[Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Santucco]

[-- Attachment #1.1: Type: text/plain, Size: 1886 bytes --]


Hello,
I’m trying to use vdispl interface from PV OS, it doesn’t work.
Configuration details:
    Xen 4.12.1
    Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel GNU/Linux
    DomU: x86 Plan9, PV
    displ_be as a backend for vdispl and vkb

when VM starts, displ_be reports about an error:
gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument (displ_be.log:221)

related Dom0 output is:
[  191.579278] Cannot provide dma-buf: use_ptemode 1 (dmesg.create.log:123)

According to the linux kernel sources use_ptemod is:
gntdev.c:1213:  use_ptemod = !xen_feature(XENFEAT_auto_translated_physmap);

But xen-features.pl script doesn't show this feature is set for Dom0:
Xen features: 000028f0
enabled: pae_pgdir_above_4gb
enabled: mmu_pt_update_preserve_ad
enabled: dom0
enabled: memory_op_vnode_supported

According to the Xen sources, auto_translated_physmap is set up in a case of :
xen/common/kernel.c:467:            if ( paging_mode_translate(d) )
xen/common/kernel.c-468-                fi.submap |= 
xen/common/kernel.c-469-                    (1U << XENFEAT_writable_page_tables) |
xen/common/kernel.c-470-                    (1U << XENFEAT_auto_translated_physmap);

where
xen/include/asm-x86/paging.h:73:#define paging_mode_translate(_d) (!!((_d)->arch.paging.mode & PG_translate))

But PG_translate is applied only for HVM, at least I have not fount any other points:
xen/arch/x86/hvm/hvm.c:606:    rc = paging_enable(d, PG_refcounts|PG_translate|PG_external);

What configuration should be for vdispl is able to work?
Or maybe I can to something to make Xen set auto_translated_physmap feature up to continue using of vdispl?
And should I make any special support for the feature in PV OS?

All logs and configs are attached.

Thanks in advance.

Best regards,
    Alexander Sychev

[-- Attachment #1.2: Type: text/html, Size: 2877 bytes --]

[-- Attachment #2: logs.tar.gz --]
[-- Type: application/x-gzip, Size: 63878 bytes --]

[-- Attachment #3: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Jürgen Groß]

On 06.01.20 08:56, Santucco wrote:
> Hello,
> 
> I’m trying to use vdispl interface from PV OS, it doesn’t work.
> Configuration details:
>      Xen 4.12.1
>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel GNU/Linux
>      DomU: x86 Plan9, PV
>      displ_be as a backend for vdispl and vkb
> 
> when VM starts, displ_be reports about an error:
> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument (displ_be.log:221)
> 
> related Dom0 output is:
> [  191.579278] Cannot provide dma-buf: use_ptemode 1 (dmesg.create.log:123)

This seems to be a limitation of the xen dma-buf driver. It was written
for being used on ARM initially where PV is not available.

CC-ing Oleksandr Andrushchenko who is the author of that driver. He
should be able to tell us what would be needed to enable PV dom0.

Depending on your use case it might be possible to use PVH dom0, but
support for this mode is "experimental" only and some features are not
yet working.


Juergen


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Oleksandr Andrushchenko]

On 1/6/20 10:38 AM, Jürgen Groß wrote:
> On 06.01.20 08:56, Santucco wrote:
>> Hello,
>>
>> I’m trying to use vdispl interface from PV OS, it doesn’t work.
>> Configuration details:
>>      Xen 4.12.1
>>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK 2019 
>> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel GNU/Linux
>>      DomU: x86 Plan9, PV
>>      displ_be as a backend for vdispl and vkb
>>
>> when VM starts, displ_be reports about an error:
>> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument 
>> (displ_be.log:221)
>>
>> related Dom0 output is:
>> [  191.579278] Cannot provide dma-buf: use_ptemode 1 
>> (dmesg.create.log:123)
>
> This seems to be a limitation of the xen dma-buf driver. It was written
> for being used on ARM initially where PV is not available.
This is true and we never tried/targeted PV domains with this 
implementation,
so if there is a need for that someone has to take a look on the proper
implementation for PV...
>
> CC-ing Oleksandr Andrushchenko who is the author of that driver. He
> should be able to tell us what would be needed to enable PV dom0.
>
> Depending on your use case it might be possible to use PVH dom0, but
> support for this mode is "experimental" only and some features are not
> yet working.
>
Well, one of the workarounds possible is to drop zero-copying use-case
(this is why display backend tries to create dmu-bufs from grants passed
by the guest domain and fails because of "Cannot provide dma-buf: 
use_ptemode 1")
So, in this case display backend will do memory copying for the incoming 
frames
and won't touch DMABUF_EXP_FROM_REFS ioctl.
To do so just disable zero-copying while building the backend [1]
>
> Juergen
>
[1] https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Santucco]

[-- Attachment #1.1: Type: text/plain, Size: 2596 bytes --]


Thank you very much for all your answers. 
>Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko < oleksandr_andrushchenko@epam.com >:
> 
>On 1/6/20 10:38 AM, Jürgen Groß wrote:
>> On 06.01.20 08:56, Santucco wrote:
>>> Hello,
>>>
>>> I’m trying to use vdispl interface from PV OS, it doesn’t work.
>>> Configuration details:
>>>      Xen 4.12.1
>>>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK 2019
>>> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel GNU/Linux
>>>      DomU: x86 Plan9, PV
>>>      displ_be as a backend for vdispl and vkb
>>>
>>> when VM starts, displ_be reports about an error:
>>> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument
>>> (displ_be.log:221)
>>>
>>> related Dom0 output is:
>>> [  191.579278] Cannot provide dma-buf: use_ptemode 1
>>> (dmesg.create.log:123)
>>
>> This seems to be a limitation of the xen dma-buf driver. It was written
>> for being used on ARM initially where PV is not available.
>This is true and we never tried/targeted PV domains with this
>implementation,
>so if there is a need for that someone has to take a look on the proper
>implementation for PV…
Have I got your right and there is no the proper implementation :-)?
>>
>> CC-ing Oleksandr Andrushchenko who is the author of that driver. He
>> should be able to tell us what would be needed to enable PV dom0.
>>
>> Depending on your use case it might be possible to use PVH dom0, but
>> support for this mode is "experimental" only and some features are not
>> yet working.
>>
>Well, one of the workarounds possible is to drop zero-copying use-case
>(this is why display backend tries to create dmu-bufs from grants passed
>by the guest domain and fails because of "Cannot provide dma-buf:
>use_ptemode 1")
>So, in this case display backend will do memory copying for the incoming
>frames
>and won't touch DMABUF_EXP_FROM_REFS ioctl.
>To do so just disable zero-copying while building the backend [1]
 
Thanks, I have just tried the workaround.  The backend has failed in an other place not corresponding with dma_buf.
Anyway it is enough to continue debugging  my frontend implementation.
 
Do you know how big is performance penalty in comparison with the zero-copy variant?
 
