| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Needed to extend the time to allow OCC to wait for our drivers to stop
using the SBEFIFO. Otherwise we get some contention on the FIFO.
Change-Id: Ia8aa50a328e45089de1ef43a2e7f685e74f96730
Signed-off-by: Edward A. James <eajames@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
| |
We need to unbind the occ-hmwon devices when either the host crashes or
we get a watchdog timeout on the host. This is so that OCC code can
start using the SBE FIFO without contention in the case of a checkstop
or other host problem.
Change-Id: I9ae3b1539a9eb4240cef74f2198c05b7a1b68380
Signed-off-by: Edward A. James <eajames@us.ibm.com>
|
|
|
|
|
| |
Change-Id: I28f21c752952f482ff605c9b48581e1d962c926f
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
| |
Bumping to pick up pflash fix
lists.ozlabs.org/pipermail/skiboot/2017-November/009601.html
Change-Id: I23eb1a32c7fde316e1d125eccdced397fb12adab
Signed-off-by: Gunnar Mills <gmills@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The Before directive had the template service mboxd-reload@.service
but because it wasn't instantiated, the dependency was not taking
place and the mboxd-reset service would not wait for the
updatesymlinks service to finish, causing the host to read old data
until the updatesymlinks completed.
Resolves openbmc/openbmc#2695
Change-Id: I3669155f1c287d5f492f3400ab9d131913a2a93f
Signed-off-by: Adriana Kobylak <anoo@us.ibm.com>
|
|
|
|
|
|
|
|
| |
Add override YAMLs for both phosphor and legacy time daemons to have
host to set BMC time.
Change-Id: I3a2b258f229ad4e110c456bd7e08c0264e85d0f6
Signed-off-by: Xo Wang <xow@google.com>
|
|
|
|
|
|
|
|
|
|
| |
The extrasdev package name was updated to packagegroup-obmc-apps-extrasdev,
update it in the barreleye platform to the new name. Also move it to the
barreleye.conf since it appears the package remove does not take effect
anymore on the local.conf.sample file.
Change-Id: Ice727a27ffb7a2328142e1e94fd1a6480ea1c6ec
Signed-off-by: Adriana Kobylak <anoo@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There should be 38 temps total that are of mode 'label',
not 32. This mode directs the code to read the label files
to determine what the tempX_input temperature values
actually represent.
The 38 are: 24 core + 8 DIMM + 6 GPU (3 GPUs, core + mem temp for each)
Resolves openbmc/openbmc#2674
Change-Id: Ic10568ace0cdd991ae02475e498f3d179a1b148b
Signed-off-by: Matt Spinler <spinler@us.ibm.com>
|
|
|
|
|
|
|
|
| |
Perform a hard power off when the defined number of fans are
nonfunctional or not present.
Change-Id: I592470dfa96e7a9cd3183f1bac39fa7693269972
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
While testing missing power supply, the bottom power supply in the MRW
is labeled as psu1, which would obviously map to powersupply1, but when
that supply was removed and a power on attempted, the call out was for
powerupply0. A simple rename of the labels does not work out right, due
to the filenames, so swizzling the GPIO and I2C information.
Resolves: openbmc/openbmc#2563
Change-Id: Iaed300becace2409e04c0097be9e2d7fa497dee3
Signed-off-by: Brandon Wyman <bjwyman@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add Witherspoon dbus-monitor rules to watch for power supplies
to become present. When that happens, invoke the callback to
resolve all errors that have that power supply as a callout.
This will cause the enclosure fault LEDs to turn off when a power
supply is replaced (if no other faults were also setting those
LEDs.)
Change-Id: I16e603d18fe3c458e1f062e12a7aa0a6a2f4300e
Signed-off-by: Matt Spinler <spinler@us.ibm.com>
|
|
|
|
|
| |
Change-Id: I6af058890fdfdf26c23dbb034f364764264f951e
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: I10a7352dfdb708bccf6af9863901226607c6121a
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
| |
Change RDEPENDS for openpower-occ-control to override syntax predicated
on "zaius" (which is added by MACHINE_OVERRIDES).
Change-Id: Iaaae449bc55d129b7102052f2440a6f78d12bcd1
Signed-off-by: Xo Wang <xow@google.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Create a new development tools package and move the obmc-pydevtools there
so they're no longer in the extrasdev development feature package.
The reason is that the extrasdev contains features that are desirable to
be removed from production systems, such as the rest-dbus. On the other
hand the pydevtools are binaries that are useful to have even in
production systems, so makes sense to put them in a separate tools
category.
Resolves openbmc/openbmc#2571
Change-Id: Ia66bf6e83d920818c8c355ec9d01a047583b82c0
Signed-off-by: Adriana Kobylak <anoo@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Increasing the allowed out-of-range time to 30sec for large fan speed
changes where the fan hardware was unable to reach the allowed minimum
target speed within 20sec.
This effectively bumps the shutdown due to fan fail/missing to 35sec
since its a 5sec delay after the fans would be marked nonfunctional.
Resolves openbmc/openbmc#2660
Change-Id: I48f83b89cfbe91e8d5a6ff9387998644c51b6174
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add Witherspoon dbus-monitor rules to watch for fans becoming
functional, which will invoke the callback that resolves all errors
that called out that fan.
This will cause the fault LEDs to turn off when the fan
is replaced and spins back up to speed.
Resolves openbmc/openbmc#2513
Change-Id: I7f787854915ca2bb1b66508b343eaa137cede6fd
Signed-off-by: Matt Spinler <spinler@us.ibm.com>
|
|
|
|
|
|
|
|
| |
Change the hardcoded power supply inventory paths to match the ones in
the MRW, which are used by the LED manager.
Change-Id: I1ba52286233e27b95c7296137aa135bbb4ef2d62
Signed-off-by: Brandon Wyman <bjwyman@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
New features:
- Hardware Trace Macro support
- Host backend
- New command: sreset
- Build and bug fixes
Alistair Popple (6):
libpdbg: Add sreset command
libpdbg/target.c: Probe all root nodes in the device-tree
Rework POWER9 device-trees to share common definitions
Add backend for host based debugfs
Add indirect SCOM access
libpdbg/device.c: Remove usage of libfdt internal functions
Cyril Bur (7):
Add *.dtsi to .gitignore
fake.dts: Whitespace cleanup
Look for 'chip-id' as opposed to 'ibm,chip-id'
Add Hardware Trace Macro (HTM) to device trees
Hardware Trace Macro (HTM)
docs: Add HTM to README.md
libpdbg: Fix incorrect restore of r1 in ram_instructions()
Joel Stanley (11):
Makefile.am: Set foreign
configure.ac: Enable silent build rules
build: Update to do both an in and out of tree build
Do not redefine _LARGEFILE64_SOURCE
Allow command line to override GIT_SHA1
kernel: Include endian.h for htobe32 and be32toh
device: remove use of ccan's endian
libpdbg: Remove unused __unused from compiler.h
Error out when no device type is specified
lipdbg/i2c: Get bus path from device tree
i2c: Update i2c device node in dt when specified
Thomas Petazzoni (1):
Rework generation of .dtb.o
Change-Id: I382bfb0d3aacdf03495450acfc0c0792835d5f36
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
| |
Change-Id: I9da4f5eab06d45e0440c275c5cbbab0cc259ad5f
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: I352a93784ce5f3c146e33dc4d0dbf0ca58a6751b
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: I3b8ff2e2ec6dcba3b19f32bbaaadcc267f75d913
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
|
|
| |
Fan control init and fan monitor init should also trigger the fan
watchdog that drives fans to full speed if they fail during poweron.
This will allow a system to power on if either of these fail or after a
previous condition persists where the fan watchdog was triggered.
