27.09.2013, 23:21, "Kirill Tkhai" <tkhai@xxxxxxxxx>: > 27.09.2013, 00:52, "David Miller" <davem@xxxxxxxxxx>: > >> Ok, this patch set builds on top of the variable PAGE_OFFSET >> patches. >> >> First we move to 8MB huge pages. The chips don't really have >> 8MB page support, but we fabricate such a thing using 4MB >> PTEs. >> >> Next we move to 64-bit pmd_t and pgd_t entries, which is the >> most trivial of the changes here. >> >> Finally, we encode huge PMDs using the plain PTE encoding of >> huge pages. All of the ugly huge PMD --> PTE translation >> code disappears, and the TLB miss path is smaller and faster. >> >> I just got this stuff fully working so caveat emptor :-) >> >> Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx> > > Hi, David, > > I've tryed the series on UltraSparc II with Spitfire MMU. Boot is ok, > and some LTP tests are passed (short tests of mm). This may confuse. Not "some are passed, some are not", but "I executed a subset and it is ok" ;) -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
