Windows Support Plan

Hi everyone!

Where do I find a high-level plan of comprehensive Windows support: design decisions, implementation order, etc?

Information on the subject is very scarce, one may think it is abandoned. Googling for "site:dpdk.org/ml/archives/dev/ windows" yields only two pages of disjoint messages. I learned about "netuio" days ago from a tiny remark in a "Minutes of Technical Board Meetings" email, and even then it took enumerating "dpdk-next-windows" branches to find the source.

The matter is, as a New Year's holiday project of mine I implemented Windows support from scratch to the point it runs in QEMU with virtio-pci [0]. It is not of production quality, cuts some corners and lacks major features (see bottom). My primary goal was fun^W making it work. Comparing it to "windpdk-v18.08" branch of "dpdk-next-windows", I can see that 1) our implementations take rather different approaches in some cases, and 2) both have severe issues and would benefit from amalgamation. I'd like to contribute to Windows support with this code, but to do so, coordination is required, because changes are significant.

Primary topics to discuss:

1. Memory management (@Anatoly)

1.1. MM changed radically since v18.08 and dpdk-next-windows does not
implement it properly anyway, it allocates segment lists in a PCI bus driver. My implementation closely follows the Linux one using VirtualAlloc2() with XXX_PLACEHOLDER flags to reserve and commit memory, but does not map hugepages to files. Is there
a consensus on MM approach in Windows?

Anyway, I think EAL private MM API would have to be changed, because memory reservation, allocation, and mapping are completely different operations. Hiding this with an mmap() shim doesn't look right, because mmap()'s behavior differs even among Unix platforms.

1.2. In Windows, there is no /dev/mem to implement rte_virt2iova(),
so a simple kernel driver is required for mapping. Moreover, Windows kernel abstracts IOMMU, so those physical addresses may be unsuitable for DMA at all (see below).

2. Kernel drivers (@Harini, @Stephen)

2.1. The most serious issue is that Windows formally prohibits using
arbitrary physical addresses with DMA in favor of allocating special buffers (presumably because IOMMU may be engaged, and there is no way to check). We can either live with it (technically, everything works with PA mode), or we could revive DMA allocation API from ethdev to ask the driver for a proper DMA buffer.

2.2. Neither netuio, nor my driver (userpci) support interrupts.
I see not inherent difficulty here, but interface should be designed carefully.

2.3. Windows allows mapping I/O ports into user-space, but there is
no API to change IOPL, which makes mapping useless and requires a syscall for every I/O port access. This demolishes virtio-legacy performance. Perhaps Microsoft could give some advice here. OTOH, PIO is all legacy, so might be much effort is not justified.

2.4. I believe GUIDs approach for identifying compatible devices
should be strictly preferred, and not DosDevices symlinks. Think of Mellanox OFED on Linux, which uses a different driver, but could provide a compatible interface. Another reason is that a single driver can implement multiple kernel interfaces with appropriate GUIDs.

2.5. DPDK Windows driver guidelines, driver review, and certification.
The quality of both netuio and userpci is below standards now (e. g. netuio does not mind its context when mapping memory, and userpci lacks synchronization), code style is a mix of Windows and DPDK, logging may be insufficient.

3. POSIX shim vs EAL wrappers (@Thomas, @Pallavi, @Ranjit)

What is the policy: to implement a POSIX shim in EAL (as the latest patches from Pallavi Kadam do), or to add dependencies (as [1] suggests)? IMO creating a shim is wrong. First, some POSIX concepts do not easily map to Windows, like poll() interface and I/O model in general. Second, there are numerous getopt, pthread, etc. implementations for Windows, no point wasting resources and repeat them, adding bugs. I can think of two exceptions:

* <sys/queue.h>, which is header-only.

* Berkeley sockets. Adding <winsock2.h> to public headers creates more trouble that its worth: definitions for a few structures and constants. May be there should be some <rte_socket.h> to abstract platform differences.

Some highlights on my implementation:

* Major features NOT supported:

* multi-process (due to limited time)
* interrupts (limited time + explained above)
* eventdev (requires access to physical memory)
* hot-plug (due to limited time and Windows knowledge)
* bbdev (see comments in config/common_windows)
* FreeBSD (trivial, I just don't use it)

* DPDK is built using MinGW-w64 with GNUmake or Meson. Drivers are built using DDK (msbuild or Visual Studio). Actually, I cross-compile DPDK and build drivers natively.

