ping?
On 01/17/18 16:13 +0800, Haozhong Zhang wrote:
> v3 is based on Eduardo's machine-next tree. Changes of
> memory-backend-file are based on my previous patches in that tree.
>
>
> Linux 4.15 introduces a new mmap flag MAP_SYNC, which can be used to
> guarantee t
On 01/18/18 19:20 -0200, Eduardo Habkost wrote:
> On Thu, Jan 18, 2018 at 03:22:05PM +, Peter Maydell wrote:
> > On 18 January 2018 at 02:09, Eduardo Habkost wrote:
> > > The following changes since commit
> > > 8e5dc9ba49743b46d955ec7dacb04e42ae7ada7c:
> > >
> > > Merge remote-tracking bra
When there are multiple memory backends in use, including the object type
name, ID and the property name in the error message can help users to
locate the error.
Signed-off-by: Haozhong Zhang
Suggested-by: "Dr. David Alan Gilbert"
---
backends/hostmem-file.c | 9 ++---
backends
Tsirkin)
* Move OS dependent definitions of MAP_SYNC and MAP_SHARED_VALIDATE
to osdep.h. (Michael S. Tsirkin)
Haozhong Zhang (3):
util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
hostmem: add hostmem type name and id in error messages
hostmem-file: add 'sync' option
back
x27;share=on', work as if
'sync=on'; otherwise, work as if 'sync=off'
Signed-off-by: Haozhong Zhang
Suggested-by: Eduardo Habkost
---
backends/hostmem-file.c | 40 +++-
docs/nvdimm.txt | 15 ++-
exec.c
- If sync == ON_OFF_AUTO_AUTO, and
* if the host OS and the backend file support MAP_SYNC, and MAP_SYNC
is not conflict with other flags, qemu_ram_mmap() will work as if
sync == ON_OFF_AUTO_ON.
* otherwise, qemu_ram_mmap() will work as if sync == ON_OFF_AUTO_OFF.
Signed-off-by: Haozhong Zhang
-
On 01/11/18 20:06 +, Dr. David Alan Gilbert wrote:
> * Haozhong Zhang (haozhong.zh...@intel.com) wrote:
> > This option controls whether QEMU mmap(2) the memory backend file with
> > MAP_SYNC flag, which can fully guarantee the guest write persistence
> > to the backend
- If sync == ON_OFF_AUTO_AUTO, and
* if the host OS and the backend file support MAP_SYNC, and MAP_SYNC
is not conflict with other flags, qemu_ram_mmap() will work as if
sync == ON_OFF_AUTO_ON.
* otherwise, qemu_ram_mmap() will work as if sync == ON_OFF_AUTO_OFF.
Signed-off-by: Haozhong Zhang
-
* Add 'sync' option to control the use of MAP_SYNC. (Eduardo Habkost)
* Remove the unnecessary set of MAP_SHARED_VALIDATE in some cases and
the retry mechanism in qemu_ram_mmap(). (Michael S. Tsirkin)
* Move OS dependent definitions of MAP_SYNC and MAP_SHARED_VALIDATE
to osdep.
x27;share=on', work as if
'sync=on'; otherwise, work as if 'sync=off'
Signed-off-by: Haozhong Zhang
Sugguested-by: Eduardo Habkost
---
backends/hostmem-file.c | 39 ++-
docs/nvdimm.txt | 15 ++-
exec.c
On 01/09/18 17:37 -0200, Marcelo Tosatti wrote:
>
> Improve hugepage allocation failure message, indicating
> whats happening to the user.
