Thread (9 messages) 9 messages, 3 authors, 2011-02-28

Re: [PATCH] cxgb{3,4}: streamline Kconfig options

From: Jan Beulich <hidden>
Date: 2011-02-24 08:00:23
Also in: linux-kbuild

quoted
quoted
On 23.02.11 at 21:27, David Miller [off-list ref] wrote:
From: "Jan Beulich" <redacted>
Date: Wed, 23 Feb 2011 09:46:10 +0000
quoted
quoted
quoted
quoted
On 22.02.11 at 19:14, David Miller [off-list ref] wrote:
From: "Jan Beulich" <redacted>
Date: Thu, 17 Feb 2011 13:29:30 +0000
quoted
The CHELSIO_T{3,4}_DEPENDS options are really awkward, and can be
easily dropped if the reverse dependencies of SCSI_CXGB{3,4}_ISCSI on
the former get converted to normal (forward) ones referring to
CHELSIO_T{3,4}.

Signed-off-by: Jan Beulich <redacted>
I think the goal of these strange rules is not to be complicated
on purpose, but rather to cause the iSCSI drivers to appear without
the user having to know that he needs to enable the networking
driver in order for that to happen.
While I realize that this might have been the reason, it's completely
contrary to how everyone else writes dependencies, and hence I
think these should be removed.
If you knew you were changing the behavior of the config option in
this way, you sure didn't think it was worth mentioning in your commit
message.
I stated in the comment what I think this is - awkward.
I definitely would never expect to have to enable a scsi option to get
some network driver visible to enable in the config, and therefore I
could see the opposite being insanely frustrating too.
The resulting dependency seems quite logical to me: Some higher
level networking functionality (iSCSI) depends on some lower level
networking functionality (an actual driver).
You can't ignore these issues and just say "that's not the normal way
so I'm going to change it anyways."
Admittedly I considered only my personal perspective.

Now, to get the whole discussion productive again - where do we
go from here? I don't think these drivers are so special that they
really need to behave backwards to how (almost?) everything else
is done... If changing it the way I did in the first try isn't deemed
acceptable, would it be at least acceptable to remove those
helper options (or, not as welcome from my perspective not the
least because of the odd dependency on INET instead of NET,
fold them into a single more generic one that others could also
benefit from)?

As to that INET vs NET dependency - is it possible that the
network drivers really just need NET, but the iSCSI ones need
INET? In which case the only common dependency would be
PCI - certainly not worth a custom helper option.

Jan
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help