Ok.
o Can you pack the tar so it unpacks into a subdir next time. I
had to pick that one out of tmp again. (Ok I should have looked but..)
On the actual code
It tries to grab as big a chunk as possible. Thats horrible if not needed.
The linux vm has problems with big chunks without someone writing a chunk
eater. A lot of drivers dont care about chunk sizes (eg the BT848 video
chip - perhaps a chunk hint is needed). Also perhaps it should look how
many big chunks are free and not use the last 'n', sort of like 2.1.x
tries to keep them free. Can I also have a GFP_DMA hint option. There
are ISA cards that could benefit here.
Its using mmap. That unfortunately means I can't drop it into shared
memory or the video ram. Thats a nasty problem for a few specific cases
I care about - firstly mmap /dev/mem, start video grabbing on it, secondly
grabbing into the image part of a sys5 shared memory MITSHM block.
Neither to my mind precludes your code being useful. Its something to
probably solve later, and mmapping pages solves so many nasties.
The other item it needs to sort somehow is limiting kernel locked pages
per user. This is something I keep pointing out to Linus that we need
for networking with large fd sets also and is critical for asynchronous
I/O scaling. Thats mostly seperate to this patch however since it comes
down to the patch calling some kind of
if(!can_i_have_locked_memory(size))
return -ENOBUFS
Otherwise it looks nice
Alan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu