From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bob Picco Date: Thu, 25 Sep 2014 17:43:09 +0000 Subject: Re: [PATCH] sparc64: swapper_tsb and swapper_4m_tsb phys correction Message-Id: <20140925174309.GE4211@zareason> List-Id: References: <1410886239-15774-1-git-send-email-bpicco@meloft.net> In-Reply-To: <1410886239-15774-1-git-send-email-bpicco@meloft.net> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: sparclinux@vger.kernel.org David Miller wrote: [Wed Sep 24 2014, 06:26:51PM EDT] > From: Bob Picco > Date: Mon, 22 Sep 2014 14:56:31 -0400 > > > I'll review more. > > > > Oh and boot with DEBUG_PAGEALLOC at some point. > > Alright, need to chase a DEBUG_PAGEALLOC issue with prom_halt() blink: > > [bpicco@zareason ~]$ cat /tmp/t5-blink.txt > > [23233.728105] IPv6: ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready > > [23233.753380] SUN4V-DTLB: Error at TPC[446930], tl 1 > > [23233.762729] SUN4V-DTLB: TPC > > [23233.773124] SUN4V-DTLB: O7[55206c] > > [23233.779910] SUN4V-DTLB: O7 > > [23233.789455] SUN4V-DTLB: vaddr[1103f23e000] ctx[0] pte[0] error[2] > > . We were nearly at the console login prompt. Hm, maybe turn off LDOMS > > config :) > > > > thanx and nice! > > Bob, here is what I have now, can you please help me test it? > > It should restore 256MB/2GB/16GB page size usage. > > The biggest bug was that I wasn't flushing the kernel TSB(s) > after building the kernel page tables and NOP'ing out that > instruction in kernel_physical_mapping_init(). > > Once I fixed that everything runs smoothly so far :-) > > I haven't gotten to testing DEBUG_PAGEALLOC yet, but I suspect it > probably works now. > > Thanks! Boots up !DEBUG_PAGEALLOC on T5-2. Review is in progress and looks good so far. This is important to have correct as we both know. I will attempt DEBUG_PAGEALLOC, hopefully later today, locally. We should give it a spin on M7. Though it is scheduled for a core upgrade today and one never knows :)