On 31.07.25 17:32, Jan Beulich wrote:
On 31.07.2025 17:16, Dmytro Prokopchuk1 wrote:
--- a/xen/arch/arm/guestcopy.c
+++ b/xen/arch/arm/guestcopy.c
@@ -109,27 +109,31 @@ static unsigned long copy_guest(void *buf, uint64_t addr,
unsigned int len,
unsigned long raw_copy_to_guest(void *to, con
On 31.07.25 11:25, Grygorii Strashko wrote:
Hi Juergen, Dario, George,
Sorry that disturbing you.
Would it be possible to take a look at this patch as it
has change in common/sched/core.c?
On 18.07.25 13:11, Grygorii Strashko wrote:
From: Grygorii Strashko
On platforms without PIRQ support e
On 31.07.25 11:10, Teddy Astie wrote:
do_sched_op(SCHEDOP_yield) just calls vcpu_yield(). Remove the indirection
through the hypercall handler and use the function directly.
Export vcpu_block_enable_events() to perform the same for SCHEDOP_block.
Not a functional change.
Signed-off-by: Teddy A
On 30.07.25 18:21, Anthony PERARD wrote:
On Fri, Jul 25, 2025 at 03:19:28PM +0200, Juergen Gross wrote:
diff --git a/tools/libs/light/libxl_dom.c b/tools/libs/light/libxl_dom.c
index a61085ca3b..2a7923533f 100644
--- a/tools/libs/light/libxl_dom.c
+++ b/tools/libs/light/libxl_dom.c
@@ -494,6 +49
On 30.07.25 16:38, Jan Beulich wrote:
On 30.07.2025 14:22, Juergen Gross wrote:
Mini-OS doesn't support using mmap() for accessing a file. In order
to support reading the live update state from a 9pfs based file, use
fread() instead of mmap().
While adding the offset member to lu_state, change
On 29.07.25 21:25, Andrew Cooper wrote:
On 29/07/2025 8:21 pm, Andrew Cooper wrote:
On 29/07/2025 12:01 pm, Juergen Gross wrote:
diff --git a/tools/xenstored/lu.c b/tools/xenstored/lu.c
index 77e0d377c5..f2c8b92d07 100644
--- a/tools/xenstored/lu.c
+++ b/tools/xenstored/lu.c
@@ -27,9 +27,11 @@
On 29.07.25 19:00, Anthony PERARD wrote:
On Tue, Jul 29, 2025 at 01:01:41PM +0200, Juergen Gross wrote:
In order to reactivate an event channel after kexec() of Mini-OS,
libxenevtchn needs to allocate the port data for the event channel
and set the handler again. Add a new interface xenevtchn_bi
On 29.07.25 11:00, Jan Beulich wrote:
On 29.07.2025 10:38, Juergen Gross wrote:
--- a/arch/x86/x86_64.S
+++ b/arch/x86/x86_64.S
@@ -33,13 +33,8 @@ _start:
stack_start:
.quad stack+(2*__STACK_SIZE)
-.globl shared_info, hypercall_page
-/* Unpleasant -- the PTE that maps thi
On 29.07.25 10:52, Jan Beulich wrote:
On 29.07.2025 10:38, Juergen Gross wrote:
--- a/arch/x86/mm.c
+++ b/arch/x86/mm.c
@@ -640,13 +640,20 @@ void change_readonly(bool readonly)
/*
* return a valid PTE for a given virtual address. If PTE does not exist,
* allocate page-table pages.
+ * P
On 28.07.25 16:21, Jan Beulich wrote:
On 08.07.2025 08:37, Juergen Gross wrote:
--- a/hypervisor.c
+++ b/hypervisor.c
@@ -27,8 +27,10 @@
#include
#include
+#include
#include
#include
+#include
#include
EXPORT_SYMBOL(hypercall_page);
@@ -37,7 +39,8 @@ EXPORT_SYMBOL(hype
On 28.07.25 16:09, Jan Beulich wrote:
On 08.07.2025 08:37, Juergen Gross wrote:
Add a small pool of statically allocated memory pages to be handed out
for very early page table allocations.
This will make it possible to do virtual allocations e.g. for mapping
the shared info page.
Signed-off-b
On 26.07.25 01:58, Jason Andryuk wrote:
posix.c and minios.c implement the same named functions serving slightly
different purposes.
