illumos

Today

04:55 PM illumos gate Bug #12498 (New): bge ring interrupt masking logic is broken

During additions of the 5719/5720 some refactoring to match a once existing Broadcom version of the bge driver. As a ... Robert Mustacchi

04:52 PM illumos gate Bug #12497 (New): bge ape locking left always disabled after 7513

In addressing 7513, it was important to determine whether or not the hardware supported ape before looking at and ini... Robert Mustacchi

04:48 PM illumos gate Bug #12496 (New): bge mac address initialization is wrong

The way that @bge@ does its mac addresses initailization and management isn't quite correct. When updates for the 571... Robert Mustacchi

2020-04-07

06:55 PM illumos gate Feature #12466: Enable IPv6 TSO Support for vioif

I tested this by going back to back in two HVM VMs in Linux using iperf. I verified that TSO was being used for IPv6.... Robert Mustacchi

06:29 PM illumos gate Bug #12467: Add support for AMD PPIN

To test this, we booted this in three different situations:
* With an AMD EPYC Rome CPU that has access to ppin.
... Robert Mustacchi

2020-04-03

01:12 AM illumos gate Bug #12468 (Closed): Remove generic_cpu -Wno-parentheses gag

This cleans up the -Wno-parentheses gag in the generic_cpu kernel module. This was tested along with 12466. As part o... Robert Mustacchi

01:10 AM illumos gate Bug #12467 (Closed): Add support for AMD PPIN

In 9747 we added support for Intel's PPIN feature to facilitate the use of CPU auto-replace. AMD added support for th... Robert Mustacchi

12:36 AM illumos gate Feature #12466 (Closed): Enable IPv6 TSO Support for vioif

While coding up 12465, I basically had written most of the code for IPv6 TSO support, but realized that the rest of t... Robert Mustacchi

12:31 AM illumos gate Bug #12465 (Closed): vioif needs length for tso checksum

For a while I'd noticed that @pkg update@ between two VMs on a Linux laptop were having pretty terrible transfer perf... Robert Mustacchi

2020-04-02

06:23 PM illumos gate Feature #12446: 3KSTAT manual pages could be clearer about data life time

I tested this with man(1). Robert Mustacchi