On Tue, 19 Jun 2007, Thomas Renninger wrote: > On Sun, 2007-06-17 at 19:49 -0300, Henrique de Moraes Holschuh wrote: > > On Sun, 17 Jun 2007, Thomas Renninger wrote: > > > Create __mod_acpi_device_table symbol for all acpi drivers. > > > > thinkpad-acpi will autoload by DMI ids in 2.6.23, which is a much better > > solution for model-specific drivers like thinkpad-acpi than autoloading by > > HID, IMHO. > Yes, I saw some dmi udev strings on Kay's machine and for the thinkpad > module dmi autoloading is much better. Yeah. The HIDs thinkpad-acpi binds to might exist on non-thinkpads, while the DMI matching should be good for all new boxes, and all of the old ones which I have some sort of data on. The few very old thinkpads I am not autoloading on are probably not going to care, wouldn't benefit much from thinkpad-acpi anyway... or they will send me email about it. > Not in general, only if the driver provides additional functions and the > ACPI device does not appear on machines that need those additional > functions. Or, if you risk too many false positives. > > Thus, I'd prefer if your patches did not make thinkpad-acpi autoload by HID. > It would not really hurt? At least thinkpad module is an acpi driver for It would, if we start trying to autoload on false positive matches. If we did not have any other way to autoload, that would be fine, but since we will have DMI matches also in 2.6.23... > those two HIDs... But your are right, it's superfluous, here with > __mod_acpi_device_table symbol ripped out of thinkpad.ko. Thank you. I did not test the patch, but I am ok with it, now. > Hmm, what do you think about splitting the Thinkpad driver into Hotkey, > PCI and rest? This one got a complicated beast... It is cooking. In fact, I was halfway through it, when Richard started getting me fired up about improving the input layer support in thinkpad-acpi, which is a damn more fun thing to do... so it stalled, but now thinkpad-acpi has proper input support, which I will be sending for acpi-test merging quite soon now. Some sort of split will be done. Probably for 2.6.23, even. -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh - To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
