om booting using the default entry. Since I'm not
aware of any distro GRUB config which puts useful entries within
conditionals, ignore them.
Signed-off-by: Ross Lagerwall
---
tools/pygrub/src/GrubConf.py | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/tools/pyg
re the one who observed this?
Yes, that would probably solve the problem.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
ng it is OK.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
SA-60
XSA-64
XSA-80
XSA-82
XSA-97
XSA-107
XSA-114
XSA-150
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
linear stack of patches with dependency management
done in userspace, despite having old code checking. What kSplice use in
practice is exactly what is implemented here; a linear stack of patches
using some sort of identifier (build-id/uuid).
--
Ross Lagerwall
__
that is determined).
There's also a different type of symbol lookup in the xsplice code:
looking up the address of the symbol to be replaced. In this case, it is
the original symbol thatr needs to be returned. This prevents having a
chain of jumps if a function is patched multiple times.
--
R
w variables. The XSAs I've listed above fall into this
category.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
o trip a 5s watchdog timeout, let alone 30ms
timeout. So what is important is not really the number of pCPUs but the
kind of workload that it being run.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
roduced in this binary patch
or whether it overrides a symbol in a previous binary patch (or the core
binary). Either way, the symbol will be used when resolving a backtrace.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
ad_lock);
+
+return rc;
+}
+
+int xsplice_op(xen_sysctl_xsplice_op_t *xsplice)
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
mbol table, if more than one payload
is loaded, there will be duplicate conflicting symbols. So, to prevent
these symbols from going into the symbol table, I disallowed STT_NOTYPE.
Perhaps not the best solution but...
--
Ross Lagerwall
___
Xen-devel
names. The problem is that these are generated
symbols, so even if you had e.g. keyhandler.c#.LC0, keyhandler.c#.LC1,
in the symbol table, they might be completed unrelated if you change the
source even slightly. Having these entries in the symbol table would not
make any sense.
Rather than ignori
sistent for core and modules.
Yes. And, as best I can tell, .L symbols are not in the core table so
this would then make it consistent for modules.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
the spinlock, so it won't have initialized the interrupt in the
meantime.
Signed-off-by: Ross Lagerwall
---
arch/x86/xen/spinlock.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 9e2ba5c..f42e78d 100644
--- a/arch/x86/xen/spin
-9]* (for assembler generated local labels)
So to match what is used by the default hypervisor build, I think we
should change the check to discard when the symbol matches (2), (5), or
(6) above and refers to a mergeable section. The above rules are defined
in _bfd_elf_is_local_label_name and elf_
On 04/25/2016 10:03 AM, Jan Beulich wrote:
On 22.04.16 at 20:06, wrote:
On Wed, Apr 20, 2016 at 02:18:48PM +0100, Ross Lagerwall wrote:
On 04/13/2016 10:09 PM, Konrad Rzeszutek Wilk wrote:
snip
+static int xsplice_action(xen_sysctl_xsplice_action_t *action)
+{
+struct payload *data
KED stated, then it doesn't
make sense to have the above case.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
ame patch.\n"
+" load upload, check and apply .\n"
Since the check command is removed, this should be "upload and apply
..."
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
return _xc_xsplice_action(xch, name, XSPLICE_ACTION_CHECK, timeout);
+}
And this?
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
) */
+/* for operation to take. */
I guess XSPLICE_ACTION_CHECK should also be removed and XSPLICE_ACTION_*
renumbered.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
mpiling with -ffunction-sections -fdata-sections, the build tool can
quite easily exceed this limit. IMO the check doesn't serve any useful
purpose and should be removed.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
ec[i].load_addr = buf + offset[i];
+
+/*
+ * Don't copy NOBITS - such as BSS. We don't memset BSS as
+ * arch_xsplice_alloc_payload has zeroed it out for us.
+ */
+ if ( elf->sec[i].sec->sh_type != SHT_NOBITS )
+
ptr < (data->ro_addr + data->ro_size)) ||
+ (ptr >= data->text_addr &&
+ ptr < (data->text_addr + data->text_size)) )
The above 3 calculations are wrong due to the use of void *.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
gion->start = payload->text_addr;
+region->end = payload->text_addr + payload->text_size;
This calculation is wrong due to the use of void *.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
}
+/* Setup the virtual region with proper data. */
+region = &payload->region;
+
+region->symbols_lookup = xsplice_symbols_lookup;
+region->start = payload->text_addr;
+region->end = payload->text_addr + payload->text_size;
This calculation is wro
On 04/26/2016 12:05 PM, Ross Lagerwall wrote:
On 04/25/2016 04:35 PM, Konrad Rzeszutek Wilk wrote:
snip
diff --git a/xen/common/xsplice.c b/xen/common/xsplice.c
index 72a3b88..11b19dd 100644
--- a/xen/common/xsplice.c
+++ b/xen/common/xsplice.c
@@ -123,6 +123,35 @@ static int verify_payload
Cc: Ross Lagerwall
---
xen/common/xsplice_elf.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/xen/common/xsplice_elf.c b/xen/common/xsplice_elf.c
index 1e1f167..918a1bf 100644
--- a/xen/common/xsplice_elf.c
+++ b/xen/common/xsplice_elf.c
@@ -397,7 +397,8 @@ static int
Here is a set of changes to make building xSplice patches easier.
Tested to boot on x86.
Compile-tested on arm.
This is probably too late to make it into 4.7, but hey, if someone wants
to put it in I've CC'd Wei.
Ross Lagerwall (7):
lib: Add a generic implementation of current
When using xsplice, use of __LINE__ can generate spurious changes in
functions due to embedded line numbers. For release builds, remove the
use of these line numbers in domain_crash*() and print the current text
address instead.
Signed-off-by: Ross Lagerwall
---
xen/include/xen/sched.h | 14
Instead of using a locking order based on line numbers which doesn't
play nicely with xSplice, statically define the locking order.
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/mm/mm-locks.h | 29 -
1 file changed, 20 insertions(+), 9 deletions(-)
diff --git
When using xsplice, use of __LINE__ can generate spurious changes in
functions due to embedded line numbers. For release builds, remove the
use of these line numbers in IOMMU_WAIT_OP() and print the current text
address instead.
Signed-off-by: Ross Lagerwall
---
xen/drivers/passthrough/vtd
x27;s special sections with an extra period.
The idea for this was taken from a similar patch series applied to the
Linux kernel by the kSplice folks.
Signed-off-by: Ross Lagerwall
---
xen/arch/arm/xen.lds.S| 14 +++---
xen/arch/x86/boot/x86_64.S| 2 +-
xen/arch/x86/hvm/
Remove the unused x86 implementation.
Signed-off-by: Ross Lagerwall
---
xen/common/lib.c| 12
xen/include/asm-x86/processor.h | 10 --
xen/include/xen/lib.h | 2 ++
3 files changed, 14 insertions(+), 10 deletions(-)
diff --git a/xen/common/lib.c
When using xsplice, use of __LINE__ can generate spurious changes in
functions due to embedded line numbers. For release builds, remove the
use of these line numbers in the ACPI code and print the current text
address instead.
Signed-off-by: Ross Lagerwall
---
xen/drivers/acpi/utilities
When using xsplice, use of __LINE__ can generate spurious changes in
functions due to embedded line numbers. For release builds, remove the
use of these line numbers in BOOT_BUG_ON() and print the current text
address instead.
Signed-off-by: Ross Lagerwall
---
xen/common/page_alloc.c | 8
Fix a declared-but-not-defined warning when building with
XEN_BALLOON_MEMORY_HOTPLUG=n. This fixes a regression introduced by
commit dfd74a1edfab
("xen/balloon: Fix crash when ballooning on x86 32 bit PAE").
Signed-off-by: Ross Lagerwall
---
drivers/xen/balloon.c | 4 ++--
1 file
When loading patches, the code prevents loading a patch containing a new
symbol that duplicates a symbol from another loaded patch. However, the
check should also prevent loading a new symbol that duplicates a symbol
from the core hypervisor.
Signed-off-by: Ross Lagerwall
---
Wei, can you
Fix a declared-but-not-defined warning when building with
XEN_BALLOON_MEMORY_HOTPLUG=n. This fixes a regression introduced by
commit dfd74a1edfab
("xen/balloon: Fix crash when ballooning on x86 32 bit PAE").
Signed-off-by: Ross Lagerwall
---
v2: Moved function instead of adding a
rom_fork+0x3f/0x70
[] ? kthread_create_on_node+0x1e0/0x1e0
Signed-off-by: Ross Lagerwall
---
drivers/xen/events/events_base.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/xen/events/events_base.c b/drivers/xen/events/events_base.c
index cb7138c..71d49a9 100644
--- a/
ation/SubmittingPatches is not clear about it. Regardless, the
build fix should be backported to 4.4.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
g boot.
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/cpu/common.c | 2 +
xen/arch/x86/microcode.c| 131 +---
xen/arch/x86/microcode_amd.c| 3 +-
xen/arch/x86/microcode_intel.c | 3 +-
xen/arch/x86/setup.c| 2 +
xen/arc
Use the return value from early_microcode_update_cpu rather than
ignoring it.
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/microcode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/xen/arch/x86/microcode.c b/xen/arch/x86/microcode.c
index 4e7dfcd..7558202 100644
--- a
EFI properties table as reserved in the E820, much like is
done with the dmi regions.
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/efi/efi-boot.h | 8
xen/common/efi/boot.c | 20
xen/common/efi/efi.h| 4
3 files changed, 32 insertions(+)
d
On 05/15/2017 02:52 PM, Julien Grall wrote:
Hi Andrew,
On 08/05/17 17:29, Andrew Cooper wrote:
On 08/05/17 17:17, Ross Lagerwall wrote:
Some EFI firmware implementations may place the EFI properties table in
RAM marked as BootServicesData, which Xen does not consider as reserved.
When dom0
Occasionally, the top three bits of MSR_IA32_LASTINTTOIP
(MSR_LER_TO_LIP) may be incorrect, as though the MSR is using the
LBR_FORMAT_EIP_FLAGS_TSX format. The MSR should contain an offset into
the current code segment according to the Intel documentation. It is not
clear why this happens. It may b
is assumed to be erratum BDF14. Workaround the issue
by sign-extending into bits 48:63 for MSR_IA32_LASTINT{FROM,TO}IP.
Signed-off-by: Ross Lagerwall
---
Changes in v2:
- Use a single check if fixup is needed.
- Rename to include the erratum name/number.
- Sign extend properly rather than just
ocess_pending_softirqs();
if ( spin_trylock(&payload_lock) )
return
The effect is that we have just effectively
taken the lock and returned without unlocking!
CC: Ross Lagerwall
CC: Boris Ostrovsky
CC: Jan Beulich
CC: Andrew Cooper
CC: Julien Grall
Signed-off
Ignore differences in discard sections. They are not included in the final xen
binary so there is no need to include them in the live patch.
Signed-off-by: Ross Lagerwall
---
create-diff-object.c | 8
1 file changed, 8 insertions(+)
diff --git a/create-diff-object.c b/create-diff
Remove the requirement that section twins have the same alignment. The
section alignment of the patched section is respected by the loader in
Xen so it shouldn't matter if the original section alignment was
different.
Signed-off-by: Ross Lagerwall
---
create-diff-object.c | 1 -
1 file ch
s not needed. I had
previously fixed this but hadn't yet sent out that patch. I've now sent
this fix along with another to xen-devel.
Regards,
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
On 06/09/2017 06:08 PM, Konrad Rzeszutek Wilk wrote:
On Fri, Jun 09, 2017 at 06:00:35PM +0100, Andrew Cooper wrote:
On 09/06/17 17:38, Konrad Rzeszutek Wilk wrote:
On Fri, Jun 09, 2017 at 05:03:36PM +0100, Ross Lagerwall wrote:
Remove the requirement that section twins have the same alignment
s are probably rarely used these days, but they are used by
GRUB 2 to write to its environment file which happens by default on
Ubuntu.
Signed-off-by: Ross Lagerwall
---
tools/firmware/rombios/rombios.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/firmware/rombios/rombio
When the guest writes to the RTC, Xen emulates it and broadcasts a
TIMEOFFSET ioreq. Emit an RTC_CHANGE QMP message when this happens
rather than ignoring it so that something useful can be done with the
information.
Signed-off-by: Ross Lagerwall
---
hw/i386/xen/xen-hvm.c | 2 ++
1 file changed
On 08/21/2017 11:30 PM, Stefano Stabellini wrote:
On Mon, 21 Aug 2017, Ross Lagerwall wrote:
When the guest writes to the RTC, Xen emulates it and broadcasts a
TIMEOFFSET ioreq. Emit an RTC_CHANGE QMP message when this happens
rather than ignoring it so that something useful can be done with
RTC.
This patch by itself doesn't affect any of the toolstacks that I
checked; the libxl toolstack doesn't currently handle this event nor
does the XAPI toolstack. If nothing handles the event, it is simply
ignored. We plan on modifying XAPI to handle it.
Signed-off-by: Ross
lying after an revert" for details.
Signed-off-by: Konrad Rzeszutek Wilk
---
Cc: Ross Lagerwall
The patch looks OK, but what is the use case for a live patch with only
.rodata?
Regards,
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lis
ot; in the warning.
Also fix one case where we would fail if the size of the .ex_table
was being zero - but that is OK.
Signed-off-by: Konrad Rzeszutek Wilk
---
Cc: Ross Lagerwall
Reviewed-by: Ross Lagerwall
___
Xen-devel mailing list
On 10/05/2017 02:51 PM, Konrad Rzeszutek Wilk wrote:
On Thu, Oct 05, 2017 at 02:47:30PM +0100, Ross Lagerwall wrote:
On 09/20/2017 11:31 PM, Konrad Rzeszutek Wilk wrote:
If the livepatch has only .rodata sections then it is OK to also
apply/revert/apply the livepatch without having to worry
titled: "livepatch/arm/x86: Rename note_depends symbol from
test-cases."
which fixes this.
Signed-off-by: Konrad Rzeszutek Wilk
---
Cc: Jan Beulich
Cc: Andrew Cooper
Cc: Ian Jackson
Cc: Wei Liu
Acked-by: Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
user_gid) < 0) {
fprintf(stderr, "Failed to setuid(%d)\n", user_pwd->pw_uid);
exit(1);
}
This last one should be user_uid, not user_gid.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
ctory
has to contain a valid /dev/null. This is a bit annoying and prevents
the chroot being on a "nodev" mount.
Regards,
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
dmop to fix this properly.
For now, document the problem.
Signed-off-by: Ian Jackson
Reported-by: Ross Lagerwall
CC: Ross Lagerwall
CC: Wei Liu
CC: Paul Durrant
---
docs/man/xl.cfg.pod.5.in | 6 ++
1 file changed, 6 insertions(+)
diff --git a/docs/man/xl.cfg.pod.5.in b/docs/man
with hardcoded rules fixes the problem
(although presumably there's a better way):
handlereg.o: $(AUTOINCS)
handlereg.opic: $(AUTOINCS)
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
hes?
Is the latter check not needed for ARM64?
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
On 10/06/2017 02:19 PM, Paul Durrant wrote:
-Original Message-
From: Xen-devel [mailto:xen-devel-boun...@lists.xen.org] On Behalf Of
Ross Lagerwall
Sent: 06 October 2017 13:58
To: Ian Jackson ; qemu-de...@nongnu.org
Cc: Anthony Perard ; xen-
de...@lists.xenproject.org; Stefano Stabellini
Return 0 to indicate success rather than whatever rc was previously set
to (-EINVAL).
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/hvm/dm.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/xen/arch/x86/hvm/dm.c b/xen/arch/x86/hvm/dm.c
index 82addea..32ade95 100644
--- a/xen/arch/x86/hvm
s its new root has a /dev/null).
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
Implement support for restricting evtchn handles to a particular domain
on Linux by calling the IOCTL_EVTCHN_RESTRICT_DOMID ioctl (support added
in Linux v4.8).
Signed-off-by: Ross Lagerwall
---
tools/include/xen-sys/Linux/evtchn.h | 15 +++
tools/libs/evtchn/Makefile
Signed-off-by: Ross Lagerwall
---
tools/Rules.mk| 2 +-
tools/libs/evtchn/Makefile| 4 ++--
tools/libs/evtchn/core.c | 13 +
tools/libs/evtchn/private.h | 3 +++
tools/libs/toolcore/include/xentoolcore.h
On 10/16/2017 11:53 AM, Ian Jackson wrote:
Ross Lagerwall writes ("[PATCH v1 1/2] tools/libs/evtchn: Add support for
restricting a handle"):
+/*
+ * Restrict this file descriptor so that it can only be used to bind
+ * new interdomain events from one domain.
Can it be used to
On 10/16/2017 12:29 PM, Ian Jackson wrote:
Ross Lagerwall writes ("Re: [PATCH v1 1/2] tools/libs/evtchn: Add support for
restricting a handle"):
No. As far as I can see, it can only be used to bind new interdomain
events, not other events.
OK, good, thanks.
This entire file (inc
On 10/16/2017 11:55 AM, Ian Jackson wrote:
Ross Lagerwall writes ("[PATCH v2 2/2] xentoolcore_restrict_all: Implement for
libxenevtchn"):
Signed-off-by: Ross Lagerwall
...
int osdep_evtchn_open(xenevtchn_handle *xce);
diff --git a/tools/libs/toolcore/include/xentoolcore.h
b/
*/
Why did all the symbols get moved to VERS_1.1 rather than adding only
the new one to VERS_1.1 and keeping the rest at VERS_1.0 (like has been
done with libxenforeignmemory)?
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https
map alone and added a new version
which simply adds the new symbol.
Fix this.
Reported-by: Ross Lagerwall
CC: Stefano Stabellini
Signed-off-by: Ian Jackson
---
Reviewed-by: Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen
Implement support for restricting evtchn handles to a particular domain
on Linux by calling the IOCTL_EVTCHN_RESTRICT_DOMID ioctl (support added
in Linux v4.8).
Signed-off-by: Ross Lagerwall
Acked-by: Ian Jackson
Release-acked-by: Julien Grall
---
No change since v1.
tools/include/xen-sys
Signed-off-by: Ross Lagerwall
---
Changed in v2:
* Keep warning about DoS and resource exhaustion being a possibility.
(Note that v1 of this patch was incorrectly titled v2.)
tools/Rules.mk| 2 +-
tools/libs/evtchn/Makefile| 4 ++--
tools/libs
Provide XEN_DMOP_add_to_physmap, a limited version of
XENMEM_add_to_physmap to allow a deprivileged QEMU to move VRAM when a
guest programs its BAR. It is equivalent to XENMEM_add_to_physmap with
space == XENMAPSPACE_gmfn.
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/hvm/dm.c | 17
Provide XEN_DMOP_pin_memory_cacheattr to allow a deprivileged QEMU to
pin the caching type of RAM after moving the VRAM. It is equivalent to
XEN_DOMCTL_pin_memory_cacheattr.
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/hvm/dm.c | 12
xen/include/public/hvm/dm_op.h | 14
Signed-off-by: Ross Lagerwall
---
tools/libs/devicemodel/core.c | 19 +++
tools/libs/devicemodel/include/xendevicemodel.h | 14 ++
tools/libs/devicemodel/libxendevicemodel.map| 1 +
3 files changed, 34 insertions(+)
diff --git a/tools/libs
Signed-off-by: Ross Lagerwall
---
tools/libs/devicemodel/Makefile | 2 +-
tools/libs/devicemodel/core.c | 17 +
tools/libs/devicemodel/include/xendevicemodel.h | 13 +
tools/libs/devicemodel/libxendevicemodel.map| 5 +
4
Make it global in preparation to be called by a new dmop.
Signed-off-by: Ross Lagerwall
---
xen/common/memory.c | 5 ++---
xen/include/xen/mm.h | 3 +++
2 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/xen/common/memory.c b/xen/common/memory.c
index ad987e0..c4f05c7 100644
--- a
The recently added support for restricting QEMU prevents use of the VGA
console. This series addresses that by adding a couple of new dmops.
A corresponding patch for QEMU is needed to make use of the new dmops.
Ross Lagerwall (5):
xen/mm: Make xenmem_add_to_physmap public
xen: Provide
r.
Based on your other comments, would it make sense to instead use
XENMAPSPACE_gmfn_range and have the caller set the size?
As it is currently, QEMU does only populate VRAM one page at a time
(using xen_xc_domain_add_to_physmap) so it is already slow but it could
be impro
Provide XEN_DMOP_pin_memory_cacheattr to allow a deprivileged QEMU to
pin the caching type of RAM after moving the VRAM. It is equivalent to
XEN_DOMCTL_pin_memory_cacheattr.
Signed-off-by: Ross Lagerwall
---
Changed in v2:
* Check pad is 0.
xen/arch/x86/hvm/dm.c | 18
Provide XEN_DMOP_add_to_physmap, a limited version of
XENMEM_add_to_physmap to allow a deprivileged QEMU to move VRAM when a
guest programs its BAR. It is equivalent to XENMEM_add_to_physmap with
space == XENMAPSPACE_gmfn_range.
