[Xenomai] Please review wip/rtnet-fixes

Jan Kiszka jan.kiszka at siemens.com
Wed Feb 28 20:36:06 CET 2018


On 2018-02-28 19:24, Philippe Gerum wrote:
> 
> Gentlemen, we have 30 patches waiting for review by RTnet gurus in
> wip/rtnet-fixes. We already had some positive feedback about them fixing
> the ugly SMAP-related issue, which we may want to merge to stable-3.0.x
> asap, but there are a couple of unrelated changes to review as well, namely:
> 
> 9a6bbc384 net/socket: forward private ioctl requests to NIC driver

Seems harmless, but do we have a user for that? I'm also skeptical if
SIOCPROTOPRIVATE should actually target a device driver. Seems more like
it's used at higher levels.

> b78817868 net/socket: align rtdev do_ioctl handler on the regular
> ndo_do_ioctl

Mostly fine, except for some unrelated whitespace breakage in
kernel/drivers/net/stack/socket.c.

> 40b9bdc44 net/socket: enforce secondary mode for SIOCETHTOOL

Looks good, just also indent the "break" of the SIOCETHTOOL block with tabs.

> 
> I don't see anything controversial in those patches, but I'd rather ask
> before merging.
> 
> 3.0.7 is currently on hold until we have wip/rtnet-fixes merged in whole
> or in part (at least the SMAP stuff), so we direly need to make some
> progress here.

I didn't look into all details, rather relying on test results as well.
Some minor findings:

There is a ping-pong debug change in "net/tcp: recvmsg: remove direct
references to user memory" vs. "net/tcp: sendmsg: remove direct
references to user memory". Please remove.

"net/udp: ioctl: fix temp arg buffer type" should get folded into the
commit that caused the regression ("net/udp: ioctl: remove direct
references to user memory").

Jan



More information about the Xenomai mailing list