For xenbus_map()
posix.c maps the local /dev/xen/xsd_kva
minios.c maps dom0 via grant and there is open coding for stub_domid in
map_interface.
Change xenbus_map(
On 26.07.25 01:58, Jason Andryuk wrote:
The dom0_domid variable is misnamed and conflates purposes. If we have
xenstored running in a Linux domain that is not dom0, this variable
controls the lookup of /proc/xen/xsd_kva and the event channel.
Rename to store_domid to better show its purpose.
O
On 26.07.25 01:58, Jason Andryuk wrote:
Usually, priv_domid == dom0_domid == 0, and that is what is expected.
If we rename s/dom0_domid/store_domid/, it seems more likely we want to
actually have the priv_domid as the owner.
That leads to follow on changes to ensure that the priv_domid is create
On 26.07.25 01:58, Jason Andryuk wrote:
Usually, priv_domid == dom0_domid == 0, and that is what is expected.
If we rename s/dom0_domid/store_domid/, it seems more likely we want to
actually have the priv_domid as the owner.
That leads to follow on changes to ensure that the priv_domid is create
On 26.07.25 01:58, Jason Andryuk wrote:
Replace dom0_init() with init_domains() which uses libxenmanage to
iterate through all existing domains, storing them in a list. The xenstore
domain is introduced first, and then all the other domains are
introduced. The xenstore domain needs to be introd
On 26.07.25 01:58, Jason Andryuk wrote:
Add helpers to lookup the event channel for a domid. This hides some
of the differences between dom0 and stubdom xenstored. Each version
defines its own.
It highlights the different meanings between get_xenbus_evtchn() in a
stubdom, where it looks up dom
On 25.07.25 12:12, Teddy Astie wrote:
xenmem_reservation_increase currently calls XENMEM_populate_physmap.
Rename the function to avoid confusion with XENMEM_increase_reservation.
Hmm, I'm not convinced this is a good move.
xenmem_reservation_increase() has the counterpart xenmem_reservation_d
On 21.06.25 16:14, Koichiro Den wrote:
When a running unit is about to be scheduled out due to a competing unit
with the highest remaining credit, the residual credit of the previous
unit is currently ignored in csched2_runtime() because it hasn't yet
been reinserted into the runqueue.
As a resu
On 25.07.25 04:28, Jason Andryuk wrote:
Update get_implicit_path to return the correct value for a non-dom0
xenstored domain.
Signed-off-by: Jason Andryuk
Reviewed-by: Juergen Gross
Juergen
OpenPGP_0xB0DE9DD628BF132F.asc
Description: OpenPGP public key
OpenPGP_signature.asc
Description
On 25.07.25 04:28, Jason Andryuk wrote:
Rename the helper that access store_domid accordingly.
While doing this, rename the is_store_domain variable.
No functional change.
Signed-off-by: Jason Andryuk
With my remark on the previous patch this one can just be dropped.
Juergen
OpenPGP_0xB
On 25.07.25 04:28, Jason Andryuk wrote:
The dom0_domid variable is misnamed and conflates purposes. If we have
xenstored running in a Linux domain that is not dom0, this variable
controls the lookup of /proc/xen/xsd_kva and the event channel.
One implication of this change is that the xenstore
On 25.07.25 04:28, Jason Andryuk wrote:
Usually, priv_domid == dom0_domid == 0, and that is what is expected.
If we rename s/dom0_domid/store_domid/, it seems more likely we want to
actually have the priv_domid as the owner.
Yes, I agree.
That leads to follow on changes to ensure that the pr
On 25.07.25 04:28, Jason Andryuk wrote:
posix.c and minios.c implement the same named functions serving slightly
different purposes.
For xenbus_map()
posix.c maps the local /dev/xen/xsd_kva
minios.c maps dom0 via grant and there is open coding for stub_domid in
map_interface.
Change xenbus_map(
On 25.07.25 04:28, Jason Andryuk wrote:
Replace dom0_init() with init_domains() which uses libxenmanage to
iterate through all existing domains, storing them in a list. The xenstore
domain is introduced first, and then all the other domains are
introduced. The xenstore domain needs to be introd
On 25.07.25 04:28, Jason Andryuk wrote:
Add a helper to lookup the event channel for a domid. This hides some
of the differences between dom0 and stubdom xenstored.
It highlights the different meanings between get_xenbus_evtchn() in a
stubdom, where it looks up dom0's event channel, and dom0, w
On 25.07.25 04:28, Jason Andryuk wrote:
Make introduce_domain() use an event channel from the the xenstore page.
It is only used if non-zero. Otherwise the passed in event channel port
is used.
The is useful for a xenstored stubdom to configure domains autonomously.
Signed-off-by: Jason Andryu
On 25.07.25 00:48, Jason Andryuk wrote:
On 2025-07-22 10:06, Juergen Gross wrote:
Add per domain server features, which are initialized by the supported
features at domain introduction, or by live update from the migration
stream. This requires to add the DOMAIN_DATA record to the migration
st
On 23.07.25 09:04, Jan Beulich wrote:
On 23.07.2025 08:55, Jürgen Groß wrote:
On 23.07.25 08:43, Jan Beulich wrote:
On 23.07.2025 08:34, Jürgen Groß wrote:
On 23.07.25 08:28, Jan Beulich wrote:
On 22.07.2025 02:19, Jason Andryuk wrote:
--- a/xen/common/domain.c
+++ b/xen/common/domain.c
On 23.07.25 08:43, Jan Beulich wrote:
On 23.07.2025 08:34, Jürgen Groß wrote:
On 23.07.25 08:28, Jan Beulich wrote:
On 22.07.2025 02:19, Jason Andryuk wrote:
--- a/xen/common/domain.c
+++ b/xen/common/domain.c
@@ -195,6 +195,14 @@ static void set_domain_state_info(struct
On 23.07.25 08:25, Jan Beulich wrote:
On 22.07.2025 14:09, Jason Andryuk wrote:
On 2025-07-21 20:19, Jason Andryuk wrote:
Expose a domain's capabilities - control, hardware or xenstore - through
stable get domain state hypercall.
The xenstore domain can use this information to assign appropria
On 23.07.25 08:28, Jan Beulich wrote:
On 22.07.2025 02:19, Jason Andryuk wrote:
--- a/xen/common/domain.c
+++ b/xen/common/domain.c
@@ -195,6 +195,14 @@ static void set_domain_state_info(struct
xen_domctl_get_domain_state *info,
info->state |= XEN_DOMCTL_GETDOMSTATE_STATE_DYING;
On 21.07.25 11:49, Bernhard Kaindl wrote:
Update vcpu_runstate_get() to return a snapshot of the accumulated
non-affine vCPU running time at the current time of this call.
We cannot change the struct vcpu_runstate_info: It is part of the
Guest shared memory area that is part of the frozen VM ABI
On 21.07.25 11:49, Bernhard Kaindl wrote:
To monitor the effectiveness of vCPU soft-affinity on NUMA hosts,
we'd like to create a vCPU metric that accumulates the amount of
vCPU time running outside of the soft affinity mask of the sched-unit:
- Add a new time counter, nonaffine_time to struct v
On 28.06.25 20:17, Julien Grall wrote:
Hi Mykola,
On 27/06/2025 11:51, Mykola Kvach wrote:
diff --git a/xen/arch/arm/include/asm/perfc_defn.h b/xen/arch/arm/include/asm/
perfc_defn.h
index effd25b69e..8dfcac7e3b 100644
--- a/xen/arch/arm/include/asm/perfc_defn.h
+++ b/xen/arch/arm/include/asm/
On 15.07.25 15:44, WangYuli wrote:
There is a spelling mistake of 'notifer' in the comment which
should be 'notifier'.
Link:
https://lore.kernel.org/all/b3c019b63c93846f+20250715071245.398846-1-wangy...@uniontech.com/
Signed-off-by: WangYuli
Reviewed-by: Juergen Gross
Juergen
OpenPGP_0x
On 17.07.25 16:29, Jakub Kicinski wrote:
On Tue, 15 Jul 2025 16:11:29 + Anthoine Bourgeois wrote:
Fixes: b27d47950e48 ("xen/netfront: harden netfront against event channel
storms")
Not entirely sure who you expect to apply this patch, but if networking
then I wouldn't classify this is a f
On 17.07.25 23:39, Jason Andryuk wrote:
On 2025-07-17 04:50, Juergen Gross wrote:
On 16.07.25 23:15, Jason Andryuk wrote:
Replace dom0_init() with init_domains() which uses libxenmanage to
iterate through all existing domains and introduce them.
dom0_domid is updated with the xenstore domain,
On 17.07.25 22:23, Jason Andryuk wrote:
On 2025-07-17 02:28, Juergen Gross wrote:
On 16.07.25 23:14, Jason Andryuk wrote:
Add an additional "caps" argument to the libxenmanage functions to
obtain a domains capabilities - control, hardware, and xenstore.
