From: Dave Hansen <dave.hansen@linux.intel.com>
To: Hanjun Guo <guohanjun@huawei.com>, Jiri Kosina <jikos@kernel.org>
Cc: Yisheng Xie <xieyisheng1@huawei.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
richard.fellner@student.tugraz.at, moritz.lipp@iaik.tugraz.at,
daniel.gruss@iaik.tugraz.at, michael.schwarz@iaik.tugraz.at,
luto@kernel.org, Linus Torvalds <torvalds@linux-foundation.org>,
keescook@google.com, hughd@google.com, x86@kernel.org,
Andrea Arcangeli <aarcange@redhat.com>
Subject: Re: [PATCH 05/23] x86, kaiser: unmap kernel from userspace page tables (core patch)
Date: Fri, 5 Jan 2018 23:55:38 -0800 [thread overview]
Message-ID: <80dbcf02-2035-94dc-ca29-4f17ad271fda@linux.intel.com> (raw)
In-Reply-To: <6b076a05-22b6-ce3e-efba-02c65dd1438d@huawei.com>
[-- Attachment #1: Type: text/plain, Size: 608 bytes --]
On 01/05/2018 10:53 PM, Hanjun Guo wrote:
>> + /*
>> + * PTI poisons low addresses in the kernel page tables in the
>> + * name of making them unusable for userspace. To execute
>> + * code at such a low address, the poison must be cleared.
>> + */
>> + pgd->pgd &= ~_PAGE_NX;
>>
>> We will have a try in a minute, and report back later.
> And it worksi 1/4 ?we can boot/reboot the system successfully, thank
> you all the quick response and debug!
I think I'll just submit the attached patch if there are no objections
(and if it works, of course!).
I just stuck the NX clearing at the bottom.
[-- Attachment #2: pti-tboot-fix.patch --]
[-- Type: text/x-patch, Size: 1449 bytes --]
From: Dave Hansen <dave.hansen@linux.intel.com>
This is another case similar to what EFI does: create a new set of
page tables, map some code at a low address, and jump to it. PTI
mistakes this low address for userspace and mistakenly marks it
non-executable in an effort to make it unusable for userspace. Undo
the poison to allow execution.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ning Sun <ning.sun@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: tboot-devel@lists.sourceforge.net
Cc: linux-kernel@vger.kernel.org
---
b/arch/x86/kernel/tboot.c | 11 +++++++++++
1 file changed, 11 insertions(+)
diff -puN arch/x86/kernel/tboot.c~pti-tboot-fix arch/x86/kernel/tboot.c
--- a/arch/x86/kernel/tboot.c~pti-tboot-fix 2018-01-05 21:50:55.755554960 -0800
+++ b/arch/x86/kernel/tboot.c 2018-01-05 23:51:41.368536890 -0800
@@ -138,6 +138,17 @@ static int map_tboot_page(unsigned long
return -1;
set_pte_at(&tboot_mm, vaddr, pte, pfn_pte(pfn, prot));
pte_unmap(pte);
+
+ /*
+ * PTI poisons low addresses in the kernel page tables in the
+ * name of making them unusable for userspace. To execute
+ * code at such a low address, the poison must be cleared.
+ *
+ * Note: 'pgd' actually gets set in p4d_alloc() _or_
+ * pud_alloc() depending on 4/5-level paging.
+ */
+ pgd->pgd &= ~_PAGE_NX;
+
return 0;
}
_
WARNING: multiple messages have this Message-ID (diff)
From: Dave Hansen <dave.hansen@linux.intel.com>
To: Hanjun Guo <guohanjun@huawei.com>, Jiri Kosina <jikos@kernel.org>
Cc: Yisheng Xie <xieyisheng1@huawei.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
richard.fellner@student.tugraz.at, moritz.lipp@iaik.tugraz.at,
daniel.gruss@iaik.tugraz.at, michael.schwarz@iaik.tugraz.at,
luto@kernel.org, Linus Torvalds <torvalds@linux-foundation.org>,
keescook@google.com, hughd@google.com, x86@kernel.org,
Andrea Arcangeli <aarcange@redhat.com>
Subject: Re: [PATCH 05/23] x86, kaiser: unmap kernel from userspace page tables (core patch)
Date: Fri, 5 Jan 2018 23:55:38 -0800 [thread overview]
Message-ID: <80dbcf02-2035-94dc-ca29-4f17ad271fda@linux.intel.com> (raw)
In-Reply-To: <6b076a05-22b6-ce3e-efba-02c65dd1438d@huawei.com>
[-- Attachment #1: Type: text/plain, Size: 604 bytes --]
On 01/05/2018 10:53 PM, Hanjun Guo wrote:
>> + /*
>> + * PTI poisons low addresses in the kernel page tables in the
>> + * name of making them unusable for userspace. To execute
>> + * code at such a low address, the poison must be cleared.
>> + */
>> + pgd->pgd &= ~_PAGE_NX;
>>
>> We will have a try in a minute, and report back later.
> And it works,we can boot/reboot the system successfully, thank
> you all the quick response and debug!
I think I'll just submit the attached patch if there are no objections
(and if it works, of course!).
I just stuck the NX clearing at the bottom.
[-- Attachment #2: pti-tboot-fix.patch --]
[-- Type: text/x-patch, Size: 1449 bytes --]
From: Dave Hansen <dave.hansen@linux.intel.com>
This is another case similar to what EFI does: create a new set of
page tables, map some code at a low address, and jump to it. PTI
mistakes this low address for userspace and mistakenly marks it
non-executable in an effort to make it unusable for userspace. Undo
the poison to allow execution.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ning Sun <ning.sun@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: tboot-devel@lists.sourceforge.net
Cc: linux-kernel@vger.kernel.org
---
b/arch/x86/kernel/tboot.c | 11 +++++++++++
1 file changed, 11 insertions(+)
diff -puN arch/x86/kernel/tboot.c~pti-tboot-fix arch/x86/kernel/tboot.c
--- a/arch/x86/kernel/tboot.c~pti-tboot-fix 2018-01-05 21:50:55.755554960 -0800
+++ b/arch/x86/kernel/tboot.c 2018-01-05 23:51:41.368536890 -0800
@@ -138,6 +138,17 @@ static int map_tboot_page(unsigned long
return -1;
set_pte_at(&tboot_mm, vaddr, pte, pfn_pte(pfn, prot));
pte_unmap(pte);
+
+ /*
+ * PTI poisons low addresses in the kernel page tables in the
+ * name of making them unusable for userspace. To execute
+ * code at such a low address, the poison must be cleared.
+ *
+ * Note: 'pgd' actually gets set in p4d_alloc() _or_
+ * pud_alloc() depending on 4/5-level paging.
+ */
+ pgd->pgd &= ~_PAGE_NX;
+
return 0;
}
_
next prev parent reply other threads:[~2018-01-06 7:55 UTC|newest]
Thread overview: 131+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-23 0:34 [PATCH 00/23] [v4] KAISER: unmap most of the kernel from userspace page tables Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 01/23] x86, kaiser: disable global pages by default with KAISER Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 02/23] x86, kaiser: prepare assembly for entry/exit CR3 switching Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 03/23] x86, kaiser: introduce user-mapped per-cpu areas Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 04/23] x86, kaiser: mark per-cpu data structures required for entry/exit Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 05/23] x86, kaiser: unmap kernel from userspace page tables (core patch) Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 4:07 ` Andy Lutomirski
2017-11-23 4:07 ` Andy Lutomirski
2017-11-26 16:10 ` Andy Lutomirski
2017-11-26 16:10 ` Andy Lutomirski
2017-11-26 16:24 ` Dave Hansen
2017-11-26 16:24 ` Dave Hansen
2017-11-26 16:29 ` Andy Lutomirski
2017-11-26 16:29 ` Andy Lutomirski
2018-01-05 4:16 ` Yisheng Xie
2018-01-05 4:16 ` Yisheng Xie
2018-01-05 5:18 ` Dave Hansen
2018-01-05 5:18 ` Dave Hansen
2018-01-05 6:16 ` Yisheng Xie
2018-01-05 6:16 ` Yisheng Xie
2018-01-05 6:29 ` Dave Hansen
2018-01-05 6:29 ` Dave Hansen
2018-01-05 11:49 ` Andrea Arcangeli
2018-01-05 11:49 ` Andrea Arcangeli
2018-01-05 18:19 ` Jiri Kosina
2018-01-05 18:19 ` Jiri Kosina
2018-01-05 19:00 ` Jiri Kosina
2018-01-05 19:00 ` Jiri Kosina
2018-01-05 19:03 ` Dave Hansen
2018-01-05 19:03 ` Dave Hansen
2018-01-05 19:17 ` Jiri Kosina
2018-01-05 19:17 ` Jiri Kosina
2018-01-05 19:18 ` Jiri Kosina
2018-01-05 19:18 ` Jiri Kosina
2018-01-05 19:55 ` Andrea Arcangeli
2018-01-05 19:55 ` Andrea Arcangeli
2018-01-05 21:07 ` Dave Hansen
2018-01-05 21:07 ` Dave Hansen
2018-01-05 21:14 ` Jiri Kosina
2018-01-05 21:14 ` Jiri Kosina
2018-01-05 21:29 ` Andy Lutomirski
2018-01-05 21:29 ` Andy Lutomirski
2018-01-05 22:48 ` Hugh Dickins
2018-01-05 22:48 ` Hugh Dickins
2018-01-06 4:54 ` Hanjun Guo
2018-01-06 4:54 ` Hanjun Guo
2018-01-06 6:06 ` Dave Hansen
2018-01-06 6:28 ` Hanjun Guo
2018-01-06 6:28 ` Hanjun Guo
2018-01-06 6:53 ` Hanjun Guo
2018-01-06 6:53 ` Hanjun Guo
2018-01-06 7:55 ` Dave Hansen [this message]
2018-01-06 7:55 ` Dave Hansen
2018-01-06 8:42 ` Hanjun Guo
2018-01-06 8:42 ` Hanjun Guo
2018-01-06 7:51 ` Dave Hansen
2018-01-06 7:51 ` Dave Hansen
2018-01-06 17:22 ` Andrea Arcangeli
2018-01-06 17:22 ` Andrea Arcangeli
2017-11-23 0:34 ` [PATCH 06/23] x86, kaiser: allow NX poison to be set in p4d/pgd Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 07/23] x86, kaiser: make sure static PGDs are 8k in size Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 08/23] x86, kaiser: map cpu entry area Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 09/23] x86, kaiser: map dynamically-allocated LDTs Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 19:42 ` Eric Biggers
2017-11-23 19:42 ` Eric Biggers
2017-11-23 20:12 ` Andy Lutomirski
2017-11-23 20:12 ` Andy Lutomirski
2017-11-23 0:34 ` [PATCH 10/23] x86, kaiser: map espfix structures Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 0:34 ` [PATCH 11/23] x86, kaiser: map entry stack variables Dave Hansen
2017-11-23 0:34 ` Dave Hansen
2017-11-23 3:31 ` Andy Lutomirski
2017-11-23 3:31 ` Andy Lutomirski
2017-11-23 15:37 ` Dave Hansen
2017-11-23 15:37 ` Dave Hansen
2017-11-23 15:55 ` Andy Lutomirski
2017-11-23 15:55 ` Andy Lutomirski
2017-11-23 0:35 ` [PATCH 12/23] x86, kaiser: map virtually-addressed performance monitoring buffers Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 13/23] x86, mm: Move CR3 construction functions Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 14/23] x86, mm: remove hard-coded ASID limit checks Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 15/23] x86, mm: put mmu-to-h/w ASID translation in one place Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 16/23] x86, pcid, kaiser: allow flushing for future ASID switches Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 17/23] x86, kaiser: use PCID feature to make user and kernel switches faster Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 18/23] x86, kaiser: disable native VSYSCALL Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 19/23] x86, kaiser: add debugfs file to turn KAISER on/off at runtime Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 20/23] x86, kaiser: add a function to check for KAISER being enabled Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-25 1:23 ` Eduardo Valentin
2017-11-25 1:23 ` Eduardo Valentin
2017-11-23 0:35 ` [PATCH 21/23] x86, kaiser: un-poison PGDs at runtime Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-25 1:17 ` Eduardo Valentin
2017-11-25 1:17 ` Eduardo Valentin
2017-11-23 0:35 ` [PATCH 22/23] x86, kaiser: allow KAISER to be enabled/disabled " Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 0:35 ` [PATCH 23/23] x86, kaiser: add Kconfig Dave Hansen
2017-11-23 0:35 ` Dave Hansen
2017-11-23 7:23 ` [PATCH 00/23] [v4] KAISER: unmap most of the kernel from userspace page tables Ingo Molnar
2017-11-23 7:23 ` Ingo Molnar
2017-11-23 7:27 ` Ingo Molnar
2017-11-23 7:27 ` Ingo Molnar
2017-11-23 7:32 ` Ingo Molnar
2017-11-23 7:32 ` Ingo Molnar
2017-11-23 15:02 ` Dave Hansen
2017-11-23 15:02 ` Dave Hansen
2017-11-23 16:20 ` Dave Hansen
2017-11-23 16:20 ` Dave Hansen
2017-11-24 6:35 ` Ingo Molnar
2017-11-24 6:35 ` Ingo Molnar
2017-11-24 6:41 ` Dave Hansen
2017-11-24 6:41 ` Dave Hansen
2017-11-24 7:33 ` Ingo Molnar
2017-11-24 7:33 ` Ingo Molnar
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=80dbcf02-2035-94dc-ca29-4f17ad271fda@linux.intel.com \
--to=dave.hansen@linux.intel.com \
--cc=aarcange@redhat.com \
--cc=daniel.gruss@iaik.tugraz.at \
--cc=guohanjun@huawei.com \
--cc=hughd@google.com \
--cc=jikos@kernel.org \
--cc=keescook@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=michael.schwarz@iaik.tugraz.at \
--cc=moritz.lipp@iaik.tugraz.at \
--cc=richard.fellner@student.tugraz.at \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.org \
--cc=xieyisheng1@huawei.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.