Does it make a sense if I make a dedicated HVM domain with linux only for the purpose of vdispl and vkbd backends? Is there a hope this approach will work?
>>
>> Juergen
>>
>[1]  https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
 
Best regards,
  Alexander Sychev

[-- Attachment #1.2: Type: text/html, Size: 4450 bytes --]

[-- Attachment #2: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Oleksandr Andrushchenko]

On 1/8/20 5:38 PM, Santucco wrote:
> Thank you very much for all your answers.
>
>     Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
>     <oleksandr_andrushchenko@epam.com
>     </compose?To=oleksandr_andrushchenko@epam.com>>:
>     On 1/6/20 10:38 AM, Jürgen Groß wrote:
>     > On 06.01.20 08:56, Santucco wrote:
>     >> Hello,
>     >>
>     >> I’m trying to use vdispl interface from PV OS, it doesn’t work.
>     >> Configuration details:
>     >>      Xen 4.12.1
>     >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK
>     2019
>     >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>     GNU/Linux
>     >>      DomU: x86 Plan9, PV
>     >>      displ_be as a backend for vdispl and vkb
>     >>
>     >> when VM starts, displ_be reports about an error:
>     >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument
>     >> (displ_be.log:221)
>     >>
>     >> related Dom0 output is:
>     >> [  191.579278] Cannot provide dma-buf: use_ptemode 1
>     >> (dmesg.create.log:123)
>     >
>     > This seems to be a limitation of the xen dma-buf driver. It was
>     written
>     > for being used on ARM initially where PV is not available.
>     This is true and we never tried/targeted PV domains with this
>     implementation,
>     so if there is a need for that someone has to take a look on the
>     proper
>     implementation for PV…
>
> Have I got your right and there is no the proper implementation :-)?
There is no
>
>     >
>     > CC-ing Oleksandr Andrushchenko who is the author of that driver. He
>     > should be able to tell us what would be needed to enable PV dom0.
>     >
>     > Depending on your use case it might be possible to use PVH dom0, but
>     > support for this mode is "experimental" only and some features
>     are not
>     > yet working.
>     >
>     Well, one of the workarounds possible is to drop zero-copying use-case
>     (this is why display backend tries to create dmu-bufs from grants
>     passed
>     by the guest domain and fails because of "Cannot provide dma-buf:
>     use_ptemode 1")
>     So, in this case display backend will do memory copying for the
>     incoming
>     frames
>     and won't touch DMABUF_EXP_FROM_REFS ioctl.
>     To do so just disable zero-copying while building the backend [1]
>
> Thanks, I have just tried the workaround.  The backend has failed 
> in an other place not corresponding with dma_buf.
> Anyway it is enough to continue debugging  my frontend implementation.
> Do you know how big is performance penalty in comparison with 
> the zero-copy variant?
Well, it solely depends on your setup, so I cannot tell what
would the numbers be in your case. Comparing to what I have doesn't
make any sense to me: one should compare apples to apples
> Does it make a sense if I make a dedicated HVM domain with linux only 
> for the purpose of vdispl and vkbd backends? Is there a hope this 
> approach will work?
You can try if this approach fits your design and requirements
>
>     >
>     > Juergen
>     >
>     [1]
>     https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$>
>
> Best regards,
>   Alexander Sychev
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Santucco]

[-- Attachment #1.1: Type: text/plain, Size: 4445 bytes --]


Hello again,
 
I have not yet made to work my drm client, so I have tried to run linux like a domU (to see how it should work), it doesn’t work too — displ_be catches SIGSEGV:

#0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
#1  0x000055723b9c5bec in Drm::DumbDrm::copy (this=0x7f4adc000e00) at /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
#2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
 
It tries to copy to mBuffer with non-accessible address.
 
For the moment I see a strange offset for mmap call of /dev/drm/card0 in the strace log — 0x100000000. Is that normal?
 
Any direction of which to dig will be very helpful.
 
Configuration details:
Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel GNU/Linux
DomU: Linux 4.20.17-gentoo
last xen-troops/libxenbe and xen-troops/displ_be
 
Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a backtrace of gdb and kernel configs are attached.
Thanks in advance.
 
On 1/8/20 5:38 PM, Santucco wrote:
> Thank you very much for all your answers.
>
> Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
> < oleksandr_andrushchenko@epam.com
> < /compose?To=oleksandr_andrushchenko@epam.com >>:
> On 1/6/20 10:38 AM, Jürgen Groß wrote:
> > On 06.01.20 08:56, Santucco wrote:
> >> Hello,
> >>
> >> I’m trying to use vdispl interface from PV OS, it doesn’t work.
> >> Configuration details:
> >>      Xen 4.12.1
> >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK
> 2019
> >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
> GNU/Linux
> >>      DomU: x86 Plan9, PV
> >>      displ_be as a backend for vdispl and vkb
> >>
> >> when VM starts, displ_be reports about an error:
> >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument
> >> (displ_be.log:221)
> >>
> >> related Dom0 output is:
> >> [  191.579278] Cannot provide dma-buf: use_ptemode 1
> >> (dmesg.create.log:123)
> >
> > This seems to be a limitation of the xen dma-buf driver. It was
> written
> > for being used on ARM initially where PV is not available.
> This is true and we never tried/targeted PV domains with this
> implementation,
> so if there is a need for that someone has to take a look on the
> proper
> implementation for PV…
>
> Have I got your right and there is no the proper implementation :-)?
There is no
>
> >
> > CC-ing Oleksandr Andrushchenko who is the author of that driver. He
> > should be able to tell us what would be needed to enable PV dom0.
> >
> > Depending on your use case it might be possible to use PVH dom0, but
> > support for this mode is "experimental" only and some features
> are not
> > yet working.
> >
> Well, one of the workarounds possible is to drop zero-copying use-case
> (this is why display backend tries to create dmu-bufs from grants
> passed
> by the guest domain and fails because of "Cannot provide dma-buf:
> use_ptemode 1")
> So, in this case display backend will do memory copying for the
> incoming
> frames
> and won't touch DMABUF_EXP_FROM_REFS ioctl.
> To do so just disable zero-copying while building the backend [1]
>
> Thanks, I have just tried the workaround.  The backend has failed
> in an other place not corresponding with dma_buf.
> Anyway it is enough to continue debugging  my frontend implementation.
> Do you know how big is performance penalty in comparison with
> the zero-copy variant?
Well, it solely depends on your setup, so I cannot tell what
would the numbers be in your case. Comparing to what I have doesn't
make any sense to me: one should compare apples to apples
> Does it make a sense if I make a dedicated HVM domain with linux only
> for the purpose of vdispl and vkbd backends? Is there a hope this
> approach will work?
You can try if this approach fits your design and requirements
>
> >
> > Juergen
> >
> [1]
>  https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$ >
>
> Best regards,
>   Alexander Sychev
 
 

[-- Attachment #1.2: Type: text/html, Size: 6468 bytes --]

[-- Attachment #2: logs.tgz --]
[-- Type: application/x-gzip, Size: 79032 bytes --]

[-- Attachment #3: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Oleksandr Andrushchenko]

On 2/1/20 4:39 PM, Santucco wrote:
> Hello again,
> I have not yet made to work my drm client, so I have tried to run 
> linux like a domU (to see how it should work), it doesn’t work too 
> — displ_be catches SIGSEGV:
>
> #0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
> #1  0x000055723b9c5bec in Drm::DumbDrm::copy (this=0x7f4adc000e00) at 
> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
> #2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy 
> (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at 
> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
> It tries to copy to mBuffer with non-accessible address.
> For the moment I see a strange offset for mmap call of /dev/drm/card0 
> in the strace log — 0x100000000. Is that normal?
> Any direction of which to dig will be very helpful.
> Configuration details:
> Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK 
> 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel GNU/Linux
> DomU: Linux 4.20.17-gentoo
> last xen-troops/libxenbe and xen-troops/displ_be
> Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a backtrace 
> of gdb and kernel configs are attached.
> Thanks in advance.
Could you please try Dom0 kernel WITHOUT the options below:
CONFIG_XEN_GNTDEV_DMABUF=y
CONFIG_XEN_GRANT_DMA_ALLOC=y

Then, just to make sure, did you build displ_be without zero-copy support?