Update the xenstored callers at the same
On 17.07.25 22:16, Jason Andryuk wrote:
On 2025-07-17 02:24, Juergen Gross wrote:
On 16.07.25 23:14, Jason Andryuk wrote:
Expose a domain's capabilities - control, hardware or xenstore - through
stable get domain state hypercall.
The xenstore domain can use this information to assign appropria
On 17.07.25 14:04, Jan Beulich wrote:
On 17.07.2025 13:58, Jürgen Groß wrote:
On 17.07.25 13:53, Jan Beulich wrote:
For the error patch to not see it uninitialized.
Fixes: 560c2a5584c4 ("mini-os: kexec: do the final kexec step")
Signed-off-by: Jan Beulich
Reviewed-by: Jue
On 17.07.25 13:53, Jan Beulich wrote:
For the error patch to not see it uninitialized.
Fixes: 560c2a5584c4 ("mini-os: kexec: do the final kexec step")
Signed-off-by: Jan Beulich
Reviewed-by: Juergen Gross
Juergen
OpenPGP_0xB0DE9DD628BF132F.asc
Description: OpenPGP public key
OpenPGP_si
On 17.07.25 08:39, Jan Beulich wrote:
On 02.07.2025 10:12, Juergen Gross wrote:
With this series applied I was able to perform Live Update of Xenstore
running in a PVH stubdom.
This series is based on a related Mini-OS patch series AND it needs
the Mini-OS commit id updated accordingly in Confi
On 17.07.25 08:25, Jan Beulich wrote:
On 17.07.2025 08:19, Juergen Gross wrote:
On 16.07.25 23:14, Jason Andryuk wrote:
Write the associated event channel into the xenstore page so xenstored
can read it. xenstored can map the grant by the reserved grant table
entry, and then read out the event
On 14.07.25 16:02, Jan Beulich wrote:
On 11.07.2025 14:03, Juergen Gross wrote:
e820_get_reserved_pfns() needs to make sure to reserve only page
aligned memory areas.
Fixes: a7d27a7332ba ("mini-os: add memory map service functions")
Signed-off-by: Juergen Gross
I've committed this, but ...
On 13.07.25 15:26, li...@treblig.org wrote:
From: "Dr. David Alan Gilbert"
Remove three uncalled functions:
xenbus_mkdir() was added in 2007 by
commit 4bac07c993d0 ("xen: add the Xenbus sysfs and virtual device hotplug
driver")
but has remained unused.
xen_get_runstate_snapshot() last u
On 10.07.25 18:11, Anthoine Bourgeois wrote:
We found at Vates that there are lot of spurious interrupts when
benchmarking the PV drivers of Xen. This issue appeared with a patch
that addresses security issue XSA-391 (see Fixes below). On an iperf
benchmark, spurious interrupts can represent up t
On 11.07.25 11:29, Teddy Astie wrote:
You also probably want to send this to linux kernel mailing list too.
Le 10/07/2025 à 18:14, Anthoine Bourgeois a écrit :
We found at Vates that there are lot of spurious interrupts when
benchmarking the PV drivers of Xen. This issue appeared with a patch
t
On 11.07.25 08:24, Jan Beulich wrote:
On 02.07.2025 10:12, Juergen Gross wrote:
Add support for a Mini-OS config file containing config items with
numerical values, e.g. CONFIG_FOO=7.
Signed-off-by: Juergen Gross
Reviewed-by: Jan Beulich
Albeit I'm puzzled by the use in the next patch. Bes
On 11.07.25 01:21, Jason Andryuk wrote:
On 2025-07-02 04:12, Juergen Gross wrote:
Especially for support of Xenstore-stubdom live update some memory must
be handed over to the new kernel without moving it around: as the
9pfs device used for storing and retrieving the state of Xenstore
needs to b
On 11.07.25 03:03, Stefano Stabellini wrote:
On Wed, 9 Jul 2025, Jürgen Groß wrote:
On 08.07.25 21:01, Stefano Stabellini wrote:
On Thu, 3 Jul 2025, Juergen Gross wrote:
When compiling the kernel with LLVM, the following warning was issued:
drivers/xen/gntdev.c:991: warning: stack frame
On 08.07.25 23:14, Jason Andryuk wrote:
On 2025-07-02 04:12, Juergen Gross wrote:
When initializing the memory allocator, don't add memory pages of
modules and the initial boot info structure to the free memory.
This is relevant only when running in PVH mode, as in PV mode only
memory above the
On 08.07.25 21:01, Stefano Stabellini wrote:
On Thu, 3 Jul 2025, Juergen Gross wrote:
When compiling the kernel with LLVM, the following warning was issued:
drivers/xen/gntdev.c:991: warning: stack frame size (1160) exceeds
limit (1024) in function 'gntdev_ioctl'
The main reason is struc
On 03.07.25 00:42, Tu Dinh wrote:
On 01/07/2025 23:53, Abinash wrote:
Hi ,
Thanks for pointing that out.
I haven’t measured the performance impact yet — my main focus was on
getting rid of the stack usage warning triggered by LLVM due to
inlining. But you're right, gntdev_ioctl_grant_copy() is
On 02.07.25 11:56, Roger Pau Monne wrote:
As a result of the addition of a new field in libxl domain build info
structure the golang types need to be regnerated, this was missing as part
of 22650d6054.
Regenerate the headers now.
Reported-by: Juergen Gross
Fixes: 22650d605462 ('x86/hvmloader:
On 30.06.25 09:57, Jan Beulich wrote:
On 26.06.2025 17:13, Juergen Gross wrote:
With the related change having been added to Mini-OS, the single V1
patch can now be expanded to a 3 patch series, especially meant to
untangle the close dependencies between grub-pv and Mini-OS internals.
Changes i
On 25.06.25 16:30, Andrew Cooper wrote:
On 25/06/2025 3:28 pm, Jürgen Groß wrote:
On 25.06.25 16:20, Jan Beulich wrote:
On 25.06.2025 13:08, Juergen Gross wrote:
@@ -432,3 +431,10 @@ out:
allocated = 0;
xc_interface_close(xc_handle );
}
+
+static void call_start_info_hook
On 25.06.25 16:20, Jan Beulich wrote:
On 25.06.2025 13:08, Juergen Gross wrote:
@@ -432,3 +431,10 @@ out:
allocated = 0;
xc_interface_close(xc_handle );
}
+
+static void call_start_info_hook(struct xc_dom_image *dom)
+{
+#undef start_info
+if ( dom->arch_hooks->start_info )
+
On 24.06.25 16:51, Andriy Sultanov wrote:
Currently, as far as I am aware, the ability of xenstore clients to properly
handle and detect batch updates is somewhat lacking. Transactions are not
directly visible to the clients watching a particular directory - they will
receive a lot of individual
On 23.06.25 10:45, Jan Beulich wrote:
On 19.06.2025 13:37, Juergen Gross wrote:
While working on xenstore-stubdom live update support I came across
some really unpleasant code using struct start_info even when running
in PVH mode.
Especially suspend/resume was doing some crazy stuff, including
On 20.06.25 02:41, jiang.pe...@zte.com.cn wrote:
From: Peng Jiang
This patch fixes a W=1 format-string warning reported by GCC 12.3.0
by annotating xenbus_switch_fatal() and xenbus_va_dev_error()
with the __printf attribute. The attribute enables compile-time
validation of printf-style format s
On 18.06.25 04:01, jiang.pe...@zte.com.cn wrote:
From: Peng Jiang
This patch fixes a W=1 format-string warning reported by GCC 12.3.0
by annotating xenbus_switch_fatal() and xenbus_va_dev_error()
with the __printf attribute. The attribute enables compile-time
validation of printf-style form
On 14.06.25 19:39, Jason Andryuk wrote:
On Fri, Mar 21, 2025 at 5:30 AM Juergen Gross wrote:
With all kexec preparations done, activate the new kernel.
Signed-off-by: Juergen Gross
---
kexec.c | 17 +++--
1 file changed, 15 insertions(+), 2 deletions(-)
diff --git a/kexec.c b
On 14.06.25 19:34, Jason Andryuk wrote:
On Fri, Mar 21, 2025 at 5:32 AM Juergen Gross wrote:
Add a kexec_call() macro which will provide the capability to register
a function for being called when doing a kexec() call. The called
functions will be called with a boolean parameter "undo" indicat
On 14.06.25 19:02, Jason Andryuk wrote:
On Fri, Mar 21, 2025 at 5:30 AM Juergen Gross wrote:
Build the parameters for the new kernel, consisting of the
hvm_start_info struct, the memory map and the command line.
Signed-off-by: Juergen Gross
@@ -212,4 +213,61 @@ void kexec_set_param_loc(co
On 14.06.25 18:41, Jason Andryuk wrote:
On Fri, Mar 21, 2025 at 5:25 AM Juergen Gross wrote:
Analyze the properties of the new kernel to be loaded by kexec. The
data needed is:
- upper boundary in final location
- copy and memory clear operations
- entry point and entry parameter
Signed-off-
On 14.06.25 18:40, Jason Andryuk wrote:
On Fri, Mar 21, 2025 at 5:25 AM Juergen Gross wrote:
Add the code and data definitions of the final kexec stage.
Put the code and related data into a dedicated section in order to be
able to copy it to another location. For this reason there must be no
Jason,
thanks for having a look at the series! I very much appreciate that!
On 14.06.25 18:40, Jason Andryuk wrote:
On Fri, Mar 21, 2025 at 5:25 AM Juergen Gross wrote:
Add a new config option CONFIG_KEXEC for support of kexec-ing into a
new mini-os kernel. Add a related kexec.c source and a
On 13.06.25 09:31, Xin Li wrote:
On 6/11/2025 5:58 AM, Juergen Gross wrote:
Here is a patch I cooked. I added an ALTERNATIVE() hack because the new
instructions can't be more than 6 bytes long. But with the patch you
just sent, it shouldn't be needed.
I have meanwhile dropped the patch copy
On 10.06.25 00:43, Chuck Zmudzinski wrote:
Hi,
I am seeing the following regression between Linux 6.14.8 and 6.15.1.
Kernel version 6.14.8 boots fine but version 6.15.1 crashes and
reboots on Xen. I don't know if 6.14.9 or 6.14.10 is affected, or
if 6.15 or the 6.15 release candidates are affec
On 03.06.25 13:18, Roger Pau Monne wrote:
The Xen PCI device (vendor ID 0x5853) exposed to x86 HVM guests doesn't
have the functionality of a traditional PCI device. The exposed MIO BAR is
s/MIO/MMIO/ ?
used by some guests (including Linux) as a safe place to map foreign
memory, including th
On 19.05.25 15:30, Anthony PERARD wrote:
On Tue, Apr 29, 2025 at 01:06:32PM +0200, Juergen Gross wrote:
diff --git a/tools/libacpi/mk_dsdt.c b/tools/libacpi/mk_dsdt.c
index 34f6753f61..227b5ceafb 100644
--- a/tools/libacpi/mk_dsdt.c
+++ b/tools/libacpi/mk_dsdt.c
@@ -105,7 +81,7 @@ int main(int a
On 17.05.25 18:57, Xin Li (Intel) wrote:
xen_read_msr_safe() currently passes an uninitialized argument err to
xen_do_read_msr(). But as xen_do_read_msr() may not set the argument,
xen_read_msr_safe() could return err with an unpredictable value.
To ensure correctness, initialize err to 0 (repr
On 15.05.25 20:11, Xin Li wrote:
On 5/15/2025 8:29 AM, Ingo Molnar wrote:
* Xin Li (Intel) wrote:
xen_read_msr_safe() currently passes an uninitialized argument err to
xen_do_read_msr(). But as xen_do_read_msr() may not set the argument,
xen_read_msr_safe() could return err with an unpredic
On 16.05.25 11:31, Teddy Astie wrote:
Hello,
This series introduce support for confidential computing along with a
AMD SEV implementation. It also bundles some of the functional
requirements (ASID scheme, ABI, ...) which could be separated if needed.
(I bundled everything in this serie to have
On 12.05.25 23:54, Stefano Stabellini wrote:
From: Stefano Stabellini
dm_op hypercalls might come from userspace and pass memory addresses as
parameters. The memory addresses typically correspond to buffers
allocated in userspace to hold extra hypercall parameters.
On ARM, when CONFIG_ARM64_SW
On 14.05.25 10:04, Roger Pau Monne wrote:
When adding extra memory regions as ballooned pages also adjust the balloon
target, otherwise when the balloon driver is started it will populate
memory to match the target value and consume all the extra memory regions
added.
This made the usage of the
On 14.05.25 11:20, Roger Pau Monne wrote:
PVH dom0 is useless without XEN_UNPOPULATED_ALLOC, as otherwise it will
very likely balloon out all dom0 memory to map foreign and grant pages.
Enable it by default as part of xen.config. This also requires enabling
MEMORY_HOTREMOVE and ZONE_DEVICE.
Si
On 13.05.25 07:55, Xin Li wrote:
On 5/12/2025 4:24 AM, Juergen Gross wrote:
Now with the mentioned patch really attached. :-)
Does it allow patching with an instruction more than 6 bytes long?
The immediate form MSR instructions are 9 bytes long.
Yes, shouldn't be a problem.
Juergen
Op
On 09.05.25 10:18, Xin Li wrote:
On 5/6/2025 2:20 AM, Juergen Gross wrote:
Instead of having callback functions for rdmsr/wrmsr on native, switch
to inline the respective instructions directly in order to avoid
overhead with the call interface.
To me, this is a beneficial addition to the exist
On 06.05.25 23:09, Jason Andryuk wrote:
Marek reported seeing a NULL pointer fault in the xenbus_thread
callstack:
BUG: kernel NULL pointer dereference, address:
RIP: e030:__wake_up_common+0x4c/0x180
Call Trace:
__wake_up_common_lock+0x82/0xd0
process_msg+0x18e/0x2f0
xen
On 30.04.25 16:29, Jason Andryuk wrote:
On 2025-04-30 06:56, Marek Marczykowski-Górecki wrote:
On Tue, Apr 29, 2025 at 08:59:45PM -0400, Jason Andryuk wrote:
Hi Marek,
On Wed, Apr 23, 2025 at 8:42 AM Marek Marczykowski-Górecki
wrote:
I've got some more report confirming it's still happenin
On 29.04.25 14:01, Jan Beulich wrote:
On 29.04.2025 13:06, Juergen Gross wrote:
--- a/tools/firmware/hvmloader/Makefile
+++ b/tools/firmware/hvmloader/Makefile
@@ -59,6 +59,7 @@ OBJS += optionroms.o 32bitbios_support.o rombios.o
CFLAGS += -DENABLE_ROMBIOS
ROMBIOS_ROM := $(ROMBIOS_DIR)/BIOS-b
On 28.04.25 17:31, Alejandro Vallejo wrote:
On Mon Apr 28, 2025 at 4:15 PM BST, Jürgen Groß wrote:
On 28.04.25 17:12, Alejandro Vallejo wrote:
On Thu Apr 24, 2025 at 10:41 AM BST, Anthony PERARD wrote:
On Tue, Apr 08, 2025 at 02:35:20PM +0200, Juergen Gross wrote:
@@ -1903,10 +1894,7 @@ it
On 28.04.25 17:12, Alejandro Vallejo wrote:
On Thu Apr 24, 2025 at 10:41 AM BST, Anthony PERARD wrote:
On Tue, Apr 08, 2025 at 02:35:20PM +0200, Juergen Gross wrote:
@@ -1903,10 +1894,7 @@ it may be useful to request a different one, like UEFI.
=item B
-Loads ROMBIOS, a 16-bit x86 compa
On 24.04.25 16:43, Anthony PERARD wrote:
On Tue, Apr 08, 2025 at 02:35:22PM +0200, Juergen Gross wrote:
diff --git a/INSTALL b/INSTALL
index 88c1464816..786e3da996 100644
--- a/INSTALL
+++ b/INSTALL
There's one "--enable-ioemu-stubdom" left in this file.
Oh, thanks for noticing.
diff --g
On 24.04.25 11:41, Anthony PERARD wrote:
On Tue, Apr 08, 2025 at 02:35:20PM +0200, Juergen Gross wrote:
diff --git a/docs/man/xl.cfg.5.pod.in b/docs/man/xl.cfg.5.pod.in
index 7339c44efd..ccf0c58895 100644
--- a/docs/man/xl.cfg.5.pod.in
+++ b/docs/man/xl.cfg.5.pod.in
@@ -895,12 +895,6 @@ is used.
On 25.04.25 14:33, Peter Zijlstra wrote:
On Wed, Apr 23, 2025 at 06:05:19PM +0200, Jürgen Groß wrote:
It's not a major change, but when it is patched to use the immediate
form MSR write instruction, it's straightforwardly streamlined.
It should be rather easy to switch the cur
On 25.04.25 10:34, Xin Li (Intel) wrote:
As pmu_msr_{read,write}() are now wrappers of pmu_msr_chk_emulated(),
remove them and use pmu_msr_chk_emulated() directly.
As pmu_msr_chk_emulated() could easily return false in the cases where
it would set *emul to false, remove the "emul" argument and u
On 25.04.25 10:34, Xin Li (Intel) wrote:
hpa found that pmu_msr_write() is actually a completely pointless
function [1]: all it does is shuffle some arguments, then calls
pmu_msr_chk_emulated() and if it returns true AND the emulated flag
is clear then does *exactly the same thing* that the calli
On 25.04.25 05:44, H. Peter Anvin wrote:
On 4/24/25 18:15, H. Peter Anvin wrote:
On 4/24/25 01:14, Jürgen Groß wrote:
Actually, that is how we get this patch with the existing alternatives
infrastructure. And we took a step further to also remove the pv_ops
MSR APIs...
And this is what I&#
On 25.04.25 03:15, H. Peter Anvin wrote:
On 4/24/25 01:14, Jürgen Groß wrote:
Actually, that is how we get this patch with the existing alternatives
infrastructure. And we took a step further to also remove the pv_ops
MSR APIs...
And this is what I'm questioning. IMHO this approa
On 22.04.25 10:21, Xin Li (Intel) wrote:
An MSR value is represented as a 64-bit unsigned integer, with existing
MSR instructions storing it in EDX:EAX as two 32-bit segments.
The new immediate form MSR instructions, however, utilize a 64-bit
general-purpose register to store the MSR value. To
On 22.04.25 10:21, Xin Li (Intel) wrote:
set_reg() is used to write the following MSRs on Xen:
MSR_FS_BASE
MSR_KERNEL_GS_BASE
MSR_GS_BASE
But none of these MSRs are written using any MSR write safe API.
Therefore there is no need to pass an error pointer argument to
set_reg() for
On 22.04.25 10:21, Xin Li (Intel) wrote:
As pmu_msr_{read,write}() are now wrappers of pmu_msr_chk_emulated(),
remove them and use pmu_msr_chk_emulated() directly.
While at it, convert the data type of MSR index to u32 in functions
called in pmu_msr_chk_emulated().
Suggested-by: H. Peter Anvin
On 24.04.25 10:06, Xin Li wrote:
On 4/23/2025 9:05 AM, Jürgen Groß wrote:
It's not a major change, but when it is patched to use the immediate form MSR
write instruction, it's straightforwardly streamlined.
It should be rather easy to switch the current wrmsr/rdmsr paravirt patching
On 23.04.25 11:03, Xin Li wrote:
On 4/22/2025 4:12 AM, Jürgen Groß wrote:
+
+static __always_inline bool __rdmsrq(u32 msr, u64 *val, int type)
+{
+ bool ret;
+
+#ifdef CONFIG_XEN_PV
+ if (cpu_feature_enabled(X86_FEATURE_XENPV))
+ return __xenpv_rdmsrq(msr, val, type);
I don&#
On 23.04.25 10:51, Xin Li wrote:
On 4/22/2025 2:57 AM, Jürgen Groß wrote:
On 22.04.25 10:22, Xin Li (Intel) wrote:
The story started from tglx's reply in [1]:
For actual performance relevant code the current PV ops mechanics
are a horrorshow when the op defaults to the n
On 22.04.25 10:22, Xin Li (Intel) wrote:
To eliminate the indirect call overhead introduced by the pv_ops API,
utilize the alternatives mechanism to read MSR:
1) When built with !CONFIG_XEN_PV, X86_FEATURE_XENPV becomes a
disabled feature, preventing the Xen code from being built
On 22.04.25 11:20, Xin Li wrote:
On 4/22/2025 1:59 AM, Jürgen Groß wrote:
On 22.04.25 10:22, Xin Li (Intel) wrote:
To eliminate the indirect call overhead introduced by the pv_ops API,
utilize the alternatives mechanism to read MSR:
1) When built with !CONFIG_XEN_PV, X86_FEATURE_XENPV
1 - 100 of 1554 matches
Mail list logo
