From: Linus Torvalds <torvalds@linux-foundation.org>
To: Dave Hansen <dave.hansen@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
tglx@linutronix.de, x86@kernel.org,
Andy Lutomirsky <luto@kernel.org>,
Borislav Petkov <bpetkov@suse.de>,
Greg KH <gregkh@linuxfoundation.org>,
keescook@google.com, hughd@google.com,
Brian Gerst <brgerst@gmail.com>,
Josh Poimboeuf <jpoimboe@redhat.com>,
Denys Vlasenko <dvlasenk@redhat.com>,
Boris Ostrovsky <boris.ostrovsky@oracle.com>,
Juergen Gross <jgross@suse.com>,
David Laight <David.Laight@aculab.com>,
Eduardo Valentin <eduval@amazon.com>,
aliguori@amazon.com, Will Deacon <will.deacon@arm.com>,
linux-mm@kvack.org, kirill.shutemov@linux.intel.com,
dan.j.williams@intel.com
Subject: Re: [PATCH v2 01/17] mm/gup: Fixup p*_access_permitted()
Date: Thu, 14 Dec 2017 22:09:24 -0800 [thread overview]
Message-ID: <CA+55aFyA1+_hnqKO11gVNTo7RV6d9qygC-p8yiAzFMb=9aR5-A@mail.gmail.com> (raw)
In-Reply-To: <8eedb9a3-0ba2-52df-58f6-3ed869d18ca3@intel.com>
[-- Attachment #1: Type: text/plain, Size: 1038 bytes --]
On Dec 14, 2017 21:04, "Dave Hansen" <dave.hansen@intel.com> wrote:
On 12/14/2017 12:54 PM, Peter Zijlstra wrote:
>> That short-circuits the page fault pretty quickly. So, basically, the
>> rule is: if the hardware says you tripped over pkey permissions, you
>> die. We don't try to do anything to the underlying page *before* saying
>> that you die.
> That only works when you trip the fault from hardware. Not if you do a
> software fault using gup().
>
> AFAIK __get_user_pages(FOLL_FORCE|FOLL_WRITE|FOLL_GET) will loop
> indefinitely on the case I described.
So, the underlying bug here is that we now a get_user_pages_remote() and
then go ahead and do the p*_access_permitted() checks against the
current PKRU. This was introduced recently with the addition of the new
p??_access_permitted() calls.
Can we please just undo that broken crap instead of trying to "fix" it?
It was wrong. We absolutely do not want to complicate the gup path.
Let's fet rid of those broken p??_access_permited() things.
Please.
Linus
[-- Attachment #2: Type: text/html, Size: 1952 bytes --]
next prev parent reply other threads:[~2017-12-15 6:09 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-14 11:27 [PATCH v2 00/17] x86/ldt: Use a VMA based read only mapping Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 01/17] mm/gup: Fixup p*_access_permitted() Peter Zijlstra
2017-12-14 12:41 ` Peter Zijlstra
2017-12-14 14:37 ` Peter Zijlstra
2017-12-14 20:44 ` Dave Hansen
2017-12-14 20:54 ` Peter Zijlstra
2017-12-14 21:18 ` Peter Zijlstra
2017-12-15 5:04 ` Dave Hansen
2017-12-15 6:09 ` Linus Torvalds [this message]
2017-12-15 7:51 ` Peter Zijlstra
2017-12-16 0:20 ` Linus Torvalds
2017-12-16 0:29 ` Dan Williams
2017-12-16 1:10 ` Linus Torvalds
2017-12-16 1:25 ` Dave Hansen
2017-12-16 2:28 ` Linus Torvalds
2017-12-16 2:48 ` Al Viro
2017-12-16 2:52 ` Linus Torvalds
2017-12-16 3:00 ` Linus Torvalds
2017-12-16 3:21 ` Dave Hansen
2017-12-16 1:29 ` Dan Williams
2017-12-16 0:31 ` Al Viro
2017-12-16 1:05 ` Linus Torvalds
2017-12-15 8:00 ` Peter Zijlstra
2017-12-15 10:25 ` Peter Zijlstra
2017-12-15 11:38 ` Peter Zijlstra
2017-12-15 16:38 ` Dan Williams
2017-12-18 11:54 ` Peter Zijlstra
2017-12-18 18:42 ` Dan Williams
2017-12-15 14:04 ` Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 02/17] mm: Exempt special mappings from mlock(), mprotect() and madvise() Peter Zijlstra
2017-12-14 16:19 ` Andy Lutomirski
2017-12-14 17:36 ` Peter Zijlstra
2018-01-02 16:44 ` Dmitry Safonov
2017-12-14 11:27 ` [PATCH v2 03/17] arch: Allow arch_dup_mmap() to fail Peter Zijlstra
2017-12-14 16:22 ` Andy Lutomirski
2017-12-14 11:27 ` [PATCH v2 04/17] x86/ldt: Rework locking Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 05/17] x86/ldt: Prevent ldt inheritance on exec Peter Zijlstra
2017-12-14 16:32 ` Andy Lutomirski
2017-12-14 11:27 ` [PATCH v2 06/17] x86/ldt: Do not install LDT for kernel threads Peter Zijlstra
2017-12-14 19:43 ` Peter Zijlstra
2017-12-14 21:27 ` Andy Lutomirski
2017-12-14 11:27 ` [PATCH v2 07/17] mm/softdirty: Move VM_SOFTDIRTY into high bits Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 08/17] mm/x86: Allow special mappings with user access cleared Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 09/17] mm: Provide vm_special_mapping::close Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 10/17] selftest/x86: Implement additional LDT selftests Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 11/17] selftests/x86/ldt_gdt: Prepare for access bit forced Peter Zijlstra
2017-12-14 16:20 ` Andy Lutomirski
2017-12-14 19:43 ` Linus Torvalds
2017-12-14 21:22 ` Andy Lutomirski
2017-12-14 21:44 ` Linus Torvalds
2017-12-14 21:48 ` Linus Torvalds
2017-12-14 22:02 ` Peter Zijlstra
2017-12-14 22:14 ` Linus Torvalds
2017-12-14 22:24 ` Peter Zijlstra
2017-12-14 22:52 ` Linus Torvalds
2017-12-14 22:11 ` Andy Lutomirski
2017-12-14 22:15 ` Linus Torvalds
2017-12-14 22:30 ` Andy Lutomirski
2017-12-14 22:23 ` Thomas Gleixner
2017-12-14 22:50 ` Linus Torvalds
2017-12-14 11:27 ` [PATCH v2 12/17] mm: Make populate_vma_page_range() available Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 13/17] x86/mm: Force LDT desc accessed bit Peter Zijlstra
2017-12-14 16:21 ` Andy Lutomirski
2017-12-14 11:27 ` [PATCH v2 14/17] x86/ldt: Reshuffle code Peter Zijlstra
2017-12-14 16:23 ` Andy Lutomirski
2017-12-14 16:31 ` Thomas Gleixner
2017-12-14 16:32 ` Thomas Gleixner
2017-12-14 16:34 ` Andy Lutomirski
2017-12-14 17:47 ` Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 15/17] x86/ldt: Prepare for VMA mapping Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 16/17] x86/ldt: Add VMA management code Peter Zijlstra
2017-12-14 11:27 ` [PATCH v2 17/17] x86/ldt: Make it read only VMA mapped Peter Zijlstra
2017-12-14 12:03 ` [PATCH v2 00/17] x86/ldt: Use a VMA based read only mapping Thomas Gleixner
2017-12-14 12:08 ` Peter Zijlstra
2017-12-14 16:35 ` Andy Lutomirski
2017-12-14 17:50 ` Peter Zijlstra
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='CA+55aFyA1+_hnqKO11gVNTo7RV6d9qygC-p8yiAzFMb=9aR5-A@mail.gmail.com' \
--to=torvalds@linux-foundation.org \
--cc=David.Laight@aculab.com \
--cc=aliguori@amazon.com \
--cc=boris.ostrovsky@oracle.com \
--cc=bpetkov@suse.de \
--cc=brgerst@gmail.com \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@intel.com \
--cc=dvlasenk@redhat.com \
--cc=eduval@amazon.com \
--cc=gregkh@linuxfoundation.org \
--cc=hughd@google.com \
--cc=jgross@suse.com \
--cc=jpoimboe@redhat.com \
--cc=keescook@google.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=will.deacon@arm.com \
--cc=x86@kernel.org \
/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).