Re: [PATCH 2/2] of/fdt: skip unflattening of disabled nodes

2017-10-09 Thread Frank Rowand
On 10/09/17 11:59, Rob Herring wrote: > On Sun, Oct 8, 2017 at 5:57 PM, Frank Rowand wrote: >> On 10/03/17 09:18, Rob Herring wrote: >>> For static DT usecases, we don't need the disabled nodes and can skip >>> unflattening. This saves a significant amount of RAM in memory constrained >>> cases. I

Re: [PATCH 2/2] of/fdt: skip unflattening of disabled nodes

2017-10-09 Thread Rob Herring
On Sun, Oct 8, 2017 at 5:57 PM, Frank Rowand wrote: > On 10/03/17 09:18, Rob Herring wrote: >> For static DT usecases, we don't need the disabled nodes and can skip >> unflattening. This saves a significant amount of RAM in memory constrained >> cases. In one example on STM32F469, the RAM usage go

Re: [PATCH 2/2] of/fdt: skip unflattening of disabled nodes

2017-10-08 Thread Frank Rowand
On 10/03/17 09:18, Rob Herring wrote: > For static DT usecases, we don't need the disabled nodes and can skip > unflattening. This saves a significant amount of RAM in memory constrained > cases. In one example on STM32F469, the RAM usage goes from 118K to 26K. > > There are a few cases in the ker

Re: [PATCH 2/2] of/fdt: skip unflattening of disabled nodes

2017-10-05 Thread Nicolas Pitre
On Tue, 3 Oct 2017, Frank Rowand wrote: > On 10/03/17 11:46, Nicolas Pitre wrote: > > On Tue, 3 Oct 2017, Rob Herring wrote: > > > >> For static DT usecases, we don't need the disabled nodes and can skip > >> unflattening. This saves a significant amount of RAM in memory constrained > >> cases. I

Re: [PATCH 2/2] of/fdt: skip unflattening of disabled nodes

2017-10-03 Thread Frank Rowand
On 10/03/17 11:46, Nicolas Pitre wrote: > On Tue, 3 Oct 2017, Rob Herring wrote: > >> For static DT usecases, we don't need the disabled nodes and can skip >> unflattening. This saves a significant amount of RAM in memory constrained >> cases. In one example on STM32F469, the RAM usage goes from 1

Re: [PATCH 2/2] of/fdt: skip unflattening of disabled nodes

2017-10-03 Thread Nicolas Pitre
On Tue, 3 Oct 2017, Rob Herring wrote: > For static DT usecases, we don't need the disabled nodes and can skip > unflattening. This saves a significant amount of RAM in memory constrained > cases. In one example on STM32F469, the RAM usage goes from 118K to 26K. > > There are a few cases in the k