Re: [PATCH 2/2] goldfish_rtc: change MemoryRegionOps endianness to DEVICE_NATIVE_ENDIAN

2020-10-12 Thread Laurent Vivier
Le 09/10/2020 à 13:38, Laurent Vivier a écrit : > The doc [1] doesn't define the endianness, but the kernel driver > uses readl() to access the registers, so we can guess it depends > on the architecture endianness. > > As riscv architecture endianness is little it might not change anything > for

Re: [PATCH 2/2] goldfish_rtc: change MemoryRegionOps endianness to DEVICE_NATIVE_ENDIAN

2020-10-09 Thread Alistair Francis
On Fri, Oct 9, 2020 at 4:40 AM Laurent Vivier wrote: > > The doc [1] doesn't define the endianness, but the kernel driver > uses readl() to access the registers, so we can guess it depends > on the architecture endianness. > > As riscv architecture endianness is little it might not change anything

[PATCH 2/2] goldfish_rtc: change MemoryRegionOps endianness to DEVICE_NATIVE_ENDIAN

2020-10-09 Thread Laurent Vivier
The doc [1] doesn't define the endianness, but the kernel driver uses readl() to access the registers, so we can guess it depends on the architecture endianness. As riscv architecture endianness is little it might not change anything for it. Moreover, android implementation uses DEVICE_NATIVE_END