Change-Id: I2bc4c5e6445406a6ffdf5d5ce5fa9dc3206abbeb
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
| |
Change-Id: If7bc1ba9bffc4cbbbebb56be2e6588b8f8cd1a00
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Recently hit a situation where an application was
core dumping consistently and this maxed out
the allowed size for BMC dumps (1MB). When this
happened, the BMC started to reach it's maximum
persistent filesystem size in /var.
The persistent journal usage in /var needs to allow
for this worst case usage by BMC dump.
This is only an issue on systems where the journal is
written to persistent storage.
Resolves openbmc/openbmc#2595
Change-Id: Iab484e59d6586677938fe20673716203dacb0846
Signed-off-by: Andrew Geissler <andrewg@us.ibm.com>
|
|
|
|
|
| |
Change-Id: Id3116f6922ebe25e3b6b698f86cf2d2ae8121edb
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
| |
Change-Id: I13b4f9067809f5e33d02df593ef87d627f365398
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch improves the obmc-flash-bios script by blocking attempts to
create and mount a pnor-patch volume when the system is in field mode.
When this scenario is reached, the current behavior is to error out
because the script tries to mount pnor-patch to /usr/local/share/pnor,
which is inaccessible in field mode. This fix simply blocks these
attempts to suppress the error message.
Change-Id: I78b5171e66f194d7b8d2b1baa639ee29a1b11794
Signed-off-by: Michael Tritz <mtritz@us.ibm.com>
|
|
|
|
|
| |
Change-Id: I6d825abdbf63bd4bca5c49a5aeab8d26123d43b0
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
|
|
| |
Mistakenly removed air cooled events from being created when
specifically enabling water cooled only. Adding these back in as fan
control will continue to be limited on air cooled systems with the floor
held at 7200 RPMs.
Change-Id: Ia910b51564fd902e0b0e8c067d5649a689fd100a
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
|
|
|
| |
Program management design change requires an orderly shutdown occur when
not enough fans are found functional or present.
Change-Id: Icbadcf6e4fd962f0b092a6686aaa88751e57fb53
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
| |
Change-Id: I76cd275eccd2ae7f076572a197a1aa6250083b5c
Signed-off-by: Joel Stanley <joel@jms.id.au>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add a dependency to the updatesymlinks.service to not run in the case
that the BMC rebooted while the host was powered on, to prevent the
PNOR version from being changed when the host is on.
This situation would happen if the user performed a PNOR update at
power on (the version is not activated), then rebooted the BMC, the
expectation is that the host continues to run the old version until
the host is powered off and powered back on.
The op-reset-chassis-running@0.service is the one that creates the
file in /run/openbmc/ to indicate that the host was on prior to
BMC reboot.
Resolves openbmc/openbmc#2533
Change-Id: I5169af00c1e992a6c72907733430087defbb48ed
Signed-off-by: Adriana Kobylak <anoo@us.ibm.com>
|
|
|
|
|
| |
Change-Id: Ie9ba0284e51e6e4f90df24fbffd8ff7e5ef82cec
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
|
| |
Water cooled thermal parameters have been verified and ok'd to integrate
by thermal team. Air cooled thermal parameters are still being
characterized, so this keeps air cooled unchanged.
Change-Id: I3d90c30c07856b9d74896625b37ccdf9764e9ed4
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
| |
Change-Id: I3042f54f3e1fe9e11f8772e93f8f7e2eabd59f7d
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: Ia580de59264ec9fa85f4b05b26698ea273131d55
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: I2f98c481b7abb2213b71b372ee6428b03076c1d8
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: I7713619a306ec6f9672895e1556f7ea418fe228c
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Moved the code to create NotPresent and Nonfunctional errors to a new
yaml, fan-errors.yaml.
Fixed the problem with duplicate errors, caused by a 2nd fan
going nonfunctional or not present which would create a 2nd error
for the 1st fan previously nonfunctional or not present.
Create a notpresent error if fan 0, 1, 2, or 3 is not present for more
than 20 seconds. Create a Nonfunctional error if fan 0, 1, 2, or 3 is not
functional for any amount of time.
The system must be powered on in both of these cases.
