Skip to contentSkip to conversation

Should GFP_ATOMIC fail when we're below low watermark?

PATCH In review Nigel Cunningham 5 participants 11 messages Aug 20, 2007 – Aug 21, 2007

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Should GFP_ATOMIC fail when we're below low watermark?

Hi all.

In current git (and for a while now), an attempt to allocate memory with GFP_ATOMIC will fail if we're below the low watermark level. The only way to access that memory that I can see (not that I've looked that hard) is to have PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is correct. Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about GFP_KERNEL?

The following patch is a potential fix for GFP_ATOMIC.

Regards,

Nigel

 page_alloc.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
mm/page_alloc.c +2 -2
diff -ruNp 995-gfp-atomic-alloc.patch-old/mm/page_alloc.c 995-gfp-atomic-alloc.patch-new/mm/page_alloc.c
@@ -1286,8 +1286,8 @@ restart:
 	/* This allocation should allow future memory freeing. */
 
 rebalance:
-	if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE)))
-			&& !in_interrupt()) {
+	if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE)) ||
+			(gfp_mask & GFP_ATOMIC)) && !in_interrupt()) {
 		if (!(gfp_mask & __GFP_NOMEMALLOC)) {
 nofail_alloc:
 			/* go through the zonelist yet again, ignoring mins */
Signature · 2 lines
See http://www.tuxonice.net for Howtos, FAQs, mailing
lists, wiki and bugzilla info.
#1 lore raw
Quoted 9 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory with GFP_ATOMIC will fail if we're below the low watermark level. The only way to access that memory that I can see (not that I've looked that hard) is to have PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is correct. Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about GFP_KERNEL?

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

#2 lore raw

Hi.

Quoted 5 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 4 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory with GFP_ATOMIC will fail if we're below the low watermark level. The only way

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is to

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 5 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

Quoted 8 lines

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated for this purpose is essentially useless 99.9% of the time?

Regards,

Nigel

Signature · 5 lines
Nigel Cunningham
Christian Reformed Church of Cobden
103 Curdie Street, Cobden 3266, Victoria, Australia
Ph. +61 3 5595 1185 / +61 417 100 574
Communal Worship: 11 am Sunday.
#3 lore raw
Quoted 29 lines · Nigel Cunningham

Hi.

Quoted 5 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 4 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory with GFP_ATOMIC will fail if we're below the low watermark level. The only way

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is to

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 5 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

Quoted 8 lines

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

Quoted 4 lines

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated for this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

#4 lore raw

Hi.

Quoted 8 lines · On Monday 20 August 2007 18:59:36 Peter Zijlstra wrote:
Quoted 7 lines · Nigel Cunningham

Hi.

Quoted 4 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory

with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GFP_ATOMIC will fail if we're below the low watermark level. The only

way

Quoted 2 lines
Quoted 2 lines

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is

to

Quoted 14 lines
Quoted 14 lines

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 9 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated

for

Quoted 8 lines
Quoted 1 line

this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

I was just trying to make hibernation more reliable in sitations where there's low amounts of memory available. I guess the amount of memory that's reserved for this has increased, because some users have been reporting issues that hadn't appeared before. No problem. I'll work around it.

Nigel

Signature · 2 lines
See http://www.tuxonice.net for Howtos, FAQs, mailing
lists, wiki and bugzilla info.
#5 lore raw
Quoted 46 lines · Nigel Cunningham

Hi.

Quoted 8 lines · On Monday 20 August 2007 18:59:36 Peter Zijlstra wrote:
Quoted 7 lines · Nigel Cunningham

Hi.

Quoted 4 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory

with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GFP_ATOMIC will fail if we're below the low watermark level. The only

way

Quoted 2 lines
Quoted 2 lines

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is

to

Quoted 14 lines
Quoted 14 lines

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 9 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated

for

Quoted 6 lines
Quoted 1 line

this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

Quoted 7 lines
Quoted 2 lines

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

I was just trying to make hibernation more reliable in sitations where there's low amounts of memory available. I guess the amount of memory that's reserved for this has increased, because some users have been reporting issues that hadn't appeared before. No problem. I'll work around it.

