lkml.org 
[lkml]   [2019]   [Mar]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC PATCH Xilinx Alveo 0/6] Xilinx PCIe accelerator driver
On Wed, Mar 27, 2019 at 12:50:14PM +0000, Sonal Santan wrote:
>
>
> > -----Original Message-----
> > From: Daniel Vetter [mailto:daniel@ffwll.ch]
> > Sent: Wednesday, March 27, 2019 1:23 AM
> > To: Sonal Santan <sonals@xilinx.com>
> > Cc: dri-devel@lists.freedesktop.org; gregkh@linuxfoundation.org; Cyril
> > Chemparathy <cyrilc@xilinx.com>; linux-kernel@vger.kernel.org; Lizhi Hou
> > <lizhih@xilinx.com>; Michal Simek <michals@xilinx.com>; airlied@redhat.com
> > Subject: Re: [RFC PATCH Xilinx Alveo 0/6] Xilinx PCIe accelerator driver
> >
> > On Wed, Mar 27, 2019 at 12:30 AM Sonal Santan <sonals@xilinx.com> wrote:
> > >
> > >
> > >
> > > > -----Original Message-----
> > > > From: Daniel Vetter [mailto:daniel.vetter@ffwll.ch] On Behalf Of
> > > > Daniel Vetter
> > > > Sent: Monday, March 25, 2019 1:28 PM
> > > > To: Sonal Santan <sonals@xilinx.com>
> > > > Cc: dri-devel@lists.freedesktop.org; gregkh@linuxfoundation.org;
> > > > Cyril Chemparathy <cyrilc@xilinx.com>; linux-kernel@vger.kernel.org;
> > > > Lizhi Hou <lizhih@xilinx.com>; Michal Simek <michals@xilinx.com>;
> > > > airlied@redhat.com
> > > > Subject: Re: [RFC PATCH Xilinx Alveo 0/6] Xilinx PCIe accelerator
> > > > driver
> > > >
> > > > On Tue, Mar 19, 2019 at 02:53:55PM -0700, sonal.santan@xilinx.com
> > wrote:
> > > > > From: Sonal Santan <sonal.santan@xilinx.com>
> > > > >
> > > > > Hello,
> > > > >
> > > > > This patch series adds drivers for Xilinx Alveo PCIe accelerator cards.
> > > > > These drivers are part of Xilinx Runtime (XRT) open source stack
> > > > > and have been deployed by leading FaaS vendors and many enterprise
> > > > customers.
> > > >
> > > > Cool, first fpga driver submitted to drm! And from a high level I
> > > > think this makes a lot of sense.
> > > >
> > > > > PLATFORM ARCHITECTURE
> > > > >
> > > > > Alveo PCIe platforms have a static shell and a reconfigurable
> > > > > (dynamic) region. The shell is automatically loaded from PROM when
> > > > > host is booted and PCIe is enumerated by BIOS. Shell cannot be
> > > > > changed till next cold reboot. The shell exposes two physical functions:
> > > > > management physical function and user physical function.
> > > > >
> > > > > Users compile their high level design in C/C++/OpenCL or RTL into
> > > > > FPGA image using SDx compiler. The FPGA image packaged as xclbin
> > > > > file can be loaded onto reconfigurable region. The image may
> > > > > contain one or more compute unit. Users can dynamically swap the
> > > > > full image running on the reconfigurable region in order to switch
> > > > > between different
> > > > workloads.
> > > > >
> > > > > XRT DRIVERS
> > > > >
> > > > > XRT Linux kernel driver xmgmt binds to mgmt pf. The driver is
> > > > > modular and organized into several platform drivers which
> > > > > primarily handle the following functionality:
> > > > > 1. ICAP programming (FPGA bitstream download with FPGA Mgr
> > > > > integration) 2. Clock scaling 3. Loading firmware container also
> > > > > called dsabin (embedded Microblaze
> > > > > firmware for ERT and XMC, optional clearing bitstream) 4.
> > > > > In-band
> > > > > sensors: temp, voltage, power, etc.
> > > > > 5. AXI Firewall management
> > > > > 6. Device reset and rescan
> > > > > 7. Hardware mailbox for communication between two physical
> > > > > functions
> > > > >
> > > > > XRT Linux kernel driver xocl binds to user pf. Like its peer, this
> > > > > driver is also modular and organized into several platform drivers
> > > > > which handle the following functionality:
> > > > > 1. Device memory topology discovery and memory management 2.
> > > > > Buffer object abstraction and management for client process 3.
> > > > > XDMA MM PCIe DMA engine programming 4. Multi-process aware
> > context management 5.
> > > > > Compute unit execution management (optionally with help of ERT) for
> > > > > client processes
> > > > > 6. Hardware mailbox for communication between two physical
> > > > > functions
> > > > >
> > > > > The drivers export ioctls and sysfs nodes for various services.
> > > > > xocl driver makes heavy use of DRM GEM features for device memory
> > > > > management, reference counting, mmap support and export/import.
> > > > > xocl also includes a simple scheduler called KDS which schedules
> > > > > compute units and interacts with hardware scheduler running ERT
> > > > > firmware. The scheduler understands custom opcodes packaged into
> > > > > command objects
> > > > and
> > > > > provides an asynchronous command done notification via POSIX poll.
> > > > >
> > > > > More details on architecture, software APIs, ioctl definitions,
> > > > > execution model, etc. is available as Sphinx documentation--
> > > > >
> > > > > https://xilinx.github.io/XRT/2018.3/html/index.html
> > > > >
> > > > > The complete runtime software stack (XRT) which includes out of
> > > > > tree kernel drivers, user space libraries, board utilities and
> > > > > firmware for the hardware scheduler is open source and available
> > > > > at https://github.com/Xilinx/XRT
> > > >
> > > > Before digging into the implementation side more I looked into the
> > > > userspace here. I admit I got lost a bit, since there's lots of
> > > > indirections and abstractions going on, but it seems like this is
> > > > just a fancy ioctl wrapper/driver backend abstractions. Not really
> > something applications would use.
> > > Sonal Santan <sonals@xilinx.com>
> > >
> > > 4:20 PM (1 minute ago)
> > >
> > > to me
> > >
> > >
> > > > -----Original Message-----
> > > > From: Daniel Vetter [mailto:daniel.vetter@ffwll.ch] On Behalf Of
> > > > Daniel Vetter
> > > > Sent: Monday, March 25, 2019 1:28 PM
> > > > To: Sonal Santan <sonals@xilinx.com>
> > > > Cc: dri-devel@lists.freedesktop.org; gregkh@linuxfoundation.org;
> > > > Cyril Chemparathy <cyrilc@xilinx.com>; linux-kernel@vger.kernel.org;
> > > > Lizhi Hou <lizhih@xilinx.com>; Michal Simek <michals@xilinx.com>;
> > > > airlied@redhat.com
> > > > Subject: Re: [RFC PATCH Xilinx Alveo 0/6] Xilinx PCIe accelerator
> > > > driver
> > > >
> > > > On Tue, Mar 19, 2019 at 02:53:55PM -0700, sonal.santan@xilinx.com
> > wrote:
> > > > > From: Sonal Santan <sonal.santan@xilinx.com>
> > > > >
> > > > > Hello,
> > > > >
> > > > > This patch series adds drivers for Xilinx Alveo PCIe accelerator cards.
> > > > > These drivers are part of Xilinx Runtime (XRT) open source stack
> > > > > and have been deployed by leading FaaS vendors and many enterprise
> > > > customers.
> > > >
> > > > Cool, first fpga driver submitted to drm! And from a high level I
> > > > think this makes a lot of sense.
> > > >
> > > > > PLATFORM ARCHITECTURE
> > > > >
> > > > > Alveo PCIe platforms have a static shell and a reconfigurable
> > > > > (dynamic) region. The shell is automatically loaded from PROM when
> > > > > host is booted and PCIe is enumerated by BIOS. Shell cannot be
> > > > > changed till next cold reboot. The shell exposes two physical functions:
> > > > > management physical function and user physical function.
> > > > >
> > > > > Users compile their high level design in C/C++/OpenCL or RTL into
> > > > > FPGA image using SDx compiler. The FPGA image packaged as xclbin
> > > > > file can be loaded onto reconfigurable region. The image may
> > > > > contain one or more compute unit. Users can dynamically swap the
> > > > > full image running on the reconfigurable region in order to switch
> > > > > between different
> > > > workloads.
> > > > >
> > > > > XRT DRIVERS
> > > > >
> > > > > XRT Linux kernel driver xmgmt binds to mgmt pf. The driver is
> > > > > modular and organized into several platform drivers which
> > > > > primarily handle the following functionality:
> > > > > 1. ICAP programming (FPGA bitstream download with FPGA Mgr
> > > > > integration) 2. Clock scaling 3. Loading firmware container also
> > > > > called dsabin (embedded Microblaze
> > > > > firmware for ERT and XMC, optional clearing bitstream) 4.
> > > > > In-band
> > > > > sensors: temp, voltage, power, etc.
> > > > > 5. AXI Firewall management
> > > > > 6. Device reset and rescan
> > > > > 7. Hardware mailbox for communication between two physical
> > > > > functions
> > > > >
> > > > > XRT Linux kernel driver xocl binds to user pf. Like its peer, this
> > > > > driver is also modular and organized into several platform drivers
> > > > > which handle the following functionality:
> > > > > 1. Device memory topology discovery and memory management 2.
> > > > > Buffer object abstraction and management for client process 3.
> > > > > XDMA MM PCIe DMA engine programming 4. Multi-process aware
> > context management 5.
> > > > > Compute unit execution management (optionally with help of ERT) for
> > > > > client processes
> > > > > 6. Hardware mailbox for communication between two physical
> > > > > functions
> > > > >
> > > > > The drivers export ioctls and sysfs nodes for various services.
> > > > > xocl driver makes heavy use of DRM GEM features for device memory
> > > > > management, reference counting, mmap support and export/import.
> > > > > xocl also includes a simple scheduler called KDS which schedules
> > > > > compute units and interacts with hardware scheduler running ERT
> > > > > firmware. The scheduler understands custom opcodes packaged into
> > > > > command objects
> > > > and
> > > > > provides an asynchronous command done notification via POSIX poll.
> > > > >
> > > > > More details on architecture, software APIs, ioctl definitions,
> > > > > execution model, etc. is available as Sphinx documentation--
> > > > >
> > > > > https://xilinx.github.io/XRT/2018.3/html/index.html
> > > > >
> > > > > The complete runtime software stack (XRT) which includes out of
> > > > > tree kernel drivers, user space libraries, board utilities and
> > > > > firmware for the hardware scheduler is open source and available
> > > > > at https://github.com/Xilinx/XRT
> > > >
> > > > Before digging into the implementation side more I looked into the
> > > > userspace here. I admit I got lost a bit, since there's lots of
> > > > indirections and abstractions going on, but it seems like this is
> > > > just a fancy ioctl wrapper/driver backend abstractions. Not really
> > something applications would use.
> > > >
> > >
> > > Appreciate your feedback.
> > >
> > > The userspace libraries define a common abstraction but have different
> > > implementations for Zynq Ultrascale+ embedded platform, PCIe based
> > > Alveo (and Faas) and emulation flows. The latter lets you run your
> > application without physical hardware.
> > >
> > > >
> > > > From the pretty picture on github it looks like there's some
> > > > opencl/ml/other fancy stuff sitting on top that applications would use. Is
> > that also available?
> > >
> > > The full OpenCL runtime is available in the same repository. Xilinx ML
> > > Suite is also based on XRT and its source can be found at
> > https://github.com/Xilinx/ml-suite.
> >
> > Hm, I did a few git grep for the usual opencl entry points, but didn't find
> > anything. Do I need to run some build scripts first (which downloads
> > additional sourcecode)? Or is there some symbol mangling going on and that's
> > why I don't find anything? Pointers very much appreciated.
>
> The bulk of the OCL runtime code can be found inside
> https://github.com/Xilinx/XRT/tree/master/src/runtime_src/xocl.
> The OCL runtime also includes https://github.com/Xilinx/XRT/tree/master/src/runtime_src/xrt.
> The OCL runtime library called libxilinxopencl.so in turn then uses XRT APIs to talk to the drivers.
> For PCIe these XRT APIs are implemented in the library libxrt_core.so the source for which is
> https://github.com/Xilinx/XRT/tree/master/src/runtime_src/driver/xclng/xrt.
>
> You can build a fully functioning runtime stack by following very simple build instructions--
> https://xilinx.github.io/XRT/master/html/build.html
>
> We do have a few dependencies on standard Linux packages including a few OpenCL packages
> bundled by Linux distros: ocl-icd, ocl-icd-devel and opencl-headers

Thanks a lot for pointers. No idea why I didn't find this stuff, I guess I
was blind.

The thing I'm really interested in is the compiler, since at least the
experience from gpus says that very much is part of the overall uapi, and
definitely needed to be able to make any chances to the implementation.
Looking at clCreateProgramWithSource there's only a lookup up cached
compiles (it looks for xclbin), and src/runtime_src/xclbin doesn't look
like that provides a compiler either. It seems like apps need to
precompile everything first. Am I again missing something, or is this how
it's supposed to work?