If a water cooled system, don't create errors for fan 1.
An error is created each time the chassis powers on if a fan is
removed or nonfunctional.
Resolves openbmc/openbmc#2472
Change-Id: Ibd71bf3a3b2381df623e7310a510a2f044c525e4
Signed-off-by: Gunnar Mills <gmills@us.ibm.com>
|
|
|
|
|
|
|
|
| |
Set power led blink on bmc_booted, and set it on when host is
powered on.
Change-Id: Ia25c1b2fa358ed508bc02b7b8545d8b0d93586a4
Signed-off-by: Lei YU <mine260309@gmail.com>
|
|
|
|
|
| |
Change-Id: I65c612edfded8749f4cff54a760a886681492c89
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
|
|
|
|
| |
According to thermal characterization, the ambient temperature sensor
requires a -2C offset applied to reflect the correct ambient
temperature.
Change-Id: I354988013d51ee57403fa68c3fa5120e59d16b74
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>
|
|
|
|
|
| |
Change-Id: Id9862afcd019817d26005cd7384ba11ac8d730dd
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: Ia814828f41083ae2cd9dd86ff9733f489ea03f0f
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: I705b73b5e7d766e695a6e3f6c5f54bb63bc1cc65
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
|
|
|
|
|
| |
Change-Id: I25e888ab410c558abb2d5e90c52f0dfcef5849a2
Signed-off-by: Gunnar Mills <gmills@us.ibm.com>
|
|
|
|
|
|
|
|
|
| |
Spelling errors found using github.com/lucasdemarchi/codespell
A tool to fix common misspellings.
This tool is licensed under GNU General Public License, version 2.
Change-Id: I649a843483394562fef2e9f1a8b3d38c0cbc6a38
Signed-off-by: Gunnar Mills <gmills@us.ibm.com>
|
|
|
|
|
|
|
|
|
| |
Create a Shutdown error log when the system shuts down due to not
enough present and functional fans.
Changed in both air cooled and water cooled yamls.
Change-Id: I1dbf13e5689bfd7a1b08eb1d1c52e3de3a084642
Signed-off-by: Gunnar Mills <gmills@us.ibm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently the journal is written to a RAM based
filesystem which is lost on BMC reboots. This commit
will move the journal to a persistent filesystem location
so it will be available after BMC reboots for debug.
This change is only being done by default on
witherspoon systems with this commit. Other systems
may use this as a reference if they wish to move.
By default, journald uses 10% of the filesystem
(default is /var/log/journal/) and it will ensure 15%
of the filesystem space is left free. Specific machine
configs in openbmc can override these defaults if
they wish. For witherspoon, we will use 3MB max and use
the 15% left free default.
Due to the way openbmc mounts its filesytems, a
systemd-journald.service file override is required to
ensure journald does not start until the /etc/ filesystem
is fully mounted.
journald uses /etc/machine-id to create a directory
in which it stores its journal data. journalctl uses
/etc/machine-id to know which directory to look into
when a user requests the data. Without this override,
journald will end up making a random machine-id
and using that because /etc/machine-id is not available
when systemd-journald.service is started. journalctl
looks up the /etc/machine-id when its run, so it will
use the correct one (but not the one journald is using).
Witherspoon uses the new UBI filesystem, and only
has a total of 4MB available within /var. Ensure the
journal only ever uses 3MB of that space to leave room
for other applications.
In testing it was found that the persistence of journald
had minor impacts to overall flash erase blocks. With
persistence enabled or disabled, it was noted that the
UBI maximum erase count value incremented by 1 for every
10-20 boots of the host. Our CI system, which gets a
lot of activity (new image flashing, REST api testing,
and host boots) average 10-12 erase counts a day. A worst
case of 15 erase blocks a day would mean 18.2 years before
hitting the 100,000 erase block limit.
Resolves openbmc/openbmc#1627
Change-Id: I30211108f26bd9cd758800f457c17ed03d13e802
Signed-off-by: Andrew Geissler <andrewg@us.ibm.com>
|