Skip to contentSkip to conversation

Memory zeroed when made available to user process

DISCUSSION Jefferson Carpenter 5 participants 6 messages Jun 27, 2018 – Jun 29, 2018

lore
Share & tools
lore
#0 lore raw

Is there a way for a user process to mark memory as 'sensitive' or 'non-sensitive' when it is allocated? That could allow it not to have to be zeroed before being allocated to another process.

#1 lore raw
Quoted 3 lines · Jefferson Carpenter

Is there a way for a user process to mark memory as 'sensitive' or 'non-sensitive' when it is allocated? That could allow it not to have to be zeroed before being allocated to another process.

Isn't this what we have Meltdown and Spectre for? ;-)

No, memory from the kernel is always zeroed. libc offers malloc() and calloc() for this purpose.

Signature · 2 lines
Thanks,
//richard
#2 lore raw
Quoted 10 lines · Richard Weinberger
Quoted 3 lines · Jefferson Carpenter

Is there a way for a user process to mark memory as 'sensitive' or 'non-sensitive' when it is allocated? That could allow it not to have to be zeroed before being allocated to another process.

Isn't this what we have Meltdown and Spectre for? ;-)

No, memory from the kernel is always zeroed. libc offers malloc() and calloc() for this purpose.

Well, except for the weird MAP_UNINITIALIZED. Anyway agreed that this is a bad idea and the flag should have never been merged. I've just mentioned it for completness.

Signature · 2 lines
Michal Hocko
SUSE Labs
#3 lore raw

Am Mittwoch, 27. Juni 2018, 15:12:48 CEST schrieb Michal Hocko:

Quoted 7 lines · Richard Weinberger
Quoted 6 lines
Quoted 3 lines · Jefferson Carpenter

Is there a way for a user process to mark memory as 'sensitive' or 'non-sensitive' when it is allocated? That could allow it not to have to be zeroed before being allocated to another process.

Quoted 8 lines
Quoted 4 lines

Isn't this what we have Meltdown and Spectre for? ;-)

No, memory from the kernel is always zeroed. libc offers malloc() and calloc() for this purpose.

Well, except for the weird MAP_UNINITIALIZED. Anyway agreed that this is a bad idea and the flag should have never been merged. I've just mentioned it for completness.

Oh, I forgot about the crazy nommu world. :-)

Thanks,
//richard

Signature · 2 lines
sigma star gmbh - Eduard-Bodem-Gasse 6 - 6020 Innsbruck - Austria
ATU66964118 - FN 374287y
#4 lore raw
Quoted 12 lines · Richard Weinberger

Am Mittwoch, 27. Juni 2018, 15:12:48 CEST schrieb Michal Hocko:

Quoted 11 lines · Richard Weinberger
Quoted 10 lines
Quoted 3 lines · Jefferson Carpenter

Is there a way for a user process to mark memory as 'sensitive' or 'non-sensitive' when it is allocated? That could allow it not to have to be zeroed before being allocated to another process.

Isn't this what we have Meltdown and Spectre for? ;-)

No, memory from the kernel is always zeroed. libc offers malloc() and calloc() for this purpose.

Interesting. Let's say

Process 1:
free(use_memory(malloc(1024)));

Then Process 2:
malloc(1024);

The physical RAM used to service Process 2's malloc call has to be zeroed to prevent it from leaking data from Process 1. However, if Process 1 could mark that memory as non-sensitive, then it would not have to be zeroed, saving the time it takes to do that. However, this would require at least a bit per memory page, so maybe it's not worth it.

Quoted 10 lines
Quoted 4 lines

Well, except for the weird MAP_UNINITIALIZED. Anyway agreed that this is a bad idea and the flag should have never been merged. I've just mentioned it for completness.

Oh, I forgot about the crazy nommu world. :-)

Thanks,
//richard

#5 lore raw

Am Freitag, 29. Juni 2018, 02:52:16 CEST schrieb Jefferson Carpenter:

Quoted 19 lines · Richard Weinberger
Quoted 12 lines

Am Mittwoch, 27. Juni 2018, 15:12:48 CEST schrieb Michal Hocko:

Quoted 11 lines · Richard Weinberger
Quoted 10 lines
Quoted 3 lines · Jefferson Carpenter

Is there a way for a user process to mark memory as 'sensitive' or 'non-sensitive' when it is allocated? That could allow it not to have to be zeroed before being allocated to another process.

Isn't this what we have Meltdown and Spectre for? ;-)

No, memory from the kernel is always zeroed. libc offers malloc() and calloc() for this purpose.

Interesting. Let's say

Process 1:
free(use_memory(malloc(1024)));

Quoted 8 lines

Then Process 2:
malloc(1024);

The physical RAM used to service Process 2's malloc call has to be zeroed to prevent it from leaking data from Process 1. However, if Process 1 could mark that memory as non-sensitive, then it would not have to be zeroed, saving the time it takes to do that. However, this would require at least a bit per memory page, so maybe it's not worth it.

Is this really a measurable overhead on your system? Do you have numbers?

Thanks,
//richard