> On 1/8/20 5:38 PM, Santucco wrote:
> > Thank you very much for all your answers.
> >
> > Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
> > <oleksandr_andrushchenko@epam.com
> > </compose?To=oleksandr_andrushchenko@epam.com>>:
> > On 1/6/20 10:38 AM, Jürgen Groß wrote:
> > > On 06.01.20 08:56, Santucco wrote:
> > >> Hello,
> > >>
> > >> I’m trying to use vdispl interface from PV OS, it doesn’t work.
> > >> Configuration details:
> > >>      Xen 4.12.1
> > >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK
> > 2019
> > >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
> > GNU/Linux
> > >>      DomU: x86 Plan9, PV
> > >>      displ_be as a backend for vdispl and vkb
> > >>
> > >> when VM starts, displ_be reports about an error:
> > >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument
> > >> (displ_be.log:221)
> > >>
> > >> related Dom0 output is:
> > >> [  191.579278] Cannot provide dma-buf: use_ptemode 1
> > >> (dmesg.create.log:123)
> > >
> > > This seems to be a limitation of the xen dma-buf driver. It was
> > written
> > > for being used on ARM initially where PV is not available.
> > This is true and we never tried/targeted PV domains with this
> > implementation,
> > so if there is a need for that someone has to take a look on the
> > proper
> > implementation for PV…
> >
> > Have I got your right and there is no the proper implementation :-)?
> There is no
> >
> > >
> > > CC-ing Oleksandr Andrushchenko who is the author of that driver. He
> > > should be able to tell us what would be needed to enable PV dom0.
> > >
> > > Depending on your use case it might be possible to use PVH dom0, but
> > > support for this mode is "experimental" only and some features
> > are not
> > > yet working.
> > >
> > Well, one of the workarounds possible is to drop zero-copying use-case
> > (this is why display backend tries to create dmu-bufs from grants
> > passed
> > by the guest domain and fails because of "Cannot provide dma-buf:
> > use_ptemode 1")
> > So, in this case display backend will do memory copying for the
> > incoming
> > frames
> > and won't touch DMABUF_EXP_FROM_REFS ioctl.
> > To do so just disable zero-copying while building the backend [1]
> >
> > Thanks, I have just tried the workaround.  The backend has failed
> > in an other place not corresponding with dma_buf.
> > Anyway it is enough to continue debugging  my frontend implementation.
> > Do you know how big is performance penalty in comparison with
> > the zero-copy variant?
> Well, it solely depends on your setup, so I cannot tell what
> would the numbers be in your case. Comparing to what I have doesn't
> make any sense to me: one should compare apples to apples
> > Does it make a sense if I make a dedicated HVM domain with linux only
> > for the purpose of vdispl and vkbd backends? Is there a hope this
> > approach will work?
> You can try if this approach fits your design and requirements
> >
> > >
> > > Juergen
> > >
> > [1]
> > 
> https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12 
> <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!gi81oZZNvWaFWUVnaZluA_mNBAItLMd4RZmnc-M_FmlpDojqeQQnS7aXSNlbo80re9uOl2wqFA$>
> > 
> <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$>
> >
> > Best regards,
> >   Alexander Sychev
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Oleksandr Andrushchenko]

Hi,

On 4/19/20 14:26, Santucco wrote:
> Hello,
> I have found a source of the problem.
> In displ_be,  BaseDump copies to the drm buffer with a size from 
> i915 drm driver, but this size a bit more than a size of my frontend 
> display buffer. I have made a quick and dirty fix, a copy a line of my 
> display buffer to a middle of a line of the drm display buffer. Patch 
> is attached.

Thank you for the patch and your efforts to fix the issue.

Could you please make a pull request to [1], so we can continue there?

Thank you in advance,

Oleksandr

> Best regards,
> Alexander
>
>     Четверг, 6 февраля 2020, 11:20 +03:00 от Oleksandr Andrushchenko
>     <oleksandr_andrushchenko@epam.com>:
>     On 2/5/20 8:59 PM, Santucco wrote:
>     > Hello,
>     > Ok, I  commented out the memcpy call and run the test.
>     > displ_be hasn’t crached, I have seen FLIP events in the log.
>     > But there hasn’t been the black screen, just a blink effect every
>     > couple of seconds.
>     > Logs are attached.
>     Ok, so I believe that frontend - backend (displ_be) communication
>     is ok
>     and there is nothing to do there.
>
>     Next, I would start debugging the following in Xen:
>     (XEN) mm.c:2223:d2v0 Bad L1 flags 80
>     and have a look at [1]. Probably, someone on Xen x86 side can tell
>     if this could be related to the flags at [2].
>
>     > Best regards,
>     > Alexander
>     >
>     > Среда, 5 февраля 2020, 9:31 +03:00 от Oleksandr Andrushchenko
>     > <oleksandr_andrushchenko@epam.com
>     </compose?To=oleksandr_andrushchenko@epam.com>>:
>     > On 2/4/20 10:28 AM, Santucco wrote:
>     > > Hello,
>     > > displ_be was compiled without zero-copy support early.
>     > > I have tried with the recompiled dom0 kernel, a result is the
>     same.
>     > > Logs and configs (+displ_be’s CMakeCache.txt ) are attached.
>     > Ok, yet another test to localize the problem.
>     > Could you please remove memcpy from
>     > #1  0x000055e5a1f28bec in Drm::DumbDrm::copy
>     (this=0x7f9338000e00) at
>     >
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>     > and just memset the destination with 0 or whatever.
>     >
>     > I expect that system won't crash, nothing will be shown (black
>     > screen), but
>     > displ_be will show page flip events in its logs.
>     > > Best regards,
>     > > Alexander
>     > >
>     > > Понедельник, 3 февраля 2020, 10:36 +03:00 от Oleksandr
>     > > Andrushchenko <oleksandr_andrushchenko@epam.com
>     </compose?To=oleksandr_andrushchenko@epam.com>
>     > </compose?To=oleksandr_andrushchenko@epam.com>>:
>     > >
>     > >
>     > > On 2/1/20 4:39 PM, Santucco wrote:
>     > > > Hello again,
>     > > > I have not yet made to work my drm client, so I have tried
>     to run
>     > > > linux like a domU (to see how it should work), it doesn’t
>     work too
>     > > > — displ_be catches SIGSEGV:
>     > > >
>     > > > #0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
>     > > > #1  0x000055723b9c5bec in Drm::DumbDrm::copy
>     > > (this=0x7f4adc000e00) at
>     > > >
>     > >
>     >
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>     > > > #2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy
>     > > > (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at
>     > > >
>     > >
>     >
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
>     > > > It tries to copy to mBuffer with non-accessible address.
>     > > > For the moment I see a strange offset for mmap call of
>     > > /dev/drm/card0
>     > > > in the strace log — 0x100000000. Is that normal?
>     > > > Any direction of which to dig will be very helpful.
>     > > > Configuration details:
>     > > > Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>     > > 11:12:24 MSK
>     > > > 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>     > > GNU/Linux
>     > > > DomU: Linux 4.20.17-gentoo
>     > > > last xen-troops/libxenbe and xen-troops/displ_be
>     > > > Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a
>     > > backtrace
>     > > > of gdb and kernel configs are attached.
>     > > > Thanks in advance.
>     > > Could you please try Dom0 kernel WITHOUT the options below:
>     > > CONFIG_XEN_GNTDEV_DMABUF=y
>     > > CONFIG_XEN_GRANT_DMA_ALLOC=y
>     > >
>     > > Then, just to make sure, did you build displ_be without zero-copy
>     > > support?
>     > >
>     > > > On 1/8/20 5:38 PM, Santucco wrote:
>     > > > > Thank you very much for all your answers.
>     > > > >
>     > > > > Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
>     > > > > <oleksandr_andrushchenko@epam.com
>     </compose?To=oleksandr_andrushchenko@epam.com>
>     > </compose?To=oleksandr_andrushchenko@epam.com>
>     > > > > </compose?To=oleksandr_andrushchenko@epam.com>>:
>     > > > > On 1/6/20 10:38 AM, Jürgen Groß wrote:
>     > > > > > On 06.01.20 08:56, Santucco wrote:
>     > > > > >> Hello,
>     > > > > >>
>     > > > > >> I’m trying to use vdispl interface from PV OS, it doesn’t
>     > work.
>     > > > > >> Configuration details:
>     > > > > >>      Xen 4.12.1
>     > > > > >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>     > 11:12:24 MSK
>     > > > > 2019
>     > > > > >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>     > > > > GNU/Linux
>     > > > > >>      DomU: x86 Plan9, PV
>     > > > > >> displ_be as a backend for vdispl and vkb
>     > > > > >>
>     > > > > >> when VM starts, displ_be reports about an error:
>     > > > > >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid
>     > > argument
>     > > > > >> (displ_be.log:221)
>     > > > > >>
>     > > > > >> related Dom0 output is:
>     > > > > >> [ 191.579278] Cannot provide dma-buf: use_ptemode 1
>     > > > > >> (dmesg.create.log:123)
>     > > > > >
>     > > > > > This seems to be a limitation of the xen dma-buf driver.
>     > It was
>     > > > > written
>     > > > > > for being used on ARM initially where PV is not available.
>     > > > > This is true and we never tried/targeted PV domains with this
>     > > > > implementation,
>     > > > > so if there is a need for that someone has to take a look
>     on the
>     > > > > proper
>     > > > > implementation for PV…
>     > > > >
>     > > > > Have I got your right and there is no the proper
>     > > implementation :-)?
>     > > > There is no
>     > > > >
>     > > > > >
>     > > > > > CC-ing Oleksandr Andrushchenko who is the author of that
>     > > driver. He
>     > > > > > should be able to tell us what would be needed to enable PV
>     > > dom0.
>     > > > > >
>     > > > > > Depending on your use case it might be possible to use PVH
>     > > dom0, but
>     > > > > > support for this mode is "experimental" only and some
>     features
>     > > > > are not
>     > > > > > yet working.
>     > > > > >
>     > > > > Well, one of the workarounds possible is to drop zero-copying
>     > > use-case
>     > > > > (this is why display backend tries to create dmu-bufs from
>     > grants
>     > > > > passed
>     > > > > by the guest domain and fails because of "Cannot provide
>     > dma-buf:
>     > > > > use_ptemode 1")
>     > > > > So, in this case display backend will do memory copying
>     for the
>     > > > > incoming
>     > > > > frames
>     > > > > and won't touch DMABUF_EXP_FROM_REFS ioctl.
>     > > > > To do so just disable zero-copying while building the
>     > backend [1]
>     > > > >
>     > > > > Thanks, I have just tried the workaround. The backend
>     has failed
>     > > > > in an other place not corresponding with dma_buf.
>     > > > > Anyway it is enough to continue debugging  my
>     > > frontend implementation.
>     > > > > Do you know how big is performance penalty in comparison with
>     > > > > the zero-copy variant?
>     > > > Well, it solely depends on your setup, so I cannot tell what
>     > > > would the numbers be in your case. Comparing to what I have
>     > doesn't
>     > > > make any sense to me: one should compare apples to apples
>     > > > > Does it make a sense if I make a dedicated HVM domain with
>     > > linux only
>     > > > > for the purpose of vdispl and vkbd backends? Is there a
>     > hope this
>     > > > > approach will work?
>     > > > You can try if this approach fits your design and requirements
>     > > > >
>     > > > > >
>     > > > > > Juergen
>     > > > > >
>     > > > > [1]
>     > > > >
>     > > >
>     > >
>     >
>     https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mhlQAPS_Ozy57xa_0OR66qc1mjlSEz7lj3MkWCyDDF91BGa7J7-BOYWYcdksplocZvxIZMirWg$>
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kZ1JQFRS2pXj_IuXBhvYhmP9Q_svcLyjCXK9465ULGB4MeiYPRz2cF7lepHggr9UxPU9zOBEUw$>
>     > >
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kvDgy3X0IuSQk7D2DdsGtsjtyGroYbNKOrPG95OpyoAkuBVbFSmzozwfor05jkRl0ita0FumBw$>
>     > > >
>     > >
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!gi81oZZNvWaFWUVnaZluA_mNBAItLMd4RZmnc-M_FmlpDojqeQQnS7aXSNlbo80re9uOl2wqFA$>
>     > > > >
>     > > >
>     > >
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$>
>     > > > >
>     > > > > Best regards,
>     > > > >   Alexander Sychev
>     > > >
>     > >
>     > >
>     >
>     ------------------------------------------------------------------------
>     >
>     [1]
>     https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c#L300
>     <https://urldefense.com/v3/__https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c*L300__;Iw!!GF_29dbcQIUBPA!mhlQAPS_Ozy57xa_0OR66qc1mjlSEz7lj3MkWCyDDF91BGa7J7-BOYWYcdksplocZvzn6flk6Q$>
>     [2]
>     https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c#L319
>     <https://urldefense.com/v3/__https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c*L319__;Iw!!GF_29dbcQIUBPA!mhlQAPS_Ozy57xa_0OR66qc1mjlSEz7lj3MkWCyDDF91BGa7J7-BOYWYcdksplocZvwjOfYJxg$>
>

