SMuRF carrier failing to configure on hammer due to DDR3 SO-DIMM failure

[swh76]

Recently, @aashrita reported that SMuRF carrier C03-45 was failing to properly configure on hammer. Looking at the low level status;

cryo@smurf-srv19:/$ amcc_dump_bsi --all shm-smrf-sp01/4
================================================================================
| BSI: shm-smrf-sp01/4/CEN (shm-smrf-sp01/4/4)                 |
BSI Ld State: 2     (BOOTING)
BSI Ld Status: 0x00000000 (SUCCESS)
 BSI Version: 0x0103 = 1.3
    MAC 0: 08:00:56:00:52:0d
    MAC 1: 08:00:56:00:52:0e
    MAC 2: 08:00:56:00:52:0f
    MAC 3: 08:00:56:00:52:10
  DDR status: 0x0002: MemErr: F, MemRdy: F, Eth Link: Up
 Enet uptime:   190737 seconds
 FPGA uptime:   190738 seconds
 FPGA version: 0x00000040
 BL start adx: 0x04000000
   Crate ID: 0x0004
  ATCA slot: 4
  AMC 0 info: Aux: 01 Ser: 9c0000018c7abb70 Type: 0a Ver: C03 BOM: 01 Tag: A01-047
  AMC 2 info: Aux: 01 Ser: 150000018ceb6a70 Type: 0a Ver: C03 BOM: 01 Tag: A01-051
   GIT hash: 7aac8b182ee866e669b14451246ae0b2c6331d00
FW bld string: 'AmcCarrierKU15PFsbl: Vivado v2018.2, rdsrv107.slac.stanford.edu (x86_64), Built Mon Aug 6 15:19:23 PDT 2018 by ruckman'
--------------------------------------------------------------------------------

this is due to a DDR failure because it never went ready. This issue may be fixable by replacing the SO-DIMM on the carrier with a new one. I have put the SO-DIMM p/n and instructions for doing the replacement on this SO confluence page - https://simonsobs.atlassian.net/wiki/spaces/PRO/pages/641466385/SMuRF+carrier+SO-DIMM+card+replacement.