From: Mike Rapoport <rppt@linux.ibm.com>
To: George Kennedy <george.kennedy@oracle.com>
Cc: David Hildenbrand <david@redhat.com>,
Andrey Konovalov <andreyknvl@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Vincenzo Frascino <vincenzo.frascino@arm.com>,
Dmitry Vyukov <dvyukov@google.com>,
Konrad Rzeszutek Wilk <konrad@darnok.org>,
Will Deacon <will.deacon@arm.com>,
Andrey Ryabinin <aryabinin@virtuozzo.com>,
Alexander Potapenko <glider@google.com>,
Marco Elver <elver@google.com>,
Peter Collingbourne <pcc@google.com>,
Evgenii Stepanov <eugenis@google.com>,
Branislav Rankov <Branislav.Rankov@arm.com>,
Kevin Brodsky <kevin.brodsky@arm.com>,
Christoph Hellwig <hch@infradead.org>,
kasan-dev <kasan-dev@googlegroups.com>,
Linux ARM <linux-arm-kernel@lists.infradead.org>,
Linux Memory Management List <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>,
Dhaval Giani <dhaval.giani@oracle.com>
Subject: Re: [PATCH] mm, kasan: don't poison boot memory
Date: Mon, 22 Feb 2021 23:55:02 +0200 [thread overview]
Message-ID: <20210222215502.GB1741768@linux.ibm.com> (raw)
In-Reply-To: <cb8564e8-3535-826b-2d42-b273a0d793fb@oracle.com>
On Mon, Feb 22, 2021 at 01:42:56PM -0500, George Kennedy wrote:
>
> On 2/22/2021 11:13 AM, David Hildenbrand wrote:
> > On 22.02.21 16:13, George Kennedy wrote:
> > >
> > > On 2/22/2021 4:52 AM, David Hildenbrand wrote:
> > > >
> > > > Let me look into the code ... I have little experience with ACPI
> > > > details, so bear with me.
> > > >
> > > > I assume that acpi_map()/acpi_unmap() map some firmware blob that is
> > > > provided via firmware/bios/... to us.
> > > >
> > > > should_use_kmap() tells us whether
> > > > a) we have a "struct page" and should kmap() that one
> > > > b) we don't have a "struct page" and should ioremap.
> > > >
> > > > As it is a blob, the firmware should always reserve that memory region
> > > > via memblock (e.g., memblock_reserve()), such that we either
> > > > 1) don't create a memmap ("struct page") at all (-> case b) )
> > > > 2) if we have to create e memmap, we mark the page PG_reserved and
> > > > *never* expose it to the buddy (-> case a) )
> > > >
> > > >
> > > > Are you telling me that in this case we might have a memmap for the HW
> > > > blob that is *not* PG_reserved? In that case it most probably got
> > > > exposed to the buddy where it can happily get allocated/freed.
> > > >
> > > > The latent BUG would be that that blob gets exposed to the system like
> > > > ordinary RAM, and not reserved via memblock early during boot.
> > > > Assuming that blob has a low physical address, with my patch it will
> > > > get allocated/used a lot earlier - which would mean we trigger this
> > > > latent BUG now more easily.
> > > >
> > > > There have been similar latent BUGs on ARM boards that my patch
> > > > discovered where special RAM regions did not get marked as reserved
> > > > via the device tree properly.
> > > >
> > > > Now, this is just a wild guess :) Can you dump the page when mapping
> > > > (before PageReserved()) and when unmapping, to see what the state of
> > > > that memmap is?
> > >
> > > Thank you David for the explanation and your help on this,
> > >
> > > dump_page() before PageReserved and before kmap() in the above patch:
> > >
> > > [ 1.116480] ACPI: Core revision 20201113
> > > [ 1.117628] XXX acpi_map: about to call kmap()...
> > > [ 1.118561] page:ffffea0002f914c0 refcount:0 mapcount:0
> > > mapping:0000000000000000 index:0x0 pfn:0xbe453
> > > [ 1.120381] flags: 0xfffffc0000000()
> > > [ 1.121116] raw: 000fffffc0000000 ffffea0002f914c8 ffffea0002f914c8
> > > 0000000000000000
> > > [ 1.122638] raw: 0000000000000000 0000000000000000 00000000ffffffff
> > > 0000000000000000
> > > [ 1.124146] page dumped because: acpi_map pre SetPageReserved
> > >
> > > I also added dump_page() before unmapping, but it is not hit. The
> > > following for the same pfn now shows up I believe as a result of setting
> > > PageReserved:
> > >
> > > [ 28.098208] BUG:Bad page state in process mo dprobe pfn:be453
> > > [ 28.098394] page:ffffea0002f914c0 refcount:0 mapcount:0
> > > mapping:0000000000000000 index:0x1 pfn:0xbe453
> > > [ 28.098394] flags: 0xfffffc0001000(reserved)
> > > [ 28.098394] raw: 000fffffc0001000 dead000000000100 dead000000000122
> > > 0000000000000000
> > > [ 28.098394] raw: 0000000000000001 0000000000000000 00000000ffffffff
> > > 0000000000000000
> > > [ 28.098394] page dumped because: PAGE_FLAGS_CHECK_AT_PREP flag(s) set
> > > [ 28.098394] page_owner info is not present (never set?)
> > > [ 28.098394] Modules linked in:
> > > [ 28.098394] CPU: 2 PID: 204 Comm: modprobe Not tainted
> > > 5.11.0-3dbd5e3 #66
> > > [ 28.098394] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),
> > > BIOS 0.0.0 02/06/2015
> > > [ 28.098394] Call Trace:
> > > [ 28.098394] dump_stack+0xdb/0x120
> > > [ 28.098394] bad_page.cold.108+0xc6/0xcb
> > > [ 28.098394] check_new_page_bad+0x47/0xa0
> > > [ 28.098394] get_page_from_freelist+0x30cd/0x5730
> > > [ 28.098394] ? __isolate_free_page+0x4f0/0x4f0
> > > [ 28.098394] ? init_object+0x7e/0x90
> > > [ 28.098394] __alloc_pages_nodemask+0x2d8/0x650
> > > [ 28.098394] ? write_comp_data+0x2f/0x90
> > > [ 28.098394] ? __alloc_pages_slowpath.constprop.103+0x2110/0x2110
> > > [ 28.098394] ? __sanitizer_cov_trace_pc+0x21/0x50
> > > [ 28.098394] alloc_pages_vma+0xe2/0x560
> > > [ 28.098394] do_fault+0x194/0x12c0
> > > [ 28.098394] ? write_comp_data+0x2f/0x90
> > > [ 28.098394] __handle_mm_fault+0x1650/0x26c0
> > > [ 28.098394] ? copy_page_range+0x1350/0x1350
> > > [ 28.098394] ? write_comp_data+0x2f/0x90
> > > [ 28.098394] ? write_comp_data+0x2f/0x90
> > > [ 28.098394] handle_mm_fault+0x1f9/0x810
> > > [ 28.098394] ? write_comp_data+0x2f/0x90
> > > [ 28.098394] do_user_addr_fault+0x6f7/0xca0
> > > [ 28.098394] exc_page_fault+0xaf/0x1a0
> > > [ 28.098394] asm_exc_page_fault+0x1e/0x30
> > > [ 28.098394] RIP: 0010:__clear_user+0x30/0x60
> >
> > I think the PAGE_FLAGS_CHECK_AT_PREP check in this instance means that
> > someone is trying to allocate that page with the PG_reserved bit set.
> > This means that the page actually was exposed to the buddy.
> >
> > However, when you SetPageReserved(), I don't think that PG_buddy is set
> > and the refcount is 0. That could indicate that the page is on the buddy
> > PCP list. Could be that it is getting reused a couple of times.
> >
> > The PFN 0xbe453 looks a little strange, though. Do we expect ACPI tables
> > close to 3 GiB ? No idea. Could it be that you are trying to map a wrong
> > table? Just a guess.
> >
> > >
> > > What would be the correct way to reserve the page so that the above
> > > would not be hit?
> >
> > I would have assumed that if this is a binary blob, that someone (which
> > I think would be acpi code) reserved via memblock_reserve() early during
> > boot.
> >
> > E.g., see drivers/acpi/tables.c:acpi_table_upgrade()->memblock_reserve().
>
> acpi_table_upgrade() gets called, but bails out before memblock_reserve() is
> called. Thus, it appears no pages are getting reserved.
acpi_table_upgrade() does not actually reserve memory but rather open
codes memblock allocation with memblock_find_in_range() +
memblock_reserve(), so it does not seem related anyway.
Do you have by chance a full boot log handy?
> 503 void __init acpi_table_upgrade(void)
> 504 {
...
> 568 if (table_nr == 0)
> 569 return; <-- bails
> out here
> "drivers/acpi/tables.c"
>
> George
>
--
Sincerely yours,
Mike.
next prev parent reply other threads:[~2021-02-22 21:55 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-17 20:56 [PATCH] mm, kasan: don't poison boot memory Andrey Konovalov
2021-02-18 8:55 ` David Hildenbrand
2021-02-18 19:40 ` Andrey Konovalov
2021-02-18 19:46 ` David Hildenbrand
2021-02-18 20:26 ` Andrey Konovalov
2021-02-19 0:06 ` George Kennedy
2021-02-19 0:09 ` Andrey Konovalov
2021-02-19 16:45 ` George Kennedy
2021-02-19 23:04 ` George Kennedy
2021-02-22 9:52 ` David Hildenbrand
2021-02-22 15:13 ` George Kennedy
2021-02-22 16:13 ` David Hildenbrand
2021-02-22 16:39 ` David Hildenbrand
2021-02-22 17:40 ` Konrad Rzeszutek Wilk
2021-02-22 18:45 ` Mike Rapoport
2021-02-22 18:42 ` George Kennedy
2021-02-22 21:55 ` Mike Rapoport [this message]
[not found] ` <9773282a-2854-25a4-9faa-9da5dd34e371@oracle.com>
2021-02-23 10:33 ` Mike Rapoport
[not found] ` <3ef9892f-d657-207f-d4cf-111f98dcb55c@oracle.com>
2021-02-23 15:47 ` Mike Rapoport
2021-02-23 18:05 ` George Kennedy
2021-02-23 20:09 ` Mike Rapoport
2021-02-23 21:16 ` George Kennedy
2021-02-23 21:32 ` Mike Rapoport
2021-02-23 21:46 ` George Kennedy
2021-02-24 10:37 ` Mike Rapoport
2021-02-24 14:22 ` George Kennedy
2021-02-25 8:53 ` Mike Rapoport
2021-02-25 12:38 ` George Kennedy
2021-02-25 14:57 ` Mike Rapoport
2021-02-25 15:22 ` George Kennedy
2021-02-25 16:06 ` George Kennedy
2021-02-25 16:07 ` Mike Rapoport
2021-02-25 16:31 ` George Kennedy
2021-02-25 17:23 ` David Hildenbrand
2021-02-25 17:41 ` Mike Rapoport
2021-02-25 17:50 ` Mike Rapoport
2021-02-25 17:33 ` George Kennedy
2021-02-26 1:19 ` George Kennedy
2021-02-26 11:17 ` Mike Rapoport
2021-02-26 16:16 ` George Kennedy
2021-02-28 18:08 ` Mike Rapoport
2021-03-01 14:29 ` George Kennedy
2021-03-02 1:20 ` George Kennedy
2021-03-02 9:57 ` Mike Rapoport
2021-02-23 21:26 ` George Kennedy
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210222215502.GB1741768@linux.ibm.com \
--to=rppt@linux.ibm.com \
--cc=Branislav.Rankov@arm.com \
--cc=akpm@linux-foundation.org \
--cc=andreyknvl@google.com \
--cc=aryabinin@virtuozzo.com \
--cc=catalin.marinas@arm.com \
--cc=david@redhat.com \
--cc=dhaval.giani@oracle.com \
--cc=dvyukov@google.com \
--cc=elver@google.com \
--cc=eugenis@google.com \
--cc=george.kennedy@oracle.com \
--cc=glider@google.com \
--cc=hch@infradead.org \
--cc=kasan-dev@googlegroups.com \
--cc=kevin.brodsky@arm.com \
--cc=konrad@darnok.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=pcc@google.com \
--cc=vincenzo.frascino@arm.com \
--cc=will.deacon@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).