Age | Commit message (Collapse) | Author | Files | Lines |
|
* Deprecate N-1 release (hardknott).
* Enable N+1 release (kirkstone).
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I39e027e02dab64b4390b46ffbd9c299c858f403e
|
|
We don't want to accidentally notify or continue on with the DHCP
netboot process until after the updated image is installed.
Change-Id: I05504632950566694e9dd60b25bd85c99b588f90
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This will allow us to have non-bridge USB devices on the system.
Change-Id: I5102fa8e1039013d9629b889350d46c6acf5cde3
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
We want to respond to neighbor solicits over NCSI for all public IPs
on the bridge.
Change-Id: I9179ca6cb82d956a6b1db82c075e803248d7248d
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This is not needed for runtime service to work correctly as the bridges
on both sides come up before the USB device is present anyway.
Change-Id: I27b6ad5b265ff18f78f41388d26b02ced39749ae
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This about doubles the performance of the network interface over USB2
Change-Id: Id8f79435b9cb7f02fa9094bf50e90023dfe188a8
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This makes it possible for our installation service to determine when
the BMC is ready to continue after DHCP.
Change-Id: I62f31c39a9d5849d18f81ab58c88e9e5ef4b9737
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
We don't want to report a successful DHCP if some of the hooks failed to
run.
Change-Id: Iaafa555fa99381f0c7ff6bd05fd7038366356fbd
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This will be picked up by the gbmc-bridge DHCP process in another
change.
Change-Id: Id80688d0d7a7872bca2eec70c31468a483f72feb
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
The MAINTAINERS files are no longer used and are out of date (references
to IRC still exist in most of them). Remove them and rely on the OWNERS
exclusively.
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I0f5d2719ad97e218ce03c4503efe1b1c92ac455e
|
|
William A. Kennington III (1):
ncsid: Fix IP / Gateway updating
Change-Id: Ia010aa536100c4e42189412510e68921108fca6d
Signed-off-by: Andrew Geissler <openbmcbump-github@yahoo.com>
|
|
William A. Kennington III (4):
libcr51sign: Temporarily ignore deprecate functions
ncsid: Fix update_ra_gw for multi-RA
dhcp-done: Add daemon
ncsid: Use per-interface gateway
Change-Id: I1205ef4bea5a2d428b932e6103cef54bb5e47f7f
Signed-off-by: Andrew Geissler <openbmcbump-github@yahoo.com>
|
|
Willy Tu (1):
google-ipmi-i2c: Remove venture from reviewer
Change-Id: I5c6312f7484607b862d8fdf15dfef479a242b001
Signed-off-by: Andrew Geissler <openbmcbump-github@yahoo.com>
|
|
A powercycle will be required in order to kick off DHCP again. If the
machine succeeds once, the power cycle service will be available for
future attempts.
Change-Id: I3b741250fd673bbe19056506f7287ef2ed734ea0
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Sometimes, the first value returned is not stable and not what is
expected from the FRU. If we perform multiple queries we can get a
stable reading and correctly detect word length.
Change-Id: I73a88870bef231566f0249d0629c751be6a5002c
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
A number of machines need network debugging and it's useful to have this
available in our dev builds.
Change-Id: I37c0a7317295fb6d75c2fcf2cc913b63dd9b20a7
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
We don't want to break the build if the compiler becomes more strict.
Change-Id: I942a18373fb154f3a4b6dea646e8cec532a5d347
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Tested: builds
Signed-off-by: Nan Zhou <nanzhoumails@gmail.com>
Change-Id: I43d7ff60626c11195c887dfb169aec72c171e1f8
|
|
We want to avoid using the dhcp relay for the primary node until our NIC
proxying code can handle relayed packets or we turn it off entirely.
Change-Id: I465edbe20d4be74c7a560f411e61e95fd89c59d1
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
We need a workaround for some well known configurations where it is safe
to use the default DHCP source port for sending relayed messages. Some
of our upstream server and relay software does not understand alternate
ports.
This will be removed when that support is fully available.
Change-Id: I24ea031c23556c62a26875dddac25708b6b8de79
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
The logic was not usable in our current environment. This makes some
tweaks to download the image correctly.
Change-Id: I86741e6029c2937d0a8106ac22a1738fa0e77cad
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Distinguish between broadcast only packets and broadcast + unicast
packets.
Change-Id: I52f7969ccd00e4142d5585a37b92cc5530b76dec
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Avoid QA warnings about fuzzy patch application.
Generated with:
devtool finish --force-patch-refresh npcm7xx-igps-native meta-nuvoton
Signed-off-by: Peter Foley <pefoley@google.com>
Change-Id: I2472c5b5f42b9f174b273ac5aa995043c4a44a25
|
|
Per [1], Yocto is going to start requiring both branch names and
`protocol=https` in GitHub SRC_URI variables. Ran the upstream tool to
automatically fix our URIs.
1. https://git.yoctoproject.org/poky/tree/documentation/migration-guides/migration-3.5.rst#n10
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I5f9619f7ff707050b469ae74c8aa16846568c5e4
|
|
This is required for OpenBMC meson projects
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
Change-Id: I80805405c1214820ef6cf26467785aa0ae5247ca
|
|
The initrd needs to be able to pet the watchdog prior to starting an
update. If this doesn't happen, we may trip the watchdog before writing
out the entire update payload.
Change-Id: I9974263585d2143e8bebb2a1522133d53e911eb7
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
William A. Kennington III (1):
libcr51sign: Sync with latest code
Change-Id: I79759b0a18c63f7b5ef49511144d919e38c8ab73
Signed-off-by: Andrew Geissler <openbmcbump-github@yahoo.com>
|
|
This service launches eStoraged for the eMMC.
