Hi Felipe I see about __devinit. There's no crash because in those logs the __init modifiers are gone.
Regards -- Pantelis On Sep 6, 2012, at 8:07 PM, Felipe Balbi wrote: > Hi, > > On Thu, Sep 06, 2012 at 05:58:27PM +0300, Pantelis Antoniou wrote: >> Hi Felipe, >> >> A proper answer required some instrumentation printks(). >> >> So what I did is that I peppered each function marked with a removed >> __init or __devinit with a >> >>> printk(KERN_INFO "%s:%d (%s) %s\n", __FILE__, __LINE__, KBUILD_MODNAME, >>> __func__); >>> >> >> Both omap2430 & musb_hdrc are compiled as module, and I issued the following >> commands on a beagleboard xm. >> >> # modprobe omap2430 >> # modprobe g_mass_storage file=backing_file >> # rmmod g_mass_storage >> # rmmod omap2430 >> # modprobe omap2430 >> >> The problem is obvious when you see that functions marked as init or devinit >> are >> called when re-loading the omap2430 module again. Those functions if marked >> as >> much can (and will as I have witnessed) be removed from memory, leading to a >> crash. >> >> What do you think? > > the problem would be with usage of __init. __devinit will only drop > .devinit.text if HOTPLUG and MODULE isn't set, AFAICT. I also can't see > a crash on below logs. > > -- > balbi -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
