"""B: per-cube trade-off (mode comparison, variant-invariant)
C: GEMM util ablation (A1 only, variant comparison)

Output: docs/report/1H-codesign-paper/figures/gqa_short_context/
"""
from __future__ import annotations

import csv
from pathlib import Path

import matplotlib.pyplot as plt
import numpy as np

ROOT = Path("/Users/lalicorne/Desktop/ahbm/kernbench2")
SWEEPS = ROOT / "docs" / "sweeps"
OUT = ROOT / "docs" / "report" / "1H-codesign-paper" / "figures" / "gqa_short_context"
OUT.mkdir(parents=True, exist_ok=True)

CONTEXTS = [8192, 16384, 32768, 65536]
CONTEXT_LABELS = ["8K", "16K", "32K", "64K"]
MODES = ["A1", "A2", "A4", "B"]
MODE_LABEL = {"A1": "1-kv-per-cube", "A2": "2-kv-per-cube",
              "A4": "4-kv-per-cube", "B":  "8-kv-per-cube"}
MODE_COLOR = {MODE_LABEL["A1"]: "#1f77b4",
              MODE_LABEL["A2"]: "#2ca02c",
              MODE_LABEL["A4"]: "#ffd000",
              MODE_LABEL["B"]:  "#d62728"}
VARIANTS = [
    ("baseline",        "(1) without composite"),
    ("composite",       "(2) with composite (GEMM-only)"),
    ("composite_fused", "(3) with composite + softmax_merge"),
]
VARIANT_COLOR = {
    "baseline":         "#1f77b4",
    "composite":        "#2ca02c",
    "composite_fused":  "#d62728",
}

CSV_MAP = {
    ("decode",  "baseline"):         SWEEPS / "short_context_decode_sweep.csv",
    ("decode",  "composite"):        SWEEPS / "short_context_decode_composite_sweep.csv",
    ("decode",  "composite_fused"):  SWEEPS / "short_context_decode_composite_fused_sweep.csv",
    ("prefill", "baseline"):         SWEEPS / "short_context_prefill_sweep.csv",
    ("prefill", "composite"):        SWEEPS / "short_context_prefill_composite_sweep.csv",
    ("prefill", "composite_fused"):  SWEEPS / "short_context_prefill_composite_fused_sweep.csv",
}


def load(phase, variant):
    """{(mode, S_kv): row dict}."""
    rows = {}
    with CSV_MAP[(phase, variant)].open() as f:
        for r in csv.DictReader(f):
            rows[(r["mode"], int(r["S_kv"]))] = r
    return rows


def _grouped_bars(ax, x_labels, groups, group_color, ylabel, title,
                  log=False, value_fmt="{:.3g}"):
    n = len(groups)
    nx = len(x_labels)
    width = 0.8 / n
    x = np.arange(nx)
    for i, (label, values) in enumerate(groups):
        offset = (i - (n - 1) / 2) * width
        bars = ax.bar(x + offset, values, width, label=label,
                      color=group_color[label], edgecolor="black",
                      linewidth=0.4)
        for bar, v in zip(bars, values):
            if v <= 0 and log:
                continue
            ax.text(bar.get_x() + bar.get_width() / 2,
                    bar.get_height(),
                    value_fmt.format(v), ha="center", va="bottom",
                    fontsize=7)
    ax.set_xticks(x)
    ax.set_xticklabels(x_labels)
    ax.set_xlabel("S_kv")
    ax.set_ylabel(ylabel)
    ax.set_title(title)
    if log:
        ax.set_yscale("log")
    ax.grid(True, axis="y", alpha=0.3, which="both")
    ax.legend(fontsize=8, loc="upper left")


# ── B: per-cube trade-off (mode comparison, variant 1 baseline) ──

def fig_b_per_cube_tradeoff():
    """3 metrics × 2 phases — mode comparison.

    Compute path doesn't affect HBM/IPCQ/KV cost (variant-invariant);
    use variant 1 baseline rows.
    """
    fig, axes = plt.subplots(2, 2, figsize=(13, 9))

    METRICS = [
        ("hbm_bw_util", "HBM BW utilization (per PE)", False, "{:.3f}"),
        ("ipcq_kb",     "IPCQ traffic (KB)",           True,  "{:.1f}"),
    ]
    for row, (key, ylabel, log, fmt) in enumerate(METRICS):
        for col, phase in enumerate(("prefill", "decode")):
            ax = axes[row][col]
            rows = load(phase, "baseline")
            groups = []
            for mode in MODES:
                values = [float(rows[(mode, s)][key]) for s in CONTEXTS]
                groups.append((MODE_LABEL[mode], values))
            _grouped_bars(ax, CONTEXT_LABELS, groups, MODE_COLOR,
                          ylabel=ylabel,
                          title=f"{phase.capitalize()}",
                          log=log, value_fmt=fmt)

    fig.suptitle(
        "Per-cube trade-off: HBM BW + IPCQ mode comparison",
        fontsize=12, fontweight="bold")
    fig.tight_layout()
    out = OUT / "per_cube_tradeoff_mode_compare.png"
    fig.savefig(out, dpi=140, bbox_inches="tight")
    plt.close(fig)
    print(f"  ✓ {out}")


# ── C: GEMM util ablation (A1, variant compare) ──

def fig_c_gemm_util_ablation():
    """A1 only; x=S_kv, grouped bars per variant; 2 panels (prefill/decode).

    NOTE annotation: gemm_util gap is largely supertile padding overhead
    (M=8 < TILE_M=32), not pure fusion gain. See ADR-0070 limitation #4.
    """
    fig, axes = plt.subplots(1, 2, figsize=(13, 5.4))
    for ax, phase in zip(axes, ("prefill", "decode")):
        groups = []
        for variant_key, _ in VARIANTS:
            rows = load(phase, variant_key)
            values = [float(rows[("A1", s)]["gemm_util"]) for s in CONTEXTS]
            groups.append((variant_key, values))
        _grouped_bars(ax, CONTEXT_LABELS, groups, VARIANT_COLOR,
                      ylabel="GEMM engine utilization (per PE)",
                      title=f"{phase.capitalize()}",
                      log=False, value_fmt="{:.4f}")
        # Replace legend with readable labels
        ax.legend([v[1] for v in VARIANTS], fontsize=8, loc="upper left")

    fig.suptitle("1-kv-per-cube: GEMM utilization variant comparison",
                 fontsize=12, fontweight="bold")
    fig.text(
        0.5, -0.04,
        "⚠ Higher gemm_util on composite/fused is largely supertile "
        "padding overhead (M=G=8 padded to TILE_M=32), not pure fusion "
        "gain. See ADR-0070 limitation #4.",
        ha="center", fontsize=9, style="italic", color="#555")
    fig.tight_layout()
    out = OUT / "gemm_util_a1_variant_ablation.png"
    fig.savefig(out, dpi=140, bbox_inches="tight")
    plt.close(fig)
    print(f"  ✓ {out}")


if __name__ == "__main__":
    print(f"Output: {OUT}")
    fig_b_per_cube_tradeoff()
    fig_c_gemm_util_ablation()
    print("\nDone.")