Messages in this thread | | | Date | Thu, 25 Nov 2010 13:04:00 -0500 | From | Kyle McMartin <> | Subject | Re: [RFT][PATCH] Poke ACPI about connected floppy drives |
| |
On Wed, Nov 24, 2010 at 10:57:22PM -0500, Len Brown wrote: > I have 45 DSDTs on-hand that include PNP0700, > but only 18 of those implement _FDE... >
Awesome, thanks for testing. Poking people to test the module on their machines I saw pretty much the same experiences. A lot of tables that do have _FDE seem to wire it to One if there's a floppy controller, regardless of whether a drive is wired.
Pretty much makes it useless for what I was hoping to use it for (avoiding probing floppies on machines without them attached, since it sometimes takes quite a while.)
> I have a Lenovo T61 which has PNP0700 and implements _FDE. > However, it doesn't physically have a floppy, and so > acpi_fde_add() will never get called -- I guess > because _CRS said it wasn't present? This seems > to make evaluating _FDE from the .add routine > somewhat redundant, no? >
I'm not sure what _CRS is, so I don't know. :/ I'll look it up.
> Note that you've got a memory leak in acpi_fde_add() > in the "if (!c)" case. Also, "rmmod floppy" takes an oops. >
Oops, sorry, it was just for testing so I didn't bother checking those.
Not sure the patch is worth the effort given how few DSDTs implement _FDE though.
regards, Kyle
> cheers, > Len Brown, Intel Open Source Technology Center >
| |