I have a puzzling question. I have an in house design using Micron MT42L256M64D4LM-25 WT
1.I am running the Stress tool to test the LPDDR2 memory.
DDR Stress Test (1.0.3) for MX6DQ
Build: Jun 25 2014, 12:09:21
Freescale Semiconductor, Inc.
2. I get consistent PASSes for ARM core speed 800MHz and DDR freq of 400MHz
I also get consistent PASSes for ARM core speed 1GHz and DDR freq of 380MHz
but
I get a consistent FAIL for ARM core speed 1GHz and DDR freq of 400MHz
here's part of what I get
Test channel 0
ABS_OFFSET=0x00000000 result[00]=0x1111
ABS_OFFSET=0x04040404 result[01]=0x1111
ABS_OFFSET=0x08080808 result[02]=0x1111
ABS_OFFSET=0x0C0C0C0C result[03]=0x1111
ABS_OFFSET=0x10101010 result[04]=0x1111
ABS_OFFSET=0x14141414 result[05]=0x1111
ABS_OFFSET=0x18181818 result[06]=0x1111
ABS_OFFSET=0x1C1C1C1C result[07]=0x1111
ABS_OFFSET=0x20202020 result[08]=0x0011
ABS_OFFSET=0x24242424 result[09]=0x0011
ABS_OFFSET=0x28282828 result[0A]=0x0011
ABS_OFFSET=0x2C2C2C2C result[0B]=0x0011
ABS_OFFSET=0x30303030 result[0C]=0x0011
ABS_OFFSET=0x34343434 result[0D]=0x0011
ABS_OFFSET=0x38383838 result[0E]=0x0011
ABS_OFFSET=0x3C3C3C3C result[0F]=0x0011
ABS_OFFSET=0x40404040 result[10]=0x0011
ABS_OFFSET=0x44444444 result[11]=0x0011
ABS_OFFSET=0x48484848 result[12]=0x0011
ABS_OFFSET=0x4C4C4C4C result[13]=0x0011
ABS_OFFSET=0x50505050 result[14]=0x0011
ABS_OFFSET=0x54545454 result[15]=0x1111
allways the lower 2 bytes.
I can't understand why it is different at 1 GHz as I understand the PLL for the memory is still originating from the 24MHz external xtal.
and the ARM PLL is separate from the System PLL.
Any ideas?