Hi Boris,
On 11/03/2017 04:28 AM, Boris Ostrovsky wrote:
> On 11/01/2017 09:19 PM, Dongli Zhang wrote:
>> Hi Boris,
>>
>> I have received from l...@intel.com that the prior version of patch hit issue
>> during compilation with aarch64-linux-gnu-gcc. I think this patch r
-conversion]
runstate_delta = kmalloc_array(num_possible_cpus(),
^
-
Thank you very much!
Dongli Zhang
On 11/02/2017 03:19 AM, Boris Ostrovsky wrote:
> On 10/31/2017 09:46 PM, Dongli Zhang wrote:
>> After g
low steal time and lead to 100% st usage in top command
for linux 4.8-4.10. A backport of this patch would fix that issue.
References:
https://0xstubs.org/debugging-a-flaky-cpu-steal-time-counter-on-a-paravirtualized-xen-guest
Signed-off-by: Dongli Zhang
---
Changed since v1:
* relocate modificat
Hi Boris,
On 10/31/2017 08:58 AM, Boris Ostrovsky wrote:
>
>
> On 10/30/2017 08:14 PM, Dongli Zhang wrote:
>> Hi Boris,
>>
>> On 10/30/2017 09:34 PM, Boris Ostrovsky wrote:
>>> On 10/30/2017 04:03 AM, Dongli Zhang wrote:
>>>> After guest
Hi Boris,
On 10/30/2017 09:34 PM, Boris Ostrovsky wrote:
> On 10/30/2017 04:03 AM, Dongli Zhang wrote:
>> After guest live migration on xen, steal time in /proc/stat
>> (cpustat[CPUTIME_STEAL]) might decrease because steal returned by
>> xen_steal_lock() might
g-a-flaky-cpu-steal-time-counter-on-a-paravirtualized-xen-guest
Signed-off-by: Dongli Zhang
---
Changed since v1:
* relocate modification to xen_get_runstate_snapshot_cpu
Changed since v2:
* accumulate runstate times before live migration
Changed since v3:
* do not accumulate times in
g-a-flaky-cpu-steal-time-counter-on-a-paravirtualized-xen-guest
Signed-off-by: Dongli Zhang
---
Changed since v1:
* relocate modification to xen_get_runstate_snapshot_cpu
Changed since v2:
* accumulate runstate times before live migration
Changed since v3:
* do not accumulate times in
Hi Juergen,
On 10/27/2017 03:31 PM, Juergen Gross wrote:
> On 27/10/17 09:16, Dongli Zhang wrote:
>> Hi Boris,
>>
>> On 10/25/2017 11:12 PM, Boris Ostrovsky wrote:
>>> On 10/25/2017 02:45 AM, Dongli Zhang wrote:
>>>> After guest live migration on x
Hi Boris,
On 10/25/2017 11:12 PM, Boris Ostrovsky wrote:
> On 10/25/2017 02:45 AM, Dongli Zhang wrote:
>> After guest live migration on xen, steal time in /proc/stat
>> (cpustat[CPUTIME_STEAL]) might decrease because steal returned by
>> xen_steal_lock() might
g-a-flaky-cpu-steal-time-counter-on-a-paravirtualized-xen-guest
Signed-off-by: Dongli Zhang
---
Changed since v1:
* relocate modification to xen_get_runstate_snapshot_cpu
Changed since v2:
* accumulate runstate times before live migration
---
drivers/xen/manage.c | 1 +
drivers/xen/tim
Hi Boris,
- boris.ostrov...@oracle.com wrote:
> On 10/19/2017 04:02 AM, Dongli Zhang wrote:
> > After guest live migration on xen, steal time in /proc/stat
> > (cpustat[CPUTIME_STEAL]) might decrease because steal returned by
> > xen_steal_lock() might be less than this_
ravirtualized-xen-guest
Signed-off-by: Dongli Zhang
---
drivers/xen/time.c | 15 ++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/drivers/xen/time.c b/drivers/xen/time.c
index ac5f23f..2b3a996 100644
--- a/drivers/xen/time.c
+++ b/drivers/xen/time.c
@@ -19,6 +19,8 @@
t;>> if (unlikely(steal <= this_rq()->prev_steal_time))
>>> return 0;
>>
>> How about you just fix up paravirt_steal_time() on migration and not
>> muck with the users ?
>
> Not just migration, either. CPU hotplug is another time to fix up
Hi Stanislaw and Peter,
On 10/10/2017 08:42 PM, Stanislaw Gruszka wrote:
> On Tue, Oct 10, 2017 at 12:59:26PM +0200, Ingo Molnar wrote:
>>
>> (Cc:-ed more gents involved in kernel/sched/cputime.c work. Full patch
>> quoted
>> below.)
