All of lore.kernel.org
 help / color / mirror / Atom feed
From: Michal Simek <michal.simek@petalogix.com>
To: Arnd Bergmann <arnd@arndb.de>
Cc: LTP <ltp-list@lists.sourceforge.net>,
	Linux Kernel list <linux-kernel@vger.kernel.org>,
	Grant Likely <grant.likely@secretlab.ca>,
	Paul Mundt <lethal@linux-sh.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Ingo Molnar <mingo@elte.hu>,
	John Williams <john.williams@petalogix.com>
Subject: Re: [LTP] mmap syscall problem
Date: Mon, 06 Jul 2009 16:43:50 +0200	[thread overview]
Message-ID: <4A520DA6.2040107@petalogix.com> (raw)
In-Reply-To: <200907061616.12220.arnd@arndb.de>

Arnd Bergmann wrote:
> On Monday 06 July 2009, Michal Simek wrote:
>   
>>> Not necessarily, even on platforms that manage aliases in hardware
>>> mappings that violate the aliasing constraints can still result in
>>> undefined behaviour, this really depends more on your cache controller
>>> and MMU than anything else. I notice that microblaze sets SHMLBA to
>>> PAGE_SIZE, you may want to see if this test still breaks after bumping it
>>> up to something like PAGE_SIZE * 4.
>>>   
>>>       
>> Yes, test still break - behavior is the same. I don't have accurate
>> information about MMU unit
>> but I will ask a question about. We are able to turn off cache
>> controller directly in HW.
>>     
>
> There may still be a problem with data being queued in some write
> buffers that don't get flushed before reading back from another
> address.
>
> What happens in a simple user space program that mmaps the same
> page to two addresses? Something like
>
> #include <sys/mman.h>
> #include <fcntl.h>
> int main(void)
> {
> 	int fd = open("existing-4k-file", O_RDWR);
> 	char *p1 = mmap(0, 4096, PROT_READ | PROT_WRITE,
> 				MAP_SHARED, fd, 0);
> 	char *p2 = mmap(p1 + 4096, 4096, PROT_READ | PROT_WRITE,
> 				MAP_SHARED | MAP_FIXED, fd, 0);
>
> 	*p1 = 0xaa; *p2 = 0x55;
>   
I closed fd too.
    close(fd);
> 	return *p1; /* returns 0xaa if broken, 0x55 if correct */
> }
>   
# ls -la existing-4k-file
-rw-rw-r--    1 monstr   monstr       4096 Jul  6  2009 existing-4k-file

# ./test-arnd
# echo $?
85
# dd if=existing-4k-file of=/dev/console count=1 2>/dev/null
U#

in file is first char U (0x55) which is IMO correct.

Michal



> 	Arnd <><
>   


-- 
Michal Simek, Ing. (M.Eng)
PetaLogix - Linux Solutions for a Reconfigurable World
w: www.petalogix.com p: +61-7-30090663,+42-0-721842854 f: +61-7-30090663


------------------------------------------------------------------------------
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

WARNING: multiple messages have this Message-ID (diff)
From: Michal Simek <michal.simek@petalogix.com>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Paul Mundt <lethal@linux-sh.org>,
	Linux Kernel list <linux-kernel@vger.kernel.org>,
	LTP <ltp-list@lists.sourceforge.net>,
	John Williams <john.williams@petalogix.com>,
	Ingo Molnar <mingo@elte.hu>,
	Andrew Morton <akpm@linux-foundation.org>,
	Grant Likely <grant.likely@secretlab.ca>,
	subrata@linux.vnet.ibm.com
Subject: Re: mmap syscall problem
Date: Mon, 06 Jul 2009 16:43:50 +0200	[thread overview]
Message-ID: <4A520DA6.2040107@petalogix.com> (raw)
In-Reply-To: <200907061616.12220.arnd@arndb.de>

Arnd Bergmann wrote:
> On Monday 06 July 2009, Michal Simek wrote:
>   
>>> Not necessarily, even on platforms that manage aliases in hardware
>>> mappings that violate the aliasing constraints can still result in
>>> undefined behaviour, this really depends more on your cache controller
>>> and MMU than anything else. I notice that microblaze sets SHMLBA to
>>> PAGE_SIZE, you may want to see if this test still breaks after bumping it
>>> up to something like PAGE_SIZE * 4.
>>>   
>>>       
>> Yes, test still break - behavior is the same. I don't have accurate
>> information about MMU unit
>> but I will ask a question about. We are able to turn off cache
>> controller directly in HW.
>>     
>
> There may still be a problem with data being queued in some write
> buffers that don't get flushed before reading back from another
> address.
>
> What happens in a simple user space program that mmaps the same
> page to two addresses? Something like
>
> #include <sys/mman.h>
> #include <fcntl.h>
> int main(void)
> {
> 	int fd = open("existing-4k-file", O_RDWR);
> 	char *p1 = mmap(0, 4096, PROT_READ | PROT_WRITE,
> 				MAP_SHARED, fd, 0);
> 	char *p2 = mmap(p1 + 4096, 4096, PROT_READ | PROT_WRITE,
> 				MAP_SHARED | MAP_FIXED, fd, 0);
>
> 	*p1 = 0xaa; *p2 = 0x55;
>   
I closed fd too.
    close(fd);
> 	return *p1; /* returns 0xaa if broken, 0x55 if correct */
> }
>   
# ls -la existing-4k-file
-rw-rw-r--    1 monstr   monstr       4096 Jul  6  2009 existing-4k-file

# ./test-arnd
# echo $?
85
# dd if=existing-4k-file of=/dev/console count=1 2>/dev/null
U#

in file is first char U (0x55) which is IMO correct.

Michal



> 	Arnd <><
>   


-- 
Michal Simek, Ing. (M.Eng)
PetaLogix - Linux Solutions for a Reconfigurable World
w: www.petalogix.com p: +61-7-30090663,+42-0-721842854 f: +61-7-30090663


  reply	other threads:[~2009-07-06 14:43 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-03 12:37 [LTP] mmap syscall problem Michal Simek
2009-07-03 12:37 ` Michal Simek
2009-07-03 15:02 ` [LTP] " Arnd Bergmann
2009-07-03 15:02   ` Arnd Bergmann
2009-07-06  6:32   ` [LTP] " Michal Simek
2009-07-06  6:32     ` Michal Simek
2009-07-06  8:05     ` [LTP] " Arnd Bergmann
2009-07-06  8:05       ` Arnd Bergmann
2009-07-06 12:07       ` [LTP] " Michal Simek
2009-07-06 12:07         ` Michal Simek
2009-07-06 12:14         ` [LTP] " Paul Mundt
2009-07-06 12:14           ` Paul Mundt
2009-07-06 12:54           ` [LTP] " Michal Simek
2009-07-06 12:54             ` Michal Simek
2009-07-06 14:16             ` [LTP] " Arnd Bergmann
2009-07-06 14:16               ` Arnd Bergmann
2009-07-06 14:43               ` Michal Simek [this message]
2009-07-06 14:43                 ` Michal Simek
2009-07-06 15:05                 ` [LTP] " Arnd Bergmann
2009-07-06 15:05                   ` Arnd Bergmann
2009-07-06 15:19                   ` [LTP] " Michal Simek
2009-07-06 15:19                     ` Michal Simek
2009-07-07  0:44               ` [LTP] " Valdis.Kletnieks
2009-07-07  0:44                 ` Valdis.Kletnieks

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4A520DA6.2040107@petalogix.com \
    --to=michal.simek@petalogix.com \
    --cc=akpm@linux-foundation.org \
    --cc=arnd@arndb.de \
    --cc=grant.likely@secretlab.ca \
    --cc=john.williams@petalogix.com \
    --cc=lethal@linux-sh.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ltp-list@lists.sourceforge.net \
    --cc=mingo@elte.hu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.