124cbd00
x29: ffffffc0124cbd00 x28: 0000000000000c30 x27: ffffff84c5e465a0
x26: ffffff84c5e46000 x25: ffffff84c5e42000 x24: 00000000000005a0
x23: ffffffc010c41000 x22: 0020000050cb0fc3 x21: ffffff84c5e45ed0
x20: ffffff84c27470a8 x19: ffffff84c2747040 x18: 0000000000000cc0
x17: 00000000b2491cf5 x16: 0000000000000014 x15: 0000000000000000
x14: 0000000000000002 x13: 00000000000b56dd x12: 0000000000000901
x11: ffffffc01116d040 x10: 00000000000038e4 x9 : 00000000ffffffff
x8 : 0000000000000000 x7 : ffffffc010cc4cd0 x6 : 0000000000000000
x5 : 000000000004028e x4 : 0000000000000000 x3 : 0000000000000000
x2 : 0000000000000001 x1 : 0000000000000000 x0 : ffffff84c08bc040
Call trace:
debug_vm_pgtable+0x6d0/0x7c4
do_one_initcall+0xec/0x278
kernel_init_freeable+0x220/0x224
kernel_init+0x10/0x108
ret_from_fork+0x10/0x18
irq event stamp: 1062694
hardirqs last enabled at (1062693): [<ffffffc0101bb030>]
free_unref_page+0x120/0x138
hardirqs last disabled at (1062694): [<ffffffc01085f31c>] el1_dbg+0x24/0x4c
softirqs last enabled at (1062636): [<ffffffc0100101c8>] _stext+0x1c8/0x43c
softirqs last disabled at (1062627): [<ffffffc010080020>]
__irq_exit_rcu+0xdc/0x124
---[ end trace 864024bcea12ea00 ]---
------------[ cut here ]------------
WARNING: CPU: 0 PID: 1 at mm/debug_vm_pgtable.c:439
debug_vm_pgtable+0x6d4/0x7c4
Modules linked in:
CPU: 0 PID: 1 Comm: swapper/0 Tainted: G W
5.13.0-salvator-x-00209-gc742199a014d #1068
Hardware name: Renesas Salvator-X 2nd version board based on r8a77951 (DT)
pstate: a0400005 (NzCv daif +PAN -UAO -TCO BTYPE=--)
pc : debug_vm_pgtable+0x6d4/0x7c4
lr : debug_vm_pgtable+0x6cc/0x7c4
sp : ffffffc0124cbd00
x29: ffffffc0124cbd00 x28: 0000000000000c30 x27: ffffff84c5e465a0
x26: ffffff84c5e46000 x25: ffffff84c5e42000 x24: 00000000000005a0
x23: ffffffc010c41000 x22: 0020000050cb0fc3 x21: ffffff84c5e45ed0
x20: ffffff84c27470a8 x19: ffffff84c2747040 x18: 0000000000000cc0
x17: 00000000b2491cf5 x16: 0000000000000014 x15: 0000000000000000
x14: 0000000000000002 x13: 00000000000b56dd x12: 0000000000000901
x11: ffffffc01116d040 x10: 00000000000038e4 x9 : 00000000ffffffff
x8 : 0000000000000000 x7 : ffffffc010cc4cd0 x6 : 0000000000000000
x5 : 000000000004028e x4 : 0000000000000000 x3 : 0000000000000000
x2 : 0000000000000001 x1 : 0000000000000000 x0 : ffffff84c08bc040
Call trace:
debug_vm_pgtable+0x6d4/0x7c4
do_one_initcall+0xec/0x278
kernel_init_freeable+0x220/0x224
kernel_init+0x10/0x108
ret_from_fork+0x10/0x18
irq event stamp: 1062706
hardirqs last enabled at (1062705): [<ffffffc01085ef80>]
exit_to_kernel_mode+0xf0/0x100
hardirqs last disabled at (1062706): [<ffffffc01085f31c>] el1_dbg+0x24/0x4c
softirqs last enabled at (1062704): [<ffffffc0100101c8>] _stext+0x1c8/0x43c
softirqs last disabled at (1062697): [<ffffffc010080020>]
__irq_exit_rcu+0xdc/0x124
---[ end trace 864024bcea12ea01 ]---
Switching to CONFIG_ARM64_VA_BITS_48=y and CONFIG_PGTABLE_LEVELS=4
makes the warning disappear.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds