Re: [PATCH v1 3/5] mm: ptdump: Provide page size to notepage()

From: Christophe Leroy
Date: Fri Apr 16 2021 - 10:41:34 EST




Le 16/04/2021 à 15:00, Steven Price a écrit :
On 16/04/2021 12:08, Christophe Leroy wrote:


Le 16/04/2021 à 12:51, Steven Price a écrit :
On 16/04/2021 11:38, Christophe Leroy wrote:


Le 16/04/2021 à 11:28, Steven Price a écrit :
On 15/04/2021 18:18, Christophe Leroy wrote:
In order to support large pages on powerpc, notepage()
needs to know the page size of the page.

Add a page_size argument to notepage().

Signed-off-by: Christophe Leroy <christophe.leroy@xxxxxxxxxx>
---
  arch/arm64/mm/ptdump.c         |  2 +-
  arch/riscv/mm/ptdump.c         |  2 +-
  arch/s390/mm/dump_pagetables.c |  3 ++-
  arch/x86/mm/dump_pagetables.c  |  2 +-
  include/linux/ptdump.h         |  2 +-
  mm/ptdump.c                    | 16 ++++++++--------
  6 files changed, 14 insertions(+), 13 deletions(-)

[...]
diff --git a/mm/ptdump.c b/mm/ptdump.c
index da751448d0e4..61cd16afb1c8 100644
--- a/mm/ptdump.c
+++ b/mm/ptdump.c
@@ -17,7 +17,7 @@ static inline int note_kasan_page_table(struct mm_walk *walk,
  {
      struct ptdump_state *st = walk->private;
-    st->note_page(st, addr, 4, pte_val(kasan_early_shadow_pte[0]));
+    st->note_page(st, addr, 4, pte_val(kasan_early_shadow_pte[0]), PAGE_SIZE);

I'm not completely sure what the page_size is going to be used for, but note that KASAN presents an interesting case here. We short-cut by detecting it's a KASAN region at a high level (PGD/P4D/PUD/PMD) and instead of walking the tree down just call note_page() *once* but with level==4 because we know KASAN sets up the page table like that.

However the one call actually covers a much larger region - so while PAGE_SIZE matches the level it doesn't match the region covered. AFAICT this will lead to odd results if you enable KASAN on powerpc.

Hum .... I successfully tested it with KASAN, I now realise that I tested it with CONFIG_KASAN_VMALLOC selected. In this situation, since https://github.com/torvalds/linux/commit/af3d0a686 we don't have any common shadow page table anymore.

I'll test again without CONFIG_KASAN_VMALLOC.


To be honest I don't fully understand why powerpc requires the page_size - it appears to be using it purely to find "holes" in the calls to note_page(), but I haven't worked out why such holes would occur.

I was indeed introduced for KASAN. We have a first commit https://github.com/torvalds/linux/commit/cabe8138 which uses page size to detect whether it is a KASAN like stuff.

Then came https://github.com/torvalds/linux/commit/b00ff6d8c as a fix. I can't remember what the problem was exactly, something around the use of hugepages for kernel memory, came as part of the series https://patchwork.ozlabs.org/project/linuxppc-dev/cover/cover.1589866984.git.christophe.leroy@xxxxxxxxxx/




Ah, that's useful context. So it looks like powerpc took a different route to reducing the KASAN output to x86.

Given the generic ptdump code has handling for KASAN already it should be possible to drop that from the powerpc arch code, which I think means we don't actually need to provide page size to notepage(). Hopefully that means more code to delete ;)


Yes ... and no.

It looks like the generic ptdump handles the case when several pgdir entries points to the same kasan_early_shadow_pte. But it doesn't take into account the powerpc case where we have regular page tables where several (if not all) PTEs are pointing to the kasan_early_shadow_page .

I'm not sure I follow quite how powerpc is different here. But could you have a similar check for PTEs against kasan_early_shadow_pte as the other levels already have?

I'm just worried that page_size isn't well defined in this interface and it's going to cause problems in the future.


I'm trying. I reverted the two commits b00ff6d8c and cabe8138.

At the moment, I don't get exactly what I expect: For linear memory I get one line for each 8M page whereas before reverting the patches I got one 16M line and one 112M line.

And for KASAN shadow area I get two lines for the 2x 8M pages shadowing linear mem then I get one 4M line for each PGDIR entry pointing to kasan_early_shadow_pte.

0xf8000000-0xf87fffff 0x07000000 8M huge rw present
0xf8800000-0xf8ffffff 0x07800000 8M huge rw present
0xf9000000-0xf93fffff 0x01430000 4M r present
0xf9400000-0xf97fffff 0x01430000 4M r present
0xf9800000-0xf9bfffff 0x01430000 4M r present
0xf9c00000-0xf9ffffff 0x01430000 4M r present
0xfa000000-0xfa3fffff 0x01430000 4M r present
0xfa400000-0xfa7fffff 0x01430000 4M r present
0xfa800000-0xfabfffff 0x01430000 4M r present
0xfac00000-0xfaffffff 0x01430000 4M r present
0xfb000000-0xfb3fffff 0x01430000 4M r present
0xfb400000-0xfb7fffff 0x01430000 4M r present
0xfb800000-0xfbbfffff 0x01430000 4M r present
0xfbc00000-0xfbffffff 0x01430000 4M r present
0xfc000000-0xfc3fffff 0x01430000 4M r present
0xfc400000-0xfc7fffff 0x01430000 4M r present
0xfc800000-0xfcbfffff 0x01430000 4M r present
0xfcc00000-0xfcffffff 0x01430000 4M r present
0xfd000000-0xfd3fffff 0x01430000 4M r present
0xfd400000-0xfd7fffff 0x01430000 4M r present
0xfd800000-0xfdbfffff 0x01430000 4M r present
0xfdc00000-0xfdffffff 0x01430000 4M r present
0xfe000000-0xfe3fffff 0x01430000 4M r present
0xfe400000-0xfe7fffff 0x01430000 4M r present
0xfe800000-0xfebfffff 0x01430000 4M r present
0xfec00000-0xfeffffff 0x01430000 4M r present

Any idea ?

Christophe