From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Xen-devel <xen-devel@lists.xen.org>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
Subject: [PATCH 0/7] Fixes to pagetable handling
Date: Mon, 27 Feb 2017 14:03:11 +0000 [thread overview]
Message-ID: <1488204198-23948-1-git-send-email-andrew.cooper3@citrix.com> (raw)
This series has been a long time in preparation (i.e. most of the 4.8 and 4.9
dev cycles). It started when I tried to make an XTF PoC for XSA-176, and
stumbled upon the the _PAGE_PAGED aliasing issue (see patch 7) which caused by
PoC to be descheduled waiting for (a non-existent) paging agent to respond.
This series is built upon:
1) The switch to using _Bool. I spent rather too long trying to debug why
CR0.WP wasn't behaving properly, and it was down to static inline bool_t
guest_wp_enabled().
2) The series to switch emulation to using system-segment relative memory
accesses (directly relevant to patches 1 and 2), which in turn resulted
in the discovery of XSA-191.
3) The CPUID improvement work to get maxphysaddr into a sensibly audited
state, and sensibly accessible location.
Outstanding hardware issues discovered include:
1) There is an observable delay in AMD Fam 10h processors between loading a
segment selector, and the results of the LDT/GDT memory access being
visible in the pagetables (via the Access bits being set).
2) SMAP provides insufficient information to the pagefault handler (via the
error code) to fully determine the nature of the access causing the
pagefault in the first place. See patch 2 for details.
Andrew Cooper (7):
x86/hvm: Correctly identify implicit supervisor accesses
x86/shadow: Try to correctly identify implicit supervisor accesses
x86/pagewalk: Helpers for reserved bit handling
x86/hvm: Adjust hvm_nx_enabled() to match how Xen behaves
x86/shadow: Use the pagewalk reserved bits helpers
x86/pagewalk: Consistently use guest_walk_*() helpers for translation
x86/pagewalk: Re-implement the pagetable walker
xen/arch/x86/hvm/emulate.c | 18 +-
xen/arch/x86/mm/guest_walk.c | 444 ++++++++++++++++++++------------------
xen/arch/x86/mm/hap/guest_walk.c | 23 +-
xen/arch/x86/mm/hap/nested_ept.c | 2 +-
xen/arch/x86/mm/p2m.c | 17 +-
xen/arch/x86/mm/shadow/multi.c | 108 +++++++---
xen/include/asm-x86/guest_pt.h | 180 +++++++++++++---
xen/include/asm-x86/hvm/hvm.h | 4 +-
xen/include/asm-x86/p2m.h | 2 +-
xen/include/asm-x86/page.h | 3 -
xen/include/asm-x86/processor.h | 3 +
xen/include/asm-x86/x86_64/page.h | 6 -
12 files changed, 489 insertions(+), 321 deletions(-)
--
2.1.4
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
next reply other threads:[~2017-02-27 14:03 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-27 14:03 Andrew Cooper [this message]
2017-02-27 14:03 ` [PATCH 1/7] x86/hvm: Correctly identify implicit supervisor accesses Andrew Cooper
2017-03-01 15:05 ` Jan Beulich
2017-03-02 16:14 ` Tim Deegan
2017-03-07 10:46 ` George Dunlap
2017-03-07 10:51 ` Andrew Cooper
2017-03-07 15:00 ` Paul Durrant
2017-02-27 14:03 ` [PATCH 2/7] x86/shadow: Try to correctly " Andrew Cooper
2017-03-01 15:11 ` Jan Beulich
2017-03-02 16:14 ` Tim Deegan
2017-03-07 11:26 ` George Dunlap
2017-03-07 11:55 ` Andrew Cooper
2017-02-27 14:03 ` [PATCH 3/7] x86/pagewalk: Helpers for reserved bit handling Andrew Cooper
2017-03-01 15:57 ` Jan Beulich
2017-03-02 12:23 ` Andrew Cooper
2017-03-02 14:12 ` Tim Deegan
2017-03-02 14:17 ` Andrew Cooper
2017-03-02 15:09 ` Tim Deegan
2017-03-02 15:14 ` Andrew Cooper
2017-03-02 16:15 ` Tim Deegan
2017-02-27 14:03 ` [PATCH 4/7] x86/hvm: Adjust hvm_nx_enabled() to match how Xen behaves Andrew Cooper
2017-03-01 16:00 ` Jan Beulich
2017-02-27 14:03 ` [PATCH 5/7] x86/shadow: Use the pagewalk reserved bits helpers Andrew Cooper
2017-03-01 16:03 ` Jan Beulich
2017-03-02 12:26 ` Andrew Cooper
2017-03-02 12:51 ` Jan Beulich
2017-03-02 12:56 ` Andrew Cooper
2017-03-02 13:19 ` Jan Beulich
2017-03-02 14:32 ` Andrew Cooper
2017-03-06 9:26 ` Tim Deegan
2017-03-02 14:33 ` Tim Deegan
2017-02-27 14:03 ` [PATCH 6/7] x86/pagewalk: Consistently use guest_walk_*() helpers for translation Andrew Cooper
2017-03-01 16:22 ` Jan Beulich
2017-03-01 16:33 ` Andrew Cooper
2017-03-01 16:41 ` Jan Beulich
2017-03-02 16:15 ` Tim Deegan
2017-03-06 18:25 ` George Dunlap
2017-02-27 14:03 ` [PATCH 7/7] x86/pagewalk: Re-implement the pagetable walker Andrew Cooper
2017-03-02 11:52 ` Jan Beulich
2017-03-02 12:00 ` Andrew Cooper
2017-03-02 12:54 ` Jan Beulich
2017-03-02 16:16 ` Tim Deegan
2017-03-06 18:28 ` George Dunlap
2017-03-06 18:33 ` Andrew Cooper
2017-03-06 18:39 ` George Dunlap
2017-03-07 12:57 ` George Dunlap
2017-03-01 16:24 ` [PATCH 0/7] Fixes to pagetable handling Jan Beulich
2017-03-01 16:32 ` Andrew Cooper
2017-03-06 16:42 ` [RFC XTF PATCH] Pagetable Emulation testing Andrew Cooper
2017-03-13 15:45 ` Jan Beulich
2017-03-13 17:48 ` Andrew Cooper
2017-03-14 11:17 ` Jan Beulich
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=1488204198-23948-1-git-send-email-andrew.cooper3@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=xen-devel@lists.xen.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 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.