On Mon, Dec 09, 2019 at 11:15:39PM +0000, Joe Hershberger wrote:

Hi Tom,

On Mon, Dec 9, 2019 at 4:49 PM Tom Rini trini@konsulko.com wrote:

...

On Mon, Dec 09, 2019 at 09:51:32AM -0600, Joe Hershberger wrote:

...

Hi Tom,

Between dropping the superceeded atheros series and applying your nfs trim patch, things appear to be all good.

Passed testing at https://travis-ci.org/jhershbe/u-boot/builds/622465407

The following changes since commit b38c3a641fc01fcd4eda5fa107ae3c247baa0196:

Merge https://gitlab.denx.de/u-boot/custodians/u-boot-x86 (2019-12-08 10:51:00 -0500)

are available in the git repository at:

https://gitlab.denx.de/u-boot/custodians/u-boot-net.git master

for you to fetch changes up to 2c4e067d493d031d1267eea1c635d1eab93ce8f7:

cmd: pxe: Increase maximum path length (2019-12-09 09:49:48 -0600)

---

Alex Marginean (18): include: phy: define XFI and USXGMII interface types include: phy: add data field for private driver data drivers: net: aquantia: use XFI, USXGMII interface types drivers: net: aquantia: add PHY generation information drivers: net: aquantia: set up SI protocol based on interface type drivers: net: aquantia: set MDI reversal based on DT property drivers: net: aquantia: set SMBus addr based on DT property drivers: net: aquantia: check system interface too when checking for link up doc: bindings: add bindings document for PHY nodes doc: bindings: Aquantia PHY node binding drivers: net: fsl_enetc: use XFI, USXGMII interface type macros net: mdio-uclass: rename arguments of dm_mdio_phy_connect for clarity net: mdio-uclass: add dm_eth_phy_connect helper function drivers: net: fsl_enetc: use the new MDIO DM helper functions drivers: net: fsl_enetc: Add 2.5Gbps to supported link speeds drivers: net: fsl_enetc: move PCS and PHY config to probe drivers: net: fsl_enetc_mdio: return with time-out if HW is stuck drivers: net: fsl_enetc: register internal MDIO bus

Anatolij Gustschin (5): net: phy: mv88e61xx: rework to enable detection of 88E6071 devices net: phy: mv88e61xx: add CPU port parameter init for 88E6071 net: phy: mv88E61xx: fix ENERGY_DET init for mv88E6071 net: phy: mv88e61xx: register phy_driver struct for 88E6071 net: phy: fix switch vendor name

Andre Przywara (1): net: tftp: Fix too small block size

Ben Wolsieffer (1): cmd: pxe: Increase maximum path length

Bin Meng (1): net: tftp: Fix tftp store address check in store_block()

Cédric Le Goater (1): net: ftgmac100: align RX/TX descriptors on ARCH_DMA_MINALIGN

Florin Chiculita (1): net: phy: aquantia: wait for phy init sequence to finish

Grygorii Strashko (8): net: phy: ti: rename ti.c to dp83867.c net: phy: dp83867: move static initialization to .probe() dt-bindings: phy: dp83867: Add documentation for disabling clock output net: phy: dp83867: Add ability to disable output clock net: phy: dp83867: rework delay rgmii delay handling net: phy: dp83867: io impedance is not dependent on RGMII delay net: phy: dp83867: refactor rgmii configuration arm: dts: k3-am654-base-board-u-boot: change cpsw2g interface mode to rgmii-rxid

Grzegorz Jaszczyk (1): net: mvpp2: mark phy as invalid in case of missing appropriate driver

Heinrich Schuchardt (1): net: avoid address-of-packed-member error

Joe Hershberger (3): net: Always build the string_to_enetaddr() helper net: Improve documentation for string_to_ip() lib: Always build support for formatting MAC and IP address

Josef Holzmayr (2): net: macb: explicitly pass phy_adr to mdio read and write net: macb: let miiphy_read/_write pass arbitrary addresses

Michael Walle (1): net: eth-uclass: ignore unavailable devices

Nevo Hed (3): arm: dts: armada-cp110-*dtsi: add xmdio nodes net: mvpp2: use new MVMDIO driver net: mvpp2: MVPP2 now needs MVMDIO

Nicolas Ferre (1): net: phy: micrel: make sure the factory test bit is cleared

Priyanka Jain (1): net/phy: Fix phy_connect() for phy addr 0

Ramon Fried (1): cmd: mdio/mii: add Kconfig help and allow break dependency

Stefan Chulski (1): net: mvpp2x: fix traffic stuck after PHY start error

Stefan Roese (1): net: phy: Increase link up delay in genphy_update_link()

Thierry Reding (1): net: rtl8169: Support RTL-8168c/8111c

Tom Rini (1): net: nfs: Only link in NFS code outside of SPL builds

Vignesh Raghavendra (6): dma: Introduce dma_get_cfg() interface dma: ti: k3-udma: Implement dma_get_cfg() interface net: ti: am65-cpsw-nuss: Rework RX flow ID handling net: ti: am65-cpsw-nuss: Add new compatible for J721e arm: dts: k3-j721e-common-proc-board: Add DMA and CPSW related DT nodes configs: j721e_evm_a72_defconfig: Enable DMA and Ethernet

arch/arm/dts/armada-cp110-master.dtsi | 9 + arch/arm/dts/armada-cp110-slave.dtsi | 9 + arch/arm/dts/k3-am654-base-board-u-boot.dtsi | 3 +- .../arm/dts/k3-j721e-common-proc-board-u-boot.dtsi | 238 ++++++++++++++++++++ arch/arm/mach-tegra/cboot.c | 2 +- board/renesas/sh7752evb/sh7752evb.c | 2 +- board/renesas/sh7753evb/sh7753evb.c | 2 +- board/renesas/sh7757lcr/sh7757lcr.c | 4 +- cmd/Kconfig | 18 +- cmd/Makefile | 4 +- cmd/ethsw.c | 2 +- cmd/nvedit.c | 2 +- cmd/pxe_utils.c | 2 +- configs/j721e_evm_a72_defconfig | 8 + doc/README.enetaddr | 4 +- doc/device-tree-bindings/net/aquantia-phy.txt | 25 +++ doc/device-tree-bindings/net/phy.txt | 24 ++ doc/device-tree-bindings/net/ti,dp83867.txt | 6 +- drivers/dma/dma-uclass.c | 12 + drivers/dma/ti/k3-udma.c | 29 +++ drivers/net/Kconfig | 2 + drivers/net/fsl_enetc.c | 86 +++----- drivers/net/fsl_enetc.h | 1 + drivers/net/fsl_enetc_mdio.c | 7 +- drivers/net/ftgmac100.c | 16 +- drivers/net/macb.c | 39 ++-- drivers/net/mvpp2.c | 241 +++++--------------- drivers/net/phy/Kconfig | 2 +- drivers/net/phy/Makefile | 2 +- drivers/net/phy/aquantia.c | 241 ++++++++++++++++++-- drivers/net/phy/{ti.c => dp83867.c} | 243 ++++++++++++++------- drivers/net/phy/micrel_ksz8xxx.c | 21 +- drivers/net/phy/mv88e61xx.c | 226 +++++++++++++++---- drivers/net/phy/phy.c | 6 +- drivers/net/rtl8169.c | 1 + drivers/net/ti/am65-cpsw-nuss.c | 16 +- include/dma-uclass.h | 11 + include/dma.h | 12 + include/dt-bindings/net/ti-dp83867.h | 3 +- include/linux/soc/ti/ti-udma.h | 19 ++ include/miiphy.h | 18 +- include/net.h | 34 +-- include/phy.h | 3 + include/phy_interface.h | 4 + lib/net_utils.c | 15 ++ lib/vsprintf.c | 4 - net/Kconfig | 2 +- net/eth-uclass.c | 19 +- net/eth_legacy.c | 2 +- net/mdio-uclass.c | 100 ++++++++- net/net.c | 36 +-- net/tftp.c | 7 +- 52 files changed, 1316 insertions(+), 528 deletions(-) create mode 100644 doc/device-tree-bindings/net/aquantia-phy.txt create mode 100644 doc/device-tree-bindings/net/phy.txt rename drivers/net/phy/{ti.c => dp83867.c} (63%)

I've reviewed this and tested it locally and things are OK. First, are you sure that a month out from release we should still take this large of a set of changes to master?

I tried to leave out the questionable changes, though I agree I wish it were in earlier... Between merge conflicts, build failures / triage, and work schedule, the PR slipped a lot more than you or I would like.

If it's ok with you, maybe we can just be willing to revert anything that causes issues, and defer that to the next release if it happens.

OK, we'll go with reverting things if issues show up.

...

Second, is there anything we can do about the growth of 'struct phy' ? We have some boards with a few possible PHY drivers and in turn a number of PHY drivers and it's a non-trivial growth to support this field.

Are you referring to all of the "registered" phy drivers based on the enabled phys? That adds to the "phy_drivers" structure when phy_register() is called? If not, what was it that you're referring to? The literal "struct phy" seems to be for generic USB phys and what not - I assume that's not what you're talking about.

Yes, sorry, I misspoke and it's phy_drivers.