Andreas Olofsson

Can you try the following: -run two scenarios, one with 0 noops inserted and one with 16 nops inserted -for each scenario, try twoo addresses, 1.) write to mailboxlo, 2)...

I want to see what happens with the exact same Epiphany code with the only difference being the addressing being written to. My theory is that you happened on a...

The following should work: idelay0=0xaaaaaaaa idleya1=0x0000000a On Tue, Jan 19, 2016 at 5:00 PM, Ola Jeppsson [email protected] wrote: > @aolofsson https://github.com/aolofsson > > Which are the right values for ELINK_RXDELAY0...

Does the following break? 00000660 : 660: 0a0b 0552 mov r0,0x5550 664: 354b 0aa2 mov r1,0xaaaa - 668: 460b 0072 mov r2,0x0000* 66c: 0aab 1552 movt r0,0x5555 670: 354b 1aa2...

Found the bug!! NASTY!!!!! Remembered that we have a long forgotten mode in the epiphany chip elink (not impemented in the fpga elink) that creates bursts when you write doubles...

How are you writing? -dma, block mem copy, or a slow for loop of e-writes? What if any is the difference in behaviour between the three cases? 2016-02-16 10:32 GMT-05:00...

Very cool! I am looking into clocking to see if we can use 1 less pll. On Jan 31, 2016 3:21 PM, "peteasa" [email protected] wrote: > Just to let you...

There is a tx_active signal. Let's see if we can tie this back to make write wait.... In the meantime, usleep(1000)..