>
> Signed-off-by: Marcelo Tosatti
>
> diff --git a/exec.c b/exec.c
> index 4722e521d4..439abedb98 100644
> --- a/exec.c
> +++ b/exec.c
> @@ -1643,7 +1643,8
On 01/03/18 11:45 -0200, Eduardo Habkost wrote:
> On Wed, Jan 03, 2018 at 11:16:39AM +0800, Haozhong Zhang wrote:
> > On 01/02/18 18:02 +0200, Michael S. Tsirkin wrote:
> > > On Wed, Dec 27, 2017 at 02:56:20PM +0800, Haozhong Zhang wrote:
> > > > When a file sup
On 01/02/18 18:02 +0200, Michael S. Tsirkin wrote:
> On Wed, Dec 27, 2017 at 02:56:20PM +0800, Haozhong Zhang wrote:
> > When a file supporting DAX is used as vNVDIMM backend, mmap it with
> > MAP_SYNC flag in addition can guarantee the persistence of guest write
> > to the
On 12/31/17 17:55 +0200, Michael S. Tsirkin wrote:
> On Mon, Dec 25, 2017 at 09:06:10AM +0800, Haozhong Zhang wrote:
> > The new function pmem_persistent() flushes the previous cached writes
> > on the specified memory buffer, which ensures the write persistence if
> > the bu
MAP_SYNC fails if MAP_SYNC is not supported by the kernel or the
backend file. On such failures, QEMU retries mmap without MAP_SYNC and
MAP_SHARED_VALIDATE.
Signed-off-by: Haozhong Zhang
---
util/mmap-alloc.c | 36 +++-
1 file changed, 35 insertions(+), 1 deletion
The new function pmem_persistent() flushes the previous cached writes
on the specified memory buffer, which ensures the write persistence if
the buffer is in persistent memory.
Signed-off-by: Haozhong Zhang
---
include/qemu/pmem.h | 25 ++
util/Makefile.objs | 1 +
util/pmem.c
QEMU intercepts guest writes to vNVDIMM labels, and then stores them
to the backend. When the backend is a real NVDIMM device, QEMU needs
to ensure the write persistence before returning to guest, so that the
guest labels will not be lost if QEMU exits abnormally.
Haozhong Zhang (2):
util/pmem
QEMU intercepts guest writes to vNVDIMM labels, and then stores them
to the backend. When the backend is a real NVDIMM device, QEMU needs
to ensure the write persistence before returning to guest, so that the
guest labels will not be lost if QEMU exits abnormally.
Signed-off-by: Haozhong Zhang
it's disabled.
Signed-off-by: Haozhong Zhang
---
hw/i386/pc.c | 18 ++
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 3fcf318a95..55686bf5d8 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -1695,9 +1695,14 @@ static
Signed-off-by: Haozhong Zhang
---
hw/i386/pc_piix.c| 15 ---
hw/i386/pc_q35.c | 13 +++--
include/hw/i386/pc.h | 3 +++
3 files changed, 26 insertions(+), 5 deletions(-)
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 5e47528993..f235ee12c2 100644
--- a/hw
CPUID_7_0_EBX_CLFLUSHOPT is missed in current "Skylake-Server" cpu
model. Add it to "Skylake-Server" cpu model on pc-i440fx-2.12 and
pc-q35-2.12. Keep it disabled in "Skylake-Server" cpu model on older
machine types.
Signed-off-by: Haozhong Zhang
---
include/hw/i3
Changes in v3:
* Rebase on QEMU 2.12.
* Add 2.12 PC machine types. Denis V. Lunev has already posted a
similar patch [1]. If that one is merged first, please ignore patch 1
in this patch series.
[1] https://lists.nongnu.org/archive/html/qemu-devel/2017-12/msg02961.html
Haozhong Zhang (2
CPUID_7_0_EBX_CLFLUSHOPT is missed in current "Skylake-Server" cpu
model. Add it to "Skylake-Server" cpu model on pc-i440fx-2.11 and
pc-q35-2.11. Keep it disabled in "Skylake-Server" cpu model on older
machine types.
Signed-off-by: Haozhong Zhang
-
The only difference from the existing CPU model "Skylake-Server" is
the add of CPUID_7_0_EBX_CLFLUSHOPT, which is missed in "Skylake-Server".
Signed-off-by: Haozhong Zhang
---
target/i386/cpu.c | 49 +
1 file changed, 49 inser
to 'memory-backend-file', so that users or
management utils, which have enough knowledge about the backend, can
specify a proper alignment via this option.
Signed-off-by: Haozhong Zhang
---
backends/hostmem-file.c | 41 -
docs/nvd
ice read-only if the unarmed flag in guest NFIT is set.
The default value of 'unarmed' option is 'off' in order to keep the
backwards compatibility.
Signed-off-by: Haozhong Zhang
---
docs/nvdimm.txt | 15 +++
hw/acpi/nvdimm.c|
Signed-off-by: Haozhong Zhang
Reviewed-by: Stefan Hajnoczi
---
hw/mem/nvdimm.c | 2 +-
include/hw/mem/nvdimm.h | 3 +++
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/hw/mem/nvdimm.c b/hw/mem/nvdimm.c
index 952fce5ec8..618c3d677b 100644
--- a/hw/mem/nvdimm.c
+++ b/hw/mem
device dax check.
* Document new options added by this patch series.
Haozhong Zhang (3):
hostmem-file: add "align" option
nvdimm: add a macro for property "label-size"
nvdimm: add 'unarmed' option
backends/hostmem-file.c | 41 ++
hvmloader memory in DSM emulation,
it has to take a different path, i.e., xen_copy_{from,to}_guest().
Signed-off-by: Haozhong Zhang
---
Cc: Xiao Guangrong
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
---
hw/acpi/nvdimm.c | 44 +---
1 file c
When QEMU is used the device model of Xen HVM domain and vNVDIMM
devices are present, enable building ACPI tables related to vNVDIMM.
Signed-off-by: Haozhong Zhang
---
Cc: Stefano Stabellini
Cc: Anthony Perard
Cc: "Michael S. Tsirkin"
Cc: Paolo Bonzini
Cc: Richard Henderson
C
Signed-off-by: Haozhong Zhang
---
Cc: Stefano Stabellini
Cc: Anthony Perard
Cc: Paolo Bonzini
Cc: Richard Henderson
Cc: Eduardo Habkost
Cc: "Michael S. Tsirkin"
---
hw/i386/xen/xen-hvm.c | 55 +++
include/hw/xen/xen.h | 3 +++
is issue, we add a 32-bit reserved field after
func_ret_status and always fill it with 0. Therefore, the minimum
length of ODAT in both ACPI 1.0 and ACPI 2.0 is always 8 bytes, so no
extra bytes will be added accidentally by the implicit conversion.
Signed-off-by: Haozhong Zhang
---
Cc: Xiao Guangrong
QEMU on KVM/TCG and Xen requires different sets of guest ACPI tables.
When QEMU builds ACPI for Xen HVM domains, the new Xen-specific ACPI
build function xen_acpi_build() is called instead of the existing path
from acpi_build().
Signed-off-by: Haozhong Zhang
---
Cc: Stefano Stabellini
Cc
x1:
-object memory-backend-xen,id=mem1,host-addr=0x1,size=4G
-device nvdimm,id=nvdimm1,memdev=mem1
Signed-off-by: Haozhong Zhang
---
Cc: Eduardo Habkost
Cc: Igor Mammedov
Cc: "Michael S. Tsirkin"
---
backends/Makefile.objs | 1 +
backends
QEMU to labels is led by a label size check,
let's not intiailize nvdimm->label_data if the label size is 0.
Signed-off-by: Haozhong Zhang
---
Cc: Xiao Guangrong
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
---
hw/mem/nvdimm.c | 10 +-
1 file changed, 9 insertions(+), 1
the Xen side support for labels is not implemented yet.
Patch 5-10 enable building ACPI tables and passing them to Xen HVM
domains.
Haozhong Zhang (10):
[01/10] xen-hvm: remove a trailing space
[02/10] xen-hvm: create the hotplug memory region on Xen
[03/10] hostmem-xen: add a host memory
Signed-off-by: Haozhong Zhang
---
Cc: Stefano Stabellini
Cc: Anthony Perard
Cc: Paolo Bonzini
Cc: Richard Henderson
Cc: Eduardo Habkost
Cc: "Michael S. Tsirkin"
---
hw/i386/xen/xen-hvm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/i386/xen/xen-hvm.c b/h
built by QEMU.
Signed-off-by: Haozhong Zhang
---
Cc: Stefano Stabellini
Cc: Anthony Perard
Cc: "Michael S. Tsirkin"
Cc: Paolo Bonzini
Cc: Richard Henderson
Cc: Eduardo Habkost
---
hw/i386/xen/xen-hvm.c | 12
1 file changed, 12 insertions(+)
diff --git a/hw/i386/xen/xen-
.
Signed-off-by: Haozhong Zhang
---
Cc: "Michael S. Tsirkin"
Cc: Paolo Bonzini
Cc: Richard Henderson
Cc: Eduardo Habkost
Cc: Stefano Stabellini
Cc: Anthony Perard
---
hw/i386/pc.c | 86 ---
hw/i386/xen/xen-hvm.c | 2 ++
include/hw
On 12/01/17 10:44 +, Stefan Hajnoczi wrote:
> On Mon, Nov 27, 2017 at 12:35:17PM +0800, Haozhong Zhang wrote:
> > @@ -312,6 +315,10 @@ nvdimm_build_structure_memdev(GArray *structures,
> > DeviceState *dev)
> >
> > /* Only one interleave for PM
[ it looks my email client had some glitch and failed to add mst in cc list ]
copy mst
On 11/27/17 12:35 +0800, Haozhong Zhang wrote:
> When mmap(2) the backend files, QEMU uses the host page size
> (getpagesize(2)) by default as the alignment of mapping address.
> However, some bac
On 11/27/17 23:07 -0200, Eduardo Habkost wrote:
> On Mon, Nov 27, 2017 at 12:35:15PM +0800, Haozhong Zhang wrote:
> > When mmap(2) the backend files, QEMU uses the host page size
> > (getpagesize(2)) by default as the alignment of mapping address.
> > However, some backends m
copy mst
On 11/27/17 12:35 +0800, Haozhong Zhang wrote:
> When mmap(2) the backend files, QEMU uses the host page size
> (getpagesize(2)) by default as the alignment of mapping address.
> However, some backends may require alignments different than the page
> size. For example, mmap
ice read-only if the unarmed flag in guest NFIT is set.
The default value of 'unarmed' option is 'off' in order to keep the
backwards compatibility.
Signed-off-by: Haozhong Zhang
---
docs/nvdimm.txt | 15 +++
hw/acpi/nvdimm.c|
to 'memory-backend-file', so that users or
management utils, which have enough knowledge about the backend, can
specify a proper alignment via this option.
Signed-off-by: Haozhong Zhang
---
backends/hostmem-file.c | 41 -
docs/nvd
Signed-off-by: Haozhong Zhang
Reviewed-by: Stefan Hajnoczi
---
hw/mem/nvdimm.c | 2 +-
include/hw/mem/nvdimm.h | 3 +++
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/hw/mem/nvdimm.c b/hw/mem/nvdimm.c
index 952fce5ec8..618c3d677b 100644
--- a/hw/mem/nvdimm.c
+++ b/hw/mem
to, because of the remove of device dax check.
* Document new options added by this patch series.
Haozhong Zhang (3):
hostmem-file: add "align" option
nvdimm: add a macro for property "label-size"
nvdimm: add 'unarmed' option
backends/hostmem-file.c |
On 10/17/17 13:45 +0200, Paolo Bonzini wrote:
> On 14/10/2017 00:46, Stefano Stabellini wrote:
> > On Fri, 13 Oct 2017, Jan Beulich wrote:
> > On 13.10.17 at 13:13, wrote:
> >>> To Jan, Andrew, Stefano and Anthony,
> >>>
> >>> what do you think about allowing QEMU to build the entire guest ACP
On 10/13/17 10:44 +0200, Igor Mammedov wrote:
> On Fri, 13 Oct 2017 15:53:26 +0800
> Haozhong Zhang wrote:
>
> > On 10/12/17 17:45 +0200, Paolo Bonzini wrote:
> > > On 12/10/2017 14:45, Haozhong Zhang wrote:
> > > > Basically, QEMU builds two ROMs
On 10/12/17 13:39 -0400, Konrad Rzeszutek Wilk wrote:
> On Thu, Oct 12, 2017 at 08:45:44PM +0800, Haozhong Zhang wrote:
> > On 10/10/17 12:05 -0400, Konrad Rzeszutek Wilk wrote:
> > > On Tue, Sep 12, 2017 at 11:15:09AM +0800, Haozhong Zhang wrote:
> > > > On
On 10/12/17 17:45 +0200, Paolo Bonzini wrote:
> On 12/10/2017 14:45, Haozhong Zhang wrote:
> > Basically, QEMU builds two ROMs for guest, /rom@etc/acpi/tables and
> > /rom@etc/table-loader. The former is unstructured to guest, and
> > contains all data of guest
On 10/10/17 12:05 -0400, Konrad Rzeszutek Wilk wrote:
> On Tue, Sep 12, 2017 at 11:15:09AM +0800, Haozhong Zhang wrote:
> > On 09/11/17 11:52 -0700, Stefano Stabellini wrote:
> > > CC'ing xen-devel, and the Xen tools and x86 maintainers.
> > >
> > >
On 09/13/17 17:28 +0200, Michal Privoznik wrote:
>
> BTW: I ran a migration from no nvdimm qemu to one that had -M nvdimm=on
> and guest migrated happily. So looks like guest ABI is stable (or at
> least stable enough not to crash). But since ACPI table is changed I
> doubt that.
One example that
On 09/11/17 11:52 -0700, Stefano Stabellini wrote:
> CC'ing xen-devel, and the Xen tools and x86 maintainers.
>
> On Mon, 11 Sep 2017, Igor Mammedov wrote:
> > On Mon, 11 Sep 2017 12:41:47 +0800
> > Haozhong Zhang wrote:
> >
> > > This is the QEMU part p
Xen currently does not support vNVDIMM hotplug and always sets QEMU
option "maxmem" to be just enough for RAM and vNVDIMM, so it's not
necessary to build _FIT method when QEMU is used as Xen device model.
Signed-off-by: Haozhong Zhang
---
Cc: "Michael S. Tsirkin"
Cc
Probe the base address and the length of guest ACPI buffer reserved
for copying ACPI from QEMU.
Signed-off-by: Haozhong Zhang
---
Cc: Stefano Stabellini
cc: Anthony Perard
Cc: "Michael S. Tsirkin"
Cc: Paolo Bonzini
Cc: Richard Henderson
Cc: Eduardo Habkost
---
hw/i386/xen/xen-
If the machine option 'nvdimm' is enabled and QEMU is used as Xen
device model, construct the guest NFIT and ACPI namespace devices of
vNVDIMM and copy them into guest memory.
Signed-off-by: Haozhong Zhang
---
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
Cc: Paolo Bonzini
C
cates it's the body of a
namespace device, and its device name is indicated by $name in
the XenStore path.
- offset offset in byte from the beginning of above guest memory region
- length size in byte of the copied ACPI
Signed-off-by: Haozhong Zhang
---
Cc: Stefano Stabellini
C
Xen relies on QEMU to build the guest NFIT.
Signed-off-by: Haozhong Zhang
---
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
Cc: Xiao Guangrong
---
hw/acpi/nvdimm.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/hw/acpi/nvdimm.c b/hw/acpi/nvdimm.c
index 9121a766c6..d9cdc5a
Xen relies on QEMU to build guest ACPI for NVDIMM. However, no fw_cfg
is created when QEMU is used as Xen device model, so QEMU should avoid
using fw_cfg on Xen.
Signed-off-by: Haozhong Zhang
---
Cc: Xiao Guangrong
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
---
hw/acpi/nv
nd a hotplug
memory region for Xen guest, in order to make the existing nvdimm
device plugging path work on Xen.
Patch 4 - 10 build and cooy NFIT from QEMU to Xen guest, when QEMU is
used as the Xen device model.
Haozhong Zhang (10):
nvdimm: do not intiailize nvdimm->label_data if label size
.
Signed-off-by: Haozhong Zhang
---
Cc: Paolo Bonzini
Cc: Richard Henderson
CC: Eduardo Habkost
Cc: "Michael S. Tsirkin"
Cc: Stefano Stabellini
Cc: Anthony Perard
---
hw/i386/pc.c | 86 ---
hw/i386/xen/xen-hvm.c | 2 ++
include/hw
QEMU to labels is led by a label size check,
let's not intiailize nvdimm->label_data if the label size is 0.
Signed-off-by: Haozhong Zhang
---
Cc: Xiao Guangrong
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
---
hw/mem/nvdimm.c | 10 +-
1 file changed, 9 insertions(+), 1
x1:
-object memory-backend-xen,id=mem1,host-addr=0x1,size=4G
-device nvdimm,id=nvdimm1,memdev=mem1
Signed-off-by: Haozhong Zhang
---
Cc: Eduardo Habkost
Cc: Igor Mammedov
Cc: "Michael S. Tsirkin"
---
backends/Makefile.objs | 1 +
backends
Xen relies on QEMU to build the ACPI namespace device of vNVDIMM for
Xen guest.
Signed-off-by: Haozhong Zhang
---
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
Cc: Xiao Guangrong
---
hw/acpi/nvdimm.c | 55 ++-
1 file changed, 38
On 07/28/17 13:45 -0600, Ross Zwisler wrote:
> On Fri, Jul 28, 2017 at 11:11:10AM -0700, Dan Williams wrote:
> > On Fri, Jul 28, 2017 at 11:04 AM, Ross Zwisler
> > wrote:
> > > I've been using the virtualized NVDIMM support in QEMU for testing, and I
> > > noticed that the physical addresses used
On 07/21/17 02:56 -0400, Pankaj Gupta wrote:
>
> Hello,
>
> We shared a proposal for 'KVM fake DAX flushing interface'.
>
> https://lists.gnu.org/archive/html/qemu-devel/2017-05/msg02478.html
>
In above link,
"Overall goal of project
is to increase the number of virtual machines that can
On 07/03/17 17:45 +0200, Paolo Bonzini wrote:
>
>
> On 03/07/2017 17:23, Haozhong Zhang wrote:
> > Otherwise, QEMU on a host with MPX support will try to set guest
> > MSR_IA32_BNDCFGS although guest MPX is not enabled, and result in
> > abort.
> >
> >
-system-x86_64: /root/qemu.git/target/i386/kvm.c:1832: kvm_put_msrs:
Assertion `ret == cpu->kvm_msr_buf->nmsrs' failed
Signed-off-by: Haozhong Zhang
---
target/i386/kvm.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/target/i386/kvm.c b/target/i386/kvm.c
index ee36502789.
On 06/28/17 11:09 +0200, Juan Quintela wrote:
> Haozhong Zhang wrote:
> > In cpu_physical_memory_sync_dirty_bitmap(rb, start, ...), the 2nd
> > argument 'start' is relative to the start of the ramblock 'rb'. When
> > it's used to a
per ramblock dirty bitmap. As a result, a guest with host
memory backend may crash after migration.
Fix it by adding the offset of ramblock when accessing the dirty memory
bitmap of ram_list in cpu_physical_memory_sync_dirty_bitmap().
Reported-by: Stefan Hajnoczi
Signed-off-by: Haozhong Zha
per ramblock dirty bitmap. As a result, a guest with host
memory backend may crash after migration.
Fix it by adding the offset of ramblock when accessing the dirty memory
bitmap of ram_list in cpu_physical_memory_sync_dirty_bitmap().
Reported-by: Stefan Hajnoczi
Signed-off-by: Haozhong Zhang
-
On 06/26/17 13:56 +0100, Stefan Hajnoczi wrote:
> On Mon, Jun 26, 2017 at 10:05:01AM +0800, Haozhong Zhang wrote:
> > On 06/23/17 10:55 +0100, Stefan Hajnoczi wrote:
> > > On Fri, Jun 23, 2017 at 08:13:13AM +0800, haozhong.zh...@intel.com wrote:
> > > > On 06/22/1
On 06/23/17 10:55 +0100, Stefan Hajnoczi wrote:
> On Fri, Jun 23, 2017 at 08:13:13AM +0800, haozhong.zh...@intel.com wrote:
> > On 06/22/17 15:08 +0100, Stefan Hajnoczi wrote:
> > > I tried live migrating a guest with NVDIMM on qemu.git/master (edf8bc984):
> > >
> > > $ qemu -M accel=kvm,nvdimm=
On 06/08/17 15:51 +0300, Michael S. Tsirkin wrote:
> On Tue, Jun 06, 2017 at 03:22:27PM +0800, Haozhong Zhang wrote:
> > Applications in Linux guest that use device-dax never trigger flush
> > that can be trapped by KVM/QEMU. Meanwhile, if the host backend is not
> > de
On 06/08/17 15:56 +0300, Michael S. Tsirkin wrote:
> On Tue, Jun 06, 2017 at 03:22:28PM +0800, Haozhong Zhang wrote:
> > If a vNVDIMM device is not backed by a DAX device and its "restrict"
> > option is enabled, bit 3 of state flags in its region mapping
> > str
/pc.c:1713
> #8 0x5581d89b in pc_machine_device_plug_cb (hotplug_dev= out>, dev=0x56705590, errp=0x7fffc6c0) at qemu/hw/i386/pc.c:2004
> #9 0x55914da6 in device_set_realized (obj=,
> value=, errp=0x7fffc7e8) at hw/core/qdev.c:926
>
> Cc: Haozhon
On 06/08/17 14:40 +0800, Xiao Guangrong wrote:
>
>
> On 06/08/2017 01:06 PM, Haozhong Zhang wrote:
> > On 06/08/17 11:07 +0800, Xiao Guangrong wrote:
> > >
> > >
> > > On 06/07/2017 04:06 PM, Haozhong Zhang wrote:
> > > > Per ACPI
On 06/07/17 16:27 +0100, Stefan Hajnoczi wrote:
> On Tue, Jun 06, 2017 at 03:22:28PM +0800, Haozhong Zhang wrote:
> > If a vNVDIMM device is not backed by a DAX device and its "restrict"
> > option is enabled, bit 3 of state flags in its region mapping
> > structure
On 06/08/17 11:07 +0800, Xiao Guangrong wrote:
>
>
> On 06/07/2017 04:06 PM, Haozhong Zhang wrote:
> > Per ACPI 6.2, section 5.2.25.6 and JEDEC Annex L Release 3, the
> > current region format interface code 0x201 indicates the block
> > addressed function inte
On 06/07/17 16:27 +0100, Stefan Hajnoczi wrote:
> On Tue, Jun 06, 2017 at 03:22:28PM +0800, Haozhong Zhang wrote:
> > If a vNVDIMM device is not backed by a DAX device and its "restrict"
> > option is enabled, bit 3 of state flags in its region mapping
> > structure
On 06/07/17 16:14 +0100, Stefan Hajnoczi wrote:
> On Tue, Jun 06, 2017 at 03:22:27PM +0800, Haozhong Zhang wrote:
> > diff --git a/util/osdep.c b/util/osdep.c
> > index a2863c8e53..02881f96bc 100644
> > --- a/util/osdep.c
> > +++ b/util/osdep.c
> > @@ -471,3 +471,
On 06/06/17 10:59 -0700, Dan Williams wrote:
> On Tue, Jun 6, 2017 at 12:22 AM, Haozhong Zhang
> wrote:
> > Applications in Linux guest that use device-dax never trigger flush
> > that can be trapped by KVM/QEMU. Meanwhile, if the host backend is not
> > device-dax, Q
.
Signed-off-by: Haozhong Zhang
---
hw/acpi/nvdimm.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/hw/acpi/nvdimm.c b/hw/acpi/nvdimm.c
index 8e7d6ec034..b5734f5897 100644
--- a/hw/acpi/nvdimm.c
+++ b/hw/acpi/nvdimm.c
@@ -338,9 +338,10 @@ static void
Signed-off-by: Haozhong Zhang
---
hw/mem/nvdimm.c | 2 +-
include/hw/mem/nvdimm.h | 3 +++
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/hw/mem/nvdimm.c b/hw/mem/nvdimm.c
index db896b0bb6..a9b0863f20 100644
--- a/hw/mem/nvdimm.c
+++ b/hw/mem/nvdimm.c
@@ -66,7 +66,7 @@ out
A DAX device may require a page size different than getpagesize().
Signed-off-by: Haozhong Zhang
---
include/qemu/osdep.h | 10 ++
util/mmap-alloc.c| 6 ++
util/osdep.c | 23 +++
3 files changed, 39 insertions(+)
diff --git a/include/qemu/osdep.h b
device-dax.
Signed-off-by: Haozhong Zhang
Message-id: capcyv4hv2-zw8smcrtd0p_86kgr3dhovne+6t5sy2u7wxg3...@mail.gmail.com
---
hw/mem/nvdimm.c | 6 ++
include/qemu/osdep.h | 9
util/osdep.c | 61
3 files changed, 76
ce as read-only.
This option is disabled by default for backwards compatibility. It's
recommended to enable for the formal usage.
Signed-off-by: Haozhong Zhang
---
hw/acpi/nvdimm.c| 16
hw/mem/nvdimm.c | 38 +-
include/
lags in guest
NVDIMM region mapping structure. This behavior is controlled by a
new QEMU nvdimm option "restrict". See details in patch 3. (Dan)
* Patch 4: drop the "align" property and get the alignment of DAX
device by QEMU itself.
* Wrap Linux specific code by __linu
On 05/30/17 10:16 +0100, Stefan Hajnoczi wrote:
> On Fri, May 26, 2017 at 10:24:38AM +0800, Haozhong Zhang wrote:
> > file_ram_alloc() currently maps the backend file via mmap to a virtual
> > address aligned to the value returned by qemu_fd_getpagesize(). When a
> > DAX dev
On 05/30/17 10:15 +0100, Stefan Hajnoczi wrote:
> On Fri, May 26, 2017 at 10:24:37AM +0800, Haozhong Zhang wrote:
> > Applications in Linux guest that use device-dax never trigger flush
> > that can be trapped by KVM/QEMU. Meanwhile, if the host backend is not
> > device-dax,
On 05/26/17 08:25 -0700, Dan Williams wrote:
> On Fri, May 26, 2017 at 7:38 AM, Daniel P. Berrange
> wrote:
> > On Thu, May 25, 2017 at 08:34:23PM -0700, Dan Williams wrote:
> >> On Thu, May 25, 2017 at 7:32 PM, Haozhong Zhang
> >> wrote:
> >> > Appli
On 05/26/17 15:55 -0300, Eduardo Habkost wrote:
> On Fri, May 26, 2017 at 07:24:26AM -0700, Dan Williams wrote:
> > On Fri, May 26, 2017 at 12:16 AM, Haozhong Zhang
> > wrote:
> > > On 05/26/17 07:05 +, Marc-André Lureau wrote:
> > >> Hi
> > >>
On 05/25/17 20:34 -0700, Dan Williams wrote:
> On Thu, May 25, 2017 at 7:32 PM, Haozhong Zhang
> wrote:
> > Applications in Linux guest that use device-dax never trigger flush
> > that can be trapped by KVM/QEMU. Meanwhile, if the host backend is not
> > device-dax, Q
On 03/23/17 08:36 +, Stefan Hajnoczi wrote:
> On Tue, Mar 21, 2017 at 01:33:57PM +0800, Xiao Guangrong wrote:
> > From: Xiao Guangrong
> >
> > My Intel mail account will be disabled soon, update the mail info
> > to my private mail
> >
> > Signed-off-by: Xiao Guangrong
> > ---
> > MAINTAIN
d as before.
Signed-off-by: Haozhong Zhang
---
Cc: Eduardo Habkost
Cc: Igor Mammedov
Cc: Paolo Bonzini
Cc: Peter Crosthwaite
Cc: Richard Henderson
Cc: Xiao Guangrong
Cc: Stefan Hajnoczi
Cc: Dan Williams
---
Resend because the wrong maintainer email address was used.
---
backends
device-dax.
Signed-off-by: Haozhong Zhang
Message-id: capcyv4hv2-zw8smcrtd0p_86kgr3dhovne+6t5sy2u7wxg3...@mail.gmail.com
---
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
Cc: Xiao Guangrong
Cc: Stefan Hajnoczi
Cc: Dan Williams
---
Resend because the wrong maintainer email address was use
device-dax.
Signed-off-by: Haozhong Zhang
Message-id: capcyv4hv2-zw8smcrtd0p_86kgr3dhovne+6t5sy2u7wxg3...@mail.gmail.com
---
Cc: "Michael S. Tsirkin"
Cc: Igor Mammedov
Cc: Xiao Guangrong
Cc: Stefan Hajnoczi
Cc: Dan Williams
---
hw/mem/nvdimm.c | 37 +++
d as before.
Signed-off-by: Haozhong Zhang
---
Cc: Eduardo Habkost
Cc: Igor Mammedov
Cc: Paolo Bonzini
Cc: Peter Crosthwaite
Cc: Richard Henderson
Cc: Xiao Guangrong
Cc: Stefan Hajnoczi
Cc: Dan Williams
---
backends/hostmem-file.c | 41 -
exec.c
On 04/07/17 14:46 +0100, Stefan Hajnoczi wrote:
> On Thu, Apr 06, 2017 at 06:46:49PM +0800, Haozhong Zhang wrote:
> > On 04/06/17 11:24 +0100, Stefan Hajnoczi wrote:
> > > On Fri, Mar 31, 2017 at 04:41:44PM +0800, Haozhong Zhang wrote:
> > > > If option 'reserv
101 - 200 of 394 matches
Mail list logo
