Hi Alex,
On 26 February 2016 at 08:44, Alexander Graf agraf@suse.de wrote:
On 02/26/2016 04:39 PM, Simon Glass wrote:
Hi Alexander,
On 26 February 2016 at 01:58, Alexander Graf agraf@suse.de wrote:
Am 26.02.2016 um 03:51 schrieb Simon Glass sjg@chromium.org:
Hi,
On 25 February 2016 at 17:49, Alexander Graf agraf@suse.de wrote: Based on the memory map we can determine a lot of hard coded fields of TCR, like the maximum VA and max PA we want to support. Calculate those dynamically to reduce the chance for pit falls.
Signed-off-by: Alexander Graf agraf@suse.de
arch/arm/cpu/armv8/cache_v8.c | 59 +++++++++++++++++++++++++++++++++++++++- arch/arm/include/asm/armv8/mmu.h | 6 +--- include/configs/thunderx_88xx.h | 3 -- 3 files changed, 59 insertions(+), 9 deletions(-)
This approach (extern variable) looks good to me.
Reviewed-by: Simon Glass sjg@chromium.org
I didn't know ARRAY_SIZE() worked on external variables, so I leaned something also.
It doesn't. Could you please point me to a case where I'm still using ARRAY_SIZE in v4? I thought I had moved everything to null-terminating entries.
I imagined some gcc linker sizeof() magic. Oh well.
See near the top of get_tcr() - the patch on this thread.
In 1/11 the mem_map is still a locally set array which gets preinitialized to a constant array. That one is supported by ARRAY_SIZE.
Later in "thunderx: Move mmu table into board file" I move the mem_map variable into board files and change all ARRAY_SIZE instances to loops that loop until they hit an empty entry. This is mostly to make review and bisectability easier, since that way only few things change at a time.
OK, makes sense. Sorry I haven't received it very closely but you seem to have got plenty of input.
Regards, Simon