Note: There's no expectation for the fully optimizing compiler, and we're
totally ok if there's an optimizing proprietary compiler and a basic open
one (amd, and bunch of other companies all have such dual stacks running
on top of drm kernel drivers). But a basic compiler that can convert basic
kernels into machine code is expected.

Thanks, Daniel

>
> Thanks,
> -Sonal
>
> >
> > > Typically end users use OpenCL APIs which are part of XRT stack. One
> > > can write an application to directly call XRT APIs defined at
> > > https://xilinx.github.io/XRT/2018.3/html/xclhal2.main.html
> >
> > I have no clue about DNN/ML unfortunately, I think I'll try to look into the ocl
> > side a bit more first.
> >
> > Thanks, Daniel
> >
> > >
> > > Thanks,
> > > -Sonal
> > > >
> > > > Thanks, Daniel
> > > >
> > > > >
> > > > > Thanks,
> > > > > -Sonal
> > > > >
> > > > > Sonal Santan (6):
> > > > > Add skeleton code: ioctl definitions and build hooks
> > > > > Global data structures shared between xocl and xmgmt drivers
> > > > > Add platform drivers for various IPs and frameworks
> > > > > Add core of XDMA driver
> > > > > Add management driver
> > > > > Add user physical function driver
> > > > >
> > > > > drivers/gpu/drm/Kconfig | 2 +
> > > > > drivers/gpu/drm/Makefile | 1 +
> > > > > drivers/gpu/drm/xocl/Kconfig | 22 +
> > > > > drivers/gpu/drm/xocl/Makefile | 3 +
> > > > > drivers/gpu/drm/xocl/devices.h | 954 +++++
> > > > > drivers/gpu/drm/xocl/ert.h | 385 ++
> > > > > drivers/gpu/drm/xocl/lib/Makefile.in | 16 +
> > > > > drivers/gpu/drm/xocl/lib/cdev_sgdma.h | 63 +
> > > > > drivers/gpu/drm/xocl/lib/libxdma.c | 4368 ++++++++++++++++++++
> > > > > drivers/gpu/drm/xocl/lib/libxdma.h | 596 +++
> > > > > drivers/gpu/drm/xocl/lib/libxdma_api.h | 127 +
> > > > > drivers/gpu/drm/xocl/mgmtpf/Makefile | 29 +
> > > > > drivers/gpu/drm/xocl/mgmtpf/mgmt-core.c | 960 +++++
> > > > > drivers/gpu/drm/xocl/mgmtpf/mgmt-core.h | 147 +
> > > > > drivers/gpu/drm/xocl/mgmtpf/mgmt-cw.c | 30 +
> > > > > drivers/gpu/drm/xocl/mgmtpf/mgmt-ioctl.c | 148 +
> > > > > drivers/gpu/drm/xocl/mgmtpf/mgmt-reg.h | 244 ++
> > > > > drivers/gpu/drm/xocl/mgmtpf/mgmt-sysfs.c | 318 ++
> > > > > drivers/gpu/drm/xocl/mgmtpf/mgmt-utils.c | 399 ++
> > > > > drivers/gpu/drm/xocl/subdev/dna.c | 356 ++
> > > > > drivers/gpu/drm/xocl/subdev/feature_rom.c | 412 ++
> > > > > drivers/gpu/drm/xocl/subdev/firewall.c | 389 ++
> > > > > drivers/gpu/drm/xocl/subdev/fmgr.c | 198 +
> > > > > drivers/gpu/drm/xocl/subdev/icap.c | 2859 +++++++++++++
> > > > > drivers/gpu/drm/xocl/subdev/mailbox.c | 1868 +++++++++
> > > > > drivers/gpu/drm/xocl/subdev/mb_scheduler.c | 3059 ++++++++++++++
> > > > > drivers/gpu/drm/xocl/subdev/microblaze.c | 722 ++++
> > > > > drivers/gpu/drm/xocl/subdev/mig.c | 256 ++
> > > > > drivers/gpu/drm/xocl/subdev/sysmon.c | 385 ++
> > > > > drivers/gpu/drm/xocl/subdev/xdma.c | 510 +++
> > > > > drivers/gpu/drm/xocl/subdev/xmc.c | 1480 +++++++
> > > > > drivers/gpu/drm/xocl/subdev/xvc.c | 461 +++
> > > > > drivers/gpu/drm/xocl/userpf/Makefile | 27 +
> > > > > drivers/gpu/drm/xocl/userpf/common.h | 157 +
> > > > > drivers/gpu/drm/xocl/userpf/xocl_bo.c | 1255 ++++++
> > > > > drivers/gpu/drm/xocl/userpf/xocl_bo.h | 119 +
> > > > > drivers/gpu/drm/xocl/userpf/xocl_drm.c | 640 +++
> > > > > drivers/gpu/drm/xocl/userpf/xocl_drv.c | 743 ++++
> > > > > drivers/gpu/drm/xocl/userpf/xocl_ioctl.c | 396 ++
> > > > > drivers/gpu/drm/xocl/userpf/xocl_sysfs.c | 344 ++
> > > > > drivers/gpu/drm/xocl/version.h | 22 +
> > > > > drivers/gpu/drm/xocl/xclbin.h | 314 ++
> > > > > drivers/gpu/drm/xocl/xclfeatures.h | 107 +
> > > > > drivers/gpu/drm/xocl/xocl_ctx.c | 196 +
> > > > > drivers/gpu/drm/xocl/xocl_drm.h | 91 +
> > > > > drivers/gpu/drm/xocl/xocl_drv.h | 783 ++++
> > > > > drivers/gpu/drm/xocl/xocl_subdev.c | 540 +++
> > > > > drivers/gpu/drm/xocl/xocl_thread.c | 64 +
> > > > > include/uapi/drm/xmgmt_drm.h | 204 +
> > > > > include/uapi/drm/xocl_drm.h | 483 +++
> > > > > 50 files changed, 28252 insertions(+) create mode 100644
> > > > > drivers/gpu/drm/xocl/Kconfig create mode 100644
> > > > > drivers/gpu/drm/xocl/Makefile create mode 100644
> > > > > drivers/gpu/drm/xocl/devices.h create mode 100644
> > > > > drivers/gpu/drm/xocl/ert.h create mode 100644
> > > > > drivers/gpu/drm/xocl/lib/Makefile.in
> > > > > create mode 100644 drivers/gpu/drm/xocl/lib/cdev_sgdma.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/lib/libxdma.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/lib/libxdma.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/lib/libxdma_api.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/Makefile
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/mgmt-core.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/mgmt-core.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/mgmt-cw.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/mgmt-ioctl.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/mgmt-reg.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/mgmt-sysfs.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/mgmtpf/mgmt-utils.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/dna.c create mode
> > > > > 100644 drivers/gpu/drm/xocl/subdev/feature_rom.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/firewall.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/fmgr.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/icap.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/mailbox.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/mb_scheduler.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/microblaze.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/mig.c create mode
> > > > > 100644 drivers/gpu/drm/xocl/subdev/sysmon.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/xdma.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/subdev/xmc.c create mode
> > > > > 100644 drivers/gpu/drm/xocl/subdev/xvc.c create mode 100644
> > > > > drivers/gpu/drm/xocl/userpf/Makefile
> > > > > create mode 100644 drivers/gpu/drm/xocl/userpf/common.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/userpf/xocl_bo.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/userpf/xocl_bo.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/userpf/xocl_drm.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/userpf/xocl_drv.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/userpf/xocl_ioctl.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/userpf/xocl_sysfs.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/version.h create mode
> > > > > 100644 drivers/gpu/drm/xocl/xclbin.h create mode 100644
> > > > > drivers/gpu/drm/xocl/xclfeatures.h
> > > > > create mode 100644 drivers/gpu/drm/xocl/xocl_ctx.c create mode
> > > > > 100644 drivers/gpu/drm/xocl/xocl_drm.h create mode 100644
> > > > > drivers/gpu/drm/xocl/xocl_drv.h create mode 100644
> > > > > drivers/gpu/drm/xocl/xocl_subdev.c
> > > > > create mode 100644 drivers/gpu/drm/xocl/xocl_thread.c
> > > > > create mode 100644 include/uapi/drm/xmgmt_drm.h create mode
> > > > > 100644 include/uapi/drm/xocl_drm.h
> > > > >
> > > > > --
> > > > > 2.17.0
> > > > > _______________________________________________
> > > > > dri-devel mailing list
> > > > > dri-devel@lists.freedesktop.org
> > > > > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> > > >
> > > > --
> > > > Daniel Vetter
> > > > Software Engineer, Intel Corporation http://blog.ffwll.ch
> >
> >
> >
> > --
> > Daniel Vetter
> > Software Engineer, Intel Corporation
> > +41 (0) 79 365 57 48 - http://blog.ffwll.ch

--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

\
 
 \ /
  Last update: 2019-03-27 15:12    [W:0.092 / U:1.272 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site