From: Peter Xu <peterx@redhat.com>
To: Jerome Glisse <jglisse@redhat.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
David Hildenbrand <david@redhat.com>,
Hugh Dickins <hughd@google.com>, Maya Gokhale <gokhale2@llnl.gov>,
Pavel Emelyanov <xemul@virtuozzo.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Martin Cracauer <cracauer@cons.org>, Shaohua Li <shli@fb.com>,
Marty McFadden <mcfadden8@llnl.gov>,
Andrea Arcangeli <aarcange@redhat.com>,
Mike Kravetz <mike.kravetz@oracle.com>,
Denis Plotnikov <dplotnikov@virtuozzo.com>,
Mike Rapoport <rppt@linux.vnet.ibm.com>,
Mel Gorman <mgorman@suse.de>,
"Kirill A . Shutemov" <kirill@shutemov.name>,
"Dr . David Alan Gilbert" <dgilbert@redhat.com>
Subject: Re: [PATCH v2 02/26] mm: userfault: return VM_FAULT_RETRY on signals
Date: Fri, 22 Feb 2019 11:51:17 +0800 [thread overview]
Message-ID: <20190222035117.GC8904@xz-x1> (raw)
In-Reply-To: <20190221152956.GB2813@redhat.com>
On Thu, Feb 21, 2019 at 10:29:56AM -0500, Jerome Glisse wrote:
> On Tue, Feb 12, 2019 at 10:56:08AM +0800, Peter Xu wrote:
> > The idea comes from the upstream discussion between Linus and Andrea:
> >
> > https://lkml.org/lkml/2017/10/30/560
> >
> > A summary to the issue: there was a special path in handle_userfault()
> > in the past that we'll return a VM_FAULT_NOPAGE when we detected
> > non-fatal signals when waiting for userfault handling. We did that by
> > reacquiring the mmap_sem before returning. However that brings a risk
> > in that the vmas might have changed when we retake the mmap_sem and
> > even we could be holding an invalid vma structure.
> >
> > This patch removes the special path and we'll return a VM_FAULT_RETRY
> > with the common path even if we have got such signals. Then for all
> > the architectures that is passing in VM_FAULT_ALLOW_RETRY into
> > handle_mm_fault(), we check not only for SIGKILL but for all the rest
> > of userspace pending signals right after we returned from
> > handle_mm_fault(). This can allow the userspace to handle nonfatal
> > signals faster than before.
> >
> > This patch is a preparation work for the next patch to finally remove
> > the special code path mentioned above in handle_userfault().
> >
> > Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
> > Suggested-by: Andrea Arcangeli <aarcange@redhat.com>
> > Signed-off-by: Peter Xu <peterx@redhat.com>
>
> See maybe minor improvement
>
> Reviewed-by: Jérôme Glisse <jglisse@redhat.com>
>
> [...]
>
> > diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
> > index 58f69fa07df9..c41c021bbe40 100644
> > --- a/arch/arm/mm/fault.c
> > +++ b/arch/arm/mm/fault.c
> > @@ -314,12 +314,12 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
> >
> > fault = __do_page_fault(mm, addr, fsr, flags, tsk);
> >
> > - /* If we need to retry but a fatal signal is pending, handle the
> > + /* If we need to retry but a signal is pending, handle the
> > * signal first. We do not need to release the mmap_sem because
> > * it would already be released in __lock_page_or_retry in
> > * mm/filemap.c. */
> > - if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) {
> > - if (!user_mode(regs))
> > + if (unlikely(fault & VM_FAULT_RETRY && signal_pending(current))) {
>
> I rather see (fault & VM_FAULT_RETRY) ie with the parenthesis as it
> avoids the need to remember operator precedence rules :)
Yes it's good practise. I've been hit by the lock_page() days ago
already so I think I'll remember (though this patch was earlier :)
I'll fix all the places in the patch. Actually I noticed that there
are four of them. And I've taken the r-b after the changes. Thanks,
>
> [...]
>
> > diff --git a/arch/nds32/mm/fault.c b/arch/nds32/mm/fault.c
> > index 68d5f2a27f38..9f6e477b9e30 100644
> > --- a/arch/nds32/mm/fault.c
> > +++ b/arch/nds32/mm/fault.c
> > @@ -206,12 +206,12 @@ void do_page_fault(unsigned long entry, unsigned long addr,
> > fault = handle_mm_fault(vma, addr, flags);
> >
> > /*
> > - * If we need to retry but a fatal signal is pending, handle the
> > + * If we need to retry but a signal is pending, handle the
> > * signal first. We do not need to release the mmap_sem because it
> > * would already be released in __lock_page_or_retry in mm/filemap.c.
> > */
> > - if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) {
> > - if (!user_mode(regs))
> > + if (fault & VM_FAULT_RETRY && signal_pending(current)) {
>
> Same as above parenthesis maybe.
>
> [...]
>
> > diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
> > index 0e8b6158f224..09baf37b65b9 100644
> > --- a/arch/um/kernel/trap.c
> > +++ b/arch/um/kernel/trap.c
> > @@ -76,8 +76,11 @@ int handle_page_fault(unsigned long address, unsigned long ip,
> >
> > fault = handle_mm_fault(vma, address, flags);
> >
> > - if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
> > + if (fault & VM_FAULT_RETRY && signal_pending(current)) {
>
> Same as above parenthesis maybe.
>
> [...]
Regards,
--
Peter Xu
next prev parent reply other threads:[~2019-02-22 3:51 UTC|newest]
Thread overview: 113+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-12 2:56 [PATCH v2 00/26] userfaultfd: write protection support Peter Xu
2019-02-12 2:56 ` [PATCH v2 01/26] mm: gup: rename "nonblocking" to "locked" where proper Peter Xu
2019-02-21 15:17 ` Jerome Glisse
2019-02-22 3:42 ` Peter Xu
2019-02-12 2:56 ` [PATCH v2 02/26] mm: userfault: return VM_FAULT_RETRY on signals Peter Xu
2019-02-21 15:29 ` Jerome Glisse
2019-02-22 3:51 ` Peter Xu [this message]
2019-02-12 2:56 ` [PATCH v2 03/26] userfaultfd: don't retake mmap_sem to emulate NOPAGE Peter Xu
2019-02-21 15:34 ` Jerome Glisse
2019-02-12 2:56 ` [PATCH v2 04/26] mm: allow VM_FAULT_RETRY for multiple times Peter Xu
2019-02-13 3:34 ` Peter Xu
2019-02-20 11:48 ` Peter Xu
2019-02-21 8:56 ` [PATCH v2.1 " Peter Xu
2019-02-21 15:53 ` Jerome Glisse
2019-02-22 4:25 ` Peter Xu
2019-02-22 15:11 ` Jerome Glisse
2019-02-25 6:19 ` Peter Xu
2019-02-12 2:56 ` [PATCH v2 05/26] mm: gup: " Peter Xu
2019-02-21 16:06 ` Jerome Glisse
2019-02-22 4:41 ` Peter Xu
2019-02-22 15:13 ` Jerome Glisse
2019-02-12 2:56 ` [PATCH v2 06/26] userfaultfd: wp: add helper for writeprotect check Peter Xu
2019-02-21 16:07 ` Jerome Glisse
2019-02-25 15:41 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 07/26] userfaultfd: wp: hook userfault handler to write protection fault Peter Xu
2019-02-21 16:25 ` Jerome Glisse
2019-02-25 15:43 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 08/26] userfaultfd: wp: add WP pagetable tracking to x86 Peter Xu
2019-02-21 17:20 ` Jerome Glisse
2019-02-25 15:48 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 09/26] userfaultfd: wp: userfaultfd_pte/huge_pmd_wp() helpers Peter Xu
2019-02-21 17:21 ` Jerome Glisse
2019-02-25 17:12 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 10/26] userfaultfd: wp: add UFFDIO_COPY_MODE_WP Peter Xu
2019-02-21 17:29 ` Jerome Glisse
2019-02-22 7:11 ` Peter Xu
2019-02-22 15:15 ` Jerome Glisse
2019-02-25 6:45 ` Peter Xu
2019-02-25 15:58 ` Mike Rapoport
2019-02-26 5:09 ` Peter Xu
2019-02-26 8:28 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 11/26] mm: merge parameters for change_protection() Peter Xu
2019-02-21 17:32 ` Jerome Glisse
2019-02-12 2:56 ` [PATCH v2 12/26] userfaultfd: wp: apply _PAGE_UFFD_WP bit Peter Xu
2019-02-21 17:44 ` Jerome Glisse
2019-02-22 7:31 ` Peter Xu
2019-02-22 15:17 ` Jerome Glisse
2019-02-25 18:00 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 13/26] mm: export wp_page_copy() Peter Xu
2019-02-21 17:44 ` Jerome Glisse
2019-02-12 2:56 ` [PATCH v2 14/26] userfaultfd: wp: handle COW properly for uffd-wp Peter Xu
