Markus Armbruster 于2018年11月19日周一 下午3:01写道：
> ÀîÇ¿ writes:
>
> > At 2018-11-17 00:52:58, "Markus Armbruster" wrote:
> >>Li Qiang writes:
> >>
> >>> Currently the user can set a negative reboot_timeout.
> >>> Also it is wrong to parse 'reboot-timeout' with qemu_opt_get() and then
> >>> convert i
The subsystem tag in the commit message's title should be "qapi: " not
"json: ".
Yuval Shaia writes:
> pvrdma requires that the same GID attached to it will be attached to the
> backend device in the host.
>
> A new QMP messages is defined so pvrdma device can broadcast any change
> made to its
On Mon, Nov 19, 2018 at 02:17:27PM +0800, Peter Xu wrote:
> On Thu, Nov 08, 2018 at 10:44:06AM +0800, Peter Xu wrote:
> > On Wed, Nov 07, 2018 at 11:21:32AM +, Peter Maydell wrote:
> > > On 7 November 2018 at 02:56, Peter Xu wrote:
> > > > Strange, "make check -j8" failed on my hosts (I tried
ÀîÇ¿ writes:
> At 2018-11-17 00:52:58, "Markus Armbruster" wrote:
>>Li Qiang writes:
>>
>>> Currently the user can set a negative reboot_timeout.
>>> Also it is wrong to parse 'reboot-timeout' with qemu_opt_get() and then
>>> convert it to number.
>>
>>Again, it's not wrong per se, just needles
On Thu, Nov 08, 2018 at 10:44:06AM +0800, Peter Xu wrote:
> On Wed, Nov 07, 2018 at 11:21:32AM +, Peter Maydell wrote:
> > On 7 November 2018 at 02:56, Peter Xu wrote:
> > > Strange, "make check -j8" failed on my hosts (I tried two) with either
> > > Markus's pull tree or qemu master:
> > >
>
On 19/11/2018 12:45, David Gibson wrote:
> On Tue, Nov 13, 2018 at 07:30:58PM +1100, Alexey Kardashevskiy wrote:
>> The current code assumes that we can address more bits on a PCI bus
>> for DMA than we really can. Limit to the known tested maximum of 55 bits
>> and assume 64K IOMMU pages.
>
>
On Mon, Nov 19, 2018 at 04:08:33PM +1100, Alexey Kardashevskiy wrote:
>
>
> On 19/11/2018 13:42, David Gibson wrote:
> > On Tue, Nov 13, 2018 at 07:31:02PM +1100, Alexey Kardashevskiy wrote:
> >> When deciding about the huge DMA window, the typical Linux pseries guest
> >> uses the maximum allowe
On 19/11/2018 14:01, David Gibson wrote:
> On Tue, Nov 13, 2018 at 07:31:04PM +1100, Alexey Kardashevskiy wrote:
>> The NVIDIA V100 GPU comes with some on-board RAM which is mapped into
>> the host memory space and accessible as normal RAM via NVLink bus.
>> The VFIO-PCI driver implements specia
On 19/11/2018 13:42, David Gibson wrote:
> On Tue, Nov 13, 2018 at 07:31:02PM +1100, Alexey Kardashevskiy wrote:
>> When deciding about the huge DMA window, the typical Linux pseries guest
>> uses the maximum allowed RAM size as the upper limit. We did the same
>> on QEMU side to match that logi
From: Seth Kintigh
I corrected these 2 memory regions based on specifications from the chip
manufacturer. The existing ranges seem to overlap and and cause odd
behavior and/or crashes when trying to set up multiple UARTs,
Signed-off-by: Seth Kintigh
---
Phil, I hope this is the right format.
P
On Tue, Nov 13, 2018 at 07:31:01PM +1100, Alexey Kardashevskiy wrote:
> The NVIDIA V100 GPUs often come in several instances on the same system
> board where they are connected directly via out of band fabric called
> "NVLink".
>
> In order to make GPUs talk to each other, NVLink has to be enabled
On Tue, Nov 13, 2018 at 07:31:02PM +1100, Alexey Kardashevskiy wrote:
> When deciding about the huge DMA window, the typical Linux pseries guest
> uses the maximum allowed RAM size as the upper limit. We did the same
> on QEMU side to match that logic. Now we are going to support GPU RAM
> pass thr
On Tue, Nov 13, 2018 at 07:30:58PM +1100, Alexey Kardashevskiy wrote:
> The current code assumes that we can address more bits on a PCI bus
> for DMA than we really can. Limit to the known tested maximum of 55 bits
> and assume 64K IOMMU pages.
This doesn't make much sense to me. It looks nothing
On Tue, Nov 13, 2018 at 07:31:04PM +1100, Alexey Kardashevskiy wrote:
> The NVIDIA V100 GPU comes with some on-board RAM which is mapped into
> the host memory space and accessible as normal RAM via NVLink bus.
> The VFIO-PCI driver implements special regions for such GPU and emulated
> NVLink brid
On Tue, Nov 13, 2018 at 07:31:00PM +1100, Alexey Kardashevskiy wrote:
> sPAPR code will use it too so move it from VFIO to the common code.
>
> Signed-off-by: Alexey Kardashevskiy
Reviewed-by: David Gibson
This looks correct to me independent of the rest of the series.
> ---
> include/hw/pci
On 11/2/18 6:25 AM, Laurent Vivier wrote:
>> If that does not work, I'm also fine if we simply deprecate the simcalls
>> (if possible).
>
> I have a patch to deprecate the interface. I will send it once the
> release will be done.
Did I ever point out that I got m68k running under your q800 emula
At 2018-11-17 00:52:58, "Markus Armbruster" wrote:
>Li Qiang writes:
>
>> Currently the user can set a negative reboot_timeout.
>> Also it is wrong to parse 'reboot-timeout' with qemu_opt_get() and then
>> convert it to number.
>
>Again, it's not wrong per se, just needlessly complicated
At 2018-11-17 00:33:34, "Markus Armbruster" wrote:
>Li Qiang writes:
>
>> Currently when the splash-time value is bigger than 0x
>> we report and correct it, when it is less than 0 we just ingore it.
>
>s/ingore/ignore/
>
>> Also we use qemu_opt_get() to get 'splash-time', then conve
Hi Dario,
On Sun, Nov 18, 2018 at 10:54 PM Dario Faggioli wrote:
>
> On Sun, 2018-11-18 at 19:47 +, Alex Bennée wrote:
> > Dario Faggioli writes:
> >
> > > Dockerfile for building an openSUSE Leap container.
> > >
> > > Tracks the latest release (at the time of writing this
> > > patch, it i
Hi Stefano,
On Fri, Nov 16, 2018 at 05:13:59PM +0100, Stefano Garzarella wrote:
> Hi,
> I'm investigating the SeaBIOS booting time, to understand if we can reduce
> the boot time in some cases (e.g. legacy hardware is not needed). I think
> this
> can be interesting also for NEMU developers.
Defin
Hi Stefan,
On Fri, Nov 16, 2018 at 04:39:20PM +, Stefan Hajnoczi wrote:
> On Fri, Nov 16, 2018 at 4:21 PM Stefano Garzarella
> wrote:
> > I'm investigating the SeaBIOS booting time, to understand if we can reduce
> > the boot time in some cases (e.g. legacy hardware is not needed). I think
>
On Sun, 2018-11-18 at 19:47 +, Alex Bennée wrote:
> Dario Faggioli writes:
>
> > Dockerfile for building an openSUSE Leap container.
> >
> > Tracks the latest release (at the time of writing this
> > patch, it is Leap 15).
> >
> > Signed-off-by: Dario Faggioli
> > ---
> > Cc: "Alex Bennée"
This image is based on Gentoo and the toolchain is built using crossdev.
Recipe from:
https://lists.gnu.org/archive/html/qemu-devel/2018-09/msg03944.html
Suggested-by: Fredrik Noring
Signed-off-by: Philippe Mathieu-Daudé
---
TODO:
- Add Fredrik Noring S-o-b in his patches
- Check patch merged
Hi,
The first patch adds a cross toolchain for the R5900 MIPS.
It is working correctly but the patches provided by Fredrik in [1] don't
have proper S-o-b, thus it is tagged RFC.
Fredrik: any update on the status of those patches upstream?
I setup this image to try Fredrik's TCG tests in [2].
The
Shippable node pool must be updated to use at lease v5.8.2:
http://docs.shippable.com/platform/runtime/machine-image/ami-v582/
Signed-off-by: Philippe Mathieu-Daudé
---
.shippable.yml | 2 ++
tests/docker/dockerfiles/gentoo-mipsr5900el-cross.docker |
Dario Faggioli writes:
> Dockerfile for building an openSUSE Leap container.
>
> Tracks the latest release (at the time of writing this
> patch, it is Leap 15).
>
> Signed-off-by: Dario Faggioli
> ---
> Cc: "Alex Bennée"
> Cc: Fam Zheng
> Cc: "Philippe Mathieu-Daudé"
> ---
> tests/docker/d
hello
I'm building qemu from source and happily using it since a bit
(2.3.0)
Since 3.1.0-rc0 (including latest 3.1.0-rc1) I'm no more able to start
qemu, getting:
8<
install:115> qemu
qemu: error: failed to set MSR 0x10a to 0x0
qemu:
/home/balducci/tmp/install-us-d/qemu-
On Sat, Oct 13, 2018 at 11:53:23AM +0200, Paolo Bonzini wrote:
> rerror=ignore was returning true from scsi_handle_rw_error but the callers
> were not
> calling scsi_req_complete when rerror=ignore returns true (this is the
> correct thing
> to do when true is returned after executing a passthrou
** Description changed:
QEMU 3.0
block/sheepdog.c: In function 'find_vdi_name':
block/sheepdog.c:1239:5: error: 'strncpy' specified bound 256 equals
destination size [-Werror=stringop-truncation]
- strncpy(buf + SD_MAX_VDI_LEN, tag, SD_MAX_VDI_TAG_LEN);
- ^~
Public bug reported:
QEMU 3.0
block/sheepdog.c: In function 'find_vdi_name':
block/sheepdog.c:1239:5: error: 'strncpy' specified bound 256 equals
destination size [-Werror=stringop-truncation]
strncpy(buf + SD_MAX_VDI_LEN, tag, SD_MAX_VDI_TAG_LEN);
^
bitmap_zero_extend is designed to work for extending, not for
shrinking.
Using g_free instead.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/rdma_rm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/rdma/rdma_rm.c b/hw/rdma/rdma_rm.c
index b7d4ebe97
Device supports only one port, let's remove a dead code that handles
more than one port.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/rdma_rm.c | 34 --
hw/rdma/rdma_rm.h | 2 +-
hw/rdma/rdma_rm_defs.h | 4 ++--
3 files changed
Guest driver enforces it, we should also.
Signed-off-by: Yuval Shaia
---
hw/rdma/vmw/pvrdma.h | 2 ++
hw/rdma/vmw/pvrdma_main.c | 12
2 files changed, 14 insertions(+)
diff --git a/hw/rdma/vmw/pvrdma.h b/hw/rdma/vmw/pvrdma.h
index b019cb843a..10a3c4fb7c 100644
--- a/hw/rdma/v
When device goes down the function fini_ports loops over all entries in
gid table regardless of the fact whether entry is valid or not. In case
that entry is not valid we'd like to skip from any further processing in
backend device.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
h
The control over the RDMA device's GID table is done by updating the
device's Ethernet function addresses.
Usually the first GID entry is determined by the MAC address, the second
by the first IPv6 address and the third by the IPv4 address. Other
entries can be added by adding more IP addresses. Th
On 13/11/2018 20:29, John Snow wrote:
> On 11/13/18 8:16 AM, Kevin Wolf wrote:
>> Am 12.11.2018 um 20:58 hat John Snow geschrieben:
>>>
>>>
>>> On 11/11/18 4:40 AM, Mark Cave-Ayland wrote:
Commit c8a35f1cf0f "fdc: use IsaDma interface instead of global DMA_*
functions" accidentally intro
In order to clean some external resources such as GIDs, QPs etc,
register to receive notification when VM is shutdown.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/vmw/pvrdma.h | 2 ++
hw/rdma/vmw/pvrdma_main.c | 12
2 files changed, 14 insertions(+)
Commit 6e7dba23af ("hw/pvrdma: Make default pkey 0x") exports
default pkey as external definition but omit the change from 0x7FFF to
0x.
