| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This introduces hub master support for the FSI driver, enabling it to initialise
and talk to the FSI address space on the second socket of P9 machines.
UBI and UBIFS support is enabled in the kernel in preparation for using them in
OpenBMC.
The P9 device trees have been brought in sync with each other, with all now
describing the mailbox and lpc drivers necessary for using mboxd for host boot
firmware communication.
Adriana Kobylak (1):
arm: configs: aspeed: Add UBI support
Christopher Bostic (4):
drivers/fsi: Add hub master support
drivers/fsi: Move common read/write code into shared utility
drivers/fsi: Cleanup and retry op on error
drivers/fsi: Initialize slave link field
Edward A. James (3):
drivers: fsi: scom: Zero out user buffer first
drivers: fsi: Fix FSI core size checking user interfaces
drivers: fsi: i2c: Fixup probe to allow multiple engines
Joel Stanley (2):
ARM: dts: aspeed: Fix RAM size in Romulus and Witherspoon
ARM: dts: aspeed: Reserve RAM on P9 machines
Lei YU (1):
ARM: dts: aspeed: Enable Romulus mailbox and LPC control nodes
Change-Id: I26855d0fcce0ba435892cc1f031a0e1b10121f5c
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Set a one-second timeout observed by the kernel after panicking. The
system will reset after the timeout is up.
The default timeout is 0, which means to wait indefinitely after panic.
Fixes openbmc/openbmc#1114
Signed-off-by: Xo Wang <xow@google.com>
Change-Id: I2eec2113fd7598f84dd91412b212ecc2a6949214
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This bumps the kernel to include a port of the FSI I2C driver from the
FSP. It drives the I2C master in the CFAM logic of the host processor
over FSI. This driver is being rewritten to be a proper i2c bus driver,
and as such will be removed once that driver is functionally equivalent.
There is also a fix for the Zaius UART.
A device tree for the Qanta Q71L, a non-openpower BMC, was added.
Christopher Bostic (3):
drivers/fsi: Add slave interrupt polling
drivers/fsi: Add Client IRQ Enable / Disable
drivers/fsi: Add sysfs file to adjust i-poll period
Edward A. James (5):
drivers: fsi: Add i2c client driver
drivers: fsi: i2c: Add engine access wrappers
drivers: fsi: i2c: probe fsi device for i2c client
drivers: fsi: i2c: add driver file operations and bus locking
drivers: fsi: i2c: boe engine
Joel Stanley (2):
drivers: fsi: Fix compilation warnings
aspeed: defconfig: add FSI_I2C driver
Rick Altherr via openbmc (2):
dts: Minimal device tree for Quanta Q71L BMC
arm: aspeed: zaius: Disable LPC reset for UART1
Change-Id: I4fd4e4e347cfe3dead1d84de8b79dc2121346b02
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Mailbox and LPC drivers:
drivers/mailbox: Add Aspeed mailbox driver
drivers/misc: Add Aspeed LPC control driver
dt-bindings: Add Aspeed LPC Control bindings
dt-bindings: Add Aspeed mailbox bindings
Enable the devices for Zaius and Witherspoon:
ARM: dts: aspeed-bmc-opp-witherspoon: Enable mailbox and LPC control
ARM: dts: aspeed-bmc-opp-zaius: Enable mailbox and LPC control node
ARM: dts: aspeed-g5: Add mailbox and LPC control node
Change-Id: I8030404c5c2f6d90935bbd6b26de7143f564f06d
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
| |
High res timers are useful in general, particularly for userspace apps
that call usleep().
Change-Id: I15be1b20141423650ed7f049a4b9ea53160260e3
Signed-off-by: Robert Lippert <rlippert@google.com>
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
U-Boot in Aspeed's SDK had a hack to manually relocate the ramdisk into
RAM because of a bug that caused the normal ramdisk relocation not to
happen if ATAGS was being used. Now that the ATAGS ramdisk relocation
bug is fixed, the load address and entrypoint of a ramdisk image are
ignored. The ramdisk is relocated to an address chosen by U-Boot and
the address is passed to the kernel.
Remove the override of oe_mkimage() in obmc-phosphor-image_types_uboot
as the default implementation is sufficient. This also removes the only
reference to a fixed ramdisk location in RAM.
Change-Id: Id92c8dc9ebeb905401e3056fcf841d982abbd3d1
Signed-off-by: Rick Altherr <raltherr@google.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- FSI changes
* Merge v1 of the FSI patchset as sent to lkml
* Merge Jeremy's raw driver for slave address-space access
* Define pins in Palmetto DTS
* Build the FSI SCOM driver for both ast2400 and ast2500 systems. The
driver has not been enabled in the device tree so it will not affect
any sytems at this stage. We will enable systems as the userspace
changes are made to support the kernel driver.
- Zaius
* Enable LPC to AHB mapping for flash access
- Update the ast2400 systems with the correct amount of RAM
* They systems have 512MB, minus the 16MB framebuffer, but our device
tree restricted them to 256MB.
* We now set them to 512MB with a 16MB no-map reservation at the top
for RAM for the PCI framebuffer
Change-Id: I7377e0f3bd075b78d923ff7295b133a75158e5d9
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The kernel driver for the iBT interface has landed upstream. In the
review process, the location of the device node was changed, as was the
naming of the ioctls and the kconfig symbol.
This change moves to the new version of btbridge, which brings in
autotools support from Matt Barth and a port to the new kernel driver
from Cédric Le Goater. The .bb file is modified to support the autotools
dependencies.
Convert build process to autotools
replace bt_host by bt_bmc
add a --device option
Corrected resulting errors from -Wall & -Werror.
We update the ast2400 and ast2500 defconfigs with the new config symbol,
and bump the kernel to the latest version. The kernel changes clean up
the old driver, and add the new one, including device tree changes.
ARM: dts: aspeed: Enable BT IPMI BMC device
ARM: aspeed: Add defconfigs for CONFIG_ASPEED_BT_IPMI_BMC
ipmi: add an Aspeed BT IPMI BMC driver
ARM: dts: aspeed: remove previous iBT definitions
ARM: aspeed: remove previous definitions in default config
Revert "misc: Add Aspeed BT IPMI host driver"
Change-Id: I8fcc43deaf5db71d587e34c4e4ed2111e7f3e182
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The rule was not creating the /dev/ttyVUART0 rule.
```
root@palmetto:~# udevadm info -q all /dev/ttyS0
P: /devices/platform/ahb/ahb:apb/1e787000.vuart/tty/ttyS0
N: ttyS0
E: DEVNAME=/dev/ttyS0
E: DEVPATH=/devices/platform/ahb/ahb:apb/1e787000.vuart/tty/ttyS0
E: MAJOR=4
E: MINOR=64
E: SUBSYSTEM=tty
E: TAGS=:systemd:
E: USEC_INITIALIZED=22674274
```
```
root@palmetto:~# systemctl status dev-ttyVUART0.device --no-pager
● dev-ttyVUART0.device
Loaded: loaded
Active: inactive (dead)
Nov 07 05:43:37 palmetto systemd[1]: dev-ttyVUART0.device: Job dev-ttyVUART0.device/start timed out.
Nov 07 05:43:37 palmetto systemd[1]: Timed out waiting for device dev-ttyVUART0.device.
Nov 07 05:43:37 palmetto systemd[1]: dev-ttyVUART0.device: Job dev-ttyVUART0.device/start failed with result 'timeout'.
```
This just sits there without returning:
```
root@palmetto:~# systemctl start dev-ttyVUART0.device
```
This is the reason udev does not trigger due to the lpc address on this system
being set to 0x417 when the rule expects 0x3f8:
```
root@palmetto:~# cat /sys/devices/platform/ahb/ahb\:apb/1e787000.vuart/lpc_address
0x417
```
- `obmc-console-server` is launched when `/dev/ttyVUART0` is created.
- `/dev/ttyVUART0` is created when a device with `lpc_address` of `0x3f8` is created
- `lpc_address` is configured by obmc-console-server.
- GOTO 10;
A fix would be to match the udev rule on the address of the peripheral. This is
what Cedric's original patch did:
```
SUBSYSTEM=="tty", ATTR{iomem_base}=="0x1E787000", SYMLINK+="tty-host"
```
Use the iomem_base in place of the lpc_address condition. The rule was
not matching with the driver==aspeed_vuart condition, so that was removed
as well.