Tested:
Installed the image on a machine and confirmed that the systemd service
started automatically.
Signed-off-by: John Wedig <johnwedig@google.com>
Change-Id: I55567da8ef491af893e1749ee05d2f39fa01758b
|
|
When running an image with a certain version of QEMU, the emulated BMC
may enter bootloop due to the Watchdog not getting petted for more than
10s.
One can check this by monitoring
s->tim[1]->watchdog_timer.base_timer.expires_ns in QEMU (where s may be
an NPCM7xxState or similar hardware state object).
Reverting the timeout to 60s seems to resolve the issue with QEMU usage.
Signed-off-by: Sui Chen <suichen@google.com>
Change-Id: I29313ff67415003d5f82f0f8a5cc1703b34fb2c0
|
|
Add distro override to avoid os-release setting effect other
not gbmc distro
Signed-off-by: Brian Ma <chma0@nuvoton.com>
Change-Id: I34afb4ce8e72f2be15a5f5668e6ff6c567548ad5
|
|
This saved 49152 bytes of compressed image space in our platform.
Signed-off-by: Josh Lehan <krellan@google.com>
Signed-off-by: Brandon Kim <brandonkim@google.com>
Change-Id: I4b5b4c7cd2bfffb9720bae02624519ee10ca73af
|
|
This saved 49152 bytes of compressed image space in our platform.
Signed-off-by: Josh Lehan <krellan@google.com>
Signed-off-by: Brandon Kim <brandonkim@google.com>
Change-Id: I40016cffd8586bbcda1cc45ec968efd35c8f4188
|
|
gBMC does not use kernel modules
Change-Id: I75d8309a787cb74f3be1f5bae08de2897ffdac0a
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
There are generally other ways to poke at the IO space of the BMC and
there isn't any evidence these are being used anymore.
Change-Id: I9d0e7187f2be8bbbf349e8cdf63c32013876260c
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Add systemd.conf options for watchdog to replace watchdog-conf
```
// interval is 60s by default
interval = 10
watchdog-device = /dev/watchdog
// The value is in pages. They are presently 4KiB each.
// Reset the BMC if it goes below 3MiB
min-memory = 768
```
Tested:
```
$ cat /lib/systemd/system.conf.d/40-system.conf
[Manager]
DefaultMemoryAccounting=yes
RuntimeWatchdogSec=10s
WatchdogDevice=/dev/watchdog
```
Change-Id: I51171746669dfeb1efe6c09952efa3457e33c86b
Signed-off-by: Willy Tu <wltu@google.com>
|
|
Change-Id: I01bfbd4022584641d8df6ee694b62eadc3cb657c
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This trims down our busybox binary to remove a bunch of unused /
duplicate utilities.
Change-Id: Iabee2b086436db800c35f5b9cd738704c493df50
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This ensures that all of the rules are processed and unexpected packets
are not allowed or blocked by the kernel at any time.
Change-Id: Ia7bb1d7f604f8ed1bd9759a23e370d20cb0c690d
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Change-Id: I5fc6db2510b9c27d8d2ce27304e0a5e821015fb3
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Patrick Williams (3):
nemora-postd: fix compile fail
sdbusplus: remove usage of deprecated alias
MAINTAINERS: fix syntax
William A. Kennington III (2):
ncsid: Only log errors a single time
ncsid: update_ra_gw: Fix SIGTERM handling
Change-Id: I6af4e21f61b4e126979cac3042bfac72516f62e3
Signed-off-by: Andrew Geissler <openbmcbump-github@yahoo.com>
|
|
The receipt of the RA is indicative of a non-NCSI configuration so we
can shut down the daemon in this instance.
Change-Id: Iff958be408cff853973bd335f2459cd3d9bca0e0
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This makes it possible for the BMC to fetch an upgraded image during
DHCP time and perform an inplace update to the new version.
Change-Id: I6d32af306180dd52d4235ba4edbb80c3e49756d7
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Change-Id: If3dacbfde966de337702b44b9b06b3eb21755546
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
The nft rules were not deleted in cases where the public addresses are
removed from the gbmc-br interface. This would create broken rules.
Change-Id: I22a88f1fb15ccbea49e586061ea8e93bbbfb1bc1
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
We were missing this address but we want neighbor discovery from outside
the BMC to work against this range.
Change-Id: I6ef139486f382df21596c460626bfe2f692c7236
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
We use functions from the network-sh library and need to include it.
Change-Id: I7f78b7dd37c4e5d38342c4625c26c4b583133bf5
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This ensures they are run prior to running other DHCP configuration
logic.
Change-Id: I0180414bbc291ea49e548a312cd2b713344613e1
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This makes it possible for a BMC to acquire a public address via DHCP
provisioning.
None of the update processes are included yet.
Change-Id: I51ca2aa4859bcd2b9e909dd5a0d9e66cfbd648af
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
This scans the gbmcbr interface for public addresses, and adds the
relevant addresses to the NCSI interface of the BMC. This is required
for neighbor discovery to work from prod over the NCSI link, when the
addresses do not already exist (BMC DHCP will not have them).
Change-Id: I27ff0cd3c4750b752b35399b8a0288db5ac9fe28
Signed-off-by: William A. Kennington III <wak@google.com>
|
|
Enabled `update-functional-on-fail` by default for gBMC.
Change-Id: I54b4fbaceb65c09b7b9cb72d7b7dcb7d048c7eba
Signed-off-by: Willy Tu <wltu@google.com>
|