Re: [RFC PATCH 2/3] arm64: Introduce IRQ stack

2015-09-08 Thread Jungseok Lee
On Sep 7, 2015, at 11:48 PM, James Morse wrote: Hi James, > On 04/09/15 15:23, Jungseok Lee wrote: >> Currently, kernel context and interrupts are handled using a single >> kernel stack navigated by sp_el1. This forces many systems to use >> 16KB stack, not 8KB one. Low memory platforms naturally

Re: [RFC PATCH 2/3] arm64: Introduce IRQ stack

2015-09-07 Thread James Morse
On 04/09/15 15:23, Jungseok Lee wrote: > Currently, kernel context and interrupts are handled using a single > kernel stack navigated by sp_el1. This forces many systems to use > 16KB stack, not 8KB one. Low memory platforms naturally suffer from > both memory pressure and performance degradation s

Re: [RFC PATCH 2/3] arm64: Introduce IRQ stack

2015-09-07 Thread Jungseok Lee
On Sep 5, 2015, at 2:12 AM, Alexnader Kuleshov wrote: > Hello Jungseok, Hello Alexnader, > On 09-04-15, Jungseok Lee wrote: >> +config IRQ_STACK >> +bool "Use separate kernel stack when handling interrupts" >> +depends on ARM64_4K_PAGES >> +help >> + Say Y here if you want to us

Re: [RFC PATCH 2/3] arm64: Introduce IRQ stack

2015-09-04 Thread Alexnader Kuleshov
Hello Jungseok, On 09-04-15, Jungseok Lee wrote: > +config IRQ_STACK > + bool "Use separate kernel stack when handling interrupts" > + depends on ARM64_4K_PAGES > + help > + Say Y here if you want to use separate kernel stack to handle both > + hard and soft interrupts. As

[RFC PATCH 2/3] arm64: Introduce IRQ stack

2015-09-04 Thread Jungseok Lee
Currently, kernel context and interrupts are handled using a single kernel stack navigated by sp_el1. This forces many systems to use 16KB stack, not 8KB one. Low memory platforms naturally suffer from both memory pressure and performance degradation simultaneously as VM page allocator falls into s