>>
>> * Dongl
10, the issue for
linux 4.11+ would only decrease but not overflow steal time after live
migration.
References:
https://0xstubs.org/debugging-a-flaky-cpu-steal-time-counter-on-a-paravirtualized-xen-guest
Signed-off-by: Dongli Zhang
---
kernel/sched/cputime.c | 13 ++---
1 file changed,
ased after live migration.
Dongli Zhang
- Original Message -
From: dario.faggi...@citrix.com
To: xen.l...@daevel.fr, xen-us...@lists.xensource.com
Cc: xen-devel@lists.xen.org
Sent: Tuesday, October 3, 2017 5:24:49 PM GMT +08:00 Beijing / Chongqing / Hong
Kong / Urumqi
Subject: Re: [Xen-
On 08/04/2017 05:13 PM, Roger Pau Monné wrote:
> On Fri, Aug 04, 2017 at 10:00:09AM +0200, Valentin Vidic wrote:
>> On Mon, Jul 31, 2017 at 09:09:19AM +0800, Dongli Zhang wrote:
>>> To verify whether the above patch would help, please check the
>>> nr_grant_frames
usage for arbitrary guest domU (including dom0)
./xen-diag gnttab_query_size [domid]
4. If your host's xen toolstack does not have xen-diag, feel free to implement
one your self via GNTTABOP_query_size hypercall and compile with -lxenctrl.
Dongli Zhang
On 07/31/2017 02:30 PM, Valentin
CCed xen-devel so that more people would be able to help.
Dongli Zhang
On 07/31/2017 09:09 AM, Dongli Zhang wrote:
> Hi Valentin,
>
> On 07/30/2017 03:42 PM, Valentin Vidic wrote:
>> I'm having a problem with a domU hang in disk IO, described here:
>>
>> https:/
you read about the book "Definitive Guide to Xen Hypervisor" and read
about linux paravirtual driver code.
Dongli Zhang
>
>
> > I gone-through kernel code and but its tittle bit confusing to me.
>
> >>Which bits do you
Hi Konrad,
In addition to Junxiao's patch on xen-blkfront, would you please help merge this
to mainline as well?
Thank you very much!
Dongli Zhang
On 06/28/2017 11:34 PM, Roger Pau Monné wrote:
> On Wed, Jun 28, 2017 at 08:57:28PM +0800, Dongli Zhang wrote:
>> This patch par
Signed-off-by: Dongli Zhang
---
.gitignore | 1 +
1 file changed, 1 insertion(+)
diff --git a/.gitignore b/.gitignore
index 068f430..e1009e0 100644
--- a/.gitignore
+++ b/.gitignore
@@ -211,6 +211,7 @@ tools/misc/xenperf
tools/misc/xenpm
tools/misc/xen-hvmctx
tools/misc/xenlockprof
+tools
Hi Wei,
I will send the patch based on staging.
Dongli Zhang
On 07/04/2017 10:10 PM, Wei Liu wrote:
> I pushed your two patches and discovered you also need to patch
> .gitignore. Could you please send a patch for that? thanks
>
> ___
This patch adds new interface for GNTTABOP_query_size in libxc to help
query the current grant table frames and maximum grant table frames for a
specific domain.
Signed-off-by: Dongli Zhang
---
Changed since v1:
* Change %d to %u in ERROR()
---
tools/libxc/include/xenctrl.h | 1 +
tools
commands for more diagnostic functions and
the framework of xen-diag.c is from xen-livepatch.c.
Signed-off-by: Dongli Zhang
---
Changed since v1:
* rewrite xen-gnttab-query.c to xen-diag.c based on livepatch.c framework
---
tools/misc/Makefile | 4 ++
tools/misc/xen-diag.c | 129
.
This utility would help the administrators to monitor the guest grant table
frame usage on dom0 side so that it is not required to debug on guest
kernel side for crash/hang analysis anymore.
Signed-off-by: Dongli Zhang
---
tools/misc/Makefile | 4
tools/misc/xen-gnttab-query.c
This patch adds new interface for GNTTABOP_query_size in libxc to help
query the current grant table frames and maximum grant table frames for a
specific domain.
Signed-off-by: Dongli Zhang
---
tools/libxc/include/xenctrl.h | 1 +
tools/libxc/xc_gnttab.c | 12
2 files
/ring.
As persistent grants management was per-queue since 73716df ("xen/blkfront:
make persistent grants pool per-queue"), we should always allocate from
persistent grants first.
Signed-off-by: Dongli Zhang
---
Changed since v1:
* use "max_grefs - rinfo->persistent_
/ring.
As persistent grants management was per-queue since 73716df ("xen/blkfront:
make persistent grants pool per-queue"), we should always allocate from
persistent grants first.
Signed-off-by: Dongli Zhang
---
drivers/block/xen-blkfront.c | 17 ++---
1 file changed,
2. Implement new SMCs from hypervsiror to TEE to track VM lifecycle.
> 3. Track VM IDs to isolated VM data.
> 4. RPCs to sleeping guests.
>
> That is basically all what I currently have on my mind. It will be
> great if all interested sides will share their
IS_ERR_VALUE() in commit 87557efc27f6a50140fb20df06a917f368ce3c66
("xen-netfront: do not cast grant table reference to signed short") would
not return true for error code unless we cast ref first to type int.
Signed-off-by: Dongli Zhang
---
drivers/net/xen-netfront.c | 4 ++--
1 file
end a new patch or an incremental
based on previous one at
https://git.kernel.org/cgit/linux/kernel/git/davem/net.git?
Thank you very much!
Dongli Zhang
- Original Message -
From: da...@davemloft.net
To: dongli.zh...@oracle.com
Cc: linux-ker...@vger.kernel.org, xen
ef won't get
> stored, preventing its later use from causing further damage).
The reason I use warning instead BUG_ON is that Linus suggested use
WARN_ON_ONCE() in a previous email:
http://lkml.iu.edu/hypermail/linux/kernel/1610.0/00878.html
I would change to BUG_ON() if it is OK to you.
Th
While grant reference is of type uint32_t, xen-netfront erroneously casts
it to signed short in BUG_ON().
This would lead to the xen domU panic during boot-up or migration when it
is attached with lots of paravirtual devices.
Signed-off-by: Dongli Zhang
---
drivers/net/xen-netfront.c | 4
live migrated).
Thank you very much!
Dongli Zhang
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
> > This patch implemented parts of TODO left in commit id
> > a902c12ee45fc9389eb8fe54eeddaf267a555c58 (More efficient TLB-flush
> > filtering in alloc_heap_pages()). It moved TLB-flush filtering out into
> > populate_physmap. Because of TLB-flush in alloc_heap_pages, it's very slow
> > to create
efficient TLB-flush filtering in alloc_heap_pages()).
"#include " is removed from xen/arch/x86/acpi/suspend.c to
avoid the compiling error after we include "" to
xen/include/xen/mm.h.
Signed-off-by: Dongli Zhang
---
Changed since v5:
* Move the if() and its body of tlbflu
has ever got unpaused by hypervisor.
MEMF_no_tlbflush can be set only during vm creation phase when
creation_finished is still false before this domain gets unpaused for the
first time.
Signed-off-by: Dongli Zhang
---
Changed since v5:
* Remove conditional check before "d->creation_f
has ever got unpaused by hypervisor.
MEMF_no_tlbflush can be set only during vm creation phase when
creation_finished is still false before this domain gets unpaused for the
first time.
Signed-off-by: Dongli Zhang
---
Changed since v4:
* Rename is_ever_unpaused to creation_finished.
* Change b
efficient TLB-flush filtering in alloc_heap_pages()).
"#include " is removed from xen/arch/x86/acpi/suspend.c to
avoid the compiling error after we include "" to
xen/include/xen/mm.h.
Signed-off-by: Dongli Zhang
---
Changed since v4:
* Wrap the filtered tlbflush mask operatio
e_physmap during vm creation
On Fri, Sep 16, 2016 at 03:47:23AM -0700, Dongli Zhang wrote:
> > > +/*
> > > + * MEMF_no_tlbflush can be set only during vm creation phase when
> > > + * is_ever_unpaused is still false before this domai
he hypervisor? In
general IMHO we should avoid providing an interface that is possible to
create a security problem.
==
Hi Wei, since it is possible to have a safety catch now in the hypervisor (the
bit is allowed only before VM creation is finished), i
*_va, struct page_info *page);
+#include
+
+static inline bool accumulate_tlbflush(bool need_tlbflush,
+ const struct page_info *page,
+ uint32_t tlbflush_timestamp)
+{
+return page->u.free.need_tlbflush &&
+ page->tlbflush_timestamp <= tlbflush_current_time() &&
+ (!need_tlbflush ||
+page->tlbflush_timestamp > tlbflush_timestamp);
+}
+
#endif /* __XEN_MM_H__ */
Thank you very much!
Dongli Zhang
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
_tlbflush, page, tlbflush_timestamp) \
+(page)->u.free.need_tlbflush && \
+(page)->tlbflush_timestamp <= tlbflush_current_time() && \
+(!need_tlbflush || \
+ (page)->tlbflush_timestamp > tlbflush_timestamp)
Thank you very much!
Dongli Zhang
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
can be set only during vm creation phase when
is_ever_unpaused is still false before this domain gets unpaused for the
first time.
Signed-off-by: Dongli Zhang
---
Changed since v3:
* Set the flag to true in domain_unpause_by_systemcontroller when
unpausing the guest domain for the first time.
This patch cleaned up the code by replacing complicated tlbflush check with
an inline function. We should use this inline function to avoid the long
and complicated to read tlbflush check when implementing TODOs left in
commit a902c12ee45fc9389eb8fe54eeddaf267a555c58.
Signed-off-by: Dongli Zhang
e of code updating the field is in function "schedule":
1411prev->last_run_time = now;
I am afraid if there would be any existing (I did not find one) or future
interfaces that can reset the entire vcpu[0] to 0.
Dongli Zhang
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
can be set only during vm creation phase when
already_scheduled is still 0 before this domain gets scheduled for the
first time.
TODO: ballooning very huge amount of memory cannot benefit from this patch
and might still be slow.
Signed-off-by: Dongli Zhang
---
Changed since v2:
* Limit this optimi
= MEMF_no_tlbflush;
int claim_enabled = dom->claim_enabled;
uint64_t total_pages;
xen_vmemrange_t dummy_vmemrange[2];
Dongli Zhang
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
This patch introduced a "MEMF_no_tlbflush" bit to memflag to indicate
whether TLB-flush should be done in alloc_heap_pages or its caller
populate_physmap. Once this bit is set in memflag, alloc_heap_pages will
ignore TLB-flush.
Signed-off-by: Dongli Zhang
---
xen/common/me
This patch introduced a "MEMF_no_tlbflush" bit to memflag to indicate
whether TLB-flush should be done in alloc_heap_pages or its caller
populate_physmap. Once this bit is set in memflag, alloc_heap_pages will
ignore TLB-flush.
Signed-off-by: Dongli Zhang
---
xen/common/me
> I suggest pasting in Olaf's exact error message here.
>
> To avoid extra round trip, I propose updating the commit message as
> followed:
>
> Initialise j to 0 to make some versions of gcc (e.g., gcc4.5/4.3)
> happy to
> avoid compilation error by commit
> beba3693f7243e68bbe31fe3794da
Initialise j to 0 to make some versions of gcc (e.g., gcc4.5/4.3) happy to
avoid compilation error by commit beba3693f7243e68bbe31fe3794da91068eeea5b.
Signed-off-by: Dongli Zhang
---
tools/misc/xen-livepatch.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/misc/xen
Local variable "j" would be used only when "i == ARRAY_SIZE(main_options)"
is true. Thus, it is not necessary to update "j" when "i ==
ARRAY_SIZE(main_options)" is false.
Signed-off-by: Dongli Zhang
---
tools/misc/xen-livepatch.c | 5 ++---
1 file change
Local variable "j" would be used only when "i == ARRAY_SIZE(main_options)"
is true. Thus, it is not necessary to update "j" when "i ==
ARRAY_SIZE(main_options)" is false.
Signed-off-by: Dongli Zhang
Reviewed-by: Konrad Rzeszutek Wilk
---
tools/misc/xe
my_old_func" and thus we have two symbols
referring the same name but different addresses now (82d0802465a4 and
82d080409070).
Are we going to use new nm symbol flag , append extra string in symbol name
(e.g., my_old_func#livepatch) or this even
Local variable "j" would be used only when "i == ARRAY_SIZE(main_options)"
is true. Thus, it is not necessary to update "j" when "i ==
ARRAY_SIZE(main_options)" is false.
Signed-off-by: Dongli Zhang
---
tools/misc/xen-xsplice.c | 3 +--
1 file changed, 1
Replace realpath with readlink since '-m' option is not supported by realpath.
Signed-off-by: Dongli Zhang
---
xsplice-build | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/xsplice-build b/xsplice-build
index 2eb7993..5852186 100755
--- a/xsplice-build
+++
Since this is the new git repo, I am not sure if I could submit
this patch to xen-devel if a patch is helpful.
Thank you very much!
Best,
Dongli Zhang
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
uld redirect to domain specific variable (e.g.,
domain->max_grant_frames) now.
Please let me know the feedback. If this is reasonable, I will work on patches.
Thank you very much!
Dongli Zhang
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
er names. I cannot easily tell whether
> it makes assumptions on settings done elsewhere in my systems,
> so don't put too high hopes into being able to use it as is.
>
> Jan
>
>
> ___
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
>
>
--
Dongli Zhang (张东立)
finallyjustice.github.io
___
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
it is
better to initialize them to 0 at the beginning.
Signed-off-by: Dongli Zhang
---
drivers/net/xen-netback/netback.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/xen-netback/netback.c
b/drivers/net/xen-netback/netback.c
index 61b97c3..32f0fbd 100644
--- a/drivers/net/xen-n
patch for xen-netfront to support feature-multicast-control?
Otherwise, I might need to make it myself.
Thank you very much!
Best,
Dongli Zhang
http://finallyjustice.github.io
___
Xen-devel mailing list
63 matches
Mail list logo
