> On Wed, Nov 17, 2010 at 12:26:42AM -0500, Kyle McMartin wrote: > > On Tue, Nov 16, 2010 at 10:22:18PM -0700, Grant Grundler wrote: > > > I agree. The kernel is expected to emulate/handle unaligned loads/stores. > > > > > > I thought we had all that misaligned access handlers working years ago and > > > somewhere probably have a user space program to test it. Oh...ldd won't be > > > user space since we don't have a PA2.0 64-bit user space. LDD support > > > can't be as well tested. > > > > > > On that note, can you try a 32-bit kernel? (SMP maybe) > > > > > > > We really shouldn't be fixing up unaligned access in the kernel, since > > we can just fix the code... If ipv6 is causing them, there must be some > > missing define that's getting defaulted to something un-padded, > > otherwise sparc wouldn't work at all (it definitely does /not/ fix > > unalignd accesses, period.) > > I agree but thought davem refused patches to use get/put_unaligned macros > in networking code. Searching for "davem ulog put_unaligned" shows him > accepting such patches. So this is clearly the right path to pursue. I also agree that the kernel shouldn't fix unaligned accesses occuring in the kernel. It allows for sloppy coding and the fixup can be a major performance hit on strict alignment machines like sparc and hppa. It would appear handle_unaligned can't always fixup kernel accesses. At least, emulate_ldd causes a hpmc attempting to fixup the ldd in init_page_buffers. This may be because this is a wrong code error and the address is in someway invalid. Maybe more address checking is needed. The unaligned access in __tcp_v6_send_check seems to be caused by saddr being misaligned. It's possible daddr is also misaligned. I was going to add a BUG_ON or two to try and get more info. Don't know if emulate_ldd can fix this access. In the case of ipv4, it's possible to pad/start packets in a manner to make it easier to access the ip header. Don't know about ipv6. This may be what Kyle is suggesting. Dave -- J. David Anglin dave.anglin@xxxxxxxxxxxxxx National Research Council of Canada (613) 990-0752 (FAX: 952-6602) -- To unsubscribe from this list: send the line "unsubscribe linux-parisc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