[1] https://github.com/xen-troops/displ_be

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Santucco]

[-- Attachment #1.1: Type: text/plain, Size: 8423 bytes --]


Ok, thank you very much. 
One more question - is that normal the /dev/drm/card0 device is mapped at offset more then 4Gb?
 
Best regards,
Alexander 
>Четверг, 6 февраля 2020, 11:20 +03:00 от Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>:
> 
>On 2/5/20 8:59 PM, Santucco wrote:
>> Hello,
>> Ok, I  commented out the memcpy call and run the test.
>> displ_be hasn’t crached, I have seen FLIP events in the log.
>> But there hasn’t been the black screen, just a blink effect every
>> couple of seconds.
>> Logs are attached.
>Ok, so I believe that frontend - backend (displ_be) communication is ok
>and there is nothing to do there.
>
>Next, I would start debugging the following in Xen:
>(XEN) mm.c:2223:d2v0 Bad L1 flags 80
>and have a look at [1]. Probably, someone on Xen x86 side can tell
>if this could be related to the flags at [2].
>
>> Best regards,
>> Alexander
>>
>> Среда, 5 февраля 2020, 9:31 +03:00 от Oleksandr Andrushchenko
>> < oleksandr_andrushchenko@epam.com >:
>> On 2/4/20 10:28 AM, Santucco wrote:
>> > Hello,
>> > displ_be was compiled without zero-copy support early.
>> > I have tried with the recompiled dom0 kernel, a result is the same.
>> > Logs and configs (+displ_be’s CMakeCache.txt ) are attached.
>> Ok, yet another test to localize the problem.
>> Could you please remove memcpy from
>> #1  0x000055e5a1f28bec in Drm::DumbDrm::copy (this=0x7f9338000e00) at
>> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>> and just memset the destination with 0 or whatever.
>>
>> I expect that system won't crash, nothing will be shown (black
>> screen), but
>> displ_be will show page flip events in its logs.
>> > Best regards,
>> > Alexander
>> >
>> > Понедельник, 3 февраля 2020, 10:36 +03:00 от Oleksandr
>> > Andrushchenko < oleksandr_andrushchenko@epam.com
>> </compose?To=oleksandr_andrushchenko@epam.com>>:
>> >
>> >
>> > On 2/1/20 4:39 PM, Santucco wrote:
>> > > Hello again,
>> > > I have not yet made to work my drm client, so I have tried to run
>> > > linux like a domU (to see how it should work), it doesn’t work too
>> > > — displ_be catches SIGSEGV:
>> > >
>> > > #0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
>> > > #1  0x000055723b9c5bec in Drm::DumbDrm::copy
>> > (this=0x7f4adc000e00) at
>> > >
>> >
>> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>> > > #2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy
>> > > (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at
>> > >
>> >
>> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
>> > > It tries to copy to mBuffer with non-accessible address.
>> > > For the moment I see a strange offset for mmap call of
>> > /dev/drm/card0
>> > > in the strace log — 0x100000000. Is that normal?
>> > > Any direction of which to dig will be very helpful.
>> > > Configuration details:
>> > > Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>> > 11:12:24 MSK
>> > > 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>> > GNU/Linux
>> > > DomU: Linux 4.20.17-gentoo
>> > > last xen-troops/libxenbe and xen-troops/displ_be
>> > > Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a
>> > backtrace
>> > > of gdb and kernel configs are attached.
>> > > Thanks in advance.
>> > Could you please try Dom0 kernel WITHOUT the options below:
>> > CONFIG_XEN_GNTDEV_DMABUF=y
>> > CONFIG_XEN_GRANT_DMA_ALLOC=y
>> >
>> > Then, just to make sure, did you build displ_be without zero-copy
>> > support?
>> >
>> > > On 1/8/20 5:38 PM, Santucco wrote:
>> > > > Thank you very much for all your answers.
>> > > >
>> > > > Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
>> > > > < oleksandr_andrushchenko@epam.com
>> </compose?To=oleksandr_andrushchenko@epam.com>
>> > > > </compose?To=oleksandr_andrushchenko@epam.com>>:
>> > > > On 1/6/20 10:38 AM, Jürgen Groß wrote:
>> > > > > On 06.01.20 08:56, Santucco wrote:
>> > > > >> Hello,
>> > > > >>
>> > > > >> I’m trying to use vdispl interface from PV OS, it doesn’t
>> work.
>> > > > >> Configuration details:
>> > > > >>      Xen 4.12.1
>> > > > >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>> 11:12:24 MSK
>> > > > 2019
>> > > > >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>> > > > GNU/Linux
>> > > > >>      DomU: x86 Plan9, PV
>> > > > >> displ_be as a backend for vdispl and vkb
>> > > > >>
>> > > > >> when VM starts, displ_be reports about an error:
>> > > > >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid
>> > argument
>> > > > >> (displ_be.log:221)
>> > > > >>
>> > > > >> related Dom0 output is:
>> > > > >> [ 191.579278] Cannot provide dma-buf: use_ptemode 1
>> > > > >> (dmesg.create.log:123)
>> > > > >
>> > > > > This seems to be a limitation of the xen dma-buf driver.
>> It was
>> > > > written
>> > > > > for being used on ARM initially where PV is not available.
>> > > > This is true and we never tried/targeted PV domains with this
>> > > > implementation,
>> > > > so if there is a need for that someone has to take a look on the
>> > > > proper
>> > > > implementation for PV…
>> > > >
>> > > > Have I got your right and there is no the proper
>> > implementation :-)?
>> > > There is no
>> > > >
>> > > > >
>> > > > > CC-ing Oleksandr Andrushchenko who is the author of that
>> > driver. He
>> > > > > should be able to tell us what would be needed to enable PV
>> > dom0.
>> > > > >
>> > > > > Depending on your use case it might be possible to use PVH
>> > dom0, but
>> > > > > support for this mode is "experimental" only and some features
>> > > > are not
>> > > > > yet working.
>> > > > >
>> > > > Well, one of the workarounds possible is to drop zero-copying
>> > use-case
>> > > > (this is why display backend tries to create dmu-bufs from
>> grants
>> > > > passed
>> > > > by the guest domain and fails because of "Cannot provide
>> dma-buf:
>> > > > use_ptemode 1")
>> > > > So, in this case display backend will do memory copying for the
>> > > > incoming
>> > > > frames
>> > > > and won't touch DMABUF_EXP_FROM_REFS ioctl.
>> > > > To do so just disable zero-copying while building the
>> backend [1]
>> > > >
>> > > > Thanks, I have just tried the workaround. The backend has failed
>> > > > in an other place not corresponding with dma_buf.
>> > > > Anyway it is enough to continue debugging  my
>> > frontend implementation.
>> > > > Do you know how big is performance penalty in comparison with
>> > > > the zero-copy variant?
>> > > Well, it solely depends on your setup, so I cannot tell what
>> > > would the numbers be in your case. Comparing to what I have
>> doesn't
>> > > make any sense to me: one should compare apples to apples
>> > > > Does it make a sense if I make a dedicated HVM domain with
>> > linux only
>> > > > for the purpose of vdispl and vkbd backends? Is there a
>> hope this
>> > > > approach will work?
>> > > You can try if this approach fits your design and requirements
>> > > >
>> > > > >
>> > > > > Juergen
>> > > > >
>> > > > [1]
>> > > >
>> > >
>> >
>>  https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kZ1JQFRS2pXj_IuXBhvYhmP9Q_svcLyjCXK9465ULGB4MeiYPRz2cF7lepHggr9UxPU9zOBEUw$ >
>> >
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kvDgy3X0IuSQk7D2DdsGtsjtyGroYbNKOrPG95OpyoAkuBVbFSmzozwfor05jkRl0ita0FumBw$ >
>> > >
>> >
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!gi81oZZNvWaFWUVnaZluA_mNBAItLMd4RZmnc-M_FmlpDojqeQQnS7aXSNlbo80re9uOl2wqFA$ >
>> > > >
>> > >
>> >
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$ >
>> > > >
>> > > > Best regards,
>> > > >   Alexander Sychev
>> > >
>> >
>> >
>> ------------------------------------------------------------------------
>>
>[1]  https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c#L300
>[2]  https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c#L319
 

[-- Attachment #1.2: Type: text/html, Size: 12521 bytes --]

[-- Attachment #2: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Oleksandr Andrushchenko]

On 2/5/20 8:59 PM, Santucco wrote:
> Hello,
> Ok, I  commented out the memcpy call and run the test.
> displ_be hasn’t crached, I have seen FLIP events in the log.
> But there hasn’t been the black screen, just a blink effect every 
> couple of seconds.
> Logs are attached.
Ok, so I believe that frontend - backend (displ_be) communication is ok
and there is nothing to do there.

Next, I would start debugging the following in Xen:
(XEN) mm.c:2223:d2v0 Bad L1 flags 80
and have a look at [1]. Probably, someone on Xen x86 side can tell
if this could be related to the flags at [2].