Fixes: 6e7dba23af ("hw/pvrdma: Make default pkey 0x")
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/vmw/pvrdma.h | 2 +-
1
Interface with the device is changed with the addition of support for
MAD packets.
Adjust documentation accordingly.
While there fix a minor mistake which may lead to think that there is a
relation between using RXE on host and the compatibility with bare-metal
peers.
Signed-off-by: Yuval Shaia
Driver checks error code let's set it.
Signed-off-by: Yuval Shaia
---
hw/rdma/vmw/pvrdma_cmd.c | 67
1 file changed, 48 insertions(+), 19 deletions(-)
diff --git a/hw/rdma/vmw/pvrdma_cmd.c b/hw/rdma/vmw/pvrdma_cmd.c
index 0d3c818c20..a326c5d470 100644
--
Add ability to pass specific WC attributes to CQE such as GRH_BIT flag.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/rdma_backend.c | 59 +++--
hw/rdma/rdma_backend.h | 4 +--
hw/rdma/vmw/pvrdma_qp_ops.c | 31 +++
Notifier will be used for signaling shutdown event to inform system is
shutdown. This will allow devices and other component to run some
cleanup code needed before VM is shutdown.
Signed-off-by: Yuval Shaia
Reviewed-by: Cornelia Huck
---
include/sysemu/sysemu.h | 1 +
vl.c|
Upon completion of incoming packet the device pushes CQE to driver's RX
ring and notify the driver (msix).
While for data-path incoming packets the driver needs the ability to
control whether it wished to receive interrupts or not, for control-path
packets such as incoming MAD the driver needs to b
pvrdma setup requires vmxnet3 device on PCI function 0 and PVRDMA device
on PCI function 1.
pvrdma device needs to access vmxnet3 device object for several reasons:
1. Make sure PCI function 0 is vmxnet3.
2. To monitor vmxnet3 device state.
3. To configure node_guid accoring to vmxnet3 device's MAC
node_guid should be set once device is load.
Make node_guid be GID format (32 bit) of PCI function 0 vmxnet3 device's
MAC.
A new function was added to do the conversion.
So for example the MAC 56:b6:44:e9:62:dc will be converted to GID
54b6:44ff:fee9:62dc.
Signed-off-by: Yuval Shaia
Reviewed-by:
User should be able to control the device by changing Ethernet function
state so if user runs 'ifconfig ens3 down' the PVRDMA function should be
down as well.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/vmw/pvrdma_cmd.c | 3 ++-
1 file changed, 2 insertions(+), 1 deleti
RDMA MAD kernel module (ibcm) disallow more than one MAD-agent for a
given MAD class.
This does not go hand-by-hand with qemu pvrdma device's requirements
where each VM is MAD agent.
Fix it by adding implementation of RDMA MAD multiplexer service which on
one hand register as a sole MAD agent with
pvrdma requires that the same GID attached to it will be attached to the
backend device in the host.
A new QMP messages is defined so pvrdma device can broadcast any change
made to its GID table. This event is captured by libvirt which in turn
will update the GID table in the backend device.
Sign
The function pvrdma_post_cqe populates CQE entry with opcode from the
given completion element. For receive operation value was not set. Fix
it by setting it to IBV_WC_RECV.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/vmw/pvrdma_qp_ops.c | 3 ++-
1 file changed, 2 inser
Function create_ah might return NULL, let's exit with an error.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/rdma_backend.c | 4
1 file changed, 4 insertions(+)
diff --git a/hw/rdma/rdma_backend.c b/hw/rdma/rdma_backend.c
index d7a4bbd91f..1e148398a2 100644
--- a/h
Device is not supporting QP0, only QP1.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/rdma_backend.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/rdma/rdma_backend.h b/hw/rdma/rdma_backend.h
index 86e8fe8ab6..3ccc9a2494 100644
--- a/hw/rdma/rdma_
opcode for WC should be set by the device and not taken from work
element.
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/vmw/pvrdma_qp_ops.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/rdma/vmw/pvrdma_qp_ops.c b/hw/rdma/vmw/pvrdma_qp_ops.c
index
Hi all.
This is a major enhancement to the pvrdma device to allow it to work with
state of the art applications such as MPI.
As described in patch #5, MAD packets are management packets that are used
for many purposes including but not limited to communication layer above IB
verbs API.
Patch 1 e
This function cannot fail - fix it to return void
Signed-off-by: Yuval Shaia
Reviewed-by: Marcel Apfelbaum
---
hw/rdma/vmw/pvrdma_main.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/hw/rdma/vmw/pvrdma_main.c b/hw/rdma/vmw/pvrdma_main.c
index 6c8c0154fa..fc2abd34af 1006
MAD (Management Datagram) packets are widely used by various modules
both in kernel and in user space for example the rdma_* API which is
used to create and maintain "connection" layer on top of RDMA uses
several types of MAD packets.
For more information please refer to chapter 13.4 in Volume 1
A
On Sat, Nov 17, 2018 at 07:27:43PM +0200, Shamir Rabinovitch wrote:
> On Tue, Nov 13, 2018 at 09:13:14AM +0200, Yuval Shaia wrote:
> > RDMA MAD kernel module (ibcm) disallow more than one MAD-agent for a
> > given MAD class.
> > This does not go hand-by-hand with qemu pvrdma device's requirements
>
Ping again.
v3.0 didn't contain 143c04c7e0639e53086519592ead15d2556bfbf2, so this commit
fixes a regression.
Le 10/11/2018 à 21:53, Hervé Poussineau a écrit :
Ping.
Le 21/10/2018 à 21:07, Hervé Poussineau a écrit :
A check for scan_enabled has been added to ps2_keyboard_event in commit
143c0
On Sat, Nov 17, 2018 at 02:06:48PM +0200, Marcel Apfelbaum wrote:
> Hi Yuval,
>
> On 11/13/18 9:12 AM, Yuval Shaia wrote:
> > MAD (Management Datagram) packets are widely used by various modules
>
> Please add a link to Spec, I sent it in the V1 mail-thread
> Please add it also as a comment in th
On Sat, Nov 17, 2018 at 02:25:55PM +0200, Marcel Apfelbaum wrote:
>
>
> On 11/13/18 9:13 AM, Yuval Shaia wrote:
> > When device goes down the function fini_ports loops over all entries in
> > gid table regardless of the fact whether entry is valid or not. In case
> > that entry is not valid we'd
On Sat, Nov 17, 2018 at 01:41:41PM +0200, Marcel Apfelbaum wrote:
>
>
> On 11/11/18 9:45 AM, Yuval Shaia wrote:
> > On Sat, Nov 10, 2018 at 08:27:44PM +0200, Marcel Apfelbaum wrote:
> > >
> > > On 11/8/18 6:08 PM, Yuval Shaia wrote:
> > > > Guest driver enforces it, we should also.
> > > >
> >
On Sat, Nov 17, 2018 at 02:22:31PM +0200, Marcel Apfelbaum wrote:
>
>
> On 11/13/18 9:13 AM, Yuval Shaia wrote:
> > Driver checks error code let's set it.
> >
> > Signed-off-by: Yuval Shaia
> > ---
> > hw/rdma/vmw/pvrdma_cmd.c | 67
> > 1 file changed
On Sat, Nov 17, 2018 at 02:48:34PM +0200, Marcel Apfelbaum wrote:
>
>
> On 11/13/18 9:13 AM, Yuval Shaia wrote:
> > The control over the RDMA device's GID table is done by updating the
> > device's Ethernet function addresses.
> > Usually the first GID entry is determine by the MAC address, the s
62 matches
Mail list logo
