From: Arianna Avanzini <avanzini.arianna@gmail.com>
To: xen-devel@lists.xen.org
Cc: Ian.Campbell@eu.citrix.com, paolo.valente@unimore.it,
keir@xen.org, stefano.stabellini@eu.citrix.com,
Ian.Jackson@eu.citrix.com, dario.faggioli@citrix.com,
tim@xen.org, julien.grall@citrix.com, etrudeau@broadcom.com,
andrew.cooper3@citrix.com, JBeulich@suse.com,
avanzini.arianna@gmail.com, viktor.kleinik@globallogic.com
Subject: [PATCH v8 00/14] Implement the XEN_DOMCTL_memory_mapping hypercall for ARM
Date: Sun, 25 May 2014 12:51:41 +0200 [thread overview]
Message-ID: <1401015115-7610-1-git-send-email-avanzini.arianna@gmail.com> (raw)
Hello,
this is a new version - the eighth - of my implementation proposal for the
memory_mapping DOMCTL for the ARM architecture. As usual, this cover letter just
lists the most relevant changes performed on the patchset, while a more detailed
description can be found in the description and changelog of each patch; also,
some more information about the patch series' origin can be found in the last
full cover letter ([1]).
First of all, the commit descriptions have been reworded to get rid of tentative
phrases that were pointless in that context, as suggested by Dario Faggioli.
Patch 0002 has been modified to revert some uncorrect changes due to my
misunderstanding of previous comments, as pointed out by Julien Grall. Also,
according to his suggestions, an error message has been added to let a failure
encountered during an unmap be less obscure. The error path has been improved
to include unmapping previously-created page-table mappings, unlocking the
p2m_lock and flushing the TLBs in case some mappings have been successfully
removed.
Patch 0004 has been changed to address issues pointed out by Julien Grall and
Ian Campbell. Now the correct DIV_ROUND_UP() macro is used instead of the
ternary operator, and the map_one_mmio() function, defined for the xgene-storm
platform, accepts pfns as arguments instead of paddr_t. Also, the function
apply_p2m_changes() is now called from map_mmio_regions() with the correct
count.
A new patch 0005 has been added to let the map_mmio_regions() function for ARM
unmap partially-mapped I/O-memory regions if mapping a page fails, as suggested
by Julien Grall. The approach which has been used includes having the function
apply_p2m_changes() return, in the case of an error, the number of pages that
have been successfully mapped, therefore allowing the caller to undo what has
been done, if necessary. The definition of unmap_mmio_regions() for ARM has been
moved to this patch and now uses the correct count as indicated by Julien Grall.
The title and description of patch 0006 have been changed to (hopefully) better
fit the patch content, as requested by Jan Beulich.
The patch factoring out map and unmap operations from the x86 version of the
memory_mapping DOMCTL has been split into two, as suggested by Jan Beulich, to
make the review easier: patch 0007 now just cleans up existing code, by letting
mfn + nr_mfns be computed only once and by changing the name of the tmp_rc local
variable to rc; patch 0009 instead actually factors out the operations.
A new patch 0008 now adds an empty stub for the ARM architecture of the MTRR-
related function memory_type_changed(), currently available only for x86, and
makes its definition common to both x86 and ARM. This was requested by Jan
Beulich and Ian Campbell.
The call to PERROR() in patch 0012 has been fixed, as it showed an error message
whose meaning was the opposite of the correct one, as pointed out by Jan
Beulich.
The patch which implemented a separation of the functions of the memory_mapping
and iomem_permission DOMCTLs has now been split into two, as suggested by Jan
Beulich: patch 0013 now contains only libxl-related preparatory changes, while
patch 0014 actually implements the separation. As for patch 0014, changes
performed on the iomem_permission DOMCTL handling code have been dropped, while
the permission checks performed by memory_mapping have been changed to be more
conservative and check both the calling domain and the grantee domain's
permissions.
As of now, patch 0014 performs the separation only for I/O memory and not for
I/O ports (as was instead suggested by Jan Beulich) because, as far as I know,
Julien Grall is working on the I/O ports patch series, and I'm afraid I might be
conflicting with his work. I am however surely willing to add a patch for that
if he is OK with it.
The code has again been tested on a cubieboard2, with Linux v3.15-rc3 as a dom0
and ERIKA Enterprise ([2]) as a domU. The x86 bits have been tested on an x86_64
machine, as I forgot to say in the last versions of the patchset and Jan Beulich
pointed out; in particular, this last version, has been tested with Linux
3.15.0-rc5 as both dom0 and domU.
Any feedback about this new version of the patchset is more than welcome,
Arianna
[1] http://lists.xen.org/archives/html/xen-devel/2014-03/msg01724.html
[2] http://erika.tuxfamily.org/drupal/
Arianna Avanzini (14):
arch/arm: domain build: let dom0 access I/O memory of mapped devices
arch/arm: add consistency check to REMOVE p2m changes
arch/arm: let map_mmio_regions() take pfn as parameters
arch/arm: let map_mmio_regions() use start and count
arch/arm: unmap partially-mapped I/O-memory regions
arch/x86: warn if to-be-removed mapping does not exist
arch/x86: cleanup memory_mapping DOMCTL
xen/common: move memory_type_changed() function to common code
xen/x86: factor out map and unmap from the memory_mapping DOMCTL
xen/common: move the memory_mapping DOMCTL hypercall to common code
tools/libxl: parse optional start gfn from the iomem config option
tools/libxl: handle the iomem parameter with the memory_mapping hcall
tools/libxl: explicitly grant access to needed I/O-memory ranges
xen/common: do not implicitly permit access to mapped I/O memory
docs/man/xl.cfg.pod.5 | 18 +++++--
tools/libxc/xc_domain.c | 12 +++++
tools/libxl/libxl.h | 10 ++++
tools/libxl/libxl_create.c | 34 +++++++++++++
tools/libxl/libxl_internal.h | 1 +
tools/libxl/libxl_pci.c | 26 ++++------
tools/libxl/libxl_types.idl | 4 ++
tools/libxl/xl_cmdimpl.c | 17 ++++---
xen/arch/arm/Makefile | 1 +
xen/arch/arm/domain_build.c | 18 +++++--
xen/arch/arm/gic.c | 15 +++---
xen/arch/arm/mtrr.c | 6 +++
xen/arch/arm/p2m.c | 99 ++++++++++++++++++++++++++++--------
xen/arch/arm/platforms/exynos5.c | 9 ++--
xen/arch/arm/platforms/omap5.c | 17 +++----
xen/arch/arm/platforms/xgene-storm.c | 16 +++---
xen/arch/x86/domctl.c | 76 ---------------------------
xen/arch/x86/mm/p2m.c | 53 +++++++++++++++++--
xen/common/domctl.c | 55 +++++++++++++++++++-
xen/include/asm-arm/mm.h | 2 +
xen/include/asm-arm/mtrr.h | 6 +++
xen/include/asm-arm/p2m.h | 11 ++--
xen/include/asm-x86/mtrr.h | 2 +-
xen/include/asm-x86/p2m.h | 3 +-
xen/include/xen/mtrr.h | 8 +++
xen/include/xen/p2m-common.h | 16 ++++++
26 files changed, 365 insertions(+), 170 deletions(-)
create mode 100644 xen/arch/arm/mtrr.c
create mode 100644 xen/include/asm-arm/mtrr.h
create mode 100644 xen/include/xen/mtrr.h
create mode 100644 xen/include/xen/p2m-common.h
--
1.9.2
next reply other threads:[~2014-05-25 10:51 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-25 10:51 Arianna Avanzini [this message]
2014-05-25 10:51 ` [PATCH v8 01/14] arch/arm: domain build: let dom0 access I/O memory of mapped devices Arianna Avanzini
2014-06-10 15:04 ` Ian Campbell
2014-05-25 10:51 ` [PATCH v8 02/14] arch/arm: add consistency check to REMOVE p2m changes Arianna Avanzini
2014-05-25 15:50 ` Julien Grall
2014-06-05 13:45 ` Ian Campbell
2014-06-05 13:50 ` Ian Campbell
2014-05-25 10:51 ` [PATCH v8 03/14] arch/arm: let map_mmio_regions() take pfn as parameters Arianna Avanzini
2014-05-25 10:51 ` [PATCH v8 04/14] arch/arm: let map_mmio_regions() use start and count Arianna Avanzini
2014-05-25 15:56 ` Julien Grall
2014-06-05 13:53 ` Ian Campbell
2014-05-25 10:51 ` [PATCH v8 05/14] arch/arm: unmap partially-mapped I/O-memory regions Arianna Avanzini
2014-05-25 16:04 ` Julien Grall
2014-06-05 14:03 ` Ian Campbell
2014-06-05 14:09 ` Julien Grall
2014-05-25 10:51 ` [PATCH v8 06/14] arch/x86: warn if to-be-removed mapping does not exist Arianna Avanzini
2014-06-05 14:06 ` Ian Campbell
2014-05-25 10:51 ` [PATCH v8 07/14] arch/x86: cleanup memory_mapping DOMCTL Arianna Avanzini
2014-05-26 9:57 ` Jan Beulich
2014-05-25 10:51 ` [PATCH v8 08/14] xen/common: move memory_type_changed() function to common code Arianna Avanzini
2014-05-25 16:15 ` Julien Grall
2014-05-26 9:58 ` Jan Beulich
2014-06-05 14:08 ` Ian Campbell
2014-05-25 10:51 ` [PATCH v8 09/14] xen/x86: factor out map and unmap from the memory_mapping DOMCTL Arianna Avanzini
2014-05-26 10:04 ` Jan Beulich
2014-05-25 10:51 ` [PATCH v8 10/14] xen/common: move the memory_mapping DOMCTL hypercall to common code Arianna Avanzini
2014-05-25 16:42 ` Julien Grall
2014-05-26 10:07 ` Jan Beulich
2014-05-26 11:03 ` Julien Grall
2014-06-05 14:21 ` Ian Campbell
2014-06-05 14:33 ` Tim Deegan
2014-06-05 14:39 ` Ian Campbell
2014-05-26 10:06 ` Jan Beulich
2014-05-25 10:51 ` [PATCH v8 11/14] tools/libxl: parse optional start gfn from the iomem config option Arianna Avanzini
2014-05-25 10:51 ` [PATCH v8 12/14] tools/libxl: handle the iomem parameter with the memory_mapping hcall Arianna Avanzini
2014-05-25 17:04 ` Julien Grall
2014-06-05 14:27 ` Ian Campbell
2014-05-25 10:51 ` [PATCH v8 13/14] tools/libxl: explicitly grant access to needed I/O-memory ranges Arianna Avanzini
2014-05-25 17:08 ` Julien Grall
2014-05-26 10:11 ` Jan Beulich
2014-05-26 10:58 ` Julien Grall
2014-05-26 11:15 ` Jan Beulich
2014-06-05 14:31 ` Ian Campbell
2014-06-05 14:37 ` Ian Campbell
2014-06-05 14:54 ` Jan Beulich
2014-06-05 14:53 ` Jan Beulich
2014-05-26 10:10 ` Jan Beulich
2014-05-25 10:51 ` [PATCH v8 14/14] xen/common: do not implicitly permit access to mapped I/O memory Arianna Avanzini
2014-07-01 10:45 ` [PATCH v8 00/14] Implement the XEN_DOMCTL_memory_mapping hypercall for ARM Julien Grall
2014-07-01 10:55 ` Arianna Avanzini
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=1401015115-7610-1-git-send-email-avanzini.arianna@gmail.com \
--to=avanzini.arianna@gmail.com \
--cc=Ian.Campbell@eu.citrix.com \
--cc=Ian.Jackson@eu.citrix.com \
--cc=JBeulich@suse.com \
--cc=andrew.cooper3@citrix.com \
--cc=dario.faggioli@citrix.com \
--cc=etrudeau@broadcom.com \
--cc=julien.grall@citrix.com \
--cc=keir@xen.org \
--cc=paolo.valente@unimore.it \
--cc=stefano.stabellini@eu.citrix.com \
--cc=tim@xen.org \
--cc=viktor.kleinik@globallogic.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 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).