Tom Rini [Tue, 5 May 2015 14:32:08 +0000 (10:32 -0400)]
Merge git://git.denx.de/u-boot-marvell
Stefan Roese [Mon, 4 May 2015 15:31:34 +0000 (17:31 +0200)]
arm: mvebu: Remove "u-boot.kwb" CONFIG_BUILD_TARGET for non-SPL targets
By removing this default build target, the "u-boot.kwb" target is not
automatically generated. This fixes a temporary build error when using
out-of-tree builds, as buildman does per default (reported by Simon).
When the full SPL support is added for these targets with the DDR training
code, the "u-boot-spl.kwb" image will be generated automatically.
Users providing a special bin_hdr binary (binary.0) file can always add
this binary file and generate the full firmware image by issuing the
"make u-boot.kwb" command directly.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Prafulla Wadaskar <prafulla@marvell.com>
Cc: Luka Perkov <luka.perkov@sartura.hr>
Stefan Roese [Wed, 22 Apr 2015 16:36:39 +0000 (18:36 +0200)]
arm: mvebu: db-mv784mp-gp: Fix ECC I2C address
The macro to select the I2C address for ECC bus-width detection
was defined incorrectly for the Marvell DB-MV784MP-GP board. This
patch changes the macro to the correct value to fix this issue.
Signed-off-by: Stefan Roese <sr@denx.de>
Stefan Roese [Sat, 25 Apr 2015 04:29:56 +0000 (06:29 +0200)]
arm: mvebu: Add Armada A38x DB-
88F6820-GP board support
This patch adds support for the Marvell DB-
88F6820-GP Armada A38x
evaluation board.
Supported peripherals are:
- UART
- Ethernet (mvneta)
- I2C
- SPI (including SPI NOR flash)
Please note that this board support right now only supports the
main U-Boot. Without the bin_hdr integration (DDR training etc). This
will be added in a few days / weeks to complete this board port. But
till then this U-Boot version can be run on the target via the
original Marvell U-Boot via this command:
tftpboot
4000000 db-
88f6820-gp/u-boot.bin;go
4000000
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:55 +0000 (06:29 +0200)]
arm: mvebu: Add d-cache invalidate before enabling the d-cache
This solves some RX problems that have been seen, when using the
mvneta ethernet driver. The cache needs to be reset into a "clean"
state before using it.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:54 +0000 (06:29 +0200)]
arm: mvebu: mv-common.h: Add CONFIG_PREBOOT
This enables the usage of the "preboot" environment variable on Marvell
boards.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:53 +0000 (06:29 +0200)]
arm: mvebu: Remove coherency configuration
We are not using the coherency feature in U-Boot at all. So lets remove
this configuration from the mbus driver.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:52 +0000 (06:29 +0200)]
arm: mvebu: Change network init code to allow a more flexible setup
With the introduction of the Armada 38x support, its necessary to change
the mvneta ethernet driver init call from always 4 times to a
configurable value. Lets make this init call more flexible by moving
the actually used devices to the config header.
Additionally this patch takes care of the slightly different base
addresses for the ethernet controllers on A38x.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Stefan Roese [Sat, 25 Apr 2015 04:29:51 +0000 (06:29 +0200)]
arm: mvebu: Add basic Armada 38x support
This patch adds support for the Marvell Armada 38x SoC family.
