>From: Alan Stern [mailto:stern@rowland.harvard.edu]For WA, when we get a buffer to be sent from a URB, it has to be split in chunks, each chunk has a header added. So we end up with a list of chunks, most of them quite small. Each requires a single URB to send. resources galore. If we could queue all those, the overhead would be reduced to allocating the headers (possibly in a continuous array) and the sg "descriptors" to describe the whole thing. However, the alignment stuff somebody mentioned in another email in this thread might cause problems. At the end it might not be all that doable (I might be missing some subtle isssues), but it is well worth a look. And the overhead of one URB per sg "node" kills it's usability for WAs. --
| Natalie Protasevich | [BUG] New Kernel Bugs |
| Greg KH | [GIT PATCH] driver core patches against 2.6.24 |
| Bart Van Assche | Integration of SCST in the mainstream Linux kernel |
| Andi Kleen | [PATCH x86] [0/16] Various i386/x86-64 changes |
git: | |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Jarek Poplawski | [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Linus Torvalds | Re: [GIT]: Networking |
| Jeff Kirsher | [net-next PATCH 1/7] e1000e: enable CRC stripping by default |
| Jukka Andberg | ata/wdc vs gcc3 on amiga |
| YAMAMOTO Takashi | Re: wd.c patch to reduce kernel stack usage |
| Jason Thorpe | Re: ksyms patches. |
| rick | NFS transport |
