This run is configured for baseline tests only.
flight 38114 xen-4.6-testing real [real]
http://osstest.xs.citrite.net/~osstest/testlogs/logs/38114/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-amd64-xl-qemuu-debianhvm-amd64-xs
flight 62544 linux-linus real [real]
http://logs.test-lab.xenproject.org/osstest/logs/62544/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-armhf-armhf-xl-cubietruck 6 xen-bootfail REGR. vs. 59254
test-armhf-armhf-xl
In case the kernel of a new pv-domU indicates it is supporting a p2m
list outside the initial kernel mapping by specifying INIT_P2M, let
the domain builder allocate the memory for the p2m list from physical
guest memory only and map it to the address the kernel is expecting.
This will enable loadi
Carve out the p2m list allocation from the .alloc_magic_pages hook of
the domain builder in order to prepare allocating the p2m list outside
of the initial kernel mapping. This will be needed to support loading
domains with huge memory (>512 GB).
Signed-off-by: Juergen Gross
Acked-by: Ian Campbel
In case the kernel of a new pv-domU indicates it is supporting an
unmapped initrd, don't waste precious virtual space for the initrd,
but allocate only guest physical memory for it.
Signed-off-by: Juergen Gross
---
tools/libxc/xc_dom_core.c | 22 --
1 file changed, 20 inserti
The Xen hypervisor supports starting a dom0 with large memory (up to
the TB range) by not including the initrd and p2m list in the initial
kernel mapping. Especially the p2m list can grow larger than the
available virtual space in the initial mapping.
The started kernel is indicating the support o
Support of an unmapped initrd is indicated by the kernel of the domain
via elf notes. In order not to have to use raw elf data in the tools
for support of an unmapped initrd add a flag to the parsed data area
to indicate the kernel supporting this feature.
Switch using this flag in the hypervisor
The allocate() callback in struct xc_dom_image is never set. Remove it.
Signed-off-by: Juergen Gross
Acked-by: Ian Jackson
---
tools/libxc/include/xc_dom.h | 2 --
tools/libxc/xc_dom_core.c| 4
2 files changed, 6 deletions(-)
diff --git a/tools/libxc/include/xc_dom.h b/tools/libxc/inc
Use a bit mask for testing of a set bit instead of test_bit in case no
atomic operation is needed, as this will lead to smaller and more
effective code.
Signed-off-by: Juergen Gross
---
xen/arch/x86/domain.c | 8
xen/arch/x86/x86_64/compat/mm.c | 4 ++--
xen/common/multicall.c
Instead of using test_bit() which is an atomic operation and limits
the compiler's choices to do optimization, use logical ANDs with
bitmasks where possible.
The possible candidates have been detected by searching definitions
of bitmasks in the form:
#define MASK 1 << _MASK
On x86 the resulting
Use a bit mask for testing of a set bit instead of test_bit in case no
atomic operation is needed, as this will lead to smaller and more
effective code.
Signed-off-by: Juergen Gross
---
xen/arch/x86/domain.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/xen/arch/x86/dom
Use a bit mask for testing of a set bit instead of test_bit in case no
atomic operation is needed, as this will lead to smaller and more
effective code.
Signed-off-by: Juergen Gross
---
xen/common/sched_rt.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/xen/common/sched
Use a bit mask for testing of a set bit instead of test_bit in case no
atomic operation is needed, as this will lead to smaller and more
effective code.
Signed-off-by: Juergen Gross
---
xen/arch/x86/domctl.c | 2 +-
xen/arch/x86/hvm/hvm.c | 4 ++--
xen/arch/x86/hvm/vpt.c | 2 +-
xen/common/d
Use a bit mask for testing of a set bit instead of test_bit in case no
atomic operation is needed, as this will lead to smaller and more
effective code.
Signed-off-by: Juergen Gross
---
xen/common/sched_credit2.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git
On 10/01/2015 02:47 PM, Ian Campbell wrote:
On Fri, 2015-09-11 at 14:32 +0200, Juergen Gross wrote:
Add an own function to allocate the p2m list in the domain builder in
I'm not sure what you mean by "own" here. Perhaps "hook" or "callback"?
Before the p2m list allocation was done in the all
flight 62540 linux-4.1 real [real]
http://logs.test-lab.xenproject.org/osstest/logs/62540/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-i386-xl-qemut-stubdom-debianhvm-amd64-xsm 13 guest-localmigrate
fail REGR. vs. 62318
test-
This is done by introducing a new guest runvar, called
'vnodes', holding, if defined, the number of vNUMA nodes
the guest should have.
>From that, and basing on the other characteristics of the
guest (namely, amount of RAM and number of vcpus), we build
a vNUMA topology that makes sense.
For now,
as a variant of regular Debian HVM jobs.
For now, the vNUMA topology is just hardcoded
here, and it is 4 vCPUs and 2 vNUMA nodes. How
to split resources is left to the test
implementation (right now, it is just evenly
split them between the 2 virtual nodes).
Note that we do not yet support save/r
Hi,
This small series implements an HVM vNUMA test case.
It is all as simple as adding (and handling) a runvar, holding the number of
vNUMA nodes we want, to the regular debian HVM test case.
For now, the resulting topology used in the test is hardcoded. This means that,
as soon as the number of
This run is configured for baseline tests only.
flight 38113 ovmf real [real]
http://osstest.xs.citrite.net/~osstest/testlogs/logs/38113/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-i3865 xen-build
flight 38112 distros-debian-wheezy real [real]
http://osstest.xs.citrite.net/~osstest/testlogs/logs/38112/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-amd64-i386-wheezy-netboot-pygrub 6 xen-boot fail REGR. vs. 38027
baselin
Hi,
Version 2 of this patch contains a couple of minor syntax changes, and
a change to return immediately if the file being checked is not a shared
image file.
This version also fixes the issue George pointed out, and the code will
now properly differentiate between devices such as 1:1, 11:1 and
During the attachment of a loopback mounted image file, the mode of all
curent instances of this device already attached to other domains must be
checked. This requires finding all loopback devices pointing to the inode
of the shared image file, and then comparing the major and minor number of
thes
flight 62535 xen-4.6-testing real [real]
http://logs.test-lab.xenproject.org/osstest/logs/62535/
Failures :-/ but no regressions.
Regressions which are regarded as allowable (not blocking):
test-armhf-armhf-xl-rtds 15 guest-start.2fail blocked in 62461
test-amd64-amd64-xl-qemut-
On Thu, Oct 01, 2015 at 10:59:05AM -0600, Tamas K Lengyel wrote:
> * Hardware:
>
> Intel(R) Core(TM) i7-2600 CPU, DQ67SW motherboard
>
> * Software:
>
> Xen 4.6 rc4, Ubuntu 14.04 LTS, Kernel 4.2 with all Xen modules built-in
>
> * Guest operating systems:
>
> Windows 7 SP1 x86 and x64, Ubuntu
On 25/09/15 03:11, Chunyan Liu wrote:
> Add pvusb commands: usb-ctrl-attach, usb-ctrl-detach, usb-list,
> usb-attach and usb-detach.
>
> To attach a usb device to guest through pvusb, one could follow
> following example:
>
> #xl usb-ctrl-attach test_vm version=1 num_ports=8
So all the way back
* Hardware:
Intel(R) Core(TM) i7-2600 CPU, DQ67SW motherboard
* Software:
Xen 4.6 rc4, Ubuntu 14.04 LTS, Kernel 4.2 with all Xen modules built-in
* Guest operating systems:
Windows 7 SP1 x86 and x64, Ubuntu 14.04, Ubuntu 15.04
* Functionality tested:
xl save/restore
pvgrub2
VT-d device pass-
On 1 Oct 2015 17:07, "Ard Biesheuvel" wrote:
>
> On 1 October 2015 at 17:58, Julien Grall wrote:
> > Hi,
> >
> > We tried today to use the UEFI binary provided by Linaro for Xen [1] and
> > noticed the guest doesn't boot anymore.
> >
>
> Thanks for reporting. My LAVA job appears to have been offl
Andrew pointed me to this one and suggested it go into 4.6 because it's
prerequisite for a bug fix patch.
On Mon, Sep 21, 2015 at 08:03:22AM -0600, Jan Beulich wrote:
> ... instead of recalculating them.
>
> At once clean up formatting of the touched code and drop a stray loop
> local variable sh
On Wed, 30 Sep 2015, Stefano Stabellini wrote:
> Hi Fu,
>
> I backported your patches to the CentOS aarch64 grub2 rpm. I am testing
> it on X-Gene with EFI firmware. It works fine when booting Xen, but
> booting native kernels (no Xen) doesn't work anymore. Grub prints
>
> error: out of memory.
On 01/10/15 11:25, Jan Beulich wrote:
> Whether the MFN changes does not depend on the new entry being valid
> (but solely on the old one), and the need to update or TLB-flush also
> depends on permission changes.
>
> Signed-off-by: Jan Beulich
Reviewed-by: George Dunlap
> ---
> v2: Split from
flight 62599 xen-unstable-smoke real [real]
http://logs.test-lab.xenproject.org/osstest/logs/62599/
Failures :-/ but no regressions.
Tests which did not succeed, but are not blocking:
test-armhf-armhf-xl 12 migrate-support-checkfail never pass
test-armhf-armhf-xl 13
On 1 October 2015 at 17:58, Julien Grall wrote:
> Hi,
>
> We tried today to use the UEFI binary provided by Linaro for Xen [1] and
> noticed the guest doesn't boot anymore.
>
Thanks for reporting. My LAVA job appears to have been offline for a
while because the Xen mustang dom0 kernel has disappe
Hi,
We tried today to use the UEFI binary provided by Linaro for Xen [1] and
noticed the guest doesn't boot anymore.
My bisector fingered the commit 41f890164bc201f69841309c6b55e24c64121960
"ArmPkg/Mmu: Fix literal number left shift bug".
I've tried to use the DEBUG firmware but didn't get any o
On Thu, Oct 1, 2015 at 11:26 AM, Jan Beulich wrote:
> Now that p2m->get_entry() always returns a valid order, utilize this
> to accelerate some of the operations in PoD code. (There are two uses
> of p2m->get_entry() left which don't easily lend themselves to this
> optimization.)
>
> Also adjust
On 01/10/15 16:16, Mike Latimer wrote:
> Hi again,
>
> On Thursday, October 01, 2015 10:51:08 AM George Dunlap wrote:
>>>
>>> - if [ "$d" = "$devmm" ]
>>> + if [[ "$devmm" == *"$d,"* ]]
>>
>> Style nit: using [[ instead of [. TBH I prefer [[, but it's probably
>> better to be consistent
Hi again,
On Thursday, October 01, 2015 10:51:08 AM George Dunlap wrote:
> >
> > - if [ "$d" = "$devmm" ]
> > + if [[ "$devmm" == *"$d,"* ]]
>
> Style nit: using [[ instead of [. TBH I prefer [[, but it's probably
> better to be consistent with the rest of the file.
I was about to cha
On 30/09/15 11:45, Julien Grall wrote:
> Hi all,
>
> ARM64 Linux is supporting both 4KB and 64KB page granularity. Although, Xen
> hypercall interface and PV protocol are always based on 4KB page granularity.
>
> Any attempt to boot a Linux guest with 64KB pages enabled will result to a
> guest c
flight 62528 linux-3.18 real [real]
http://logs.test-lab.xenproject.org/osstest/logs/62528/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-amd64-xl-pvh-intel 11 guest-start fail REGR. vs. 58581
Tests which are failin
On Thu, Oct 01, 2015 at 04:25:01AM -0600, Jan Beulich wrote:
> Whether the MFN changes does not depend on the new entry being valid
> (but solely on the old one), and the need to update or TLB-flush also
> depends on permission changes.
>
> Signed-off-by: Jan Beulich
Release-acked-by: Wei Liu
On 01/10/15 15:36, Jan Beulich wrote:
On 01.10.15 at 15:34, wrote:
>> On 01/10/15 11:25, Jan Beulich wrote:
>>> @@ -645,11 +665,12 @@ p2m_pt_set_entry(struct p2m_domain *p2m,
>>> && (gfn + (1UL << page_order) - 1 > p2m->max_mapped_pfn) )
>>> p2m->max_mapped_pfn = gfn + (1UL
>>> On 01.10.15 at 15:34, wrote:
> On 01/10/15 11:25, Jan Beulich wrote:
>> @@ -645,11 +665,12 @@ p2m_pt_set_entry(struct p2m_domain *p2m,
>> && (gfn + (1UL << page_order) - 1 > p2m->max_mapped_pfn) )
>> p2m->max_mapped_pfn = gfn + (1UL << page_order) - 1;
>>
>> -if ( iomm
Hi George,
On Thursday, October 01, 2015 10:51:08 AM George Dunlap wrote:
> >then
> > -echo 'local'
> > +echo "local $d"
> > return
> >fi
> > fi
> > @@ -90,13 +107,13 @@ check_sharing()
> > do
> >d=$(xenstore_read_default "$base_path/$dom/
This run is configured for baseline tests only.
flight 38110 xen-unstable real [real]
http://osstest.xs.citrite.net/~osstest/testlogs/logs/38110/
Failures and problems with tests :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-i386-pair
On 01/10/15 11:25, Jan Beulich wrote:
> Whether the MFN changes does not depend on the new entry being valid
> (but solely on the old one), and the need to update or TLB-flush also
> depends on permission changes.
>
> Signed-off-by: Jan Beulich
> ---
> v2: Split from larger patch. Fix logic determ
On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
> Hi all,
>
> Only patch #7 is related to the subject of the cover letter. The rest is
> clean up of code I looked while I was working on this patch series.
>
> For all the changes, see in each patch.
>
> Sincerely yours,
>
> Julien Grall (
On Tue, 2015-09-29 at 15:44 +0100, Julien Grall wrote:
> Julien Grall (3):
> xen/arm: io: Extend write/read handler to pass private data
> xen/arm: vgic-v3: Correctly retrieve the vCPU associated to a
> re-distributor
> xen/arm: io: Shorten the name of the fields and clean up
All applie
On Thu, 2015-10-01 at 13:43 +0100, Julien Grall wrote:
> On 01/10/15 13:08, Ian Campbell wrote:
> > On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
> > > @@ -641,7 +643,30 @@ static int __init gicv2_init(void)
> > > panic("GICv2: Cannot find the maintenance IRQ");
> > > gicv2_i
On Fri, 2015-09-11 at 14:32 +0200, Juergen Gross wrote:
> Add an own function to allocate the p2m list in the domain builder in
I'm not sure what you mean by "own" here. Perhaps "hook" or "callback"?
> order to prepare allocating the p2m list outside of the initial kernel
> mapping. This will be
On 01/10/15 13:08, Ian Campbell wrote:
> On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
>> @@ -641,7 +643,30 @@ static int __init gicv2_init(void)
>> panic("GICv2: Cannot find the maintenance IRQ");
>> gicv2_info.maintenance_irq = res;
>>
>> -/* TODO: Add check on distri
On Tue, 2015-09-29 at 15:44 +0100, Julien Grall wrote:
> The field names in the IO emulation are really long and use repeatedly
> the term handler which make some line cumbersome to read:
>
> mmio_handler->mmio_handler_ops->write_handler
>
> Also take the opportunity to do some clean up:
> -
On 01/10/15 13:01, Ian Campbell wrote:
> On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
>> Xen is unconditionally using certain device tree paths to create DOM0
>> specific node (for instance /psci, /memory and /hypervisor).
>>
>> Print a warning message on the console to let the user know
On Tue, 2015-09-29 at 15:44 +0100, Julien Grall wrote:
> Some handlers may require to use private data in order to get quickly
> information related to the region emulated.
>
> Signed-off-by: Julien Grall
Acked-by: Ian Campbell
___
Xen-devel mailing
flight 62541 linux-mingo-tip-master real [real]
http://logs.test-lab.xenproject.org/osstest/logs/62541/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-i386-pvops 5 kernel-build fail REGR. vs. 60684
build-amd6
On Tue, 2015-09-29 at 16:59 +0100, Stefano Stabellini wrote:
> Free the memory used for the compressed kernel and update the relative
> mod->start and mod->size parameters with the uncompressed ones.
>
> To decompress the kernel, allocate memory from dommheap, because freeing
> the modules is done
On Tue, 2015-09-29 at 17:55 +0100, Julien Grall wrote:
> On 29/09/15 17:21, Julien Grall wrote:
> > All the quirks has been replaced by proper detection. Lets drop the
> > callback and hope that no one will need new quirks.
> >
> > At the same time, remove the definition platform_dom0_evtchn_ppi w
On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
> We are currently using a per-platform quirk to know if the 2 4KB region
> of
> the GIC CPU interface are each aligned to 64KB. Although, it may be
> possible to have different layout on a same platform (depending on the
> firmware version).
>
On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
> @@ -641,7 +643,30 @@ static int __init gicv2_init(void)
> panic("GICv2: Cannot find the maintenance IRQ");
> gicv2_info.maintenance_irq = res;
>
> -/* TODO: Add check on distributor, cpu size */
> +/* TODO: Add check o
On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
> Xen is unconditionally using certain device tree paths to create DOM0
> specific node (for instance /psci, /memory and /hypervisor).
>
> Print a warning message on the console to let the user know if we
> re-use one of these nodes.
>
> Note
On Thu, 2015-10-01 at 02:03 -0600, Jan Beulich wrote:
> > > > On 29.09.15 at 18:55, wrote:
> > In case of credit1, remove_vcpu() handling needs some
> > fixing remove_vcpu() too, i.e.:
> > - it manipulates runqueues, so the runqueue lock must
> >be acquired;
>
> Is this really a fix needed o
On Tue, 2015-09-29 at 17:21 +0100, Julien Grall wrote:
> The functions dt_n_*_cells return the number of cells for a "reg"
> property of a given node. So those numbers won't be correct if the
> parent of a given node is passed.
>
> This is fine today because the parent is always the root node whic
On 25/09/15 03:11, Chunyan Liu wrote:
> Add API for listing assignable USB devices info.
> Assignable USB device means the USB device type is assignable and
> it's not assigned to any guest yet.
>
> Signed-off-by: Chunyan Liu
> ---
> This could be squashed with previous patch. Split because there
On Thu, 2015-10-01 at 12:14 +0100, Ian Jackson wrote:
> The logic in try_fetch_logs for setting $ok was wrong. $ok would be
> set if we reached the end of any outer (pattern) loop iteration. If
> the host is actually dead all the pattern expansions would fail, but
> some of the patterns are liter
On 01/10/15 11:26, Jan Beulich wrote:
> Now that p2m->get_entry() always returns a valid order, utilize this
> to accelerate some of the operations in PoD code. (There are two uses
> of p2m->get_entry() left which don't easily lend themselves to this
> optimization.)
>
> Also adjust a few types as
The logic in try_fetch_logs for setting $ok was wrong. $ok would be
set if we reached the end of any outer (pattern) loop iteration. If
the host is actually dead all the pattern expansions would fail, but
some of the patterns are literals and do not need expansion. The
inner (logfile) loop would
On Thu, 2015-10-01 at 11:45 +0100, Ian Campbell wrote:
> On Tue, 2015-09-29 at 10:44 +0100, Ian Campbell wrote:
> > prepareguest has already assigned this so we should use it instead of
> > replicating (perhaps wrongly since target_guest_lv_name and
> > target_choose_vg can behave differently if mu
>>> On 01.10.15 at 12:35, wrote:
> On Thu, 2015-10-01 at 04:12 -0600, Jan Beulich wrote:
>> Okay. The thing is that looking at schedule_cpu_switch() alone
>> (and also considering its name) it is not clear that all callers
>> either
>> move the CPU into unusable state (from scheduling pov) or out
On Tue, 2015-09-29 at 10:44 +0100, Ian Campbell wrote:
> prepareguest has already assigned this so we should use it instead of
> replicating (perhaps wrongly since target_guest_lv_name and
> target_choose_vg can behave differently if multiple vgs are present).
This is wrong in practice because
On Thu, 2015-10-01 at 04:12 -0600, Jan Beulich wrote:
> Okay. The thing is that looking at schedule_cpu_switch() alone
> (and also considering its name) it is not clear that all callers
> either
> move the CPU into unusable state (from scheduling pov) or out
> of it, but never between CPUs from us
On Thu, 1 Oct 2015, Ian Campbell wrote:
> On Wed, 2015-09-09 at 11:31 +0100, Ian Campbell wrote:
> > On Thu, 2015-09-03 at 23:29 +, osstest service owner wrote:
> > > flight 61295 linux-linus real [real]
> > > http://logs.test-lab.xenproject.org/osstest/logs/61295/
> > >
> > > Regressions :-(
Now that p2m->get_entry() always returns a valid order, utilize this
to accelerate some of the operations in PoD code. (There are two uses
of p2m->get_entry() left which don't easily lend themselves to this
optimization.)
Also adjust a few types as needed and remove stale comments from
p2m_pod_cac
Whether the MFN changes does not depend on the new entry being valid
(but solely on the old one), and the need to update or TLB-flush also
depends on permission changes.
Signed-off-by: Jan Beulich
---
v2: Split from larger patch. Fix logic determining whether to update/
flush IOMMU mappings.