> Best regards,
> Alexander
>
>     Среда, 5 февраля 2020, 9:31 +03:00 от Oleksandr Andrushchenko
>     <oleksandr_andrushchenko@epam.com>:
>     On 2/4/20 10:28 AM, Santucco wrote:
>     > Hello,
>     > displ_be was compiled without zero-copy support early.
>     > I have tried with the recompiled dom0 kernel, a result is the same.
>     > Logs and configs (+displ_be’s CMakeCache.txt ) are attached.
>     Ok, yet another test to localize the problem.
>     Could you please remove memcpy from
>     #1  0x000055e5a1f28bec in Drm::DumbDrm::copy (this=0x7f9338000e00) at
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>     and just memset the destination with 0 or whatever.
>
>     I expect that system won't crash, nothing will be shown (black
>     screen), but
>     displ_be will show page flip events in its logs.
>     > Best regards,
>     > Alexander
>     >
>     > Понедельник, 3 февраля 2020, 10:36 +03:00 от Oleksandr
>     > Andrushchenko <oleksandr_andrushchenko@epam.com
>     </compose?To=oleksandr_andrushchenko@epam.com>>:
>     >
>     >
>     > On 2/1/20 4:39 PM, Santucco wrote:
>     > > Hello again,
>     > > I have not yet made to work my drm client, so I have tried to run
>     > > linux like a domU (to see how it should work), it doesn’t work too
>     > > — displ_be catches SIGSEGV:
>     > >
>     > > #0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
>     > > #1  0x000055723b9c5bec in Drm::DumbDrm::copy
>     > (this=0x7f4adc000e00) at
>     > >
>     >
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>     > > #2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy
>     > > (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at
>     > >
>     >
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
>     > > It tries to copy to mBuffer with non-accessible address.
>     > > For the moment I see a strange offset for mmap call of
>     > /dev/drm/card0
>     > > in the strace log — 0x100000000. Is that normal?
>     > > Any direction of which to dig will be very helpful.
>     > > Configuration details:
>     > > Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>     > 11:12:24 MSK
>     > > 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>     > GNU/Linux
>     > > DomU: Linux 4.20.17-gentoo
>     > > last xen-troops/libxenbe and xen-troops/displ_be
>     > > Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a
>     > backtrace
>     > > of gdb and kernel configs are attached.
>     > > Thanks in advance.
>     > Could you please try Dom0 kernel WITHOUT the options below:
>     > CONFIG_XEN_GNTDEV_DMABUF=y
>     > CONFIG_XEN_GRANT_DMA_ALLOC=y
>     >
>     > Then, just to make sure, did you build displ_be without zero-copy
>     > support?
>     >
>     > > On 1/8/20 5:38 PM, Santucco wrote:
>     > > > Thank you very much for all your answers.
>     > > >
>     > > > Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
>     > > > <oleksandr_andrushchenko@epam.com
>     </compose?To=oleksandr_andrushchenko@epam.com>
>     > > > </compose?To=oleksandr_andrushchenko@epam.com>>:
>     > > > On 1/6/20 10:38 AM, Jürgen Groß wrote:
>     > > > > On 06.01.20 08:56, Santucco wrote:
>     > > > >> Hello,
>     > > > >>
>     > > > >> I’m trying to use vdispl interface from PV OS, it doesn’t
>     work.
>     > > > >> Configuration details:
>     > > > >>      Xen 4.12.1
>     > > > >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>     11:12:24 MSK
>     > > > 2019
>     > > > >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>     > > > GNU/Linux
>     > > > >>      DomU: x86 Plan9, PV
>     > > > >> displ_be as a backend for vdispl and vkb
>     > > > >>
>     > > > >> when VM starts, displ_be reports about an error:
>     > > > >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid
>     > argument
>     > > > >> (displ_be.log:221)
>     > > > >>
>     > > > >> related Dom0 output is:
>     > > > >> [ 191.579278] Cannot provide dma-buf: use_ptemode 1
>     > > > >> (dmesg.create.log:123)
>     > > > >
>     > > > > This seems to be a limitation of the xen dma-buf driver.
>     It was
>     > > > written
>     > > > > for being used on ARM initially where PV is not available.
>     > > > This is true and we never tried/targeted PV domains with this
>     > > > implementation,
>     > > > so if there is a need for that someone has to take a look on the
>     > > > proper
>     > > > implementation for PV…
>     > > >
>     > > > Have I got your right and there is no the proper
>     > implementation :-)?
>     > > There is no
>     > > >
>     > > > >
>     > > > > CC-ing Oleksandr Andrushchenko who is the author of that
>     > driver. He
>     > > > > should be able to tell us what would be needed to enable PV
>     > dom0.
>     > > > >
>     > > > > Depending on your use case it might be possible to use PVH
>     > dom0, but
>     > > > > support for this mode is "experimental" only and some features
>     > > > are not
>     > > > > yet working.
>     > > > >
>     > > > Well, one of the workarounds possible is to drop zero-copying
>     > use-case
>     > > > (this is why display backend tries to create dmu-bufs from
>     grants
>     > > > passed
>     > > > by the guest domain and fails because of "Cannot provide
>     dma-buf:
>     > > > use_ptemode 1")
>     > > > So, in this case display backend will do memory copying for the
>     > > > incoming
>     > > > frames
>     > > > and won't touch DMABUF_EXP_FROM_REFS ioctl.
>     > > > To do so just disable zero-copying while building the
>     backend [1]
>     > > >
>     > > > Thanks, I have just tried the workaround. The backend has failed
>     > > > in an other place not corresponding with dma_buf.
>     > > > Anyway it is enough to continue debugging  my
>     > frontend implementation.
>     > > > Do you know how big is performance penalty in comparison with
>     > > > the zero-copy variant?
>     > > Well, it solely depends on your setup, so I cannot tell what
>     > > would the numbers be in your case. Comparing to what I have
>     doesn't
>     > > make any sense to me: one should compare apples to apples
>     > > > Does it make a sense if I make a dedicated HVM domain with
>     > linux only
>     > > > for the purpose of vdispl and vkbd backends? Is there a
>     hope this
>     > > > approach will work?
>     > > You can try if this approach fits your design and requirements
>     > > >
>     > > > >
>     > > > > Juergen
>     > > > >
>     > > > [1]
>     > > >
>     > >
>     >
>     https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kZ1JQFRS2pXj_IuXBhvYhmP9Q_svcLyjCXK9465ULGB4MeiYPRz2cF7lepHggr9UxPU9zOBEUw$>
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kvDgy3X0IuSQk7D2DdsGtsjtyGroYbNKOrPG95OpyoAkuBVbFSmzozwfor05jkRl0ita0FumBw$>
>     > >
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!gi81oZZNvWaFWUVnaZluA_mNBAItLMd4RZmnc-M_FmlpDojqeQQnS7aXSNlbo80re9uOl2wqFA$>
>     > > >
>     > >
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$>
>     > > >
>     > > > Best regards,
>     > > >   Alexander Sychev
>     > >
>     >
>     >
>     ------------------------------------------------------------------------
>
[1] https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c#L300
[2] https://elixir.bootlin.com/linux/v5.5/source/drivers/xen/gntdev.c#L319
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Santucco]

[-- Attachment #1.1: Type: text/plain, Size: 6964 bytes --]


Hello,
 
Ok, I  commented out the memcpy call and run the test.
displ_be hasn’t crached, I have seen FLIP events in the log.
But there hasn’t been the black screen, just a blink effect every couple of seconds.
 
Logs are attached.
 
