lib<foo>_stubs.a was linking against $(LIBS_<foo>) which only includes
dynamic linkage information which is not useful for a static library.
$(LIBS_<foo>) is still used when linking <foo>.cmxa
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
Add an ocaml binding for libxl_get_topologyinfo in the style of
libxl_get_physinfo.
The function returns an array of optional records, one per possible pCPU.
Signed-off-by: David Scott <dave.scott@eu.citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
hashtable.c has already been copied, forked and modified, there
doesn't seem much point in avoiding the same for the header until
someone feels motivated to properly refactor.
Add comments to the various duplicated files cross-referencing each
other for future reference and as a barrier to forking again...
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
22166:251694a87f1d changed the libxl interface to remove the need for
users to understand BDF encoding but did not propagate the change to
the ocaml bindings. Do that now.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
The structure of the BDF argument used with PCI passthrough related
hypercalls was taken from the structure of the PCI config_address
register (I/O port 0xCF8) which allows I/O mapped access to PCI
configuration space but these fields have no meaning in the context of
PCI passthrough configuration and hence do not need to be exposed via
libxl.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
current check includes ingress and pfifo_fast.
Add mq to the list of allowed qdiscs already installed
on ifb. This patch fixes cases where remus fails to start,
due to an mq qdisc already present on the vif.
Signed-off-by: Shriram Rajagopalan <rshriram@cs.ubc.ca>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
Most users can use _PAGE_NX_BIT directly.
The few genuine users in mm.c can do the cpu_has_nx check more clearly
in other ways.
Signed-off-by: Keir Fraser <keir@xen.org>
When submitting the patch that became 23101:dd386a4b6595 I forgot that
I actually intended to remove the pointless IRQ disabling and
restoring (and the then pointless comment). Pointless because the
code, other than its pre-23013:65d26504e843 original, now runs before
interrupts get enabled for the first time.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Was broken by 23099:612171ff82ea.
A bool_t is a single byte, and needs a 'q' register constraint. Avoid
the whole issue by changing the variable to an int, and explicitly
specify the operand suffix as 'l' for good measure.
Signed-off-by: Keir Fraser <keir@xen.org>
TXT code calls acpi_parse_dmar() with a transient copy of the DMAR
table, and hence storing the pointer for later reference was wrong.
Obtain the pointer in acpi_dmar_init() instead.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Tested-by: Gang Wei <gang.wei@intel.com>
Add a barrier-appropriate consistency check to spinlock.c, and add
code comments to explain why barrier operations are more relaxed than
lock-acquisition operations.
Signed-off-by: Keir Fraser <keir@xen.org>
Also rework to match the 'trylock' style of raw function used for
spinlocks.
Inspired by Jan Beulich's patch to do similar improved scaling.
Signed-off-by: Keir Fraser <keir@xen.org>
Build has been broken since at least 18969:d6889b3b6423 (early
2009) and it has been unhooked from the top level build since forever
AFAICT. The last actual development (as opposed to tree wide
cleanups and build fixes) appears to have been 11594:6d7bba6443ef in
2006. The functionality of vnet has apparently been superceded by
VLANs, ebtables, Ethernet-over-IP etc all of which are well integrated
with upstream kernels and distros.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: Keir Fraser <keir@xen.org>
According to the (now getting removed) comment in struct
hpet_event_channel, this was to prevent accessing a CPU's
timer_deadline after it got cleared from cpumask. This can be done
without a lock altogether - hpet_broadcast_exit() can simply clear
the bit, and handle_hpet_broadcast() can read timer_deadline before
looking at the mask a second time (the cpumask bit was already
found set by the surrounding loop).
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Acked-by: Gang Wei <gang.wei@intel.com>
Print number of pages to allocate for per-cpu tracebuffer and metadata
to ease debugging when allocation fails.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
This isolates users of libxenlight from the need to know about the
different OS schemes for UUIDs, in particular the linkage
requirements.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
19025:bd78714b8594 switched to "-MMD -MF" in preference, adding the
correct runes to tools/Rules.mk but appears to have missed some other
uses of -Wp,...
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
All external users pass 0 anyway so make the version of
libxl_domain_rename which takes a transaction internal and provide an
external facing function which does not expose a transaction.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
It's not needed unless you are including headers in the current
directory _and_ you can't use "" instead of <> because the headers are
to be installed.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
Ian Campbell
David Scott
Ian Jackson
Shriram Rajagopalan
Keir Fraser
Jan Beulich
Olaf Hering