RE: [PATCH 3/4] DaVinci SPI: Fix SPI clock prescale factor computation

2010-04-26 Thread Koeller, T.
From: Kevin Hilman [mailto:khil...@deeprootsystems.com]: > What platforms was this tested on? dm365_evm > > This looks like a fix that should be queued for 2.6.34-rc, but would > be helpful to have a review/ack from Sandeep on this. > > Also, this fix needs to go upstream through the SPI maintai

FW: [PATCH 4/4] DM365: Add platform resource management

2010-04-22 Thread Koeller, T.
From: Nori, Sekhar [mailto:nsek...@ti.com]: > Okay. Sorry, but I have not seen any platforms doing this as well. > (of course, it doesn't help that I am familiar only with DaVinci). > > Can you quote some example platform where this is being done? See arch/mips/basler/excite, where I implemented

RE: [PATCH 4/4] DM365: Add platform resource management

2010-04-22 Thread Koeller, T.
From: Nori, Sekhar [mailto:nsek...@ti.com] > I haven't seen any in-kernel driver where DMA and IRQ resources > are reserved using linux/ioport.h. Can you point to any such code? No, of course not. My very point was that drivers are not supposed to do resource management, this is a responsibility

RE: [PATCH 4/4] DM365: Add platform resource management

2010-04-21 Thread Koeller, T.
From: Nori, Sekhar [mailto:nsek...@ti.com] > IRQ resources are managed by request_irq(). EDMA does its own resource > management, but ideally should be using drivers/dma/ resource manager. This is not the kind of resource management I am talking about. You could call request_irq() with any bogus

RE: [PATCH 1/4] DM365: Make all SPI units SPI0..SPI4 available

2010-04-21 Thread Koeller, T.
Hi Kevin, > > --- /dev/null > > +++ b/arch/arm/mach-davinci/board_dm365_evm_resources.h > > @@ -0,0 +1,17 @@ > > +#ifndef _DAVINCI_EVM_RESOURCES_H > > +#define _DAVINCI_EVM_RESOURCES_H > > + > > +#include > > +#include > > + > > +/* IRQs */ > > +#define IRQ_SPI0 IRQ_DM365_SPIINT0_0 >

RE: [PATCH 4/4] DM365: Add platform resource management

2010-04-21 Thread Koeller, T.
Hi Kevin, > > Actually, the driver is the place where the resources are used and > therefore where they should be reserved and released. This patch > removes this from the drivers and puts it in SoC code which I don't > like. > > There are really two main things going on here that I see: > > 1

Which linux kernel tree to use for development?

2010-02-18 Thread Koeller, T.
Hi, being in the early stages of porting the kernel to a new DaVinci-based platform, I found that there are two git repositories specifically targeted at DaVinci development, namely the one at arago-project.org/git/projects/linux-davinci.git, and another one which is at git.kernel.org/pub/scm/li