Supported peripherals are:
- UART
- Ethernet (mvneta)
- I2C
- SPI (including SPI NOR flash)
Tested on Marvell DB-
88F6820-GP evaluation board.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:50 +0000 (06:29 +0200)]
arm: mvebu: Move CONFIG_SPL_LDSCRIPT to common header
This way, new MVEBU boards don't need to specifiy the common location
for the SPL linker script.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:49 +0000 (06:29 +0200)]
arm: mvebu: Only define MV88F78X60 for Armada XP
This define is used by the DDR training code for Armada XP. With the
upcoming addition of Armada 38x support, lets only define it for
Armada XP in this common header.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:48 +0000 (06:29 +0200)]
arm: mvebu: Remove unreferenced define
MAX_MVNETA_DEVS is not used anywhere in U-Boot. So lets remove it.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:47 +0000 (06:29 +0200)]
arm: mvebu: Change header macros from ARMADA_XP to MVEBU
Since these files will be used for other MVEBU SoC's, lets reflect this
in the headers marcos as well.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:46 +0000 (06:29 +0200)]
arm: mvebu: Move mvebu-common into mach-mvebu
Now that the mach-mvebu directory exists and is used by Armada XP we can
move the mvebu-common files into this directory as well.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:45 +0000 (06:29 +0200)]
arm: armada-xp: Move SoC headers to mach-mvebu/include/mach
Move arch/arm/include/asm/arch-armada-xp/*
-> arch/arm/mach-mvebu/include/mach/*
Additionally the SYS_SOC is renamed from "armada-xp" to "mvebu". With this
change all these files can better be shared with other, newer Mavell
MVEBU SoC's. Like the upcoming Armada 38x support.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Stefan Roese [Sat, 25 Apr 2015 04:29:44 +0000 (06:29 +0200)]
arm: armada-xp: Move SoC sources to mach-mvebu
Move arch/arm/cpu/armv7armada-xp/* -> arch/arm/mach-mvebu/*
Since this platform will be extended to support other Marvell SoC's as
well, lets rename it directly to mvebu.
This will be used by the upcoming Armada 38x suport (A38x).
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Tested-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Tested-by: Dirk Eibach <dirk.eibach@gdsys.cc>
Tom Rini [Tue, 5 May 2015 11:00:38 +0000 (07:00 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-mmc
Tom Rini [Tue, 5 May 2015 11:00:11 +0000 (07:00 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-arm
Marek Vasut [Mon, 4 May 2015 20:54:36 +0000 (22:54 +0200)]
ARM: mmc: bcm283x: Remove get_timer_us() from mmc driver
The get_timer_us() function is something which is no longer
existing in case we use generic timer framework, so replace
it with get_timer().
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Stephen Warren <swarren@wwwdotorg.org>
Cc: Tyler Baker <tyler.baker@linaro.org>
Simon Glass [Mon, 4 May 2015 17:31:16 +0000 (11:31 -0600)]
tegra: mmc: Set the removable flag correctly
If the mmc device is non-removable (as indicated by the device tree), set
the flag so that users of the device know.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
Alexander Stein [Fri, 17 Apr 2015 15:33:17 +0000 (17:33 +0200)]
mmc: bcm2835_sdhci: Use calloc to allocate bcm2835_sdhci_host
We need to clear the allocated memory explicitly as the included
struct sdhci_host has function pointers. Those are compared to NULL to
test if this (optional) feature is supported. Leaving them undefined let
u-boot jump to arbitrary memory.
Signed-off-by: Alexander Stein <alexanders83@web.de>
Yangbo Lu [Wed, 15 Apr 2015 02:13:12 +0000 (10:13 +0800)]
mmc: fsl_esdhc: update eMMC44 adapter card erase timeout
Freescale eMMC44 adapter card uses Micron N2M400FDB311A3CF eMMC
memory. According to the silicon datasheet, secure erase timeout
is 600ms. So increase erase timeout value from 250ms to 600ms.
Signed-off-by: Yangbo Lu <yangbo.lu@freescale.com>
Cc: York Sun <yorksun@freescale.com>
Kevin Liu [Mon, 23 Mar 2015 22:57:00 +0000 (17:57 -0500)]
mmc: sdhci: add timeout setting for response busy command
Timeout interrupt also work for response busy command(R1b) like
cmd38/cmd6. So need to set it accordingly. Current code only
set timeout for data command.
Signed-off-by: Kevin Liu <kliu5@marvell.com>
Signed-off-by: Rob Herring <robh@kernel.org>
Cc: Pantelis Antoniou <panto@antoniou-consulting.com>
Rob Herring [Mon, 23 Mar 2015 22:56:59 +0000 (17:56 -0500)]
mmc: remove the MMC_MODE_HC flag
High capacity support is not a host capability, but a device capability
that is queried via the OCR. The flag in the operating conditions
request argument can just be set unconditionally. This matches the Linux
implementation.
[panto] Hand merged and renumbering MMC_MODE_DDR_52MHz.
Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Pantelis Antoniou <pantelis.antoniou@konsulko.com>
Cc: Pantelis Antoniou <pantelis.antoniou@konsulko.com>
Andrew Gabbasov [Thu, 19 Mar 2015 12:44:07 +0000 (07:44 -0500)]
mmc: Fix splitting device initialization
Starting part of device initialization sets the init_in_progress flag
only if the MMC card did not yet come to ready state and needs to continue
polling. If the card is SD or if the MMC card became ready quickly,
the flag is not set and (if using pre-initialization) the starting
phase will be re-executed from mmc_init function.
Set the init_in_progress flag in all non-error cases. Also, move flags
setting statements around so that the flags are not set in error paths.
Also, IN_PROGRESS return status becomes unnecessary, so get rid of it.
Signed-off-by: Andrew Gabbasov <andrew_gabbasov@mentor.com>
Andrew Gabbasov [Thu, 19 Mar 2015 12:44:06 +0000 (07:44 -0500)]
mmc: Restructure polling loops to avoid extra delays
The polling loops in sd_send_op_cond and mmc_complete_op_cond functions
check the ready flag state at the end of the loop, that is after executing
a delay inside the loop, which, in case of exiting with no error,
is not needed. Also, one of these loops, as well as the loop
in mmc_send_status, have the delay just before exiting on timeout
conditions.
Restructure all these loops to check the respective conditions before making
a delay for the next loop pass, and to appropriately exit without the delay.
Signed-off-by: Andrew Gabbasov <andrew_gabbasov@mentor.com>
Andrew Gabbasov [Thu, 19 Mar 2015 12:44:05 +0000 (07:44 -0500)]
mmc: Continue polling MMC card for OCR only if it is still not ready
Some MMC cards come to ready state quite quickly, so that the respective
flag appears to be set in mmc_send_op_cond already. In this case trying
to continue polling the card with CMD1 in mmc_complete_op_cond is incorrect
and may lead to unpredictable results. So check the flag before polling
and skip it appropriately.
Signed-off-by: Andrew Gabbasov <andrew_gabbasov@mentor.com>
Andrew Gabbasov [Thu, 19 Mar 2015 12:44:04 +0000 (07:44 -0500)]
mmc: Do not pass external mmc_cmd structure to mmc_send_op_cond_iter()
The previous change to use 'ocr' structure field for storing send_op_cond
command response also stopped using command response directly
outside of mmc_send_op_cond_iter(). Now it becomes possible to use
command structure in mmc_send_op_cond_iter() locally, removing a necessity
to pass it as an argument from the caller.
Signed-off-by: Andrew Gabbasov <andrew_gabbasov@mentor.com>
Andrew Gabbasov [Thu, 19 Mar 2015 12:44:03 +0000 (07:44 -0500)]
mmc: Avoid extra duplicate entry in mmc device structure
The 'op_cond_response' field in mmc structure contains the response
from the last SEND_OP_COND MMC command while making iterational
polling of the card. Later it is copied to 'ocr' field, designed
to contain the OCR register value, which is actually the same
response from the same command. So, these fields have actually
the same data, just in different time periods. It's easier to use
the same 'ocr' field in both cases at once, without temporary using
of the 'op_cond_response' field.
Signed-off-by: Andrew Gabbasov <andrew_gabbasov@mentor.com>
Andrew Gabbasov [Thu, 19 Mar 2015 12:44:02 +0000 (07:44 -0500)]
mmc: Fix typo in MMC type checking macro
The version flag constant name used in IS_MMC macro is incorrect/undefined.
Signed-off-by: Andrew Gabbasov <andrew_gabbasov@mentor.com>
Albert ARIBAUD [Tue, 5 May 2015 08:09:06 +0000 (10:09 +0200)]
Merge branch 'u-boot/master' into 'u-boot-arm/master'
Tom Rini [Mon, 4 May 2015 21:50:20 +0000 (17:50 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-sunxi
Hans de Goede [Mon, 27 Apr 2015 14:57:54 +0000 (16:57 +0200)]
sunxi: usb: Protect phy-init and phy-power-on against multiple calls
Once we add support for the ohci controller the phy-init and phy-power-on
functions may be called twice (once by the ehci code and once by the ohci
code) protect them against this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 27 Apr 2015 14:50:04 +0000 (16:50 +0200)]
sunxi: usb: Do not call phy_probe from hcd code
The 2/3 usb-phys on the sunxi SoCs are really a single separate functional
block, and are modelled as such in devicetree. So once we've moved all the
sunxi usb code to the driver-model then phy_probe will be called once
for the entire block from the driver-model enumeration code.
Move to this now as this also avoids problems with phy_probe being called
multiple times once we introduce ohci support. This also allows us to get rid
of the sunxi_usb_phy_enabled_count variable as phy_probe now is guaranteed
to be called only once.
Since we're effectively rewriting the probe / remove functions, move them
to the end of the file while we are at it, as that is the most logical place
for them.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 27 Apr 2015 13:05:10 +0000 (15:05 +0200)]
sunxi: usb: Rename the usbc.? files to usb_phy.?
The usbc.? files now only contain usb-phy related code, rename them to make
this clear.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 27 Apr 2015 12:54:47 +0000 (14:54 +0200)]
sunxi: usb: Rename sunxi_usbc_foo functions to sunxi_usb_phy_bar
Rename the sunxi_usbc_foo functions to sunxi_usb_phy_bar to make it clear
that these are usb-phy functions. Also change the verbs & nouns in the suffix
to match the verbs & nouns used in the Linux kernels generic phy framework.
This patch purely renames things, it contains no functional changes.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 27 Apr 2015 12:36:23 +0000 (14:36 +0200)]
sunxi: usb: Remove sunxi_usbc_get_io_base function
This is the only function left in sunxi/usbc.c which is not phy related,
so remove it.
This is a preparation patch for turning the usbc.c code into a proper
usb phy driver.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 27 Apr 2015 09:44:22 +0000 (11:44 +0200)]
sunxi: usb: Move setup of host controller clocks to the host controller drivers
The sunxi "usbc" code is mostly about phy setup, but currently also sets up
the host controller clocks, which is something which really belongs in the
host controller drivers, so move it there.
This is a preparation patch for moving the sunxi ehci code to the driver
model and for adding ohci support.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sun, 26 Apr 2015 09:19:37 +0000 (11:19 +0200)]
sunxi: axp: Remove non driver-model support from the axp gpio code
Now that all sunxi boards are using driver-model for gpio (*), we can remove
the non driver-model support from the axp gpio code, and the glue to call
into the axp gpio code from the sunxi_gpio non driver-model code.
*) For the regular u-boot build, SPL still uses non driver-model gpio for
now, but the SPL never uses axp gpios support and we were already not building
axp-gpio support for the SPL.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Thu, 23 Apr 2015 16:47:47 +0000 (18:47 +0200)]
sunxi: emac: Remove non driver-model code
All sunxi boards now use the driver-model, so remove the non driver-model
code.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 15 Apr 2015 18:46:48 +0000 (20:46 +0200)]
sunxi: Move all boards to the driver-model
Now that we've everything prepared for it remove the DM settings from the
defconfig(s) and simply always set them for sunxi.
This makes all sunxi boards allways use the driver model for gpios and
ethernet, and allows us to move over more bits to the driver-model without
the need to introduce #ifdef-ery for boards which are not yet using DM.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sat, 18 Apr 2015 21:32:23 +0000 (23:32 +0200)]
sunxi: dts: Add a CONFIG_DEFAULT_DEVICE_TREE setting to all sunxi boards
This is a preparation patch for switching all sunxi boards over to using
the driver model.
Note that rather then defining both CONFIG_DEFAULT_DEVICE_TREE (for u-boot)
and CONFIG_FDTFILE (for the kernel), this commit simply replaces all
CONFIG_FDTFILE defconfig settings with CONFIG_DEFAULT_DEVICE_TREE and
uses CONFIG_DEFAULT_DEVICE_TREE for setting the default fdtfile env value
in sunxi-common.h .
