Hi Ted,
what's the problem with the table growing ?
I would rather trade some bytes for simplicity
(one data structure for registering PCI devices)
and readability of the source (don't have to grok
different methods to do the same thing).
Theodore Ts'o wrote:
> ...
> The real problem with the patch is that you're exploding the size of the
> PCI table. It's rather unfortunate that Timeida chose such an
> nasty method of assigning PCI device id's. Why are there 29 different
> PCI subdevice ID's for a dual port serial board, whose properties seem
> to be all identical.
In fact the boards come in different flavors: 16c550,650,750, RS-232/RS-422,
with and w/o surge protection, with and without parallel ports...
>
>
> So I'll be sending a patch to linus to replace all of those entries with
> this:
>
> /*
> * Timedia has an explosion of boards, and to avoid the PCI table from
> * growing *huge*, we use this function to collapse some 70 entries
> * in the PCI table into one, for sanity's and compactness's sake.
> */
> ...
I will re-test the patch when it appears :-)
>
> Umm... I didn't get your patch from before. Are you sure you sent it to
> the right e-mail address?
I sent this patch only to linux-serial and Linus. On June,17th I had sent a previous
version
to your right address ([EMAIL PROTECTED]) and linux-serial.
Regards, Gunther
��칻 �&�~�&���+-��ݶ��w��˛���m�b��lz���)���w* jg��� �����ݢj/���� �ǝ
