From: Ian Campbell <ian.campbell@citrix.com>
To: xen-devel@lists.xensource.com
Cc: Ian Campbell <ian.campbell@citrix.com>
Subject: [PATCH 00 of 27 v2] libxl: rationalise libxl_device_* APIs
Date: Thu, 13 Oct 2011 10:53:25 +0100 [thread overview]
Message-ID: <patchbomb.1318499605@localhost.localdomain> (raw)
The following series overhauls the libxl_device_* APIs in an attempt
to make them more consistent across the types of devices and move
towards something we can support as a stable API longterm.
This is broadly speaking the changes I discussed in [0]
One of the early patches adds a big comment describing the API. It
would be useful if this got a particularly close eye with a view to
supporting it long term -- especially from actual and potential
consumers of the API (of who I've CC a few who sprang to mind).
Along the way I filed some rough edges of the internal implementation
of this stuff but my primary concern is to make the public facing API
one that we can commit to keeping stable.
One aspect which is missing is the ability to do asynchronous
add/remove etc. This requires the overhaul of the libxl event system
which Ian Jackson described at [1]. I did bear this in mind so
hopefully I have provided the majority of the necessary moving parts
internally.
Changes since last time:
- Rename type "destructor" functions to "dispose" freeing up the
verb "destroy" to mean, uh, destroying things, such as a forcible
remove of a device or destroying a domain.
- Use this new found freedom to s/force_remove/destroy/.
- Added a flags parameter to libxl_ctx_alloc. Might as well have one
for future flexibility
- Updated language bindings to use new scheme.
[0] http://www.gossamer-threads.com/lists/xen/devel/204668
[1] http://www.gossamer-threads.com/lists/xen/devel/212580
& http://www.gossamer-threads.com/lists/xen/devel/212578
& http://www.gossamer-threads.com/lists/xen/devel/212579
& http://www.gossamer-threads.com/lists/xen/devel/212581
next reply other threads:[~2011-10-13 9:53 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-10-13 9:53 Ian Campbell [this message]
2011-10-13 9:53 ` [PATCH 01 of 27 v2] libxl: add missing "break; " to do_pci_remove Ian Campbell
2011-10-13 9:53 ` [PATCH 02 of 27 v2] libxl: remove generated file testidl.c on clean Ian Campbell
2011-10-13 9:53 ` [PATCH 03 of 27 v2] libxl: idl: use "dispose" rather than "destroy" for function to free IDL types Ian Campbell
2011-10-13 11:26 ` Roger Pau Monné
2011-10-13 11:39 ` Ian Campbell
2011-10-13 11:47 ` Roger Pau Monné
2011-10-13 9:53 ` [PATCH 04 of 27 v2] libxl: libxl_destroy_cpumap becomes libxl_cpumap_destroy Ian Campbell
2011-10-13 9:53 ` [PATCH 05 of 27 v2] libxl: add a comment describing the device interfaces Ian Campbell
2011-10-13 9:53 ` [PATCH 06 of 27 v2] libxl: various fixes to libxl_device_disk_list (and internals) Ian Campbell
2011-10-13 9:53 ` [PATCH 07 of 27 v2] libxl: refactor code to construct disk from xenstore backend Ian Campbell
2011-10-13 9:53 ` [PATCH 08 of 27 v2] libxl: do not read f.e. xenstore dir in disk list function Ian Campbell
2011-10-17 14:20 ` Ian Jackson
2011-10-17 14:23 ` Ian Campbell
2011-10-17 14:26 ` Ian Jackson
2011-10-13 9:53 ` [PATCH 09 of 27 v2] libxl: update nic list API to use common device API style Ian Campbell
2011-10-13 9:53 ` [PATCH 10 of 27 v2] libxl: reimplement devid->nic in terms of from_xs_be function Ian Campbell
2011-10-13 9:53 ` [PATCH 11 of 27 v2] libxl: reimplement devid->disk " Ian Campbell
2011-10-13 9:53 ` [PATCH 12 of 27 v2] libxl: libxl_devid_to_* should take an integer device id Ian Campbell
2011-10-13 9:53 ` [PATCH 13 of 27 v2] libxl: separate forced and non-forced device remove Ian Campbell
2011-10-13 9:53 ` [PATCH 14 of 27 v2] libxl: split forced and non-forced uses of libxl__device_del Ian Campbell
2011-10-13 9:53 ` [PATCH 15 of 27 v2] libxl: use IDL to define device front- and back-end kinds Ian Campbell
2011-10-13 9:53 ` [PATCH 16 of 27 v2] libxl: do not remove unidentified frontend paths in libxl__devices_destroy Ian Campbell
2011-10-13 9:53 ` [PATCH 17 of 27 v2] libxl: use libxl__device in libxl_devices_destroy and libxl__device_pci_remove_xenstore Ian Campbell
2011-10-13 9:53 ` [PATCH 18 of 27 v2] libxl: merge libxl__device_del into libxl__device_remove Ian Campbell
2011-10-17 14:34 ` Roger Pau Monné
2011-10-17 14:43 ` Ian Campbell
2011-10-13 9:53 ` [PATCH 19 of 27 v2] libxl: use more descriptive variable names in libxl__devices_destroy Ian Campbell
2011-10-13 9:53 ` [PATCH 20 of 27 v2] libxl: convert disk handling to device API Ian Campbell
2011-10-13 9:53 ` [PATCH 21 of 27 v2] libxl: convert NIC " Ian Campbell
2011-10-13 9:53 ` [PATCH 22 of 27 v2] libxl: remove libxl_device_console_add Ian Campbell
2011-10-13 9:53 ` [PATCH 23 of 27 v2] libxl: convert VKB handling to device API Ian Campbell
2011-10-13 9:53 ` [PATCH 24 of 27 v2] libxl: convert VFB " Ian Campbell
2011-10-13 9:53 ` [PATCH 25 of 27 v2] libxl: reorder device functions to put functions for each device together Ian Campbell
2011-10-13 9:53 ` [PATCH 26 of 27 v2] libxl: convert PCI device handling to device API Ian Campbell
2011-10-13 9:53 ` [PATCH 27 of 27 v2] imported patch libxl-flags-on-ctx-alloc.patch Ian Campbell
2011-10-13 9:55 ` Ian Campbell
2011-10-13 10:44 ` libxl stable API work left (Was: Re: [PATCH 00 of 27 v2] libxl: rationalise libxl_device_* APIs) Ian Campbell
2011-10-18 12:55 ` [PATCH] libxl: fix device removal handling of wait_for_dev_destroy Roger Pau Monne
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=patchbomb.1318499605@localhost.localdomain \
--to=ian.campbell@citrix.com \
--cc=xen-devel@lists.xensource.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.