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Thu, 23 Apr 2015 15:04:19 +0000 (17:04 +0200)]
sunxi: dts: Add minimal dts files for board which lack a dts sofar
u-boot has support for a number of boards for which a dts file still needs
to be written, add minimal dts files for these boards so that we can switch
them over to driver-model / fdt.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Hans de Goede [Thu, 23 Apr 2015 12:27:01 +0000 (14:27 +0200)]
sunxi: dts: Add dts files which have been submitted but not yet merged upstream
We need dts files for all boards we support, so bring in a few unmerged ones,
these will be replaced with the upstream merged versions the next time we
sync dts files.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 15 Apr 2015 17:03:49 +0000 (19:03 +0200)]
sunxi: dts: Sync all dts files with upstream kernel
Bring all the sunxi dts files (and update existing ones) from
mripard/sunxi/dt-for-4.1 (which will be merged into upstream master any
day now). This is necessary so that we can move all sunxi boards over to
the driver model.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sun, 19 Apr 2015 09:48:19 +0000 (11:48 +0200)]
sunxi: emac: Add driver model support
Modify the sunxi-emac eth driver to support driver model.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Reviewed-by: Stefan Roese <sr@denx.de>
Hans de Goede [Sat, 25 Apr 2015 11:46:28 +0000 (13:46 +0200)]
sunxi: emac: Rename DMA_CPU_TRRESHOLD to EMAC_RX_BUFSIZE
Besides being spelled wrong, the DMA_CPU_TRRESHOLD define actually has
nothing to do with DMA as we only use mmio fifo access. Rename it to
EMAC_RX_BUFSIZE to properly reflect what it does.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Reviewed-by: Stefan Roese <sr@denx.de>
Hans de Goede [Sat, 18 Apr 2015 12:44:38 +0000 (14:44 +0200)]
sunxi: emac: Prepare for driver-model support
Split all the core functionality out into functions taking a
struct emac_eth_dev *priv argument as preparation for adding driver-model
support.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Reviewed-by: Stefan Roese <sr@denx.de>
Hans de Goede [Thu, 16 Apr 2015 19:47:06 +0000 (21:47 +0200)]
sunxi: emac: port to phylib
This is a preparation-patch for adding device-model support to the emac
driver.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Reviewed-by: Stefan Roese <sr@denx.de>
Hans de Goede [Wed, 22 Apr 2015 15:55:10 +0000 (17:55 +0200)]
sunxi: gmac: Move sunxi_gmac_initialize proto out of netdev.h
netdev.h should not be included in driver-model enabled builds (doing so
causes compiler warnings about struct eth_driver not being declared), but
we do use sunxi_gmac_initialize in the driver-model case, so move it out of
netdev.h .
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 22 Apr 2015 09:31:22 +0000 (11:31 +0200)]
sunxi: axp: Add driver-model support to the axp_gpio code
Add driver-model support to the axp_gpio code, note that this needs a small
tweak to the driver-model version of sunxi_name_to_gpio to deal with the
vbus detect and enable pins which are not standard numbered gpios.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sat, 25 Apr 2015 15:25:14 +0000 (17:25 +0200)]
sunxi: axp: Move axp gpio code to a separate axpi-gpio driver
Move the axp-gpio code out of the drivers/power/axp*.c code, and into
a new separate axpi-gpio driver.
This change drops supports for the gpio3 pin on the axp209, as that requires
special handling, and no boards are using it.
Besides cleaning things up by moving the code to a separate driver, as
a bonus this change also adds support for the (non vusb) gpio pins on the
axp221 and the gpio pins on the axp152.
The new axp-gpio driver gets its own Kconfig option, and is only enabled
on boards which need it. Besides that it only gets enabled in the regular
u-boot build and not for the SPL as we never need it in the SPL.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sat, 25 Apr 2015 20:18:09 +0000 (22:18 +0200)]
sunxi: axp: Add support for i2c based PMICs to the pmic-bus helpers
Add support for the axp152 and axp209 PMICs to the pmic register access
helpers. This is a preparation patch for moving the axp gpio code to a
separate gpio driver.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sat, 25 Apr 2015 12:07:37 +0000 (14:07 +0200)]
sunxi: axp: Move axp pmic register helpers to a separate file
Move the register helpers used to access the registers via p2wi resp.
rsb bus on the otherwise identical axp221 and axp223 pmics to a separate
file, so that they can be used by the upcoming standalone axp gpio driver
too.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 22 Apr 2015 14:27:01 +0000 (16:27 +0200)]
sunxi: axp: Change axp_gpio_foo prototypes to match gpio uclass ops
Change the axp_gpio_foo function prototypes to match the gpio uclass op
prototypes, this is a preparation patch for moving the axp gpio code to
a separate driver-model gpio driver.
