Thread (6 messages) 6 messages, 2 authors, 2015-02-16

RE: chelsio: Use a more common const struct pci_device_id foo[] style

From: Casey Leedom <hidden>
Date: 2015-02-16 19:30:54
Also in: linux-scsi, lkml

  Okay, thanks for your patience with my lack of understanding.  I'll work with Hariprasad tomorrow to get a revised patch out with the "const" change and eliminate the redundant check of CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN to guard the file contents.  Thanks for caching these improvements!

  And given that it's pretty empty at work here on this holiday and how beautiful it is outside, I think it's time to head home to enjoy lunch outside!

Casey

________________________________________
From: Joe Perches [joe@perches.com]
Sent: Monday, February 16, 2015 11:18 AM
To: Casey Leedom
Cc: Hariprasad S; James E.J. Bottomley; netdev@vger.kernel.org; linux-kernel@vger.kernel.org; linux-scsi; David Miller
Subject: Re: chelsio: Use a more common const struct pci_device_id foo[] style

On Mon, 2015-02-16 at 19:07 +0000, Casey Leedom wrote:
  I understand that OS-independence issues aren't something which are
normally accommodated, but as long as definitions don't introduce
unnecessary "foreign intrusion" I would hope that it would be okay.
As I noted, our t4_regs.h file is also OS-independent and used by six
other OS device drivers.  Putting Linux definitions into
t4_pci_id_tbl.h would be somewhat akin to injecting Linux dependencies
into t4_regs.h.

  But, if the change must be made, then we'll just maintain a
translation between our Common Code and the kernel.org code.  If
that's the case, probably the best documentation for the proposed
CH_PCI_ID_TABLE_ENTRY_DATA might be something like:

 * CH_PCI_ID_TABLE_ENTRY_DATA
 *   -- Used for the individual PCI Device ID entries for the
PCI_VDEVICE() "dev"
 *   -- parameter.

  So it sounds like Chelsio would be required to make this change
then?  I'm still unclear on the likes of responsibility/authority
here.  We're being told that we must do this but we have to be the
ones requesting it?  Sorry for my confusion.
It's just a suggested patch.
It's your code, you don't _have_ to do anything.
(Which is doubly apparent since I came into work this morning only to
realize that it's a company holiday.  Color me a moron today.)
Yay for holidays.

It can be personally beneficial to take the day off.
It could also be very productive to work with no distractions.

As always, your choice...

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