Skip to contentSkip to conversation

Overcommit_memory logic fails when swap is off

DISCUSSION Martin Egholm Nielsen 2 participants 3 messages Apr 5, 2005 – Apr 12, 2005

lore
Share & tools
lore
#0 lore raw

Hello there,

After 2½ year I stumbled over this thread:

Quoted 11 lines · Alan Cox
Quoted 4 lines

2.5 propses including the ability to set the %age between the 0% of mode 3, the 50 of mode 2 and upwards to things relevant in some embedded system cases. So for 2.6 you will be able to tune it in different ways according to precise understanding of workload.

Alan, hch or someone asked if it would be possible to merge the 2.5 behavior into 2.4-ac ... are you interested or do you not want to break compatibility?

Note for "mode 2" the behavior is identical. For "mode 3" they would also need to set vm_memory_ratio to "0".

If you are willing, I'll send you a diff...
Robert Love

As I can see, the patch never made it in 2.4.x, right?! So, I would like a diff - if still possible :-)

I know it's been a while, but I'm stuck with a 2.4 kernel for my embedded device and have problems with the OOM...

Best regards,
Martin Egholm

#1 lore raw
Quoted 2 lines · Martin Egholm Nielsen

As I can see, the patch never made it in 2.4.x, right?! So, I would like a diff - if still possible :-)

I submitted it ages ago but it was rejected and punted to 2.6. The 2.4 support is available in Red Hat Enterprise Linux 3, so you can pull it out of the source rpms on the ftp site or out of Centos etc

Alan

#2 lore raw

Hi Alan,

Quoted 5 lines
Quoted 2 lines

As I can see, the patch never made it in 2.4.x, right?! So, I would like a diff - if still possible :-)

I submitted it ages ago but it was rejected and punted to 2.6. The 2.4 support is available in Red Hat Enterprise Linux 3, so you can pull it out of the source rpms on the ftp site or out of Centos etc

Bingo. I seem to have found what looks like it in:
linux-2.4.21-selected-ac-bits.patch linux-2.4.21-tunable-overcommit.patch

But many of the other patches messes with mmap.c, as well :-), so I guess there is some tracking assignment ahead...

But nevertheless, thanks for pointing me to the source...

BR,
Martin Egholm