On Sun, Oct 14, 2007 at 04:12:20PM -0700, Jeremy Fitzhardinge wrote:Every vunmap() cal causes a global TLB sync, and the region lists are globl with a spin lock protecting them. I thin kNick has shown a 64p altix with ~60 cpus spinning on the vmap locks under a simple workload.... A directory traversal when using large directory block sizes with large directories.... The pages are still attached to the blockdev address space mapping, but there's nothing stopping them from being reclaimed before they are unmapped. It's all "freed memory". At the time we pull the buffer down, there are no further references to the buffer. the pages are released and the mapping is never used again until it is torn down. it is torn down either on the next xfsbufd run (either memory pressure or every 15s) or every 64th new vmap() call to map new buffers. We've already got a memroy shrinker hook that triggers this reclaim. Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group -
| Borislav Petkov | 2.6.23-rc1: no setup signature found... |
| Andrew Morton | Re: [PATCH] Memory management livelock |
| Greg KH | [GIT PATCH] driver core patches against 2.6.24 |
| Steven King | ti_usb_3410_5052 breakage in 2.6.24-rc1 |
git: | |
| Linus Torvalds | cleaner/better zlib sources? |
| Theodore Ts'o | [PATCH] Add --no-reuse-delta, --window, and --depth options to git-gc |
| Karl | Re: [PATCH] Add a birdview-on-the-source-code section to the user manual |
| Yossi Leybovich | corrupt object on git-gc |
| GVG GVG | ssh_exchange_identification: Connection closed by remote host |
| Richard Stallman | Re: Real men don't attack straw men |
| David Newman | setting dscp or tos bits |
| Stijn | Re: [i386/Thinkpad T41]USB mouse + Xorg obsd 4.1 |
| Quentin Garnier | [cube@cubidou.net: Re: Moving ethfoo in the main tree] |
| cube | Moving ethfoo in the main tree |
| Blair Sadewitz | PCI latency timer values |
| Oliver Gould | kqemu: driver(9)/filedesc(9) questions |