Best regards,
Alexander
 
  
>Среда, 5 февраля 2020, 9:31 +03:00 от Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>:
> 
>On 2/4/20 10:28 AM, Santucco wrote:
>> Hello,
>> displ_be was compiled without zero-copy support early.
>> I have tried with the recompiled dom0 kernel, a result is the same.
>> Logs and configs (+displ_be’s CMakeCache.txt ) are attached.
>Ok, yet another test to localize the problem.
>Could you please remove memcpy from
>#1  0x000055e5a1f28bec in Drm::DumbDrm::copy (this=0x7f9338000e00) at
>/home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>and just memset the destination with 0 or whatever.
>
>I expect that system won't crash, nothing will be shown (black screen), but
>displ_be will show page flip events in its logs.
>> Best regards,
>> Alexander
>>
>> Понедельник, 3 февраля 2020, 10:36 +03:00 от Oleksandr
>> Andrushchenko < oleksandr_andrushchenko@epam.com >:
>>
>>
>> On 2/1/20 4:39 PM, Santucco wrote:
>> > Hello again,
>> > I have not yet made to work my drm client, so I have tried to run
>> > linux like a domU (to see how it should work), it doesn’t work too
>> > — displ_be catches SIGSEGV:
>> >
>> > #0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
>> > #1  0x000055723b9c5bec in Drm::DumbDrm::copy
>> (this=0x7f4adc000e00) at
>> >
>> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>> > #2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy
>> > (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at
>> >
>> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
>> > It tries to copy to mBuffer with non-accessible address.
>> > For the moment I see a strange offset for mmap call of
>> /dev/drm/card0
>> > in the strace log — 0x100000000. Is that normal?
>> > Any direction of which to dig will be very helpful.
>> > Configuration details:
>> > Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>> 11:12:24 MSK
>> > 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>> GNU/Linux
>> > DomU: Linux 4.20.17-gentoo
>> > last xen-troops/libxenbe and xen-troops/displ_be
>> > Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a
>> backtrace
>> > of gdb and kernel configs are attached.
>> > Thanks in advance.
>> Could you please try Dom0 kernel WITHOUT the options below:
>> CONFIG_XEN_GNTDEV_DMABUF=y
>> CONFIG_XEN_GRANT_DMA_ALLOC=y
>>
>> Then, just to make sure, did you build displ_be without zero-copy
>> support?
>>
>> > On 1/8/20 5:38 PM, Santucco wrote:
>> > > Thank you very much for all your answers.
>> > >
>> > > Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
>> > > < oleksandr_andrushchenko@epam.com
>> > > </compose?To=oleksandr_andrushchenko@epam.com>>:
>> > > On 1/6/20 10:38 AM, Jürgen Groß wrote:
>> > > > On 06.01.20 08:56, Santucco wrote:
>> > > >> Hello,
>> > > >>
>> > > >> I’m trying to use vdispl interface from PV OS, it doesn’t work.
>> > > >> Configuration details:
>> > > >>      Xen 4.12.1
>> > > >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK
>> > > 2019
>> > > >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>> > > GNU/Linux
>> > > >>      DomU: x86 Plan9, PV
>> > > >> displ_be as a backend for vdispl and vkb
>> > > >>
>> > > >> when VM starts, displ_be reports about an error:
>> > > >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid
>> argument
>> > > >> (displ_be.log:221)
>> > > >>
>> > > >> related Dom0 output is:
>> > > >> [ 191.579278] Cannot provide dma-buf: use_ptemode 1
>> > > >> (dmesg.create.log:123)
>> > > >
>> > > > This seems to be a limitation of the xen dma-buf driver. It was
>> > > written
>> > > > for being used on ARM initially where PV is not available.
>> > > This is true and we never tried/targeted PV domains with this
>> > > implementation,
>> > > so if there is a need for that someone has to take a look on the
>> > > proper
>> > > implementation for PV…
>> > >
>> > > Have I got your right and there is no the proper
>> implementation :-)?
>> > There is no
>> > >
>> > > >
>> > > > CC-ing Oleksandr Andrushchenko who is the author of that
>> driver. He
>> > > > should be able to tell us what would be needed to enable PV
>> dom0.
>> > > >
>> > > > Depending on your use case it might be possible to use PVH
>> dom0, but
>> > > > support for this mode is "experimental" only and some features
>> > > are not
>> > > > yet working.
>> > > >
>> > > Well, one of the workarounds possible is to drop zero-copying
>> use-case
>> > > (this is why display backend tries to create dmu-bufs from grants
>> > > passed
>> > > by the guest domain and fails because of "Cannot provide dma-buf:
>> > > use_ptemode 1")
>> > > So, in this case display backend will do memory copying for the
>> > > incoming
>> > > frames
>> > > and won't touch DMABUF_EXP_FROM_REFS ioctl.
>> > > To do so just disable zero-copying while building the backend [1]
>> > >
>> > > Thanks, I have just tried the workaround. The backend has failed
>> > > in an other place not corresponding with dma_buf.
>> > > Anyway it is enough to continue debugging  my
>> frontend implementation.
>> > > Do you know how big is performance penalty in comparison with
>> > > the zero-copy variant?
>> > Well, it solely depends on your setup, so I cannot tell what
>> > would the numbers be in your case. Comparing to what I have doesn't
>> > make any sense to me: one should compare apples to apples
>> > > Does it make a sense if I make a dedicated HVM domain with
>> linux only
>> > > for the purpose of vdispl and vkbd backends? Is there a hope this
>> > > approach will work?
>> > You can try if this approach fits your design and requirements
>> > >
>> > > >
>> > > > Juergen
>> > > >
>> > > [1]
>> > >
>> >
>>  https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kvDgy3X0IuSQk7D2DdsGtsjtyGroYbNKOrPG95OpyoAkuBVbFSmzozwfor05jkRl0ita0FumBw$ >
>> >
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!gi81oZZNvWaFWUVnaZluA_mNBAItLMd4RZmnc-M_FmlpDojqeQQnS7aXSNlbo80re9uOl2wqFA$ >
>> > >
>> >
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$ >
>> > >
>> > > Best regards,
>> > >   Alexander Sychev
>> >
>>
>> ------------------------------------------------------------------------
 
 

[-- Attachment #1.2: Type: text/html, Size: 10002 bytes --]

[-- Attachment #2: logs.tgz --]
[-- Type: application/x-gzip, Size: 96796 bytes --]

[-- Attachment #3: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Oleksandr Andrushchenko]

On 2/4/20 10:28 AM, Santucco wrote:
> Hello,
> displ_be was compiled without zero-copy support early.
> I have tried with the recompiled dom0 kernel, a result is the same.
> Logs and configs (+displ_be’s CMakeCache.txt ) are attached.
Ok, yet another test to localize the problem.
Could you please remove memcpy from
#1  0x000055e5a1f28bec in Drm::DumbDrm::copy (this=0x7f9338000e00) at 
/home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
and just memset the destination with 0 or whatever.

I expect that system won't crash, nothing will be shown (black screen), but
displ_be will show page flip events in its logs.
> Best regards,
> Alexander
>
>     Понедельник, 3 февраля 2020, 10:36 +03:00 от Oleksandr
>     Andrushchenko <oleksandr_andrushchenko@epam.com>:
>
>
>     On 2/1/20 4:39 PM, Santucco wrote:
>     > Hello again,
>     > I have not yet made to work my drm client, so I have tried to run
>     > linux like a domU (to see how it should work), it doesn’t work too
>     > — displ_be catches SIGSEGV:
>     >
>     > #0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
>     > #1  0x000055723b9c5bec in Drm::DumbDrm::copy
>     (this=0x7f4adc000e00) at
>     >
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>     > #2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy
>     > (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at
>     >
>     /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
>     > It tries to copy to mBuffer with non-accessible address.
>     > For the moment I see a strange offset for mmap call of
>     /dev/drm/card0
>     > in the strace log — 0x100000000. Is that normal?
>     > Any direction of which to dig will be very helpful.
>     > Configuration details:
>     > Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28
>     11:12:24 MSK
>     > 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>     GNU/Linux
>     > DomU: Linux 4.20.17-gentoo
>     > last xen-troops/libxenbe and xen-troops/displ_be
>     > Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a
>     backtrace
>     > of gdb and kernel configs are attached.
>     > Thanks in advance.
>     Could you please try Dom0 kernel WITHOUT the options below:
>     CONFIG_XEN_GNTDEV_DMABUF=y
>     CONFIG_XEN_GRANT_DMA_ALLOC=y
>
>     Then, just to make sure, did you build displ_be without zero-copy
>     support?
>
>     > On 1/8/20 5:38 PM, Santucco wrote:
>     > > Thank you very much for all your answers.
>     > >
>     > > Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
>     > > <oleksandr_andrushchenko@epam.com
>     > > </compose?To=oleksandr_andrushchenko@epam.com>>:
>     > > On 1/6/20 10:38 AM, Jürgen Groß wrote:
>     > > > On 06.01.20 08:56, Santucco wrote:
>     > > >> Hello,
>     > > >>
>     > > >> I’m trying to use vdispl interface from PV OS, it doesn’t work.
>     > > >> Configuration details:
>     > > >>      Xen 4.12.1
>     > > >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK
>     > > 2019
>     > > >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>     > > GNU/Linux
>     > > >>      DomU: x86 Plan9, PV
>     > > >> displ_be as a backend for vdispl and vkb
>     > > >>
>     > > >> when VM starts, displ_be reports about an error:
>     > > >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid
>     argument
>     > > >> (displ_be.log:221)
>     > > >>
>     > > >> related Dom0 output is:
>     > > >> [ 191.579278] Cannot provide dma-buf: use_ptemode 1
>     > > >> (dmesg.create.log:123)
>     > > >
>     > > > This seems to be a limitation of the xen dma-buf driver. It was
>     > > written
>     > > > for being used on ARM initially where PV is not available.
>     > > This is true and we never tried/targeted PV domains with this
>     > > implementation,
>     > > so if there is a need for that someone has to take a look on the
>     > > proper
>     > > implementation for PV…
>     > >
>     > > Have I got your right and there is no the proper
>     implementation :-)?
>     > There is no
>     > >
>     > > >
>     > > > CC-ing Oleksandr Andrushchenko who is the author of that
>     driver. He
>     > > > should be able to tell us what would be needed to enable PV
>     dom0.
>     > > >
>     > > > Depending on your use case it might be possible to use PVH
>     dom0, but
>     > > > support for this mode is "experimental" only and some features
>     > > are not
>     > > > yet working.
>     > > >
>     > > Well, one of the workarounds possible is to drop zero-copying
>     use-case
>     > > (this is why display backend tries to create dmu-bufs from grants
>     > > passed
>     > > by the guest domain and fails because of "Cannot provide dma-buf:
>     > > use_ptemode 1")
>     > > So, in this case display backend will do memory copying for the
>     > > incoming
>     > > frames
>     > > and won't touch DMABUF_EXP_FROM_REFS ioctl.
>     > > To do so just disable zero-copying while building the backend [1]
>     > >
>     > > Thanks, I have just tried the workaround. The backend has failed
>     > > in an other place not corresponding with dma_buf.
>     > > Anyway it is enough to continue debugging  my
>     frontend implementation.
>     > > Do you know how big is performance penalty in comparison with
>     > > the zero-copy variant?
>     > Well, it solely depends on your setup, so I cannot tell what
>     > would the numbers be in your case. Comparing to what I have doesn't
>     > make any sense to me: one should compare apples to apples
>     > > Does it make a sense if I make a dedicated HVM domain with
>     linux only
>     > > for the purpose of vdispl and vkbd backends? Is there a hope this
>     > > approach will work?
>     > You can try if this approach fits your design and requirements
>     > >
>     > > >
>     > > > Juergen
>     > > >
>     > > [1]
>     > >
>     >
>     https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!kvDgy3X0IuSQk7D2DdsGtsjtyGroYbNKOrPG95OpyoAkuBVbFSmzozwfor05jkRl0ita0FumBw$>
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!gi81oZZNvWaFWUVnaZluA_mNBAItLMd4RZmnc-M_FmlpDojqeQQnS7aXSNlbo80re9uOl2wqFA$>
>     > >
>     >
>     <https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$>
>     > >
>     > > Best regards,
>     > >   Alexander Sychev
>     >
>
> ------------------------------------------------------------------------
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] PV DRM doesn't work without auto_translated_physmap feature in Dom0