Fixes: openbmc/openbmc#749
Fixes: openbmc/openbmc#718
Fixes: openbmc/openbmc#706
Change-Id: Ia55f47a84d657ab9572a8b0495420b1f56e16c02
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Minimaise the differences between the ast2400 and ast2500 defconifgs.
ast2500 changes:
- Enable IPv6
- Disable unused compression algorithms
- Enable SYN cookies
* resolves https://github.com/openbmc/openbmc/issues/504
- Enable kenrel hardening features
- Disable unused USB support
- Enable earlyprintk
- Disable support for ancient libc
ast2400 changes:
- Remove unused configfs support
- Disable IPv6 IPSec support
Change-Id: Id1e388723160541de80b26c378b87a1a2da8091e
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Modified SERIAL_CONSOLES variable in the hardware
systems so that we can get rid of the hacks file.
The hacks file was used to manually stop the issue
but now that is taken out and the issue is handled
in the include files so the process doesn't start
in the first place. This fixes issue number 586.
Resolves openbmc/openbmc#586
Change-Id: I1eb54dae6cbec661f4b8865875e0fe5bd734c70c
Signed-off-by: Josh King <jdking@us.ibm.com>
|
|
|
|
|
| |
Change-Id: Id1f7d22b63493d16a56fbcafe8b33ae971076f5a
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
|
|
|
|
|
|
|
|
|
|
|
| |
Added 61-aspeed-vuart.rules with the AST2400/AST2500 VUART
udev rules. Chose '61' because the basic serial support from
upstream is in '60'.
Fixes openbmc/openbmc#528.
Change-Id: Id7832bd3b421d6ffa6918e12654f9b74833384d6
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The commit message in the kernel patch applied by this commit outlines
the critical information on how it was generated and the limitations
that apply to using GPIO hogs to produce a bootable host.
Please refer to the kernel patch commit message.
Note the patch is best applied during the OpenBMC build process, rather
than being carried by the OpenBMC kernel fork. The OpenBMC kernel fork is
used to test Aspeed SoC support at large, often with non-OpenBMC
userspaces. While the OpenBMC userspace is sometimes not used in kernel
testing, the tests often take place on OpenBMC supported machines such as
the Palmetto. As kernel development and maintenance often includes
exporting various GPIOs to poke at, it's best that we minimise the
number of GPIOs tied up in hogs.
It's acknowledged that this opens an opportunity for predictable flaws
if a kernel is booted with the OpenBMC userspace without this patch
applied, but it's assumed the user knows what they are doing in this
case*.
* Having said that, it will no doubt trip me up.
Change-Id: I08f5f5b29cc81257655bb9a35dbb6c5ce3f11292
Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
|
|
|
|
|
| |
Change-Id: I945d5bc50490193d31c24cf1a691c26377aff625
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
|
|
|
|
|
|
| |
Change-Id: Ifa807e8ecf9d0b94e4681dd4c7af7243fef04536
Signed-off-by: Jaghathiswari Rankappagounder Natarajan <jaghu@google.com>
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
| |
Change-Id: Iffaa780c72f91e8feda6aaa17751b122c2a60aeb
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
| |
Change-Id: I8b1dc64b8f53b3f90c6a24b690507384912908a4
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This moves all machines to a new u-boot branch. The u-boot branch is
based on this month's upstream v2016.07 release, and contains cleanups
that get us closer to submitting upstream.
Importantly for the machine configurations, we now have four new
defconfigs that replace the previous targets:
- ast_g4_ncsi_defconfig: ast2400 with NCSI networking
- ast_g4_phy_defconfig: ast2400 with directly attached PHY networking
- ast_g5_ncsi_defconfig: ast2500 with NCSI networking
- ast_g5_phy_defconfig: ast2500 with directly attached PHY networking
We now support Firestone again with the ast_g4_phy_defconfig.
This commit updates all machines to the new branch and defconfig. It
also removes unused patches that were being made to u-boot, in the case
of barreleye (a patch to aspeednic.c) and palmetto (a modification to
the u-boot environment).
Change-Id: Ic4baf9ae0d5f4046e69cab99d54865771b01cd2a
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
| |
This was left over from when the ast2400 systems used a different kernel
than the ast2500 systems. Now that they are both on 4.6, we can drop
this PREFERRED_VERSION.
Change-Id: I683fdfdc804789fe1353ef09154c92714009cd07
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
| |
The defconfig is regnerated with the newer kernel base. A large number
of unused options are disabled to reduce the kernel binary size.
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
| |
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
|
|
|
|
|
|
|
| |
Move 2016.05 tree to point at the latest openbmc commit and point
all AST2400 machines to it.
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
|
|
|
|
|
|
|
|
|
| |
The AST2500 is an ARM SOC made by Aspeed.
This is only a stub; there are a couple missing bits of support
from the kernel/uboot.
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
|
|
|
|
|
|
| |
In preparation for supporting AST2500 with a different branch.
Signed-off-by: Rick Altherr <raltherr@google.com>
|
|
|
|
|
|
|
| |
An evaluation board target would be nice here but at least Palmetto is
an improvement over the generic ARM target.
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
|
|
|
|
|
|
| |
In preparation for supporting AST2500 with a different branch.
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
|
|
|
|
|
|
| |
Enable system specific modules in their respective layers.
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
|
|
|
|
|
|
|
| |
These patches aren't needed anymore since they are in the openbmc
u-boot fork.
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
|
|
|
|
|
|
|
| |
Barreleye requires adm1278 driver be built into kernel.
Add config to defconfig, instead of hwmon.cfg.
Signed-off-by: Yi Li <adamliyi@msn.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Enable the vuart driver in the defconfig.
Based on the update to aspeed_defconfig in the linux repository.
Won't actually be enabled until the base recipe has the driver
merged again.
Signed-off-by: Milton Miller <miltonm@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The fw_setenv is not working with the kernel mtd driver. It works
on a plain flat file, and flashcp which is used in /update works.
Find the u-boot-env mtd, link it in run and copy the contents to a
plain file in /run during init. In shutdown, check for the link,
the file having data, no current image update for the u-boot-env,
and that the content differs from the mtd via the link. If all
are true create a symlink to the file and trigger the update.
Along wtih changing the fw_env.config this causes the fw_setenv
and fw_printenv utilities to set and update this cached file which
will be synced during a normal bmc reboot. This will allow us to
set the ethaddr variable which already requires a reboot to activate.
Note: the links are directly in /run because the fw_{set,print}env
utilities limit the env file name to 16 characters.
Signed-off-by: Milton Miller <miltonmm@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit creates a class and recipie based on the upstream 2015
branch of u-boot and patches it with our fw_env.config.
Having these tools for the initfs will allow us to modify the u-boot
enviornment to do things like boot from network for update and set the
bmc mac addresses in the environment for u-boot and use by the kernel.
This commit is based on the upstream fw-utils commit. However, the
initial commit was not creating target binaries. After commenting
out the cross class support for creating host binaries to prepare
an environment, a study of the u-boot Makefile showed that HOSTCC
etc needed to be set. It is anticipated something to the commented
lines will be necessary for merging upstream but this addresses our
immediate use case.
Signed-off-by: Milton Miller <miltonm@us.ibm.com>
|
|
|
|
|
|
|
|
| |
Add the fw_env.config file to SRC_URI to include it in the build
tree, and update it to point to /dev/mtd2 as 128k, which the
current flash layout.
Signed-off-by: Milton Miller <miltonm@us.ibm.com>
|
|
|
|
|
|
|
| |
Turn on the new spi-nor and occ drivers. Also add file system
support for the initial filesystem layout.
Signed-off-by: Milton Miller <miltonm@us.ibm.com>
|
|
|
|
|
|
|
|
|
| |
Pull in the fixes that allows netbooting to work well
Without this change netbooting was impossible due to
timeout problems downloading images for netbooting.
Sometimes I would get timeouts right away, other times
it would take longer but every time it fail to copy
the file completely before timeouts would hit.
|
|
|
|
|
|
|
| |
We have an update for the kernel, containing fixes for the i2c subsystem.
This includes a defconfig update.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Missing a 'file://' in linux-obmc SRC_URI
|
|
|
|
| |
hwmon core, nct7904, lm75, and at24.
|
|
|
|
|
|
| |
I finally figured out how to get kernel config fragments working.
This allows removal of some earlier kernel recipe hacks.
Part 2 of 2 ( meta-phosphor, meta-openbmc-bsp )
|
| |
|
|
|
|
| |
Part 1 of 2 (meta-openbmc-bsp, meta-openbmc-machines)
|
| |
|
|
'940e097002829605e2b5ec36f02f59676b33aedc'
git-subtree-dir: meta-openbmc-bsp/meta-aspeed
git-subtree-mainline: ef552f65c3d54126b5ecc813bdd7265ed8f33e4b
git-subtree-split: 940e097002829605e2b5ec36f02f59676b33aedc
|