fix: ERR-0026 SDRAM tMRD datasheet-unit mismatch at 64MHz

Real Alliance Memory AS4C4M16SA-6TIN datasheet (Rev 5.0, Table 17)
specifies tMRD as a fixed 2-tCK cycle count, not an ns value.
sdram_controller.v modeled it via ns_to_cycles(12), which rounded to
2 cycles by coincidence at every previously-tested frequency
(100/133/166MHz) but rounds to only 1 cycle at the real 64MHz board
target -- an under-provisioned one-time init sequence. Fixed by
hardcoding T_MRD=2, matching how CAS_LATENCY is already modeled.

Verified zero regression: full 9-config legacy sweep + a new
dedicated 64MHz config (461/461 PASS each), N=2/N=4 D-Stress
(identical cycle counts), board-level smoke test (11/11 PASS).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013xXuuRUWZScuo1DeYJxs3v

hardware/v2/logs/decisions.log

@@ -2140,3 +2140,24 @@ and the new host-arb arbitration layer) -- this was deliberately not
 attempted before now, since running the real toolchain against RTL
 known to compute wrong answers would not have been a meaningful
 result. That reason no longer applies.
+
+DEC-0038
+
+DATE: 2026-09-06
+
+DECISION:
+The SDRAM datasheet-parameter audit (PRE-PCB VERIFICATION FREEZE,
+section 8) is CLOSED. Every timing parameter in hardware/v2/nms/rtl/
+sdram_controller.v was cross-checked against the real Alliance Memory
+AS4C4M16SA-6TIN datasheet (Rev 5.0, Table 17) at the real 64MHz target
+frequency, not merely re-cited from a previous, uncross-checked
+baseline. This found and closed ERR-0026 (T_MRD unit mismatch -- see
+errors.log), the only genuine discrepancy found. tRCD, tRP, tRAS,
+tWR, tMRD, tREFI are all now confirmed correct at 64MHz with zero
+regression to the existing bit-exact regression suite. See
+hardware/v2/docs/PRE_PCB_VERIFICATION.md for the full parameter-by-
+parameter table (DATASHEET PARAMETER -> REQUIRED VALUE -> RTL VALUE ->
+STATUS).
+This REVISES CHIP_READINESS.md/OPEN_ITEMS.md's own prior "SDRAM
+datasheet-parameter cross-check: OPEN (sim-level only)" status to
+CLOSED.

hardware/v2/logs/errors.log

@@ -1244,3 +1244,47 @@ STATUS: ERR-0025 (Parts A and B) fully RESOLVED. The physical SPI host
 interface is now verified correct end-to-end (SPI -> dependency_
 manager -> compute -> SDRAM -> result) under both tight and realistic-
 gap job pacing, with zero regression to the STEP19 baseline.
+
+ERR-0026 -- T_MRD datasheet-unit mismatch (SDRAM init sequence, 64MHz)
+
+DATE: 2026-09-06
+FOUND DURING: PRE-PCB VERIFICATION FREEZE, section-8 SDRAM datasheet-
+level audit (real Alliance Memory AS4C4M16SA-6TIN datasheet, Rev 5.0,
+Oct 2018, Table 17 "Electrical Characteristics/AC Operating
+Conditions", -6 speed grade).
+ROOT CAUSE: hardware/v2/nms/rtl/sdram_controller.v modeled tMRD (LOAD
+MODE REGISTER -> any command spacing) as `ns_to_cycles(12)`, i.e. as
+if it were a nanosecond-based timing spec like tRCD/tRP. The real
+datasheet specifies tMRD as a FIXED CYCLE COUNT, "2 tCK", independent
+of clock frequency -- the same category of spec as CAS_LATENCY, which
+this same file already correctly models as a fixed value two lines
+below. `ns_to_cycles(12)` rounds to exactly 2 cycles at every
+frequency this design had previously been tested at (100/133/166MHz),
+so the wrong unit model was silently masked by coincidence. At the
+real V2 board's own 64MHz operating point it rounds to only 1 cycle --
+one cycle short of the real, fixed 2-tCK minimum -- a genuine,
+datasheet-violating under-provisioning of the one-time SDRAM power-up/
+mode-register-set sequence. Confirmed NOT a bug in tRCD/tRP (correctly
+ns-based, correctly use ns_to_cycles()), tRAS (satisfied by construction:
+the fixed tRCD+CAS_LATENCY+BURST_LEN dispatch sequence is always >=6
+cycles, 93.75ns >= the real 42ns minimum at 64MHz), or tWR (folded in
+conservatively via T_RP+1, giving 3 cycles >= the real 2-tCK minimum).
+FIX: hardcoded `localparam T_MRD = 2;` (matching how CAS_LATENCY is
+already modeled), replacing `localparam T_MRD = ns_to_cycles(12);`.
+Only affects the one-time SDRAM init sequence, not per-transaction
+timing.
+VERIFICATION (all via Verilator, the trusted tool per DEC-0004):
+  - tb_sdram_controller.v: full 9-config legacy sweep (100/133/166MHz
+    x BURST_LEN 1/4/8), 461/461 PASS each config, zero regression.
+  - tb_sdram_controller.v: NEW dedicated 64MHz/BURST_LEN=4 config (the
+    real board target, not covered by the legacy sweep) -- 461/461
+    PASS, confirming the fix is correct at the frequency where the
+    bug actually manifested.
+  - tb_nms_dstress_sdram_unified.v N=2: 49788 cycles, 256/256
+    bit-exact -- IDENTICAL to pre-fix (expected: T_MRD only affects
+    one-time init, not steady-state per-transaction timing).
+  - tb_nms_dstress_sdram_unified.v N=4: 49771 cycles, 256/256
+    bit-exact -- IDENTICAL to pre-fix.
+  - tb_fpga_neural_v2_top_smoke.v: 11/11 PASS, unaffected.
+STATUS: RESOLVED. Files changed: hardware/v2/nms/rtl/sdram_
+controller.v (single localparam, plus explanatory comment).

hardware/v2/nms/rtl/sdram_controller.v

@@ -99,7 +99,21 @@ module sdram_controller #(
     // (tRCD + CAS_LATENCY + BURST_LEN data cycles, always >= 3+3+1=7
     // even at the narrowest BURST_LEN=1) already comfortably exceeds
     // it by construction before auto-precharge can begin internally.
-    localparam T_MRD    = ns_to_cycles(12);   // LOAD MODE REGISTER -> any command
+    // tMRD is specified by the real AS4C4M16SA-6TIN datasheet (Table 17)
+    // as a FIXED CYCLE COUNT ("2 tCK"), not a nanosecond value -- unlike
+    // tRCD/tRP, which genuinely are ns-based and correctly belong behind
+    // ns_to_cycles(). A previous draft modeled tMRD as ns_to_cycles(12),
+    // an assumed-equivalent ns figure that happened to round up to
+    // exactly 2 cycles at every frequency this design had been tested at
+    // (100/133/166MHz), silently masking the wrong unit model. At the
+    // real V2 board's own 64MHz operating point, ns_to_cycles(12) rounds
+    // to only 1 cycle -- one cycle short of the real, fixed 2-tCK
+    // minimum -- found via this step's own fresh datasheet-level audit
+    // (real Alliance Memory AS4C4M16SA-6TIN datasheet Rev.5.0, Table 17).
+    // Fixed by hardcoding the real, frequency-independent requirement
+    // directly, matching how CAS_LATENCY (also a real fixed-cycle spec)
+    // is already modeled two lines below.
+    localparam T_MRD    = 2;                  // LOAD MODE REGISTER -> any command (tMRD = 2 tCK, fixed)
     localparam T_INIT_US= 200;                // power-up wait, real datasheet value
     localparam T_INIT   = T_INIT_US * CLK_FREQ_MHZ;
     localparam CAS_LATENCY = 3;               // fixed for this part/speed grade