I think the last time the default reserves were changed was 2.6.12 or there about.

Perhaps Mel fiddled with it in .23-rc ?

#6 lore raw
Quoted 55 lines · On Mon, 20 Aug 2007 13:06:01 +0200 Peter Zijlstra wrote:
Quoted 46 lines · Nigel Cunningham

Hi.

Quoted 8 lines · On Monday 20 August 2007 18:59:36 Peter Zijlstra wrote:
Quoted 7 lines · Nigel Cunningham

Hi.

Quoted 4 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory

with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GFP_ATOMIC will fail if we're below the low watermark level. The only

way

Quoted 2 lines
Quoted 2 lines

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is

to

Quoted 14 lines
Quoted 14 lines

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 9 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated

for

Quoted 6 lines
Quoted 1 line

this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

Quoted 7 lines
Quoted 2 lines

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

I was just trying to make hibernation more reliable in sitations where there's low amounts of memory available. I guess the amount of memory that's reserved for this has increased, because some users have been reporting issues that hadn't appeared before. No problem. I'll work around it.

Quoted 4 lines

I think the last time the default reserves were changed was 2.6.12 or there about.

Perhaps Mel fiddled with it in .23-rc ?

Could there be a slab vs. slub difference?

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

#7 lore raw
In reply to Peter Zijlstra

Hi.

Quoted 4 lines · On Monday 20 August 2007 21:06:01 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi.

On Monday 20 August 2007 18:59:36 Peter Zijlstra wrote:

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines · Nigel Cunningham
Quoted 7 lines

Hi.

Quoted 4 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate

memory

Quoted 2 lines
Quoted 2 lines

with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GFP_ATOMIC will fail if we're below the low watermark level. The

only

Quoted 3 lines
Quoted 3 lines

way

Quoted 2 lines
Quoted 2 lines

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard)

is

Quoted 3 lines
Quoted 3 lines

to

Quoted 2 lines
Quoted 2 lines

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour

is

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 6 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has

__GFP_WAIT

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

and hence can sleep and wait for reclaim so that should not be a

problem

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

(usually).

GFP_ATOMIC may not access the reserves because the reserves are

needed

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines

to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is

the

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

only purpose, particularly considering that whatever memory is

allocated

Quoted 11 lines
Quoted 11 lines

for

Quoted 8 lines
Quoted 1 line

this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

I was just trying to make hibernation more reliable in sitations where

there's

Quoted 1 line
Quoted 1 line

low amounts of memory available. I guess the amount of memory that's

reserved

Quoted 7 lines
Quoted 2 lines

for this has increased, because some users have been reporting issues that hadn't appeared before. No problem. I'll work around it.

I think the last time the default reserves were changed was 2.6.12 or there about.

Perhaps Mel fiddled with it in .23-rc ?

Nothing jumps out at me from a quick browse through gitk. I guess I'll just shrug my shoulders and get on with life.

Regards,

Nigel

Signature · 5 lines
Nigel Cunningham
Christian Reformed Church of Cobden
103 Curdie Street, Cobden 3266, Victoria, Australia
Ph. +61 3 5595 1185 / +61 417 100 574
Communal Worship: 11 am Sunday.
#8 lore raw
In reply to Peter Zijlstra

On (20/08/07 13:06), Peter Zijlstra didst pronounce:

Quoted 55 lines · Nigel Cunningham
Quoted 46 lines

Hi.

Quoted 8 lines · On Monday 20 August 2007 18:59:36 Peter Zijlstra wrote:
Quoted 7 lines · Nigel Cunningham

Hi.

Quoted 4 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory

with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GFP_ATOMIC will fail if we're below the low watermark level. The only

way

Quoted 2 lines
Quoted 2 lines

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is

to

Quoted 14 lines
Quoted 14 lines

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 9 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated

for

Quoted 6 lines
Quoted 1 line

this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

Quoted 7 lines
Quoted 2 lines

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

I was just trying to make hibernation more reliable in sitations where there's low amounts of memory available. I guess the amount of memory that's reserved for this has increased, because some users have been reporting issues that hadn't appeared before. No problem. I'll work around it.

Quoted 5 lines

I think the last time the default reserves were changed was 2.6.12 or there about.

Perhaps Mel fiddled with it in .23-rc ?

I haven't changed how this works as such. When ZONE_MOVABLE is configured, then the reserves will be higher than normal because of how sysctl_lowmem_reserve_ratio[] works. But in the normal cases like we're dealing with here, the ratios remain unchanged.

Signature · 3 lines
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab
#9 lore raw
In reply to Nigel Cunningham
Quoted 46 lines · Nigel Cunningham

Hi.

Quoted 8 lines · On Monday 20 August 2007 18:59:36 Peter Zijlstra wrote:
Quoted 7 lines · Nigel Cunningham

Hi.

Quoted 4 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory

with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GFP_ATOMIC will fail if we're below the low watermark level. The only

way

Quoted 2 lines
Quoted 2 lines

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is

to

Quoted 14 lines
Quoted 14 lines

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 9 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated

for

Quoted 6 lines
Quoted 1 line

this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

Quoted 7 lines
Quoted 2 lines

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

I was just trying to make hibernation more reliable in sitations where there's low amounts of memory available. I guess the amount of memory that's reserved for this has increased, because some users have been reporting issues that hadn't appeared before. No problem. I'll work around it.

Can you please point me to these reports, BTW?

Greetings,
Rafael

#10 lore raw
In reply to Nigel Cunningham

On (20/08/07 20:55), Nigel Cunningham didst pronounce:

Quoted 46 lines

Hi.

Quoted 8 lines · On Monday 20 August 2007 18:59:36 Peter Zijlstra wrote:
Quoted 7 lines · Nigel Cunningham

Hi.

Quoted 4 lines · On Monday 20 August 2007 12:43:50 Peter Zijlstra wrote:
Quoted 3 lines · Nigel Cunningham

Hi all.

In current git (and for a while now), an attempt to allocate memory

with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GFP_ATOMIC will fail if we're below the low watermark level. The only

way

Quoted 2 lines
Quoted 2 lines

to

Quoted 1 line
Quoted 1 line

access that memory that I can see (not that I've looked that hard) is

to

Quoted 14 lines
Quoted 14 lines

have

Quoted 1 line
Quoted 1 line

PF_MEMALLOC set (ie from kswapd). I'm wondering if this behaviour is

correct.

Quoted 1 line
Quoted 1 line

Shouldn't GFP_ATOMIC allocations ignore watermarks too? How about

GFP_KERNEL?

Quoted 9 lines
Quoted 2 lines

The following patch is a potential fix for GFP_ATOMIC.

Sorry, no.

GFP_ATOMIC must fail when below the watermark. GFP_KERNEL has __GFP_WAIT and hence can sleep and wait for reclaim so that should not be a problem (usually).

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

GFP_ATOMIC may not access the reserves because the reserves are needed to get out of OOM deadlocks within the VM. Consider the fact that freeing memory needs memory - if there is no memory free, you cannot free memory and you're pretty much stuck.

I guess, then, the question should be whether the watermark values are appropriate. Do we need high order allocations watermarked if this is the only purpose, particularly considering that whatever memory is allocated

for

Quoted 6 lines
Quoted 1 line

this purpose is essentially useless 99.9% of the time?

Could you perhaps explain what you're trying to do? No matter what we do, GFP_ATOMIC will fail eventually, there is only so much one can do without blocking.

Quoted 7 lines
Quoted 2 lines

As for higher order allocations, until we have a full online defrag solution those too can fail at any moment (even with __GFP_WAIT).

I was just trying to make hibernation more reliable in sitations where there's low amounts of memory available. I guess the amount of memory that's reserved for this has increased, because some users have been reporting issues that hadn't appeared before. No problem. I'll work around it.

Where are these reports? I'm not familiar with how hibernation works but why does it need a large number of GFP_ATOMIC allocations?

Signature · 3 lines
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab