Jeff Garzik <jeff@garzik.org> writes:Thanks. They (#ifdef __ARMEB__) are indeed very ugly but I don't know how to clean them, except by splitting the driver into two (which is worse). The most used things are #defined and then used without those __ARMEB__ checks. Duplicating those functions is effectively splitting into two :-( Well, perhaps Intel could give us firmware which could convert LE-BE on the fly while TX/RX, that would make the LE code go away. Binary-only :-( OTOH I could convert #if* DEBUG_*, I will look at it. -- Krzysztof Halasa --
| Greg KH | [GIT PATCH] driver core patches against 2.6.24 |
| Ingo Molnar | Re: [BUG] New Kernel Bugs |
| Tony Lindgren | [PATCH 42/90] ARM: OMAP: Tabify mux.c |
| Roland Dreier | Re: Integration of SCST in the mainstream Linux kernel |
git: | |
| Martin Langhoff | Re: pack operation is thrashing my server |
| Andreas Ericsson | Re: VCS comparison table |
| Ingo Molnar | [OT] Your branch is ahead of the tracked remote branch 'origin/master' by 50 commi... |
| Nicolas Vilz 'niv' | git + ssh + key authentication feature-request |
| Richard Stallman | Real men don't attack straw men |
| Darren Spruell | Re: About Xen: maybe a reiterative question but .. |
| Nick Holland | Re: 4.1 on ALIX.1C - recommendations? |
| Lord Sporkton | Re: low-MHz server |
| KOSAKI Motohiro | [bug?] tg3: Failed to load firmware "tigon/tg3_tso.bin" |
| Mark Lord | Re: 2.6.25-rc8: FTP transfer errors |
| Jarek Poplawski | [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Brandeburg, Jesse | RE: e1000 full-duplex TCP performance well below wire speed |
