♟ hselasky

I think the bit-perfect parameters should be used, if possible.

What about providing the default sample bit depth used?

Implement xa_init() in the LinuxKPI as a wrapper for xa_init_flags().

More style fixes (partial revert of r366994).

Fix order of header files:

Run code through "clang-format -style=file" with some additional fixes.

Fix for loading cuse.ko via rc.d . Make sure we declare the cuse(3)

Compile fix for MIPS, MIPS64, POWERPC and POWERPC64.

Fix for colliding change (r366917).

Fix for monotolithic kernel builds using device lagg(4).

Add support for IP over infiniband, IPoIB, to lagg(4). Currently only

Implement mbuf hashing routines for IP over infiniband, IPoIB.

Factor out generic IP over infiniband, IPoIB, definitions and code

Allow if_infiniband to be unloaded.

Implement suggestions from @melifaro .

Implement suggestion from @melfario .

@melfario

require lagg creation to specify laggtype in the creation params, assuming ethernet by default. ifconfig(8) needs to be updated (changes similar to vxlan_create() in sbin/ifconfig/ifvxlan.c).

In D26254#599468, @rpokala wrote:

In D26254#599454, @melifaro wrote:

The only real question/concern I have is the name. I'm a bit afraid that we're saying that "bond" is an IB port-channel and "lagg" is an ethernet port-channel. It's a bit confusing, given one cannot distinguish which is which by name and the fact that Linux folks are used to "bond" as the ethernet port-channel. If we really want to have the different names (to keep some applications happy?), would it be possible to have a name other than "bond"?

I agree with @melifaro ; Link Aggregation Group covers Ethernet (failover, lacp, loadbalance, roundrobin); why can't it also cover InfiniBand failover as part of the same?

Add comment for LAGG's mutex, as suggested by @ae.

Factor hashing functions as suggested by @ae .

MFC r349277 and r366669:

MFC r347596 and r366432:

Remove ifdefs around IS_ALIGNED() definition in the LinuxKPI.

Implement more RCU list functions in the LinuxKPI.

If you MFC it too it is fine. I'll rebase my patch.

This piece of code is about to change with D26254 . I've added the patch there instead.

D26733: Allow IPoIB to work with mutiple FIBs

Need to set VNET before the lladdr event.

Fix for double free of mbuf.

Generate proper lladdr change event.

Implement full support for failover.

Looks good from mlx5_en_main.c's point of view.
@gallatin: Did you test this with mlx5en?

Rebase patch mostly.

Try a bit harder to get the USB device descriptor in case the initial read…

The ethernet header structure is read-only. Add const keyword.

Allow evdev's rcpt_mask and sysmouse_t_axis parameters to be specified in

Properly cleanup driver during remove_one() in mlx5core.

Fix rc.d status command for virtual_oss.

Fix path for virtual_oss when checking pidfile in rc.d file.

Looks good - approved!

Populate the acquire context field of a ww_mutex in the LinuxKPI.

Add support for Google Cr50 (GSC) Closed Case Debugging UART interfaces to

Looks good. You need someone to commit this?

Add missing LD_PRELOAD= environment variable to the virtual_oss rc.d .

Format should use "J" instead, I guess, because it allows -1 as ignore value.

Also sadly add subvendor, etc. fields as comments at the end

to the descriptor_string seem to prevent autoloading and that

Make sure this code compiles for all platforms we have before committing. Not only amd64 and aarch64.

This change looks good. Just address @kib's comments.

hselasky (Hans Petter Selasky)
User

Projects

User Details

Recent Activity
View All

Yesterday

Thu, Oct 29

Wed, Oct 28

Sat, Oct 24

Fri, Oct 23

Thu, Oct 22

Wed, Oct 21

Tue, Oct 20

Mon, Oct 19

Sun, Oct 18

Fri, Oct 16

Tue, Oct 13

Sun, Oct 11

Sat, Oct 10

Thu, Oct 8

Wed, Oct 7

Tue, Oct 6

Sun, Oct 4

Sat, Oct 3

Fri, Oct 2

Thu, Oct 1

Sep 30 2020

hselasky (Hans Petter Selasky)User