public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: Marcin Wojtas <mw@semihalf.com>
To: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: edk2-devel-01 <edk2-devel@lists.01.org>,
	"Gao, Liming" <liming.gao@intel.com>,
	"Ni, Ruiyu" <ruiyu.ni@intel.com>,
	"Kinney, Michael D" <michael.d.kinney@intel.com>,
	afish@apple.com,  Leif Lindholm <leif.lindholm@linaro.org>
Subject: Re: [PATCH v3 0/5] MdeModulePkg: add support for non-discoverable devices
Date: Thu, 17 Nov 2016 05:36:53 +0100	[thread overview]
Message-ID: <CAPv3WKdxZzCBuQDL80+B0Pg9m-C63_p__1FJB+jTbYQYRmLiSg@mail.gmail.com> (raw)
In-Reply-To: <1479315571-14953-1-git-send-email-ard.biesheuvel@linaro.org>

Hi Ard,

Series tested on Marvell Armada 7040 board (XHCI and SDHCI). You can add:
Tested-by: Marcin Wojtas <mw@semihalf.com>

Best regards,
Marcin

2016-11-16 17:59 GMT+01:00 Ard Biesheuvel <ard.biesheuvel@linaro.org>:
> The rationale for this series is the fact that many ARM platforms implement
> some form of PCI 'emulation', to allow non-discoverable devices that implement
> standardized host controller interfaces (i.e., EHCI, AHCI) to be controlled by
> the generic EDK2 drivers, which are layered on top of the PCI I/O protocols
> (even though the respective host controller specifications don't mandate that)
>
> There are a couple of problems with that approach:
> - Most implementations are based on the original code created for BeagleBoard,
>   which is a 32-bit platform. Unlike x86, which usually does not perform PCI
>   DMA above 4 GB, the ARM ecosystem is much more heterogeneous, and platforms
>   that have memory both above and below the 4 GB mark may ship with, e.g., EHCI
>   controllers that do not implement 64-bit DMA addressing.
> - Implementations depend on the DmaLib library class in EmbeddedPkg, of which
>   coherent and non-coherent implementations exists. However, both types of
>   devices may appear on a single platform, requiring several instances of the
>   same driver.
> - Existing implementations do not follow the UEFI driver model, but instantiate
>   a fixed number of PCI I/O protocol handles, and bring up all the devices when
>   doing so. However, the UEFI philosophy is to only instantiate (and thus
>   initialize) devices that are involved in booting.
>
> So instead, let's define a base protocol that simply asserts the presence of
> a certain kind of device at a certain memory offset, allowing platforms to
> instantiate any number of these statically, and leave it to post-DXE driver
> dispatch to actually bind the drivers as usual. This is implemented in patch #1.
> Note that it includes an AMBA device type, which we intend to use in the future
> to move ARM AMBA drivers to the UEFI driver model as well (i.e., LCD controller,
> SD/MMC controller)
>
> Patch #2 implements a utility library to register non-discoverable devices.
>
> Patch #3 implements the UEFI driver that instantiates PCI I/O protocol handles
> for non-discoverable devices that we know can be driven by a generic driver in
> EDK2. The initial version implements coherent DMA only.
>
> Patch #4 implements non-coherent DMA for the driver added in patch #3.
>
> Patch #5 is included for reference. It ports the BeagleBoard platform to the
> new driver stack.
>
> Changes in v3:
> - Modified the base protocol to describe any number of resources using ACPI
>   resource descriptors. The current crop of targeted peripherals does not
>   require that, but it allows for much more flexibility in the future. This
>   also means that the MMIO region sizes are no longer hard-coded base on the
>   controller type (EHCI, AHCI, etc) but need to be declared when the devices
>   are registered.
> - Update the registration library API to allow multiple MMIO regions to be
>   declared.
> - When performing uncached allocations, record the original memory region
>   attributes so we can restore them correctly on free.
> - Some fixes in the 4GB boundary checks for Map/Unmap
> - Rename gNonDiscoverableDeviceProtocolGuid to
>   gEdkiiNonDiscoverableDeviceProtocolGuid
>
> Ard Biesheuvel (5):
>   MdeModulePkg: introduce non-discoverable device protocol
>   MdeModule: introduce helper library to register non-discoverable
>     devices
>   MdeModulePkg: implement generic PCI I/O driver for non-discoverable
>     devices
>   MdeModulePkg/NonDiscoverablePciDeviceDxe: add support for non-coherent
>     DMA
>   Omap35xxPkg/PciEmulation: port to new non-discoverable device
>     infrastructure
>
>  BeagleBoardPkg/BeagleBoardPkg.dsc                                                                  |    2 +
>  BeagleBoardPkg/BeagleBoardPkg.fdf                                                                  |    1 +
>  MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/ComponentName.c                                   |   75 ++
>  MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceDxe.c                     |  226 ++++
>  MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceDxe.inf                   |   44 +
>  MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceIo.c                      | 1187 ++++++++++++++++++++
>  MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceIo.h                      |  113 ++
>  MdeModulePkg/Include/Library/NonDiscoverableDeviceRegistrationLib.h                                |   49 +
>  MdeModulePkg/Include/Protocol/NonDiscoverableDevice.h                                              |   90 ++
>  MdeModulePkg/Library/NonDiscoverableDeviceRegistrationLib/NonDiscoverableDeviceRegistrationLib.c   |  161 +++
>  MdeModulePkg/Library/NonDiscoverableDeviceRegistrationLib/NonDiscoverableDeviceRegistrationLib.inf |   34 +
>  MdeModulePkg/MdeModulePkg.dec                                                                      |    7 +
>  MdeModulePkg/MdeModulePkg.dsc                                                                      |    3 +
>  Omap35xxPkg/Omap35xxPkg.dsc                                                                        |    2 +-
>  Omap35xxPkg/PciEmulation/PciEmulation.c                                                            |  578 +---------
>  Omap35xxPkg/PciEmulation/PciEmulation.h                                                            |  292 -----
>  Omap35xxPkg/PciEmulation/PciEmulation.inf                                                          |   16 +-
>  Omap35xxPkg/PciEmulation/PciRootBridgeIo.c                                                         |  306 -----
>  18 files changed, 2025 insertions(+), 1161 deletions(-)
>  create mode 100644 MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/ComponentName.c
>  create mode 100644 MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceDxe.c
>  create mode 100644 MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceDxe.inf
>  create mode 100644 MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceIo.c
>  create mode 100644 MdeModulePkg/Bus/Pci/NonDiscoverablePciDeviceDxe/NonDiscoverablePciDeviceIo.h
>  create mode 100644 MdeModulePkg/Include/Library/NonDiscoverableDeviceRegistrationLib.h
>  create mode 100644 MdeModulePkg/Include/Protocol/NonDiscoverableDevice.h
>  create mode 100644 MdeModulePkg/Library/NonDiscoverableDeviceRegistrationLib/NonDiscoverableDeviceRegistrationLib.c
>  create mode 100644 MdeModulePkg/Library/NonDiscoverableDeviceRegistrationLib/NonDiscoverableDeviceRegistrationLib.inf
>  delete mode 100644 Omap35xxPkg/PciEmulation/PciEmulation.h
>  delete mode 100644 Omap35xxPkg/PciEmulation/PciRootBridgeIo.c
>
> --
> 2.7.4
>


  parent reply	other threads:[~2016-11-17  4:36 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-16 16:59 [PATCH v3 0/5] MdeModulePkg: add support for non-discoverable devices Ard Biesheuvel
2016-11-16 16:59 ` [PATCH v3 1/5] MdeModulePkg: introduce non-discoverable device protocol Ard Biesheuvel
2016-11-16 17:48   ` Leif Lindholm
2016-11-17  2:53     ` Ni, Ruiyu
2016-11-17  6:07       ` Ard Biesheuvel
2016-11-17  7:52         ` Ni, Ruiyu
2016-11-17 10:43           ` Ard Biesheuvel
2016-11-18  2:11             ` Ni, Ruiyu
2016-11-18  4:59               ` Ard Biesheuvel
2016-11-18  5:24                 ` Tian, Feng
2016-11-18  6:57                   ` Ard Biesheuvel
2016-11-18  8:39                     ` Tian, Feng
2016-11-18  8:52                       ` Ard Biesheuvel
2016-11-18  6:13                 ` Ni, Ruiyu
2016-11-18  7:04                   ` Ard Biesheuvel
2016-11-18 13:39                     ` Ni, Ruiyu
2016-11-18 13:50                       ` Ard Biesheuvel
2016-11-25 15:21                         ` Ard Biesheuvel
2016-11-16 16:59 ` [PATCH v3 2/5] MdeModule: introduce helper library to register non-discoverable devices Ard Biesheuvel
2016-11-16 16:59 ` [PATCH v3 3/5] MdeModulePkg: implement generic PCI I/O driver for " Ard Biesheuvel
2016-11-17  3:29   ` Ni, Ruiyu
2016-11-18 12:30     ` Ard Biesheuvel
2016-11-24 18:14       ` Ard Biesheuvel
2016-11-16 16:59 ` [PATCH v3 4/5] MdeModulePkg/NonDiscoverablePciDeviceDxe: add support for non-coherent DMA Ard Biesheuvel
2016-11-16 16:59 ` [PATCH v3 5/5] Omap35xxPkg/PciEmulation: port to new non-discoverable device infrastructure Ard Biesheuvel
2016-11-17  4:36 ` Marcin Wojtas [this message]
2016-11-23 14:31   ` [PATCH v3 0/5] MdeModulePkg: add support for non-discoverable devices Marcin Wojtas

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-list from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAPv3WKdxZzCBuQDL80+B0Pg9m-C63_p__1FJB+jTbYQYRmLiSg@mail.gmail.com \
    --to=devel@edk2.groups.io \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox