Re: [PATCH] ppc32/64: Map prefetchable PCI without guarded bit

2005-03-25 Thread Benjamin Herrenschmidt
On Thu, 2005-03-24 at 19:20 +0100, Segher Boessenkool wrote: > > While experimenting with framebuffer access performances, we noticed a > > very significant improvement in write access to it when not setting > > the "guarded" bit on the MMU mappings. This bit basically says that > > reads and write

Re: [PATCH] ppc32/64: Map prefetchable PCI without guarded bit

2005-03-25 Thread Benjamin Herrenschmidt
On Thu, 2005-03-24 at 08:54 -0800, Jesse Barnes wrote: > On Wednesday, March 23, 2005 10:24 pm, Benjamin Herrenschmidt wrote: > > While experimenting with framebuffer access performances, we noticed a > > very significant improvement in write access to it when not setting > > the "guarded" bit on t

Re: [PATCH] ppc32/64: Map prefetchable PCI without guarded bit

2005-03-24 Thread Segher Boessenkool
While experimenting with framebuffer access performances, we noticed a very significant improvement in write access to it when not setting the "guarded" bit on the MMU mappings. This bit basically says that reads and writes won't have side effects (it allows speculation). Unless the data is already

Re: [PATCH] ppc32/64: Map prefetchable PCI without guarded bit

2005-03-24 Thread Jesse Barnes
On Thursday, March 24, 2005 8:54 am, Jesse Barnes wrote: > On Wednesday, March 23, 2005 10:24 pm, Benjamin Herrenschmidt wrote: > > While experimenting with framebuffer access performances, we noticed a > > very significant improvement in write access to it when not setting > > the "guarded" bit on

Re: [PATCH] ppc32/64: Map prefetchable PCI without guarded bit

2005-03-24 Thread Jesse Barnes
On Wednesday, March 23, 2005 10:24 pm, Benjamin Herrenschmidt wrote: > While experimenting with framebuffer access performances, we noticed a > very significant improvement in write access to it when not setting > the "guarded" bit on the MMU mappings. This bit basically says that > reads and write