Skip to contentSkip to conversation

sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()

PATCH Guenter Roeck 2 participants 3 messages May 23, 2020 – May 24, 2020 Stable branch

lore
Share & tools
lore
#0 lore raw

Hi,

Quoted 43 lines · Mike Rapoport

The kbuild test robot reported the following warning:

arch/sparc/mm/srmmu.c: In function 'srmmu_nocache_init': arch/sparc/mm/srmmu.c:300:9: error: variable 'pud' set but not used [-Werror=unused-but-set-variable] 300 | pud_t *pud;

This warning is caused by misprint in the page table traversal in srmmu_nocache_init() function which accessed a PMD entry using PGD rather than PUD.

Since sparc32 has only 3 page table levels, the PGD and PUD are essentially the same and usage of __nocache_fix() removed the type checking.

Use PUD for the consistency and to silence the compiler warning.

Fixes 7235db268a2777bc38 ("sparc32: use pgtable-nopud instead of 4level-fixup")
Cc David S. Miller <[email protected]>
Cc Anatoly Pugachev <[email protected]>
---
 arch/sparc/mm/srmmu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/sparc/mm/srmmu.c +1 -1
index b7c94de70cca..e9f7af32da07 100644
@@ -333,7 +333,7 @@ static void __init srmmu_nocache_init(void)
 		pgd = pgd_offset_k(vaddr);
 		p4d = p4d_offset(__nocache_fix(pgd), vaddr);
 		pud = pud_offset(__nocache_fix(p4d), vaddr);
-		pmd = pmd_offset(__nocache_fix(pgd), vaddr);
+		pmd = pmd_offset(__nocache_fix(pud), vaddr);
 		pte = pte_offset_kernel(__nocache_fix(pmd), vaddr);
 
 		pteval = ((paddr >> 4) | SRMMU_ET_PTE | SRMMU_PRIV);
Signature · 1 line
2.17.1

This quite innocent looking patch crashes all my sparc32 boot tests. Crash log and bisect results below. Reverting it fixes the problem.

Guenter

---
# bad: [423b8baf18a8c03f2d6fa99aa447ed0da189bb95] Merge branch 'akpm' (patches from Andrew)
# good: [051143e1602d90ea71887d92363edd539d411de5] Merge tag 'apparmor-pr-2020-05-21' of git://git.kernel.org/pub/scm/linux/kernel/git/jj/linux-apparmor
git bisect start 'HEAD' '051143e1602d'
# good: [e644645abf4788e919beeb97925fb6bf43e890a2] Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
git bisect good e644645abf4788e919beeb97925fb6bf43e890a2
# good: [8f261041b18ee80ad8afdd1621c909c4df9f6cc3] Merge tag 'staging-5.7-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
git bisect good 8f261041b18ee80ad8afdd1621c909c4df9f6cc3
# good: [23f0dac848412dafd197566b62d831d5a68b5b6b] Merge tag 'driver-core-5.7-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect good 23f0dac848412dafd197566b62d831d5a68b5b6b
# good: [33cd65e73abd693c00c4156cf23677c453b41b3b] kasan: disable branch tracing for core runtime
git bisect good 33cd65e73abd693c00c4156cf23677c453b41b3b
# bad: [c2bc26f7ca1ff1165bb6669a7a4cccc20ffd2ced] sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()
git bisect bad c2bc26f7ca1ff1165bb6669a7a4cccc20ffd2ced
# good: [f7fa1876af81512444631d324adb77383f56c37a] MAINTAINERS: update email address for Naoya Horiguchi
git bisect good f7fa1876af81512444631d324adb77383f56c37a

# first bad commit: [c2bc26f7ca1ff1165bb6669a7a4cccc20ffd2ced] sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()

---
Configuration device id QEMU version 1 machine id 36
Probing SBus slot 0 offset 0
Probing SBus slot 1 offset 0
Probing SBus slot 2 offset 0
Probing SBus slot 3 offset 0
Probing SBus slot 4 offset 0
Probing SBus slot 5 offset 0
Invalid FCode start byte
CPUs: 1 x FMI,MB86904
UUID: 00000000-0000-0000-0000-000000000000
Welcome to OpenBIOS v1.1 built on Oct 28 2019 17:08
Type 'help' for detailed information
[sparc] Kernel already loaded
switching to new context:
PROMLIB: obio_ranges 1
PROMLIB: Sun Boot Prom Version 3 Revision 2
Linux version 5.7.0-rc6-00161-g423b8baf18a8 ([email protected]) (gcc version 6.5.0 (Buildroot 2018.11-rc2-00071-g4310260), GNU ld (GNU Binutils) 2.31.1) #1 Sat May 23 11:57:22 PDT 2020
printk: bootconsole [earlyprom0] enabled
ARCH: SUN4M
TYPE: SPARCstation Classic
Ethernet address: 52:54:00:12:34:56

Unable to handle kernel paging request at virtual address d9440000
tsk->{mm,active_mm}->context = ffffffff
tsk->{mm,active_mm}->pgd = fc000000
              \|/ ____ \|/
              "@'/ ,. \`@"
              /_| \__/ |_\
                 \__U_/
swapper(0): Oops [#1]
CPU: 0 PID: 0 Comm: swapper Not tainted 5.7.0-rc6-00161-g423b8baf18a8 #1
PSR: 04901fc3 PC: f06592a0 NPC: f06592a4 Y: 00000000    Not tainted
PC: <srmmu_paging_init+0x450/0xba0>
%G: 00000008 f06c0000  00000000 cd4403f0  fc000000 006c0000  f0622000 ec000000
%O: 0006c200 000003f0  0c000000 f06c0000  00000000 00001000  f0623d48 0c000000
RPC: <0xc000000>
%L: f067f000 f067f1d4  f067f000 f067f344  0fffffff 0000000f  14000000 fc0003f0
%I: f0000000 f0636698  04000000 f067f000  f067f000 f067f000  f0623e68 f06585dc

Disabling lock debugging due to kernel taint
Caller[f06585dc]: paging_init+0x4/0x24
Caller[f06565b8]: setup_arch+0x430/0x468
Caller[f0652b98]: start_kernel+0x48/0x520
Caller[f065243c]: continue_boot+0x324/0x334
Caller[00000000]: 0x0
Instruction DUMP:
8600c001
8600c001
8600c001
<c600c00a>
8088c018
12800007
9a103fff
8728e004
9a21c002

Kernel panic - not syncing: Attempted to kill the idle task!
Press Stop-A (L1-A) from sun keyboard or send break
twice on console to return to the boot prom
---[ end Kernel panic - not syncing: Attempted to kill the idle task! ]---
#1 lore raw
Quoted 3 lines · Guenter Roeck

This quite innocent looking patch crashes all my sparc32 boot tests. Crash log and bisect results below. Reverting it fixes the problem.

It should also fixed by 0cfc8a8d70dc ("sparc32: fix page table traversal in srmmu_nocache_init()"). No?

Linus

#2 lore raw
Quoted 8 lines · Linus Torvalds
Quoted 3 lines · Guenter Roeck

This quite innocent looking patch crashes all my sparc32 boot tests. Crash log and bisect results below. Reverting it fixes the problem.

It should also fixed by 0cfc8a8d70dc ("sparc32: fix page table traversal in srmmu_nocache_init()"). No?

Yes, it is. Sorry for the noise. 0cfc8a8d70dc wasn't available when I started the bisect, and I didn't re-check afterwards.

Guenter