From mboxrd@z Thu Jan  1 00:00:00 1970
From: Alexander Graf <agraf@suse.de>
Subject: HV KVM fails on 970 due to HTAB allocation
Date: Thu, 12 Jan 2012 19:16:51 +0100
Message-ID: <4F0F2393.6010102@suse.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "paulus@samba.org" <paulus@samba.org>, DValeev@suse.com,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>
To: kvm-ppc <kvm-ppc@vger.kernel.org>
Return-path: <kvm-owner@vger.kernel.org>
Received: from cantor2.suse.de ([195.135.220.15]:44687 "EHLO mx2.suse.de"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751988Ab2ALSQw (ORCPT <rfc822;kvm@vger.kernel.org>);
	Thu, 12 Jan 2012 13:16:52 -0500
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

Hi Paul,

While trying to run HV KVM for something useful on 970, we stumbled over 
the following code path:

         /* Allocate guest's hashed page table */
         hpt = 
__get_free_pages(GFP_KERNEL|__GFP_ZERO|__GFP_REPEAT|__GFP_NOWARN,
                                HPT_ORDER - PAGE_SHIFT);
         if (!hpt) {
                 pr_err("kvm_alloc_hpt: Couldn't alloc HPT\n");
                 return -ENOMEM;
         }
         kvm->arch.hpt_virt = hpt;

We're most of the time running into the !hpt case, because we simply 
don't have 16MB of contiguous memory lying around.

I was trying to check if we could maybe allocate a huge_tlb page from 
within kernel space, since that usually matches the 16MB pretty well. 
However that seems to be very tricky. Maybe something similar to the RMA 
thing would be a good idea?


Alex