Note that the ugly calls with a NULL udev pointer in drivers/gpio/sunxi_gpio.c
this adds are removed in a later patch.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 22 Apr 2015 15:59:01 +0000 (17:59 +0200)]
sunxi: gpio: Build sunxi_name_to_gpio_bank for driver-model code too
When doing a driver-model enabled build we still need sunxi_name_to_gpio_bank
(for now) for the mmc pinmux code in board/sunxi/board.c, so build it for
driver-model enabled builds too.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 22 Apr 2015 09:29:38 +0000 (11:29 +0200)]
sunxi: gpio: Add compatible strings for all supported SoCs
We want to use driver-model/fdt with other model SoCs too, so add
compatible strings for the other SoCs to the dm sunxi gpio code.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Simon Glass [Sat, 18 Apr 2015 17:33:43 +0000 (11:33 -0600)]
sunxi: gpio: Add temporary implementation of name_to_gpio()
Until sunxi moves to device tree (e.g. for USB) we need to convert named
GPIOs to numbers. Add a function to do this.
This fixes the USB / EHCI support not working on the LinkSprite pcDuino3
(which uses devicemodel).
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Simon Glass [Sat, 18 Apr 2015 17:33:44 +0000 (11:33 -0600)]
sunxi: gpio: Rename GPIOs to include a 'P' prefix
By convention, sunxi GPIOs are named PA1, PA2 instead of A1, A2. Change
the driver model GPIO driver for sunxi to use these names.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Hans de Goede [Thu, 23 Apr 2015 21:23:50 +0000 (23:23 +0200)]
sunxi: soft-i2c: Fix gpio handling to work with the driver-model
i2c_init_board() gets called before the driver-model (gpio) code is
initialized, so move the setup of the soft-i2c pins out of i2c_init_board()
and into board_init(), at which time the driver-model setup has been done.
Also add proper error checking and properly request the gpios as that is
mandatory with the driver-model.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 22 Apr 2015 15:45:59 +0000 (17:45 +0200)]
sunxi: display: Fix gpio handling to work with the driver-model
The driver-model gpio functions may return another value then -1 as error,
make the sunxi display code properly handle this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 22 Apr 2015 15:39:59 +0000 (17:39 +0200)]
sunxi: usbc: Fix vbus gpio handling to work with the driver-model
The driver-model gpio functions may return another value then -1 as error,
make the sunxi usbc properly handle this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Wed, 22 Apr 2015 15:03:17 +0000 (17:03 +0200)]
sunxi: mmc: Fix card-detect gpio handling to work with the driver-model
The driver-model gpio functions may return another value then -1 as error,
make the sunxi mmc code properly handle this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Thu, 23 Apr 2015 15:47:22 +0000 (17:47 +0200)]
sunxi: Do not build i2c support when we've no i2c controllers
This fixes the following errors being printed during boot:
Error, wrong i2c adapter 0 max 0 possible
Error, wrong i2c adapter 0 max 0 possible
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Tue, 7 Apr 2015 07:06:39 +0000 (09:06 +0200)]
sunxi: Add Ippo_q8h_v1_2_a33_1024x600 defconfig
Add a defconfig for generic 7" tablets using the Ippo q8h v1.2 pcb,
with an A33 SoC (the pcb can take an A23 or an A33), and a 1024x600 LCD.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Vishnu Patekar [Sun, 1 Mar 2015 18:24:48 +0000 (23:54 +0530)]
sunxi: Add softwinner astar mid756 A33 tablet board defconfig
The Astar MID756 is a 7" tablet using the A33 SoC with a 800x480 LCD screen,
512M RAM, 8G ROM and integrated sdio wifi.
Also see: http://linux-sunxi.org/Softwinner_astar-rda
Signed-off-by: Vishnu Patekar <vishnupatekar0510@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Vishnu Patekar [Sun, 1 Mar 2015 18:17:48 +0000 (23:47 +0530)]
sunxi: Add basic A33 basic support
Enable full support for the A33 SoC including display, otg-usb, etc.
Signed-off-by: Vishnu Patekar <vishnupatekar0510@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Vishnu Patekar [Sun, 1 Mar 2015 18:19:39 +0000 (23:49 +0530)]
sunxi: Add a33 dram init code
Based on Allwinner dram init code from the a33 bsp:
https://github.com/allwinner-zh/bootloader/blob/master/basic_loader/bsp/bsp_for_a33/init_dram/mctl_hal.c
Initial u-boot port by Vishnu Patekar, major cleanup / rewrite by
Hans de Goede.
