Re: [RFC] [PATCH] vmemmap fixes to use smaller pages

2008-05-15 Thread Geoff Levand
Benjamin Herrenschmidt wrote: > This patch changes vmemmap to use a different region (region 0xf) of the > address space whose page size can be dynamically configured at boot. This doesn't seem to cause any problems, and users successfully used it with the ubuntu hardy kernel, so I think it is OK

Re: [RFC] [PATCH] vmemmap fixes to use smaller pages

2008-05-02 Thread Geoff Levand
Benjamin Herrenschmidt wrote: > On Thu, 2008-05-01 at 14:46 -0700, Geoff Levand wrote: >> >> It doesn't seem to cause problems on PS3, and I added it into >> ps3-linux.git >> as other/powerpc-vmemmap-variable-page-size.diff, but I couldn't get >> it to >> fail without the patch... >> >> Could yo

Re: [RFC] [PATCH] vmemmap fixes to use smaller pages

2008-05-01 Thread Benjamin Herrenschmidt
On Thu, 2008-05-01 at 14:46 -0700, Geoff Levand wrote: > > It doesn't seem to cause problems on PS3, and I added it into > ps3-linux.git > as other/powerpc-vmemmap-variable-page-size.diff, but I couldn't get > it to > fail without the patch... > > Could you send me your kernel .config? ps3_def

Re: [RFC] [PATCH] vmemmap fixes to use smaller pages

2008-05-01 Thread Geoff Levand
Benjamin Herrenschmidt wrote: > This patch changes vmemmap to use a different region (region 0xf) of the > address space whose page size can be dynamically configured at boot. > > The problem with the current approach of always using 16M pages is that > it's not well suited to machines that have s

Re: [RFC] [PATCH] vmemmap fixes to use smaller pages

2008-04-30 Thread Benjamin Herrenschmidt
On Wed, 2008-04-30 at 12:06 -0700, Geoff Levand wrote: > > booting 2.6.25 with a ps3_defconfig and that doesn't work neither > > (though at least when doing the later, I do get a black screen & no > > sync, like of ps3fb failed monitor detection, while with current > > upstream, I just get the las

Re: [RFC] [PATCH] vmemmap fixes to use smaller pages

2008-04-30 Thread Geoff Levand
Benjamin Herrenschmidt wrote: > This patch changes vmemmap to use a different region (region 0xf) of the > address space whose page size can be dynamically configured at boot. > > The problem with the current approach of always using 16M pages is that > it's not well suited to machines that have s

[RFC] [PATCH] vmemmap fixes to use smaller pages

2008-04-29 Thread Benjamin Herrenschmidt
This patch changes vmemmap to use a different region (region 0xf) of the address space whose page size can be dynamically configured at boot. The problem with the current approach of always using 16M pages is that it's not well suited to machines that have small amounts of memory such as small par