2019-02-21 18:04 ` Jerome Glisse
2019-02-22 8:46 ` Peter Xu
2019-02-22 15:35 ` Jerome Glisse
2019-02-25 7:13 ` Peter Xu
2019-02-25 15:32 ` Jerome Glisse
2019-02-12 2:56 ` [PATCH v2 15/26] userfaultfd: wp: drop _PAGE_UFFD_WP properly when fork Peter Xu
2019-02-21 18:06 ` Jerome Glisse
2019-02-22 9:09 ` Peter Xu
2019-02-22 15:36 ` Jerome Glisse
2019-02-25 18:19 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 16/26] userfaultfd: wp: add pmd_swp_*uffd_wp() helpers Peter Xu
2019-02-21 18:07 ` Jerome Glisse
2019-02-25 18:20 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 17/26] userfaultfd: wp: support swap and page migration Peter Xu
2019-02-21 18:16 ` Jerome Glisse
2019-02-25 7:48 ` Peter Xu
2019-02-25 18:28 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 18/26] khugepaged: skip collapse if uffd-wp detected Peter Xu
2019-02-21 18:17 ` Jerome Glisse
2019-02-25 18:50 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 19/26] userfaultfd: introduce helper vma_find_uffd Peter Xu
2019-02-21 18:19 ` Jerome Glisse
2019-02-25 20:48 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 20/26] userfaultfd: wp: support write protection for userfault vma range Peter Xu
2019-02-21 18:23 ` Jerome Glisse
2019-02-25 8:16 ` Peter Xu
2019-02-25 20:52 ` Mike Rapoport
2019-02-26 6:06 ` Peter Xu
2019-02-26 6:43 ` Mike Rapoport
2019-02-26 7:20 ` Peter Xu
2019-02-26 7:46 ` Mike Rapoport
2019-02-26 7:54 ` Peter Xu
2019-02-12 2:56 ` [PATCH v2 21/26] userfaultfd: wp: add the writeprotect API to userfaultfd ioctl Peter Xu
2019-02-21 18:28 ` Jerome Glisse
2019-02-25 8:31 ` Peter Xu
2019-02-25 21:03 ` Mike Rapoport
2019-02-26 6:30 ` Peter Xu
2019-02-12 2:56 ` [PATCH v2 22/26] userfaultfd: wp: enabled write protection in userfaultfd API Peter Xu
2019-02-21 18:29 ` Jerome Glisse
2019-02-25 8:34 ` Peter Xu
2019-02-12 2:56 ` [PATCH v2 23/26] userfaultfd: wp: don't wake up when doing write protect Peter Xu
2019-02-21 18:36 ` Jerome Glisse
2019-02-25 8:58 ` Peter Xu
2019-02-25 21:15 ` Mike Rapoport
2019-02-25 21:09 ` Mike Rapoport
2019-02-26 6:24 ` Peter Xu
2019-02-26 7:29 ` Mike Rapoport
2019-02-26 7:41 ` Peter Xu
2019-02-26 8:00 ` Mike Rapoport
2019-02-28 2:47 ` Peter Xu
2019-02-26 8:00 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 24/26] userfaultfd: wp: UFFDIO_REGISTER_MODE_WP documentation update Peter Xu
2019-02-21 18:38 ` Jerome Glisse
2019-02-25 21:19 ` Mike Rapoport
2019-02-26 6:53 ` Peter Xu
2019-02-26 7:04 ` Mike Rapoport
2019-02-26 7:42 ` Peter Xu
2019-02-12 2:56 ` [PATCH v2 25/26] userfaultfd: selftests: refactor statistics Peter Xu
2019-02-26 6:50 ` Mike Rapoport
2019-02-12 2:56 ` [PATCH v2 26/26] userfaultfd: selftests: add write-protect test Peter Xu
2019-02-26 6:58 ` Mike Rapoport
2019-02-26 7:52 ` Peter Xu
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=20190222035117.GC8904@xz-x1 \
--to=peterx@redhat.com \
--cc=aarcange@redhat.com \
--cc=cracauer@cons.org \
--cc=david@redhat.com \
--cc=dgilbert@redhat.com \
--cc=dplotnikov@virtuozzo.com \
--cc=gokhale2@llnl.gov \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=jglisse@redhat.com \
--cc=kirill@shutemov.name \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mcfadden8@llnl.gov \
--cc=mgorman@suse.de \
--cc=mike.kravetz@oracle.com \
--cc=rppt@linux.vnet.ibm.com \
--cc=shli@fb.com \
--cc=xemul@virtuozzo.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).