Signed-off-by: Vishnu Patekar <vishnupatekar0510@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sun, 12 Apr 2015 09:46:41 +0000 (11:46 +0200)]
sunxi: Add support for A33 PLL11 (second DRAM pll)
Add support for the new second DRAM PLL found on the A33 SoC.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 6 Apr 2015 18:55:39 +0000 (20:55 +0200)]
sunxi: s/sun8i/sun8i_a23/
This is a preparation patch for adding A33 support, which will have a
mach name of sun8i-a33.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 6 Apr 2015 18:33:34 +0000 (20:33 +0200)]
sunxi: Introduce a hidden SUNXI_GEN_SUNxI Kconfig bool
sun6i and newer (derived) SoCs such as the sun8i-a23, sun8i-a33 and sun9i
have a various things in common, like having separate ahb reset control
registers, the SID living inside the pmic, custom pmic busses, new style
watchdog, etc.
This commit introduces a new hidden SUNXI_GEN_SUN6I Kconfig bool which can be
used to check for these features avoiding the need for an ever growing list
of "#if defined CONFIG_MACH_SUN?I" conditionals as we add support for more
"new style" sunxi SoCs.
Note that this commit changes the behavior of the gmac and hdmi code for
sun8i and the upcoming sun9i devices. This does not matter as sun8i does
not have gmac nor hdmi, and sun9i has new hardware-blocks for these so
the old code will not work there.
Also this is intentional as if a sun8i / sun9i variant which does use the
old hwblocks shows up then the GEN_SUN6I code paths will be the right ones
to use.
For completeness this also adds a SUNXI_GEN_SUN4I bool for A10/A13/A20.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 6 Apr 2015 18:17:46 +0000 (20:17 +0200)]
sunxi: usbc: Remove unused irq field
We do not use irqs in u-boot so remove the unused irq field, and all the
#ifdef-ery around the irq initialization.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Mon, 6 Apr 2015 18:16:36 +0000 (20:16 +0200)]
sunxi: Also set Auxiliary Ctl SMP bit in SPL
There is no reason not to and this make the #ifdef-ery easier to read.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Ian Campbell <ijc@hellion.org.uk>
Jan Kiszka [Tue, 21 Apr 2015 05:18:25 +0000 (07:18 +0200)]
sun7i: Remove duplicate call to psci_arch_init
This is already invoked a few cycles later in monitor mode by
_secure_monitor (_sunxi_cpu_entry calls _do_nonsec_entry which triggers
_secure_monitor via smc #0). Drop it here, it serves no purpose.
CC: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Tested-by: Thierry Reding <treding@nvidia.com>
Tested-by: Ian Campbell <ijc@hellion.org.uk>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Simon Glass [Thu, 30 Apr 2015 04:26:03 +0000 (22:26 -0600)]
x86: Enable multi-core init for Minnowboard MAX
Enable the CPU uclass and Simple Firmware interface for Minnowbaord MAX. This
enables multi-core support in Linux.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 30 Apr 2015 04:26:02 +0000 (22:26 -0600)]
x86: Add a CPU driver for baytrail
This driver supports multi-core init and sets up the CPU frequencies
correctly.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 30 Apr 2015 04:26:01 +0000 (22:26 -0600)]
x86: Allow CPUs to be set up after relocation
This permits init of additional CPU cores after relocation and when driver
model is ready.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 30 Apr 2015 04:26:00 +0000 (22:26 -0600)]
x86: Add functions to set and clear bits on MSRs
Since we do these sorts of operations a lot, it is useful to have a simpler
API, similar to clrsetbits_le32().
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 30 Apr 2015 04:25:59 +0000 (22:25 -0600)]
x86: Add multi-processor init
Most modern x86 CPUs include more than one CPU core. The OS normally requires
that these 'Application Processors' (APs) be brought up by the boot loader.
Add the required support to U-Boot to init additional APs.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:16 +0000 (20:25 -0600)]
x86: Provide access to the IDT
Add a function to return the address of the Interrupt Descriptor Table.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:15 +0000 (20:25 -0600)]
x86: Store the GDT pointer in global_data
When we start up additional CPUs we want them to use the same Global
Descriptor Table. Store the address of this in global_data so we can
reference it later.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:14 +0000 (20:25 -0600)]
x86: Add an mfence macro
Provide access to this x86 instruction from C code.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:13 +0000 (20:25 -0600)]
x86: Add defines for fixed MTRRs
Add MSR numbers for the fixed MTRRs.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:12 +0000 (20:25 -0600)]
x86: Add atomic operations
Add a subset of this header file from Linux 4.0 to support atomic operations
in U-Boot.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:11 +0000 (20:25 -0600)]
Add a 'cpu' command to print CPU information
Add a simple command which provides access to a list of available CPUs along
with descriptions and basic information.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:10 +0000 (20:25 -0600)]
x86: Add support for the Simple Firmware Interface (SFI)
This provides a way of passing information to Linux without requiring the
full ACPI horror. Provide a rudimentary implementation sufficient to be
recognised and parsed by Linux.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:09 +0000 (20:25 -0600)]
dm: Implement a CPU uclass
It is useful to be able to keep track of the available CPUs in a multi-CPU
system. This uclass is mostly intended for use with SMP systems.
The uclass provides methods for getting basic information about each CPU.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 13:56:43 +0000 (07:56 -0600)]
Add print_freq() to display frequencies nicely
Add a function similar to print_size() that works for frequencies. It can
handle from Hz to GHz.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:07 +0000 (20:25 -0600)]
Move display_options functions to their own header
Before adding one more function, create a separate header to help reduce
the size of common.h. Add the missing function comments and tidy up.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:06 +0000 (20:25 -0600)]
x86: Disable -Werror
This is annoying during development and serves no useful purpose since
warnings are clearly displayed now that we are using Kbuild. Remove this
option.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:05 +0000 (20:25 -0600)]
x86: Remove unwanted MMC debugging
This printf() should not have made it into the code.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:25:04 +0000 (20:25 -0600)]
dm: core: Add a function to bind a driver for a device tree node
Some device tree nodes do not have compatible strings but do require
drivers. This is pretty rare, and somewhat unfortunate. Add a function
to permit creation of a driver for any device tree node.
Signed-off-by: Simon Glass <sjg@chromium.org>
Simon Glass [Wed, 29 Apr 2015 02:25:03 +0000 (20:25 -0600)]
Fix comment nits in board_f.c
Try to make it a little clearer.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:11:32 +0000 (20:11 -0600)]
x86: fsp: Use reset_cpu()
Now that reset_cpu() functions correctly, use it instead of directly
accessing the port on boards that use a Firmware Support Package (FSP).
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:11:31 +0000 (20:11 -0600)]
x86: quark: Use reset_cpu()
Now that reset_cpu() functions correctly, use it instead of directly
accessing the port.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:11:30 +0000 (20:11 -0600)]
x86: ivybridge: Use reset_cpu()
Now that reset_cpu() functions correctly, use it instead of directly
accessing the port.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 29 Apr 2015 02:11:29 +0000 (20:11 -0600)]
x86: Implement reset_cpu() correctly for modern CPUs
The existing code is pretty ancient and is unreliable on modern hardware.
Generally it will hang.
We can use port 0xcf9 to initiate reset on more modern hardware (say in the
last 10 years). Update the reset_cpu() function to do this, and add a new
'full reset' function to perform a full power cycle.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Mon, 20 Apr 2015 13:07:03 +0000 (07:07 -0600)]
x86: link: Add PCH driver to support SPI Flash
U-Boot on coreboot does not have a driver for the PCH so cannot see the
SPI peripheral now that it has moved inside the PCH. Add a simple driver so
that SPI flash works again.
Signed-off-by: Simon Glass <sjg@chromium.org>
Simon Glass [Mon, 20 Apr 2015 04:05:37 +0000 (22:05 -0600)]
x86: Update chromebook_link instructions for binary blob
The MRC image is incorrect, or at least this one now does not seem to
work. Fix it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Gabriel Huau [Sat, 25 Apr 2015 15:13:11 +0000 (08:13 -0700)]
x86: minnowmax: use the correct NOR in the configuration
The SPI NOR on the minnowboard max is a MICRON N25Q064A
Signed-off-by: Gabriel Huau <contact@huau-gabriel.fr>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>