[PATCH 1/2] mm: do not limit VmData with RLIMIT_DATA

[PATCH 1/2] mm: do not limit VmData with RLIMIT_DATA

[Konstantin Khlebnikov]

This partially reverts 84638335900f ("mm: rework virtual memory accounting")

Before that commit RLIMIT_DATA have control only over size of the brk region.
But that change have caused problems with all existing versions of valgrind
because they set RLIMIT_DATA to zero for some reason.

More over, current check has a major flaw: RLIMIT_DATA in bytes,
not pages. So, some problems might have slipped through testing.
Let's revert it for now and put back in next release.

Signed-off-by: Konstantin Khlebnikov <koct9i@gmail.com>
Link: http://lkml.kernel.org/r/20151228211015.GL2194@uranus
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
---
 mm/mmap.c |    4 ----
 1 file changed, 4 deletions(-)

diff --git a/mm/mmap.c b/mm/mmap.c
index 84b12624ceb0..e0cd98c510ba 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -2982,10 +2982,6 @@ bool may_expand_vm(struct mm_struct *mm, vm_flags_t flags, unsigned long npages)
 	if (mm->total_vm + npages > rlimit(RLIMIT_AS) >> PAGE_SHIFT)
 		return false;
 
-	if ((flags & (VM_WRITE | VM_SHARED | (VM_STACK_FLAGS &
-				(VM_GROWSUP | VM_GROWSDOWN)))) == VM_WRITE)
-		return mm->data_vm + npages <= rlimit(RLIMIT_DATA);
-
 	return true;
 }
 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH 2/2] mm: limit VmData with RLIMIT_DATA

[Konstantin Khlebnikov]

This adds is correct version of RLIMIT_DATA check.
And kernel boot option "ignore_rlimit_data" for reverting old behavior.
Also could be set by /sys/module/kernel/parameters/ignore_rlimit_data.

Signed-off-by: Konstantin Khlebnikov <koct9i@gmail.com>
---
 Documentation/kernel-parameters.txt |    5 +++++
 mm/mmap.c                           |    8 ++++++++
 2 files changed, 13 insertions(+)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index cfb2c0f1a4a8..850239102e86 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1461,6 +1461,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			could change it dynamically, usually by
 			/sys/module/printk/parameters/ignore_loglevel.
 
+	ignore_rlimit_data
+			Ignore setrlimit(RLIMIT_DATA) setting for private
+			mappings (as it was before). Could be changed by
+			/sys/module/kernel/parameters/ignore_rlimit_data.
+
 	ihash_entries=	[KNL]
 			Set number of hash buckets for inode cache.
 
diff --git a/mm/mmap.c b/mm/mmap.c
index e0cd98c510ba..af272025b1b9 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -42,6 +42,7 @@
 #include <linux/memory.h>
 #include <linux/printk.h>
 #include <linux/userfaultfd_k.h>
+#include <linux/moduleparam.h>
 
 #include <asm/uaccess.h>
 #include <asm/cacheflush.h>
@@ -69,6 +70,8 @@ const int mmap_rnd_compat_bits_max = CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MAX;
 int mmap_rnd_compat_bits __read_mostly = CONFIG_ARCH_MMAP_RND_COMPAT_BITS;
 #endif
 
+static bool ignore_rlimit_data = false;
+core_param(ignore_rlimit_data, ignore_rlimit_data, bool, 0644);
 
 static void unmap_region(struct mm_struct *mm,
 		struct vm_area_struct *vma, struct vm_area_struct *prev,
@@ -2982,6 +2985,11 @@ bool may_expand_vm(struct mm_struct *mm, vm_flags_t flags, unsigned long npages)
 	if (mm->total_vm + npages > rlimit(RLIMIT_AS) >> PAGE_SHIFT)
 		return false;
 
+	if (!ignore_rlimit_data && (flags & (VM_WRITE | VM_SHARED |
+		(VM_STACK_FLAGS & (VM_GROWSUP | VM_GROWSDOWN)))) == VM_WRITE &&
+	    mm->data_vm + npages > rlimit(RLIMIT_DATA) >> PAGE_SHIFT)
+		return false;
+
 	return true;
 }
 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 1/2] mm: do not limit VmData with RLIMIT_DATA

[Cyrill Gorcunov]

On Sat, Jan 23, 2016 at 10:39:40AM +0300, Konstantin Khlebnikov wrote:
> This partially reverts 84638335900f ("mm: rework virtual memory accounting")
> 
> Before that commit RLIMIT_DATA have control only over size of the brk region.
> But that change have caused problems with all existing versions of valgrind
> because they set RLIMIT_DATA to zero for some reason.
> 
> More over, current check has a major flaw: RLIMIT_DATA in bytes,
> not pages. So, some problems might have slipped through testing.
> Let's revert it for now and put back in next release.
> 
> Signed-off-by: Konstantin Khlebnikov <koct9i@gmail.com>
> Link: http://lkml.kernel.org/r/20151228211015.GL2194@uranus
> Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>

Looks great for me. Thanks a lot, Kostya!
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 2/2] mm: limit VmData with RLIMIT_DATA

[Cyrill Gorcunov]

On Sat, Jan 23, 2016 at 10:39:47AM +0300, Konstantin Khlebnikov wrote:
> This adds is correct version of RLIMIT_DATA check.
> And kernel boot option "ignore_rlimit_data" for reverting old behavior.
> Also could be set by /sys/module/kernel/parameters/ignore_rlimit_data.
> 
> Signed-off-by: Konstantin Khlebnikov <koct9i@gmail.com>
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>