| | 1 | '''1/13/06'' |
| | 2 | |
| | 3 | Our setup: Fedora Core 3 linux, kernel 2.6.12-1.1378_FC3. |
| | 4 | gcc version 3.4.4 20050721 (Red Hat 3.4.4-2). |
| | 5 | Dell motherboard USB (1.1) |
| | 6 | |
| | 7 | Hooked up to Spartan-3 eval board with 6" ribbon cable. Connections: |
| | 8 | |
| | 9 | {{{ |
| | 10 | QuickUSB Signal Spartan-A2 pin (FPGA) |
| | 11 | J2 8 IFCLK 35 (D9) |
| | 12 | J2 19, 21...33 FD0-7 19..33 (B5...A13) |
| | 13 | J2 22 REN 22 (B7) |
| | 14 | J2 24 WEN 24 (B8) |
| | 15 | J3 4 SLRD 4 (E6) |
| | 16 | J3 6 SLWR 5 (D5) |
| | 17 | (plus GNDs) |
| | 18 | |
| | 19 | }}} |
| | 20 | |
| | 21 | Simple FPGA firmware which just monitors REN, WEN and sends counter values for reading. |
| | 22 | Read data checked superficially. Write data not checked (just word count). |
| | 23 | Results: |
| | 24 | |
| | 25 | * Reads, writes work for 128, 256, 512 bytes at a time using master mode |
| | 26 | |
| | 27 | Other byte counts cause mysterious problems. After a failure, it seems that |
| | 28 | re-enumeration is required to restore order. |
| | 29 | |
| | 30 | Slave mode doesn''t work at all (yet). Tried loop-back: |
| | 31 | |
| | 32 | {{{ |
| | 33 | slwr <= not ren |
| | 34 | slrd <= not wen |
| | 35 | |
| | 36 | }}} |
| | 37 | With and without the ''not'. Doesn't work. |
| | 38 | |
