lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Wed, 10 Feb 2016 16:24:16 -0800
From:	Tim Chen <tim.c.chen@...ux.intel.com>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	Andrey Ryabinin <aryabinin@...tuozzo.com>, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org, Andi Kleen <ak@...ux.intel.com>,
	Mel Gorman <mgorman@...hsingularity.net>,
	Vladimir Davydov <vdavydov@...tuozzo.com>,
	Konstantin Khlebnikov <koct9i@...il.com>
Subject: Re: [RFC PATCH 3/3] mm: increase scalability of global memory
 commitment accounting

On Wed, 2016-02-10 at 13:28 -0800, Andrew Morton wrote:

> 
> If a process is unmapping 4MB then it's pretty crazy for us to be
> hitting the percpu_counter 32 separate times for that single operation.
> 
> Is there some way in which we can batch up the modifications within the
> caller and update the counter less frequently?  Perhaps even in a
> single hit?

I think the problem is the batch size is too small and we overflow
the local counter into the global counter for 4M allocations.
The reason for the small batch size was because we use
percpu_counter_read_positive in __vm_enough_memory and it is not precise
and the error could grow with large batch size.

Let's switch to the precise __percpu_counter_compare that is 
unaffected by batch size.  It will do precise comparison and only add up
the local per cpu counters when the global count is not precise
enough.  

So maybe something like the following patch with a relaxed batch size.
I have not tested this patch much other than compiling and booting
the kernel.  I wonder if this works for Andrey. We could relax the batch
size further, but that will mean that we will incur the overhead
of summing the per cpu counters earlier when the global count get close
to the allowed limit.

Thanks.

Tim

Signed-off-by: Tim Chen <tim.c.chen@...ux.intel.com>
---
diff --git a/mm/mm_init.c b/mm/mm_init.c
index fdadf91..996c332 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -151,8 +151,8 @@ static void __meminit mm_compute_batch(void)
 	s32 nr = num_present_cpus();
 	s32 batch = max_t(s32, nr*2, 32);
 
-	/* batch size set to 0.4% of (total memory/#cpus), or max int32 */
-	memsized_batch = min_t(u64, (totalram_pages/nr)/256, 0x7fffffff);
+	/* batch size set to 3% of (total memory/#cpus), or max int32 */
+	memsized_batch = min_t(u64, (totalram_pages/nr)/32, 0x7fffffff);
 
 	vm_committed_as_batch = max_t(s32, memsized_batch, batch);
 }
diff --git a/mm/mmap.c b/mm/mmap.c
index 79bcc9f..eec9dfd 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -131,7 +131,7 @@ struct percpu_counter vm_committed_as ____cacheline_aligned_in_smp;
  */
 unsigned long vm_memory_committed(void)
 {
-	return percpu_counter_read_positive(&vm_committed_as);
+	return percpu_counter_sum(&vm_committed_as);
 }
 EXPORT_SYMBOL_GPL(vm_memory_committed);
 
@@ -155,10 +155,6 @@ int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
 {
 	long free, allowed, reserve;
 
-	VM_WARN_ONCE(percpu_counter_read(&vm_committed_as) <
-			-(s64)vm_committed_as_batch * num_online_cpus(),
-			"memory commitment underflow");
-
 	vm_acct_memory(pages);
 
 	/*
@@ -224,7 +220,7 @@ int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
 		allowed -= min_t(long, mm->total_vm / 32, reserve);
 	}
 
-	if (percpu_counter_read_positive(&vm_committed_as) < allowed)
+	if (__percpu_counter_compare(&vm_committed_as, allowed, vm_committed_as_batch) < 0)
 		return 0;
 error:
 	vm_unacct_memory(pages);
diff --git a/mm/nommu.c b/mm/nommu.c
index ab14a20..53e4cae 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -70,7 +70,7 @@ atomic_long_t mmap_pages_allocated;
  */
 unsigned long vm_memory_committed(void)
 {
-	return percpu_counter_read_positive(&vm_committed_as);
+	return percpu_counter_sum(&vm_committed_as);
 }
 
 EXPORT_SYMBOL_GPL(vm_memory_committed);
@@ -1914,7 +1914,7 @@ int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
 		allowed -= min_t(long, mm->total_vm / 32, reserve);
 	}
 
-	if (percpu_counter_read_positive(&vm_committed_as) < allowed)
+	if (__percpu_counter_compare(&vm_committed_as, allowed, vm_committed_as_batch) < 0)
 		return 0;
 
 error:


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