>>> On 01.10.15 at 11:26, wrote:
> On Thu, 2015-10-01 at 02:17 -0600, Jan Beulich wrote:
>> > > > On 29.09.15 at 18:55, wrote:
>> > This happens because, right now, the scheduler of the
>> > target pool remaps the runqueue lock during (rt_)alloc_pdata,
>> > which is called at the very beginning o
On Thu, Oct 1, 2015 at 10:51 AM, George Dunlap
wrote:
> CC'ing relevant maintainers, and someone who's worked with block
> scripts before...
>
> On Wed, Sep 30, 2015 at 10:10 PM, Mike Latimer wrote:
>> Create the list of shared loopback devices from within check_sharing, rather
>> than calling ch
CC'ing relevant maintainers, and someone who's worked with block
scripts before...
On Wed, Sep 30, 2015 at 10:10 PM, Mike Latimer wrote:
> Create the list of shared loopback devices from within check_sharing, rather
> than calling check_sharing for every loopback device using the same shared
> im
On Thu, 2015-10-01 at 10:32 +0100, Andrew Cooper wrote:
> On 01/10/15 08:43, Juergen Gross wrote:
> > On 10/01/2015 08:33 AM, Dario Faggioli wrote:
> > > On Thu, 2015-10-01 at 07:21 +0200, Juergen Gross wrote:
> > > #define SCHED_OP(opsptr, fn,
> > > ...) \
On 01/10/15 08:43, Juergen Gross wrote:
> On 10/01/2015 08:33 AM, Dario Faggioli wrote:
>> On Thu, 2015-10-01 at 07:21 +0200, Juergen Gross wrote:
>>> On 09/29/2015 06:55 PM, Dario Faggioli wrote:
>>
--- a/xen/common/schedule.c
+++ b/xen/common/schedule.c
@@ -1407,6 +1407,9 @@ static
On Thu, Oct 01, 2015 at 12:19:59PM +0530, Shivangi Dhir wrote:
> Hi,
>
> On Mon, Sep 28, 2015 at 9:40 PM, Ian Campbell
> wrote:
> >
> > I had a quick glance at coverity and found:
> >
> > CID1055958: use of mkstemp in libxl_vncviewer_exec without first setting a
> > sensible umask.
> >
> > CID131
On Thu, 2015-10-01 at 02:17 -0600, Jan Beulich wrote:
> > > > On 29.09.15 at 18:55, wrote:
> > This happens because, right now, the scheduler of the
> > target pool remaps the runqueue lock during (rt_)alloc_pdata,
> > which is called at the very beginning of schedule_cpu_switch().
> > Credit2's c
On Wed, 2015-09-30 at 11:14 -0400, Konrad Rzeszutek Wilk wrote:
> On Tue, Sep 29, 2015 at 11:01:44AM +0100, Lars Kurth wrote:
> > Hi,
> >
> > I agree to the conditions in the XenProject Coverity contribution
> > guidelines [1].
> >
> > I have been community manager for the Xen Project since 2011
Hi,
At 15:05 + on 30 Sep (1443625549), Xu, Quan wrote:
> >> >>> On September 29, 2015, at 5:12 PM, wrote:
> > So you'll need to do something else to make the unmap safe.
> >The usual
> > method in Xen is to hold a reference to the page (for read-only
> > mappings)
>
>
> Read-only mapping re
On Thu, 2015-10-01 at 00:44 -0600, Jan Beulich wrote:
> > > > On 30.09.15 at 19:58, wrote:
> > On 9/18/15 1:41 PM, Doug Goldstein wrote:
> > > This variable appears to be unused throughout the code base.
> > >
> > > Signed-off-by: Doug Goldstein
> > > ---
> > > Makefile | 3 +--
> > > 1 file ch
On Wed, 2015-09-09 at 11:31 +0100, Ian Campbell wrote:
> On Thu, 2015-09-03 at 23:29 +, osstest service owner wrote:
> > flight 61295 linux-linus real [real]
> > http://logs.test-lab.xenproject.org/osstest/logs/61295/
> >
> > Regressions :-(
> >
> > Tests which did not succeed and are blockin
On Wed, 2015-09-30 at 19:11 +0100, Julien Grall wrote:
> On 29/09/15 15:23, Ian Campbell wrote:
> > On Tue, 2015-09-29 at 14:36 +0100, Julien Grall wrote:
> > > On 29/09/15 14:07, Ian Campbell wrote:
> > > > On Fri, 2015-09-25 at 15:51 +0100, Julien Grall wrote:
> > > > > Xen is currently directly
On Thu, 2015-10-01 at 01:51 -0600, Jan Beulich wrote:
> > > > On 29.09.15 at 18:55, wrote:
> > --- a/xen/common/sched_credit2.c
> > +++ b/xen/common/sched_credit2.c
> > @@ -162,7 +162,7 @@
> > #define CSFLAG_runq_migrate_request
> > (1<<__CSFLAG_runq_migrate_request)
> >
> >
> > -int opt_migr
>>> On 29.09.15 at 18:55, wrote:
> This happens because, right now, the scheduler of the
> target pool remaps the runqueue lock during (rt_)alloc_pdata,
> which is called at the very beginning of schedule_cpu_switch().
> Credit2's csched2_free_pdata(), however, wants to find the spin
> lock the sa
On Thu, 2015-10-01 at 01:49 -0600, Jan Beulich wrote:
> > > > On 29.09.15 at 18:56, wrote:
> > @@ -1501,8 +1501,8 @@ int schedule_cpu_switch(unsigned int cpu,
> > struct cpupool *c)
> > return 0;
> >
> > idle = idle_vcpu[cpu];
> > -ppriv = SCHED_OP(new_ops, alloc_pdata, cpu);
>
>>> On 29.09.15 at 18:55, wrote:
> In case of credit1, remove_vcpu() handling needs some
> fixing remove_vcpu() too, i.e.:
> - it manipulates runqueues, so the runqueue lock must
>be acquired;
Is this really a fix needed only as a result of the insert side
changes? If not, since the insert s
>>> On 29.09.15 at 18:55, wrote:
> --- a/xen/common/sched_credit2.c
> +++ b/xen/common/sched_credit2.c
> @@ -162,7 +162,7 @@
> #define CSFLAG_runq_migrate_request (1<<__CSFLAG_runq_migrate_request)
>
>
> -int opt_migrate_resist=500;
> +static int __read_mostly opt_migrate_resist = 500;
> int
On 01/10/2015 07:49, Shivangi Dhir wrote:
Hi,
Hi Shivangi,
On Mon, Sep 28, 2015 at 9:40 PM, Ian Campbell mailto:ian.campb...@citrix.com>> wrote:
>
> I had a quick glance at coverity and found:
>
> CID1055958: use of mkstemp in libxl_vncviewer_exec without first
setting a
> sensible um
>>> On 29.09.15 at 18:56, wrote:
> @@ -1501,8 +1501,8 @@ int schedule_cpu_switch(unsigned int cpu, struct
> cpupool *c)
> return 0;
>
> idle = idle_vcpu[cpu];
> -ppriv = SCHED_OP(new_ops, alloc_pdata, cpu);
> -if ( ppriv == NULL )
> +if ( (new_ops->alloc_pdata != NULL)
flight 62525 qemu-mainline real [real]
http://logs.test-lab.xenproject.org/osstest/logs/62525/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-armhf 4 host-build-prep fail REGR. vs. 62339
test-armhf-armhf-xl
On 10/01/2015 09:23 AM, Dario Faggioli wrote:
On Thu, 2015-10-01 at 07:48 +0200, Juergen Gross wrote:
On 09/29/2015 06:56 PM, Dario Faggioli wrote:
In fact, credit2 uses CPU topology to decide how to arrange
its internal runqueues. Before this change, only 'one runqueue
per socket' was allowed.
On 10/01/2015 08:33 AM, Dario Faggioli wrote:
On Thu, 2015-10-01 at 07:21 +0200, Juergen Gross wrote:
On 09/29/2015 06:55 PM, Dario Faggioli wrote:
--- a/xen/common/schedule.c
+++ b/xen/common/schedule.c
@@ -1407,6 +1407,9 @@ static int cpu_schedule_callback(
switch ( action )
On Thu, 2015-10-01 at 07:48 +0200, Juergen Gross wrote:
> On 09/29/2015 06:56 PM, Dario Faggioli wrote:
> > In fact, credit2 uses CPU topology to decide how to arrange
> > its internal runqueues. Before this change, only 'one runqueue
> > per socket' was allowed. However, experiments have shown tha
* Andy Lutomirski wrote:
> > These could still be open coded in an inlined fashion, like the scheduler
> > usage.
>
> We could have a raw_rdmsr for those.
>
> OTOH, I'm still not 100% convinced that this warn-but-don't-die behavior is
> worth the effort. This isn't a frequent source of bugs
96 matches
Mail list logo
