Age | Commit message (Collapse) | Author |
|
On a UP system, there is no "physical-id" in /proc/cpuinfo.
|
|
Rather than reinventing existing code, use library.
|
|
Some single-queue NICs use the multiqueue IRQ naming format
(e.g. eth0-rx-0). We need to select the CPU assignements for these
IRQs using the same algorithm that we use for single-queue NICs using
the standard naming format (e.g. eth0).
|
|
|
|
|
|
|
|
For routing and other applications it is helpful to provide some
mechanism to reserve some set of CPU's and not assign interface
IRQ's to them.
Uses environment variable VYATTA_IRQAFFINITY_BANNED_CPUS
as mechanism similar to irqbalance(8).
|
|
|
|
left over from conversion to new grub.
|
|
|
|
|
|
Multiqueue setup was broken ixgbe because of s/assign/assing/
|
|
|
|
/opt/vyatta/etc/config
|
|
|
|
The script was looking up hostname and setting $LOGFILE, but
these parts are never used in current code.
|
|
|
|
|
|
Bug 7063
Need to convert MAC octets from hex to numeric before doing
math operations.
|
|
Bug 7062
The irq's on Netxen NIC are named:
eth0[0] eth0[1] eth0[2] ...
This confuses the auto IRQ affinity script.
|
|
|
|
The 'terminal' config statement is replaced by terminal_input and
terminal_output.
|
|
Need to pass correct offset for start of root partition
|
|
On tiny disks, then it doesn't make sense to waste space by offsetting
root partition. On big disks, it doesn't hurt and it allows for running
on SSD.
|
|
Bug 7032 (reprise)
Since there are various forms of multi-queue naming, it is better
to just go with the simplest pattern which is to take all the irq's
of form ethX-... and sort them.
|
|
Bug 7032
Fix matchinging of irq's named 'eth0-TxRx-0'. And change the code to
handle any form of IRQ naming of multiqueue that is eth0-xxx-0.
|
|
Parted has magic value '0%' which causes it to choose optimum
alignment (2048 on most disks).
|
|
Redo Bug 7038
The previous change did nothing; the alignment option only reports
alignment, it does not change values. Need to manually reserve space
before root partition.
|
|
Bug 7038
Align the created disk partition for both install-system and initial
install-image.
Note: can't fix existing systems where disk is already partitioned.
They will just have to suffer with poor performance.
|
|
* added /config/scripts/vyatta-postconfig-bootup.script that gets called from
/etc/rc.local. Users/Support can put in any workarounds in the script for
unfixed bugs and enhancements that are not in Vyatta yet. Placing the
script under /config makes sure that changes to the script will be
preserved for the new image
|
|
This script executes scripts lying in the following directory:
/opt/vyatta/bin/sudo-users/check-params-on-reboot.d
Scripts in check-params-on-reboot.d are intended for parameters
that require a reboot to take effect after they are set/changed.
Because of the reboot requirement, underlying value and related
config paramter value can become out-of-sync in couple of cases:
1. change parameter value, don't save config, reboot. Underlying
value is different than the value in config file.
2. install new image; when new image boots, underlying value is
the default Vyatta shipped value not the value in config file.
More info: http://bugzilla.vyatta.com/show_bug.cgi?id=6915#c1
This check-in also adds ipv6_disable_blacklist script in
check-params-on-reboot.d to fix the above mentioned problem
for 'system ipv6 disable' and 'system ipv6 blacklist'
|
|
to push users towards putting scripts in /config/scripts
|
|
|
|
The mislabeled commit ddce08161907797fe914ba609b362d812e23fc8a
Fix wrong name in get_irq_affinity
Was some code to handle Broadcom device IRQ naming convention
that was untested. The part that built regex was incorrectly
expanding a string with regex characters.
|
|
Was using 'smp_affinity' where code meant 'get_irq_affinity'
|
|
|
|
|
|
Bug 6947
When adding device to bond-group, restore smp affinity
|
|
If speed can not be determined then ospeed is undef.
|
|
Bug 6845, 6821
Need to avoid using /sys/class/net/ethX/device/irq (or /sys/bus/pci/.../irq)
because these don't handle MSI or multiqueue. This also resolves issues
with Vmware/ESX.
|
|
Bug 6906
1. Fix parsing of 'Unknown' Speed in regex
2. Don't error if device can not get current speed/duplex
3. Do error if ethtool fails when setting speed/duplex
|
|
is found.
|
|
meant to allow any upper (or lower case) in domain name.
|
|
Bug 6432
Since IPv6 can be disabled several ways and the config
may not match the system, use socket to test if IPv6 is disabled.
|
|
Bug 6853
Handle OID format when ifIndex is 10 or greater.
Needed to fix regex parsing.
Also allow symbolic OID.
|
|
Bug 6818
The regex to extract wireless name was incorrect.
Recode for clarity rather than adding more complex regex.
|
|
Bug 6861
Fix submitted by devolish <ole.myhre@dataoppdrag.no>
|
|
Bug 6845
Warn (and ignore) attempt to assign IRQ directly on multiqueuej
NIC.
|
|
Bug 6837
Xen does not behave like normal hardware. It assigns a random MAC
address (with local bit set).
|
|
Can't do it early in boot because ipv6 module may not be loaded.
|