* Only tested on Windows 10 in QEMU with virtio-legacy.

* No docs, but there's nothing unusual for those familiar with Windows. Bind virt2phys driver to Root\virt2phys, bind userpci driver to device(s).

* Commit history is squashed, because it was a mess from experiments. There also may be some leftover changes, but those commits are not proper patches anyway.

References:

[0]: https://github.com/PlushBeaver/dpdk/commits/windows
[1]: http://mails.dpdk.org/archives/dev/2015-February/014245.html

--
Dmitry Kozlyuk

Hi,

I haven't really been following the Windows port much so i have no idea of how it works for now.

The main reason DPDK memory management works the way it does is because of need to support multiprocess. In order to map memory in all processes, we need that space reserved (otherwise there's no guarantee that the newly mapped memory segment will be mapped in all processes, and it'll cause runtime failure). If it wasn't for that, we could allocate memory arbitrarily and as needed. Windows should either follow this model, or drop secondary support and go its own way - the internals are OS-specific anyway.

If there are changes needed to private memalloc API to support the above, that's completely fine - that's why all of this stuff is internal-only :) As long as public API stays roughly the same, we should be good. Bear in mind that DPDK also supports external memory, you might need to make some allowances for that too.

As for IOMMU - we don't support IOVA as VA addressing on FreeBSD, so if Windows port can only work with IOVA as PA, that's fine too. The question of IOVA mode really boils down to, do we control the DMA addresses (IOVA as VA mode), or does the system (IOVA as PA). I'm not familiar with how IOMMU works on Windows, but as long as it fits into that model and we keep the API, it should also be OK :)

Thanks,
Anatoly

You should talk to the Windows DPDK developers.
They have been presenting regularly at dpdk summits. Look up videos for more info.

The initial port is focused on running DPDK on bare metal with Intel NIC. Your version looks more aligned with Windows as guest in KVM.

Get Outlook for Android<https://aka.ms/ghei36>
________________________________
From: Dmitry Kozliuk <[email protected]>
Sent: Sunday, February 2, 2020 9:37:36 PM
To: [email protected] <[email protected]>
Cc: Thomas Monjalon <[email protected]>; Pallavi Kadam <[email protected]>; Anatoly Burakov <[email protected]>; Ranjit Menon <[email protected]>; Harini Ramakrishnan <[email protected]>; Stephen Hemminger <[email protected]>
Subject: [EXTERNAL] Windows Support Plan

Hi everyone!

Where do I find a high-level plan of comprehensive Windows support: design decisions, implementation order, etc?

Information on the subject is very scarce, one may think it is abandoned. Googling for "site:dpdk.org/ml/archives/dev/ windows" yields only two pages of disjoint messages. I learned about "netuio" days ago from a tiny remark in a "Minutes of Technical Board Meetings" email, and even then it took enumerating "dpdk-next-windows" branches to find the source.

The matter is, as a New Year's holiday project of mine I implemented Windows support from scratch to the point it runs in QEMU with virtio-pci [0]. It is not of production quality, cuts some corners and lacks major features (see bottom). My primary goal was fun^W making it work. Comparing it to "windpdk-v18.08" branch of "dpdk-next-windows", I can see that 1) our implementations take rather different approaches in some cases, and 2) both have severe issues and would benefit from amalgamation. I'd like to contribute to Windows support with this code, but to do so, coordination is required, because changes are significant.

Primary topics to discuss:

1. Memory management (@Anatoly)

1.1. MM changed radically since v18.08 and dpdk-next-windows does not
implement it properly anyway, it allocates segment lists in a PCI bus driver. My implementation closely follows the Linux one using VirtualAlloc2() with XXX_PLACEHOLDER flags to reserve and commit memory, but does not map hugepages to files. Is there
a consensus on MM approach in Windows?

Anyway, I think EAL private MM API would have to be changed, because memory reservation, allocation, and mapping are completely different operations. Hiding this with an mmap() shim doesn't look right, because mmap()'s behavior differs even among Unix platforms.

1.2. In Windows, there is no /dev/mem to implement rte_virt2iova(),
so a simple kernel driver is required for mapping. Moreover, Windows kernel abstracts IOMMU, so those physical addresses may be unsuitable for DMA at all (see below).

2. Kernel drivers (@Harini, @Stephen)

2.1. The most serious issue is that Windows formally prohibits using
arbitrary physical addresses with DMA in favor of allocating special buffers (presumably because IOMMU may be engaged, and there is no way to check). We can either live with it (technically, everything works with PA mode), or we could revive DMA allocation API from ethdev to ask the driver for a proper DMA buffer.