Signed-off-by: Ross Lagerwall
---
Changed in v2:
* Make it operate
Signed-off-by: Ross Lagerwall
---
Changed in v2:
* Make it operate on a range.
tools/libs/devicemodel/Makefile | 2 +-
tools/libs/devicemodel/core.c | 21 +
tools/libs/devicemodel/include/xendevicemodel.h | 15 +++
tools/libs
Signed-off-by: Ross Lagerwall
---
Acked-by: Ian Jackson
Reviewed-by: Paul Durrant
---
tools/libs/devicemodel/core.c | 19 +++
tools/libs/devicemodel/include/xendevicemodel.h | 14 ++
tools/libs/devicemodel/libxendevicemodel.map| 1 +
3 files
Make it global in preparation to be called by a new dmop.
Signed-off-by: Ross Lagerwall
---
Reviewed-by: Paul Durrant
---
xen/common/memory.c | 5 ++---
xen/include/xen/mm.h | 3 +++
2 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/xen/common/memory.c b/xen/common/memory.c
on a
range.
Ross Lagerwall (5):
xen/mm: Make xenmem_add_to_physmap global
xen: Provide XEN_DMOP_add_to_physmap
xen: Provide XEN_DMOP_pin_memory_cacheattr
tools: libxendevicemodel: Provide xendevicemodel_add_to_physmap
tools: libxendevicemodel: Provide xendevicemodel_pin_memory_cach
above size rather than below
(since IIRC we only need pad up to the next 4 byte boundary).
Nope, the build fails unless I pad it to an 8 byte boundary. This is
also why I added padding to struct xen_dm_op_pin_memory_cacheattr...
--
Ross Lagerwall
__
n a range.
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
andle()
_before_ calling osdep_evtchn_close() to avoid trying to restrict a
closed fd or some other fd that happens to have the same number.
I think all the other libs need to be fixed as well, unless there was a
reason it was done this way.
--
Ross Lagerwall
On 11/14/2017 11:51 AM, Ian Jackson wrote:
Ross Lagerwall writes ("Re: [PATCH for-4.10] libs/evtchn: Remove active handler on
clean-up or failure"):
On 11/10/2017 05:10 PM, Julien Grall wrote:
Commit 89d55473ed16543044a31d1e0d4660cf5a3f49df "xentoolcore_restrict_all
close(h->watch_pipe[1]);
}
-close(h->fd);
xentoolcore__deregister_active_handle(&h->tc_ah);
+close(h->fd);
Since the rest of this file uses tabs, you may as well use tabs for this
line as well.
Reviewed-by: Ross Lagerwall
_
Gross
Thanks,
--
Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
#x27;m looking at this code...
This doesn't work if access->width > sizeof(val) (4 bytes). The same
value (access->address) is always passed into acpi_cpumap_access_common
for 'port' and this is used as an offset into the avail_cpus array. So
the offset is unchanged and only
On 07/31/2017 03:29 PM, Boris Ostrovsky wrote:
On 07/31/2017 10:12 AM, Andrew Cooper wrote:
On 31/07/17 14:55, Boris Ostrovsky wrote:
On 07/31/2017 09:20 AM, Ross Lagerwall wrote:
Hi Boris,
I've modified your PVH VCPU hotplug support v6 patch series [1] to
support HVM guests running _wi
.
Additionally, perform a safety check on elf->sym[symndx].sym before
derefencing it, to avoid tripping over a NULL pointer when calculating val.
Signed-off-by: Andrew Cooper
Reviewed-by: Ross Lagerwall
___
Xen-devel mailing list
Xen-devel@lists.xen.
On 06/22/2017 07:15 PM, Andrew Cooper wrote:
Each of these arrays is sparse. Use zeroed allocations to cause uninitialised
array elements to contain deterministic values, most importantly for the
embedded pointers.
Signed-off-by: Andrew Cooper
Reviewed-by: Ross Lagerwall
.
Additionally, fix an off-by-one error while range checking symndx, and perform
a safety check on elf->sym[symndx].sym before derefencing it, to avoid
tripping over a NULL pointer when calculating val.
Signed-off-by: Andrew Cooper
Reviewed-by: Ross Lagerw
1 - 100 of 357 matches
Mail list logo
