From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jiang Liu Date: Wed, 19 Aug 2015 07:18:59 +0000 Subject: Re: [Patch V3 2/9] kernel/profile.c: Replace cpu_to_mem() with cpu_to_node() Message-Id: <55D42DE3.2040506@linux.intel.com> List-Id: References: <1439781546-7217-1-git-send-email-jiang.liu@linux.intel.com> <1439781546-7217-3-git-send-email-jiang.liu@linux.intel.com> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: David Rientjes Cc: Andrew Morton , Mel Gorman , Mike Galbraith , Peter Zijlstra , "Rafael J . Wysocki" , Tang Chen , Tejun Heo , Tony Luck , linux-mm@kvack.org, linux-hotplug@vger.kernel.org, linux-kernel@vger.kernel.org, x86@kernel.org On 2015/8/18 8:31, David Rientjes wrote: > On Mon, 17 Aug 2015, Jiang Liu wrote: > >> Function profile_cpu_callback() allocates memory without specifying >> __GFP_THISNODE flag, so replace cpu_to_mem() with cpu_to_node() >> because cpu_to_mem() may cause suboptimal memory allocation if >> there's no free memory on the node returned by cpu_to_mem(). >> > > Why is cpu_to_node() better with regard to free memory and NUMA locality? Hi David, Thanks for review. This is a special case pointed out by Tejun. For the imagined topology, A<->B<->X<->C<->D, where A, B, C, D has memory and X is memoryless. Possible fallback lists are: B: [ B, A, C, D] X: [ B, C, A, D] C: [ C, D, B, A] cpu_to_mem(X) will either return B or C. Let's assume it returns B. Then we will use "B: [ B, A, C, D]" to allocate memory for X, which is not the optimal fallback list for X. And cpu_to_node(X) returns X, and "X: [ B, C, A, D]" is the optimal fallback list for X. Thanks! Gerry > >> It's safe to use cpu_to_mem() because build_all_zonelists() also >> builds suitable fallback zonelist for memoryless node. >> > > Why reference that cpu_to_mem() is safe if you're changing away from it? Sorry, it should be cpu_to_node() instead of cpu_to_mem(). > >> Signed-off-by: Jiang Liu >> --- >> kernel/profile.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/kernel/profile.c b/kernel/profile.c >> index a7bcd28d6e9f..d14805bdcc4c 100644 >> --- a/kernel/profile.c >> +++ b/kernel/profile.c >> @@ -336,7 +336,7 @@ static int profile_cpu_callback(struct notifier_block *info, >> switch (action) { >> case CPU_UP_PREPARE: >> case CPU_UP_PREPARE_FROZEN: >> - node = cpu_to_mem(cpu); >> + node = cpu_to_node(cpu); >> per_cpu(cpu_profile_flip, cpu) = 0; >> if (!per_cpu(cpu_profile_hits, cpu)[1]) { >> page = alloc_pages_exact_node(node,