Messages in this thread | | | Date | Mon, 15 Aug 2005 13:20:03 +0530 | From | Noor <> | Subject | slab cache size-32 objects growing forever |
| |
Hi,
I noticed that my system's (kernel 2.4.22) cache memory is fast reducing. Its normal value is around 225M always. its reduced to 175M over about 30 days and still going low.
I notice this unusual behaviour /proc/slabinfo.
total size-32 objects are very large. about 300 objects of some type are being added to size-32 slab every 10min. There is no sign of any reduction.
size-32(DMA) 0 0 32 0 0 1 size-32 1979925 1979986 32 17522 17522
1) What kind of driver/kernel functions need to allocate such size-32 objects. ( for eg, if inode is for file inodes, size-32 is for ???)
2) Would it be possible to view the objects (any tool/utility for this)
3) What does this uncontrolled growth imply? is someone requesting size-32 objects uncontrollably or the Virtual Mem/slab is failing to release the cache. or could it be both ?
3) If this is some driver's fault, what kind of flags/code do i grep in driver dir, so that it will help me figure a faulty driver (eg; SLAB_NO_REAP ?)
Thanks in advance. Noor (NB: newbie post) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |