On Thu, 27 Mar 2008, Ivan Kokshaysky wrote:Adding an alignment field should be a non-issue: the size of this structure is not likely to be a big deal (yeah, we have something like 12 of them in each PCI device etc, so smaller is better, but it's still not going to be something anybody really notices). And yeah, it might be nice to have "size" instead of "end", but the real problem with that one is actually that on 32-bit (without the 64-bit resource configuration) we want it to be "size-1" in order to be able to fit a whole 0-0xffffffff resource into a resource. And *that* would be really ugly with "size": at least right now it makes a certain amount of sense with "end" pointing to the last entry. So while I can understand your wish for "start+len" rather than "start+end", I don't think it's really practical. Linus --
| Con Kolivas | Re: [ANNOUNCE] RSDL completely fair starvation free interactive cpu scheduler |
| debian developer | Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3 |
| Steven Rostedt | Major regression on hackbench with SLUB |
| Bart Van Assche | Integration of SCST in the mainstream Linux kernel |
git: | |
| Junio C Hamano | [0/4] What's not in 1.5.2 (overview) |
| Jan Hudec | Smart fetch via HTTP? |
| Peter Oberndorfer | Re: [StGIT PATCH] Don't use patches/<branch>/current |
| Nicolas Pitre | Re: git-index-pack really does suck.. |
| GVG GVG | ssh_exchange_identification: Connection closed by remote host |
| Reyk Floeter | Re: hoststated(8): DNS Relay uses unexpected source IP address |
| Mark Zimmerman | Re: Samba(SMB) or Netatalk(AFP)? |
| Anselm R. Garbe | OpenBSD 4.0 / Xorg -> vesa 1920x1200 widescreen resolution |
| hh | Hardware compatibility list |
| Doug Evans | Re: Stabilizing Linux |
| Linus Benedict Torvalds | More answers |
| Dave `geek' Gymer | WARNING (was Re: New afio release) |
