Re: [Patch] Updated version of emutramp_enabled_check() for PaX kernels

Re: [Patch] Updated version of emutramp_enabled_check() for PaX kernels

[Magnus Granberg]

>Hi 
>Magnus, Reading /proc/self/status is a little more heavy weight than checking 
an environment variable. Do we really have to do this every time we call 
dlmmap(), or can we do it once and cache the result? 
>Thanks, 
>AG

The selinux check read /proc/mount for selinux stuff and even check if it have 
the selinux dir. I think the selinux check have the same heavy weight as the 
pax check and then the selinux need to make a temp file in some dir the app
have acces to make file in. For me is fine to read /proc/... on every time.

/Magnus

Re: [Patch] Updated version of emutramp_enabled_check() for PaX kernels

[Anthony Green]

Magnus Granberg <zorry@gentoo.org> writes:

>>Hi 
>>Magnus, Reading /proc/self/status is a little more heavy weight than checking 
> an environment variable. Do we really have to do this every time we call 
> dlmmap(), or can we do it once and cache the result? 
>>Thanks, 
>>AG
>
> The selinux check read /proc/mount for selinux stuff and even check if it have 
> the selinux dir. I think the selinux check have the same heavy weight as the 
> pax check and then the selinux need to make a temp file in some dir the app
> have acces to make file in. For me is fine to read /proc/... on every time.
>
> /Magnus


Thanks Magnus,  I am applying your patch.

AG

Re: [Patch] Updated version of emutramp_enabled_check() for PaX kernels

[Magnus Granberg]

onsdag 22 maj 2013 07.50.57 skrev  Anthony Green:
> Hi Magnus,
> 
>   Reading /proc/self/status is a little more heavy weight than
> checking an environment variable.   Do we really have to do this every
> time we call dlmmap(), or can we do it once
> and cache the result?
> 
> Thanks,
> 
> AG
Would be fine to cache the result.

/Magnus

Re: [Patch] Updated version of emutramp_enabled_check() for PaX kernels

[Anthony Green]

Hi Magnus,

  Reading /proc/self/status is a little more heavy weight than
checking an environment variable.   Do we really have to do this every
time we call dlmmap(), or can we do it once
and cache the result?

Thanks,

AG


On Mon, Apr 29, 2013 at 5:44 PM, Magnus Granberg <zorry@gentoo.org> wrote:
> Hi
> This patch check the PaX status in /proc/self/status
> and it will return true or fales depend on the status of
> emulate trampolines in Pax.
>
> /Magnus
>
> 2013-04-29  Magnus Granberg <zorry@gentoo.org>
>
>         * src/cloures.c (emutramp_enabled_check): Check /proc/self/status for PaX
>         status.
>

Re: [Patch] Updated version of emutramp_enabled_check() for PaX kernels

[Magnus Granberg]

måndag 29 april 2013 23.44.27 skrev  Magnus Granberg:
> Hi
> This patch check the PaX status in /proc/self/status
> and it will return true or fales depend on the status of
> emulate trampolines in Pax.
> 
> /Magnus
> 
> 2013-04-29  Magnus Granberg <zorry@gentoo.org>
> 
> 	* src/cloures.c (emutramp_enabled_check): Check /proc/self/status for PaX
> 	status.
Can this patch be included with next release?
/Magnus

[Patch] Updated version of emutramp_enabled_check() for PaX kernels

[Magnus Granberg]

[-- Attachment #1: Type: text/plain, Size: 286 bytes --]

Hi
This patch check the PaX status in /proc/self/status
and it will return true or fales depend on the status of
emulate trampolines in Pax.

/Magnus

2013-04-29  Magnus Granberg <zorry@gentoo.org>

	* src/cloures.c (emutramp_enabled_check): Check /proc/self/status for PaX 	
	status.


[-- Attachment #2: libffi-3.0.13-emutramp_pax_log.patch --]
[-- Type: text/x-patch, Size: 791 bytes --]

--- a/src/closures.c	2013-03-17 23:27:11.000000000 +0100
+++ b/src/closures.c	2013-04-29 23:26:02.279022022 +0200
@@ -181,10 +181,26 @@ static int emutramp_enabled = -1;
 static int
 emutramp_enabled_check (void)
 {
-  if (getenv ("FFI_DISABLE_EMUTRAMP") == NULL)
-    return 1;
-  else
+  char *buf = NULL;
+  size_t len = 0;
+  FILE *f;
+  int ret;
+  f = fopen ("/proc/self/status", "r");
+  if (f == NULL)
     return 0;
+  ret = 0;
+
+  while (getline (&buf, &len, f) != -1)
+    if (!strncmp (buf, "PaX:", 4))
+      {
+        char emutramp;
+        if (sscanf (buf, "%*s %*c%c", &emutramp) == 1)
+          ret = (emutramp == 'E');
+        break;
+      }
+  free (buf);
+  fclose (f);
+  return ret;
 }
 
 #define is_emutramp_enabled() (emutramp_enabled >= 0 ? emutramp_enabled \