On Wed, 25 Nov 2015, Markus Rechberger wrote: > Seems like it's getting stuck again, since we're entering devices with > a higher bandwidth need now it seems to be necessary to really fix it > in the kernel. > > I propose add it as it is, it's proven that it works and there are 2 > testcases out there right now and it decreases the CPU load on intel > Atoms as well. > Otherwise it might take ages until things are getting forward here. > Apple is just running fine for ages and they're not struggling on > mailinglists about it. Aside of that the Mac USB 2.0 through USB 3.0 > support also works rock solid while linux is just a shame in that area > .. it never worked and even totally crashes some systems. > While that linux issue is known for a long time it just seems like > it's accepted already. > Either come up with a solution or just add this patch so we can make use of it! I want to see a modified version of your patch. Several things need to be changed or fixed, but the major change needs to be the way memory is allocated. It should be done as part of the mmap system call, not as a separate ioctl. There are numerous smaller issues: The allocated memory needs to be charged against usbfs_memory_usage, the memory should be allocated using dma_zalloc_coherent instead of kmalloc, proc_do_submiturb should check that the buffer starts anywhere within the allocated memory rather than just at the beginning, and so on. I'd make these changes myself, but I just don't have enough free time now. If anybody else would like to work on them, please feel free. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
