Hi, >> The problem that I am facing is not consistent. Out of some 30 runs >> of few of the applications, the OOM occurs, though there is request >> for just one page. Why the behavior is not consistent? > >Can you explain in detail, in those 30 runs, what kind of result did you >get, how the application did memory allocation, the output of >/proc/buddyinfo at that moment and other valuable informations? Armed >with these data, perhaps we can reveal the reason of that >"inconsistency".... > I tried to run the attached code, compiled for an arm target, on the board. I didn't checked for /proc/buddyinfo but checked for /proc/meminfo. It showed out of memory (actually page allocation failure) after some 30 runs. This was a test application. With actual applications running on the target, /proc/buddyinfo gives the following information... cat /proc/buddyinfo Node 0, zone DMA 1 30 13 4 0 0 1 0 0 In my opinion when I run the actual application there is OOM as page was reserved beforehand but the actual allocation from Memory Manager was done at the point of actual usage of the page and since there is no free page, out of memory condition appeared. As IMHO the page is actually allocated when the page is used, means some read/write operation id done on that. Regards, Abu.
Attachment:
try-sound.c
Description: Binary data
