RE: Is the future MmProbeAndLockPages() guarranted to suc ceed if the memory has been locked before?

So for the following situation:

The number of bytes is just for example, we may need go as big as 256MB for
pVa1 and pVa2.


| pVa1(locked) | pVa2(locked) |

pVa3

pVa = malloc(1024 * 10);

pVa1 = pVa;
pVa2 = pVa + 1024 * 5;
pVa3 = pVa + 1024;

You see, pVa1 and pVa2 are virtually continuous memory address. But I locked
pVa1 with size 5K bytes and pVa2 with size 5K bytes. I assume all these
memory can be locked successfully.

pVa3 is just another user virtual address in the range of pVa1, however, the
memory pointed by pVa3 goes beyound pVa1.

Therefore, my question is, can I lock pVa3 with guarrantted success? What is
the overhead for multiple MmProbeAndLockPages()?

Thanks,

zhong

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Peter Wieland
Sent: Friday, October 01, 2004 5:08 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Is the future MmProbeAndLockPages() guarranted to
succeed if the memory has been locked before?

yep

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
xxxxx@exgate.tek.com
Sent: Friday, October 01, 2004 3:45 PM
To: Windows System Software Devs Interest List
Subject: [ntdev] Is the future MmProbeAndLockPages()
guarranted to succeed if the memory has been locked before?

In my program, we want to MmProbeAndLockPages() a certain
amount of memory in powerup, but in the runtime, we may only
need a part of pre-locked memory to do DMA operation. I am
wondering whether the future MmProbeAndLockPages() can be
ensured to work if the memory has already been locked. I
think it does, but I just want to make sure.

Thanks,

zhongsheng


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as:
xxxxx@windows.microsoft.com To unsubscribe send a blank
email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: unknown lmsubst tag argument: ''
To unsubscribe send a blank email to xxxxx@lists.osr.com

Yes, pVa3 should succeed. The cost shouldn't be too bad since there
won't be any actual paging I/O.

-p

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
xxxxx@exgate.tek.com
Sent: Friday, October 01, 2004 5:36 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Is the future MmProbeAndLockPages() guarranted to
suc ceed if the memory has been locked before?

So for the following situation:

The number of bytes is just for example, we may need go as big as 256MB
for
pVa1 and pVa2.


| pVa1(locked) | pVa2(locked) |

pVa3

pVa = malloc(1024 * 10);

pVa1 = pVa;
pVa2 = pVa + 1024 * 5;
pVa3 = pVa + 1024;

You see, pVa1 and pVa2 are virtually continuous memory address. But I
locked
pVa1 with size 5K bytes and pVa2 with size 5K bytes. I assume all these
memory can be locked successfully.

pVa3 is just another user virtual address in the range of pVa1, however,
the memory pointed by pVa3 goes beyound pVa1.

Therefore, my question is, can I lock pVa3 with guarrantted success?
What is the overhead for multiple MmProbeAndLockPages()?

Thanks,

zhong

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Peter Wieland
Sent: Friday, October 01, 2004 5:08 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Is the future MmProbeAndLockPages() guarranted to
succeed if the memory has been locked before?

yep

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
xxxxx@exgate.tek.com
Sent: Friday, October 01, 2004 3:45 PM
To: Windows System Software Devs Interest List
Subject: [ntdev] Is the future MmProbeAndLockPages() guarranted to
succeed if the memory has been locked before?

In my program, we want to MmProbeAndLockPages() a certain amount of
memory in powerup, but in the runtime, we may only need a part of
pre-locked memory to do DMA operation. I am wondering whether the
future MmProbeAndLockPages() can be ensured to work if the memory has
already been locked. I think it does, but I just want to make sure.

Thanks,

zhongsheng


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as:
xxxxx@windows.microsoft.com To unsubscribe send a blank email to
xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: unknown lmsubst tag argument:
''
To unsubscribe send a blank email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@windows.microsoft.com
To unsubscribe send a blank email to xxxxx@lists.osr.com