Mike (mwester) wrote: > This is "The Question That Must Be Asked": > > What's involved in backporting the work you've done to fix the wifi to > the "stable" branch (2.6.24)?
In fact, this question has been asked already ... http://lists.openmoko.org/pipermail/openmoko-kernel/2008-November/006269.html ... and answered ;-) http://lists.openmoko.org/pipermail/openmoko-kernel/2008-November/006270.html (And no rotten tomatoes have been thrown ;-) When working on "stable", I compared the old and the new S3C MCI driver, but couldn't find any obvious bug in the old one. In fact, the old one looked as if it had improvements for SDIO the new one was lacking at that time. Nevertheless, it didn't work. Since there was a pile of other bug fixes for the new driver on linux-arm, I didn't poke around the old driver much, and just moved to stable-tracking. For all I know, there might also be relevant differences in the SDIO stack, but I think the main problems is the driver. The Atheros code and my HIF should be the same. You'd also have to transplant arch/arm/plat-s3c24xx/gta02_pm_wlan.c back into 2.6.24 for unload/resume power control. Considering that we seem to be making good progress towards switching to stable-tracking anyway, I think it may be easier to just wait for that to happen. BTW, Andy, what's your prognosis for the switch ? - Werner