[Santucco]

[-- Attachment #1.1: Type: text/plain, Size: 5635 bytes --]


Hello,
 
displ_be was compiled without zero-copy support early.
I have tried with the recompiled dom0 kernel, a result is the same.
 
Logs and configs (+displ_be’s CMakeCache.txt ) are attached.
 
 
Best regards,
Alexander
 
 
>Понедельник, 3 февраля 2020, 10:36 +03:00 от Oleksandr Andrushchenko < oleksandr_andrushchenko@epam.com >:
> 
>
>
>On 2/1/20 4:39 PM, Santucco wrote:
>> Hello again,
>> I have not yet made to work my drm client, so I have tried to run
>> linux like a domU (to see how it should work), it doesn’t work too
>> — displ_be catches SIGSEGV:
>>
>> #0  0x00007f4afed1c161 in ?? () from /lib64/libc.so.6
>> #1  0x000055723b9c5bec in Drm::DumbDrm::copy (this=0x7f4adc000e00) at
>> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/drm/Dumb.cpp:149
>> #2  0x000055723b9a8f51 in BuffersStorage::getFrameBufferAndCopy
>> (this=0x7f4ae00010e0, fbCookie=18446612682295083264) at
>> /home/santucco/tmp/xen-troops/displ_be/src/displayBackend/BuffersStorage.cpp:165
>> It tries to copy to mBuffer with non-accessible address.
>> For the moment I see a strange offset for mmap call of /dev/drm/card0
>> in the strace log — 0x100000000. Is that normal?
>> Any direction of which to dig will be very helpful.
>> Configuration details:
>> Xen 4.12.1 Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK
>> 2019 x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel GNU/Linux
>> DomU: Linux 4.20.17-gentoo
>> last xen-troops/libxenbe and xen-troops/displ_be
>> Logs (dmesg, xl dmesg, displ_be, strace log of displ_be), a backtrace
>> of gdb and kernel configs are attached.
>> Thanks in advance.
>Could you please try Dom0 kernel WITHOUT the options below:
>CONFIG_XEN_GNTDEV_DMABUF=y
>CONFIG_XEN_GRANT_DMA_ALLOC=y
>
>Then, just to make sure, did you build displ_be without zero-copy support?
>
>> On 1/8/20 5:38 PM, Santucco wrote:
>> > Thank you very much for all your answers.
>> >
>> > Среда, 8 января 2020, 10:54 +03:00 от Oleksandr Andrushchenko
>> > < oleksandr_andrushchenko@epam.com
>> > < /compose?To=oleksandr_andrushchenko@epam.com >>:
>> > On 1/6/20 10:38 AM, Jürgen Groß wrote:
>> > > On 06.01.20 08:56, Santucco wrote:
>> > >> Hello,
>> > >>
>> > >> I’m trying to use vdispl interface from PV OS, it doesn’t work.
>> > >> Configuration details:
>> > >>      Xen 4.12.1
>> > >>      Dom0: Linux 4.20.17-gentoo #13 SMP Sat Dec 28 11:12:24 MSK
>> > 2019
>> > >> x86_64 Intel(R) Celeron(R) CPU N3050 @ 1.60GHz GenuineIntel
>> > GNU/Linux
>> > >>      DomU: x86 Plan9, PV
>> > >>      displ_be as a backend for vdispl and vkb
>> > >>
>> > >> when VM starts, displ_be reports about an error:
>> > >> gnttab: error: ioctl DMABUF_EXP_FROM_REFS failed: Invalid argument
>> > >> (displ_be.log:221)
>> > >>
>> > >> related Dom0 output is:
>> > >> [  191.579278] Cannot provide dma-buf: use_ptemode 1
>> > >> (dmesg.create.log:123)
>> > >
>> > > This seems to be a limitation of the xen dma-buf driver. It was
>> > written
>> > > for being used on ARM initially where PV is not available.
>> > This is true and we never tried/targeted PV domains with this
>> > implementation,
>> > so if there is a need for that someone has to take a look on the
>> > proper
>> > implementation for PV…
>> >
>> > Have I got your right and there is no the proper implementation :-)?
>> There is no
>> >
>> > >
>> > > CC-ing Oleksandr Andrushchenko who is the author of that driver. He
>> > > should be able to tell us what would be needed to enable PV dom0.
>> > >
>> > > Depending on your use case it might be possible to use PVH dom0, but
>> > > support for this mode is "experimental" only and some features
>> > are not
>> > > yet working.
>> > >
>> > Well, one of the workarounds possible is to drop zero-copying use-case
>> > (this is why display backend tries to create dmu-bufs from grants
>> > passed
>> > by the guest domain and fails because of "Cannot provide dma-buf:
>> > use_ptemode 1")
>> > So, in this case display backend will do memory copying for the
>> > incoming
>> > frames
>> > and won't touch DMABUF_EXP_FROM_REFS ioctl.
>> > To do so just disable zero-copying while building the backend [1]
>> >
>> > Thanks, I have just tried the workaround.  The backend has failed
>> > in an other place not corresponding with dma_buf.
>> > Anyway it is enough to continue debugging  my frontend implementation.
>> > Do you know how big is performance penalty in comparison with
>> > the zero-copy variant?
>> Well, it solely depends on your setup, so I cannot tell what
>> would the numbers be in your case. Comparing to what I have doesn't
>> make any sense to me: one should compare apples to apples
>> > Does it make a sense if I make a dedicated HVM domain with linux only
>> > for the purpose of vdispl and vkbd backends? Is there a hope this
>> > approach will work?
>> You can try if this approach fits your design and requirements
>> >
>> > >
>> > > Juergen
>> > >
>> > [1]
>> >
>>  https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt#L12
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!gi81oZZNvWaFWUVnaZluA_mNBAItLMd4RZmnc-M_FmlpDojqeQQnS7aXSNlbo80re9uOl2wqFA$ >
>> >
>> < https://urldefense.com/v3/__https://github.com/xen-troops/displ_be/blob/master/CMakeLists.txt*L12__;Iw!!GF_29dbcQIUBPA!mz3gn1wQMX2DXeNuAV-1_dI7nxFYYZOgdPiJNSFMesCz9lAzOKlwVPlddbxbcLmUO44NOy0TFA$ >
>> >
>> > Best regards,
>> >   Alexander Sychev
>>
 
   
----------------------------------------------------------------------

[-- Attachment #1.2: Type: text/html, Size: 8809 bytes --]

[-- Attachment #2: logs.tgz --]
[-- Type: application/x-gzip, Size: 73345 bytes --]

[-- Attachment #3: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel