- User Since
- Jan 27 2015, 7:34 PM (173 w, 4 d)
Include contents of/update to D15558 again, since relying on dependent revisions
in Phabricator is a PITA.
Pulled out TCP saving changes since those are in D15558
Update due to D15575 being committed.
Fri, May 25
- iflib: Move stats increment in IPv6 case to match IPv4 case
- iflib: Remove comment and restore/fix setting CSUM_IP6_TCP flag during TSO6
- Make style fix.
Replaced by https://reviews.freebsd.org/D15577
Remove testing config files and make spelling fixes.
Thu, May 24
- iflib: Fixup changes made to record TCP checksum info in iflib for ixl
Mon, May 14
Mon, May 7
Apr 17 2018
Update converted version of ixl(4) to 1.9.8-k; includes some other bugfixes and updates.
Apr 8 2018
I've reviewed this patch-by-patch; it all looks good to me. We just need to make sure validation says it compiles and does what it's supposed to, too.
Apr 6 2018
I don't see the console hanging on the version of ixl-iflib where I did see them when watchdogs happened, so this appears to fix the deadlock in that situation.
Mar 19 2018
Committed in rS331224.
Feb 14 2018
You mentioned polling in a previous comment -- did you make changes to the kernel to enable that? I'm assuming you've also made changes to the driver as well, since we don't have polling support in the driver.
Feb 13 2018
As I look at this more, I don't think that last section of the patch is right. ixl_rxeof() makes that same DD bit check in the for-loop; it's supposed to have processed as many descriptors as it can by the time it hits the end of the function.
Feb 9 2018
Until Chris's questions get answered.
Feb 7 2018
These changes look good to me, but I'm going to add a co-worker to get his opinion, too. We may want to delay committing this until we get our internal updated version of ixl(4) committed.
Feb 6 2018
Jan 22 2018
Jan 11 2018
The changes to the Intel ethernet drivers look fine to me.
Is this going to be MFC'd into stable/10 and/or stable/11?
Jan 10 2018
Why didn't they provide a calloc(), to follow the malloc()/free() pattern?
Jan 4 2018
Jan 3 2018
Jan 2 2018
We don't have anything for the Ice Lake hardware, but the Cannon Lake stuff looks like it's good.
Dec 21 2017
Dec 20 2017
Dec 5 2017
Nov 28 2017
Oct 17 2017
I'm working on this on a github branch here:
Sep 23 2017
Aug 24 2017
I guess formally recognize the current state the patch is in.
Aug 22 2017
Can we also get a description of what this fixes, for future reference?
Aug 11 2017
Aug 7 2017
Jul 31 2017
Jul 27 2017
@sbruno - I think Bartosz needed an update to the in-kernel ixl(4) for this to work properly.
Jul 24 2017
Jul 13 2017
With the answer from @sbruno, I approve of the change to em(4).
This should be abandoned, right?
This probably needs to be regenerated right? Or is it going to be dropped due to the upcoming conversion to iflib?
There's still only one entry before the patch. But this looks more correct.
Jul 12 2017
@rlibby , after reading the linked bug report, I think it would be a good idea to go the extern route. Could you regenerate the patch and define ixl_bcast_addr and ixl_fc_string in ixl_pf_main.c? (The latter would only ever get used in the PF driver; only the PF driver uses the former). Or just commit everything but the changes to those two, and I could make those changes.
Is gcc requiring the additional const keywords?
Jul 11 2017
Jul 8 2017
Jul 5 2017
Jun 27 2017
I should add that we're going to update the ixl(4) driver soon because as @bartosz.sobczak_intel.com mentioned, it will fix a crash.