2.2. Neither netuio, nor my driver (userpci) support interrupts.
I see not inherent difficulty here, but interface should be designed carefully.

2.3. Windows allows mapping I/O ports into user-space, but there is
no API to change IOPL, which makes mapping useless and requires a syscall for every I/O port access. This demolishes virtio-legacy performance. Perhaps Microsoft could give some advice here. OTOH, PIO is all legacy, so might be much effort is not justified.

2.4. I believe GUIDs approach for identifying compatible devices
should be strictly preferred, and not DosDevices symlinks. Think of Mellanox OFED on Linux, which uses a different driver, but could provide a compatible interface. Another reason is that a single driver can implement multiple kernel interfaces with appropriate GUIDs.

2.5. DPDK Windows driver guidelines, driver review, and certification.
The quality of both netuio and userpci is below standards now (e. g. netuio does not mind its context when mapping memory, and userpci lacks synchronization), code style is a mix of Windows and DPDK, logging may be insufficient.

3. POSIX shim vs EAL wrappers (@Thomas, @Pallavi, @Ranjit)

What is the policy: to implement a POSIX shim in EAL (as the latest patches from Pallavi Kadam do), or to add dependencies (as [1] suggests)? IMO creating a shim is wrong. First, some POSIX concepts do not easily map to Windows, like poll() interface and I/O model in general. Second, there are numerous getopt, pthread, etc. implementations for Windows, no point wasting resources and repeat them, adding bugs. I can think of two exceptions:

* <sys/queue.h>, which is header-only.

* Berkeley sockets. Adding <winsock2.h> to public headers creates more trouble that its worth: definitions for a few structures and constants. May be there should be some <rte_socket.h> to abstract platform differences.

Some highlights on my implementation:

* Major features NOT supported:

* multi-process (due to limited time)
* interrupts (limited time + explained above)
* eventdev (requires access to physical memory)
* hot-plug (due to limited time and Windows knowledge)
* bbdev (see comments in config/common_windows)
* FreeBSD (trivial, I just don't use it)

* DPDK is built using MinGW-w64 with GNUmake or Meson. Drivers are built using DDK (msbuild or Visual Studio). Actually, I cross-compile DPDK and build drivers natively.

* Only tested on Windows 10 in QEMU with virtio-legacy.

* No docs, but there's nothing unusual for those familiar with Windows. Bind virt2phys driver to Root\virt2phys, bind userpci driver to device(s).

* Commit history is squashed, because it was a mess from experiments. There also may be some leftover changes, but those commits are not proper patches anyway.

References:

[0]: https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FPlushBeaver%2Fdpdk%2Fcommits%2Fwindows&amp;data=02%7C01%7Csthemmin%40microsoft.com%7C485559de220c43a1fe2408d7a81fd5e9%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637162727454625299&amp;sdata=W%2BrqF4EWaBmwEOb7t3fRrKfmu7GkHpIyNJ2us6Dx6QU%3D&amp;reserved=0
[1]: https://nam06.safelinks.protection.outlook.com/?url=http%3A%2F%2Fmails.dpdk.org%2Farchives%2Fdev%2F2015-February%2F014245.html&amp;data=02%7C01%7Csthemmin%40microsoft.com%7C485559de220c43a1fe2408d7a81fd5e9%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637162727454625299&amp;sdata=Hb%2FCD99bjzhDlfrcbKdBN%2FlFkqQyN3F%2BvYlPl1VIz8w%3D&amp;reserved=0

--
Dmitry Kozlyuk

Dmitry...
There is a DPDK Windows community meeting every second Wednesday at 8:00am (Pacific Time). If this time works for you, we can have Harini add you to this meeting series. thanks,

ranjit m.

From: Stephen Hemminger <[email protected]>
Sent: Monday, February 3, 2020 1:16 AM
To: Dmitry Kozliuk <[email protected]>; [email protected]
Cc: Thomas Monjalon <[email protected]>; Kadam, Pallavi <[email protected]>; Burakov, Anatoly <[email protected]>; Menon, Ranjit <[email protected]>; Harini Ramakrishnan <[email protected]>
Subject: Re: [EXTERNAL] Windows Support Plan

You should talk to the Windows DPDK developers.
They have been presenting regularly at dpdk summits. Look up videos for more info.

The initial port is focused on running DPDK on bare metal with Intel NIC. Your version looks more aligned with Windows as guest in KVM.

Get Outlook for Android<https://aka.ms/ghei36>
________________________________
From: Dmitry Kozliuk <[email protected]<mailto:[email protected]>>
Sent: Sunday, February 2, 2020 9:37:36 PM
To: [email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>>
Cc: Thomas Monjalon <[email protected]<mailto:[email protected]>>; Pallavi Kadam <[email protected]<mailto:[email protected]>>; Anatoly Burakov <[email protected]<mailto:[email protected]>>; Ranjit Menon <[email protected]<mailto:[email protected]>>; Harini Ramakrishnan <[email protected]<mailto:[email protected]>>; Stephen Hemminger <[email protected]<mailto:[email protected]>>
Subject: [EXTERNAL] Windows Support Plan

Hi everyone!

Where do I find a high-level plan of comprehensive Windows support: design decisions, implementation order, etc?

Information on the subject is very scarce, one may think it is abandoned. Googling for "site:dpdk.org/ml/archives/dev/ windows" yields only two pages of disjoint messages. I learned about "netuio" days ago from a tiny remark in a "Minutes of Technical Board Meetings" email, and even then it took enumerating "dpdk-next-windows" branches to find the source.

The matter is, as a New Year's holiday project of mine I implemented Windows support from scratch to the point it runs in QEMU with virtio-pci [0]. It is not of production quality, cuts some corners and lacks major features (see bottom). My primary goal was fun^W making it work. Comparing it to "windpdk-v18.08" branch of "dpdk-next-windows", I can see that 1) our implementations take rather different approaches in some cases, and 2) both have severe issues and would benefit from amalgamation. I'd like to contribute to Windows support with this code, but to do so, coordination is required, because changes are significant.

Primary topics to discuss:

1. Memory management (@Anatoly)

1.1. MM changed radically since v18.08 and dpdk-next-windows does not
implement it properly anyway, it allocates segment lists in a PCI bus driver. My implementation closely follows the Linux one using VirtualAlloc2() with XXX_PLACEHOLDER flags to reserve and commit memory, but does not map hugepages to files. Is there
a consensus on MM approach in Windows?

Anyway, I think EAL private MM API would have to be changed, because memory reservation, allocation, and mapping are completely different operations. Hiding this with an mmap() shim doesn't look right, because mmap()'s behavior differs even among Unix platforms.

1.2. In Windows, there is no /dev/mem to implement rte_virt2iova(),
so a simple kernel driver is required for mapping. Moreover, Windows kernel abstracts IOMMU, so those physical addresses may be unsuitable for DMA at all (see below).

2. Kernel drivers (@Harini, @Stephen)

2.1. The most serious issue is that Windows formally prohibits using
arbitrary physical addresses with DMA in favor of allocating special buffers (presumably because IOMMU may be engaged, and there is no way to check). We can either live with it (technically, everything works with PA mode), or we could revive DMA allocation API from ethdev to ask the driver for a proper DMA buffer.

2.2. Neither netuio, nor my driver (userpci) support interrupts.
I see not inherent difficulty here, but interface should be designed carefully.

2.3. Windows allows mapping I/O ports into user-space, but there is
no API to change IOPL, which makes mapping useless and requires a syscall for every I/O port access. This demolishes virtio-legacy performance. Perhaps Microsoft could give some advice here. OTOH, PIO is all legacy, so might be much effort is not justified.

2.4. I believe GUIDs approach for identifying compatible devices
should be strictly preferred, and not DosDevices symlinks. Think of Mellanox OFED on Linux, which uses a different driver, but could provide a compatible interface. Another reason is that a single driver can implement multiple kernel interfaces with appropriate GUIDs.

2.5. DPDK Windows driver guidelines, driver review, and certification.
The quality of both netuio and userpci is below standards now (e. g. netuio does not mind its context when mapping memory, and userpci lacks synchronization), code style is a mix of Windows and DPDK, logging may be insufficient.

3. POSIX shim vs EAL wrappers (@Thomas, @Pallavi, @Ranjit)

What is the policy: to implement a POSIX shim in EAL (as the latest patches from Pallavi Kadam do), or to add dependencies (as [1] suggests)? IMO creating a shim is wrong. First, some POSIX concepts do not easily map to Windows, like poll() interface and I/O model in general. Second, there are numerous getopt, pthread, etc. implementations for Windows, no point wasting resources and repeat them, adding bugs. I can think of two exceptions:

* <sys/queue.h>, which is header-only.

* Berkeley sockets. Adding <winsock2.h> to public headers creates more trouble that its worth: definitions for a few structures and constants. May be there should be some <rte_socket.h> to abstract platform differences.

Some highlights on my implementation:

* Major features NOT supported:

* multi-process (due to limited time)
* interrupts (limited time + explained above)
* eventdev (requires access to physical memory)
* hot-plug (due to limited time and Windows knowledge)
* bbdev (see comments in config/common_windows)
* FreeBSD (trivial, I just don't use it)

* DPDK is built using MinGW-w64 with GNUmake or Meson. Drivers are built using DDK (msbuild or Visual Studio). Actually, I cross-compile DPDK and build drivers natively.

* Only tested on Windows 10 in QEMU with virtio-legacy.

* No docs, but there's nothing unusual for those familiar with Windows. Bind virt2phys driver to Root\virt2phys, bind userpci driver to device(s).

* Commit history is squashed, because it was a mess from experiments. There also may be some leftover changes, but those commits are not proper patches anyway.

References:

[0]: https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FPlushBeaver%2Fdpdk%2Fcommits%2Fwindows&amp;data=02%7C01%7Csthemmin%40microsoft.com%7C485559de220c43a1fe2408d7a81fd5e9%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637162727454625299&amp;sdata=W%2BrqF4EWaBmwEOb7t3fRrKfmu7GkHpIyNJ2us6Dx6QU%3D&amp;reserved=0
[1]: https://nam06.safelinks.protection.outlook.com/?url=http%3A%2F%2Fmails.dpdk.org%2Farchives%2Fdev%2F2015-February%2F014245.html&amp;data=02%7C01%7Csthemmin%40microsoft.com%7C485559de220c43a1fe2408d7a81fd5e9%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637162727454625299&amp;sdata=Hb%2FCD99bjzhDlfrcbKdBN%2FlFkqQyN3F%2BvYlPl1VIz8w%3D&amp;reserved=0

--
Dmitry Kozlyuk

Thanks Ranjit, Harini's sent me an invitation for this Wednesday.

Dmitry Kozlyuk

02/02/2020 21:37, Dmitry Kozliuk:

Please help documenting design decisions in the DPDK doc. For implementation order, we'll discuss it soon together.

I agree.
I think Harini will address this lack of information.

You are very welcome.
The work you already did looks amazing and it is very well presented.

[...]

You are right, we should think about adding new dependencies which are easily and generally available.

I do not like the shim layer either.

[...]

As Anatoly said, multi-process is not a priority.
This feature has a high cost, so we should think twice before deciding to support it on Windows.

[...]

Thanks a lot

I think Windows should support multi-process, because there is a demand and an ongoing design effort for multi-tenancy and resource arbitration [0]. Until Windows kernel implements "secure API" for the architecture proposed by
[0] (if it does at all), DPDK multi-process model can to some point support the features desired. For example, a primary process may be a service performing resource arbitration for applications being secondary processes.

I haven't considered external memory yet. Does it need anything beyond mapping VA to IOVA?

AFAIK, Windows doesn't expose IOMMU either to applications or drivers. Do I understand correctly that implies only IOVA as PA can be supported, because mappings can't be set up?

The trouble is, PA cannot generally be used if IOMMU is present, but there is no way to tell if it is. Windows kernel offers API to allocate buffers for DMA [1], but MM doesn't know if it allocates memory for DMA or not, even if that kernel API would be exposed. If I got it right, DPDK just can't be used on Windows with IOMMU enabled (can't tell for VMs that don't see IOMMU).

[0]:
https://www.dpdk.org/wp-content/uploads/sites/35/2018/12/RMenonOCardona_Improving-Security-in-Windows-DPDK.pdf
[1]:
https://docs.microsoft.com/en-us/windows-hardware/drivers/ddi/wdm/nc-wdm-pallocate_common_buffer_ex

Dmitry Kozlyuk

No, just a few checks here and there. The brunt of the mapping is on the shoulders of the user, and "external memory API" is really just registering this memory with DPDK so that calls like rte_virt2memseg() work correctly.

In general, yes. The name probably doesn't match very well, admittedly, but that's the equivalent as far as software is concerned. That's not a problem - FreeBSD doesn't support IOVA as VA either :)

Thanks,
Anatoly