Merge branch 'dev' into test-fp8

.github/workflows/on_pull_request.yaml

@@ -20,7 +20,7 @@ jobs:
           #     not to have GHA download an (at the time of writing) 4 GB cache
           #     of PyTorch and other dependencies.
       - name: Install Ruff
-        run: pip install ruff==0.0.272
+        run: pip install ruff==0.1.6
       - name: Run Ruff
         run: ruff .
   lint-js:

README.md

@@ -121,7 +121,9 @@ Alternatively, use online services (like Google Colab):
 # Debian-based:
 sudo apt install wget git python3 python3-venv libgl1 libglib2.0-0
 # Red Hat-based:
-sudo dnf install wget git python3
+sudo dnf install wget git python3 gperftools-libs libglvnd-glx 
+# openSUSE-based:
+sudo zypper install wget git python3 libtcmalloc4 libglvnd
 # Arch-based:
 sudo pacman -S wget git python3
 ```
@@ -174,5 +176,6 @@ Licenses for borrowed code can be found in `Settings -> Licenses` screen, and al
 - TAESD - Ollin Boer Bohan - https://github.com/madebyollin/taesd
 - LyCORIS - KohakuBlueleaf
 - Restart sampling - lambertae - https://github.com/Newbeeer/diffusion_restart_sampling
+- Hypertile - tfernd - https://github.com/tfernd/HyperTile
 - Initial Gradio script - posted on 4chan by an Anonymous user. Thank you Anonymous user.
 - (You)

extensions-builtin/hypertile/hypertile.py

@@ -0,0 +1,345 @@
+"""
+Hypertile module for splitting attention layers in SD-1.5 U-Net and SD-1.5 VAE
+Warn: The patch works well only if the input image has a width and height that are multiples of 128
+Original author: @tfernd Github: https://github.com/tfernd/HyperTile
+"""
+
+from __future__ import annotations
+
+import functools
+from dataclasses import dataclass
+from typing import Callable
+
+from functools import wraps, cache
+
+import math
+import torch.nn as nn
+import random
+
+from einops import rearrange
+
+
+@dataclass
+class HypertileParams:
+    depth = 0
+    layer_name = ""
+    tile_size: int = 0
+    swap_size: int = 0
+    aspect_ratio: float = 1.0
+    forward = None
+    enabled = False
+
+
+
+# TODO add SD-XL layers
+DEPTH_LAYERS = {
+    0: [
+        # SD 1.5 U-Net (diffusers)
+        "down_blocks.0.attentions.0.transformer_blocks.0.attn1",
+        "down_blocks.0.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.3.attentions.0.transformer_blocks.0.attn1",
+        "up_blocks.3.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.3.attentions.2.transformer_blocks.0.attn1",
+        # SD 1.5 U-Net (ldm)
+        "input_blocks.1.1.transformer_blocks.0.attn1",
+        "input_blocks.2.1.transformer_blocks.0.attn1",
+        "output_blocks.9.1.transformer_blocks.0.attn1",
+        "output_blocks.10.1.transformer_blocks.0.attn1",
+        "output_blocks.11.1.transformer_blocks.0.attn1",
+        # SD 1.5 VAE
+        "decoder.mid_block.attentions.0",
+        "decoder.mid.attn_1",
+    ],
+    1: [
+        # SD 1.5 U-Net (diffusers)
+        "down_blocks.1.attentions.0.transformer_blocks.0.attn1",
+        "down_blocks.1.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.2.attentions.0.transformer_blocks.0.attn1",
+        "up_blocks.2.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.2.attentions.2.transformer_blocks.0.attn1",
+        # SD 1.5 U-Net (ldm)
+        "input_blocks.4.1.transformer_blocks.0.attn1",
+        "input_blocks.5.1.transformer_blocks.0.attn1",
+        "output_blocks.6.1.transformer_blocks.0.attn1",
+        "output_blocks.7.1.transformer_blocks.0.attn1",
+        "output_blocks.8.1.transformer_blocks.0.attn1",
+    ],
+    2: [
+        # SD 1.5 U-Net (diffusers)
+        "down_blocks.2.attentions.0.transformer_blocks.0.attn1",
+        "down_blocks.2.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.1.attentions.0.transformer_blocks.0.attn1",
+        "up_blocks.1.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.1.attentions.2.transformer_blocks.0.attn1",
+        # SD 1.5 U-Net (ldm)
+        "input_blocks.7.1.transformer_blocks.0.attn1",
+        "input_blocks.8.1.transformer_blocks.0.attn1",
+        "output_blocks.3.1.transformer_blocks.0.attn1",
+        "output_blocks.4.1.transformer_blocks.0.attn1",
+        "output_blocks.5.1.transformer_blocks.0.attn1",
+    ],
+    3: [
+        # SD 1.5 U-Net (diffusers)
+        "mid_block.attentions.0.transformer_blocks.0.attn1",
+        # SD 1.5 U-Net (ldm)
+        "middle_block.1.transformer_blocks.0.attn1",
+    ],
+}
+# XL layers, thanks for GitHub@gel-crabs for the help
+DEPTH_LAYERS_XL = {
+    0: [
+        # SD 1.5 U-Net (diffusers)
+        "down_blocks.0.attentions.0.transformer_blocks.0.attn1",
+        "down_blocks.0.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.3.attentions.0.transformer_blocks.0.attn1",
+        "up_blocks.3.attentions.1.transformer_blocks.0.attn1",
+        "up_blocks.3.attentions.2.transformer_blocks.0.attn1",
+        # SD 1.5 U-Net (ldm)
+        "input_blocks.4.1.transformer_blocks.0.attn1",
+        "input_blocks.5.1.transformer_blocks.0.attn1",
+        "output_blocks.3.1.transformer_blocks.0.attn1",
+        "output_blocks.4.1.transformer_blocks.0.attn1",
+        "output_blocks.5.1.transformer_blocks.0.attn1",
+        # SD 1.5 VAE
+        "decoder.mid_block.attentions.0",
+        "decoder.mid.attn_1",
+    ],
+    1: [
+        # SD 1.5 U-Net (diffusers)
+        #"down_blocks.1.attentions.0.transformer_blocks.0.attn1",
+        #"down_blocks.1.attentions.1.transformer_blocks.0.attn1",
+        #"up_blocks.2.attentions.0.transformer_blocks.0.attn1",
+        #"up_blocks.2.attentions.1.transformer_blocks.0.attn1",
+        #"up_blocks.2.attentions.2.transformer_blocks.0.attn1",
+        # SD 1.5 U-Net (ldm)
+        "input_blocks.4.1.transformer_blocks.1.attn1",
+        "input_blocks.5.1.transformer_blocks.1.attn1",
+        "output_blocks.3.1.transformer_blocks.1.attn1",
+        "output_blocks.4.1.transformer_blocks.1.attn1",
+        "output_blocks.5.1.transformer_blocks.1.attn1",
+        "input_blocks.7.1.transformer_blocks.0.attn1",
+        "input_blocks.8.1.transformer_blocks.0.attn1",
+        "output_blocks.0.1.transformer_blocks.0.attn1",
+        "output_blocks.1.1.transformer_blocks.0.attn1",
+        "output_blocks.2.1.transformer_blocks.0.attn1",
+        "input_blocks.7.1.transformer_blocks.1.attn1",
+        "input_blocks.8.1.transformer_blocks.1.attn1",
+        "output_blocks.0.1.transformer_blocks.1.attn1",
+        "output_blocks.1.1.transformer_blocks.1.attn1",
+        "output_blocks.2.1.transformer_blocks.1.attn1",
+        "input_blocks.7.1.transformer_blocks.2.attn1",
+        "input_blocks.8.1.transformer_blocks.2.attn1",
+        "output_blocks.0.1.transformer_blocks.2.attn1",
+        "output_blocks.1.1.transformer_blocks.2.attn1",
+        "output_blocks.2.1.transformer_blocks.2.attn1",
+        "input_blocks.7.1.transformer_blocks.3.attn1",
+        "input_blocks.8.1.transformer_blocks.3.attn1",
+        "output_blocks.0.1.transformer_blocks.3.attn1",
+        "output_blocks.1.1.transformer_blocks.3.attn1",
+        "output_blocks.2.1.transformer_blocks.3.attn1",
+        "input_blocks.7.1.transformer_blocks.4.attn1",
+        "input_blocks.8.1.transformer_blocks.4.attn1",
+        "output_blocks.0.1.transformer_blocks.4.attn1",
+        "output_blocks.1.1.transformer_blocks.4.attn1",
+        "output_blocks.2.1.transformer_blocks.4.attn1",
+        "input_blocks.7.1.transformer_blocks.5.attn1",
+        "input_blocks.8.1.transformer_blocks.5.attn1",
+        "output_blocks.0.1.transformer_blocks.5.attn1",
+        "output_blocks.1.1.transformer_blocks.5.attn1",
+        "output_blocks.2.1.transformer_blocks.5.attn1",
+        "input_blocks.7.1.transformer_blocks.6.attn1",
+        "input_blocks.8.1.transformer_blocks.6.attn1",
+        "output_blocks.0.1.transformer_blocks.6.attn1",
+        "output_blocks.1.1.transformer_blocks.6.attn1",
+        "output_blocks.2.1.transformer_blocks.6.attn1",
+        "input_blocks.7.1.transformer_blocks.7.attn1",
+        "input_blocks.8.1.transformer_blocks.7.attn1",
+        "output_blocks.0.1.transformer_blocks.7.attn1",
+        "output_blocks.1.1.transformer_blocks.7.attn1",
+        "output_blocks.2.1.transformer_blocks.7.attn1",
+        "input_blocks.7.1.transformer_blocks.8.attn1",
+        "input_blocks.8.1.transformer_blocks.8.attn1",
+        "output_blocks.0.1.transformer_blocks.8.attn1",
+        "output_blocks.1.1.transformer_blocks.8.attn1",
+        "output_blocks.2.1.transformer_blocks.8.attn1",
+        "input_blocks.7.1.transformer_blocks.9.attn1",
+        "input_blocks.8.1.transformer_blocks.9.attn1",
+        "output_blocks.0.1.transformer_blocks.9.attn1",
+        "output_blocks.1.1.transformer_blocks.9.attn1",
+        "output_blocks.2.1.transformer_blocks.9.attn1",
+    ],
+    2: [
+        # SD 1.5 U-Net (diffusers)
+        "mid_block.attentions.0.transformer_blocks.0.attn1",
+        # SD 1.5 U-Net (ldm)
+        "middle_block.1.transformer_blocks.0.attn1",
+        "middle_block.1.transformer_blocks.1.attn1",
+        "middle_block.1.transformer_blocks.2.attn1",
+        "middle_block.1.transformer_blocks.3.attn1",
+        "middle_block.1.transformer_blocks.4.attn1",
+        "middle_block.1.transformer_blocks.5.attn1",
+        "middle_block.1.transformer_blocks.6.attn1",
+        "middle_block.1.transformer_blocks.7.attn1",
+        "middle_block.1.transformer_blocks.8.attn1",
+        "middle_block.1.transformer_blocks.9.attn1",
+    ],
+    3 : [] # TODO - separate layers for SD-XL
+}
+
+
+RNG_INSTANCE = random.Random()
+
+
+def random_divisor(value: int, min_value: int, /, max_options: int = 1) -> int:
+    """
+    Returns a random divisor of value that
+        x * min_value <= value
+    if max_options is 1, the behavior is deterministic
+    """
+    min_value = min(min_value, value)
+
+    # All big divisors of value (inclusive)
+    divisors = [i for i in range(min_value, value + 1) if value % i == 0] # divisors in small -> big order
+
+    ns = [value // i for i in divisors[:max_options]]  # has at least 1 element # big -> small order
+
+    idx = RNG_INSTANCE.randint(0, len(ns) - 1)
+
+    return ns[idx]
+
+
+def set_hypertile_seed(seed: int) -> None:
+    RNG_INSTANCE.seed(seed)
+
+
+@functools.cache
+def largest_tile_size_available(width: int, height: int) -> int:
+    """
+    Calculates the largest tile size available for a given width and height
+    Tile size is always a power of 2
+    """
+    gcd = math.gcd(width, height)
+    largest_tile_size_available = 1
+    while gcd % (largest_tile_size_available * 2) == 0:
+        largest_tile_size_available *= 2
+    return largest_tile_size_available
+
+
+def iterative_closest_divisors(hw:int, aspect_ratio:float) -> tuple[int, int]:
+    """
+    Finds h and w such that h*w = hw and h/w = aspect_ratio
+    We check all possible divisors of hw and return the closest to the aspect ratio
+    """
+    divisors = [i for i in range(2, hw + 1) if hw % i == 0] # all divisors of hw
+    pairs = [(i, hw // i) for i in divisors] # all pairs of divisors of hw
+    ratios = [w/h for h, w in pairs] # all ratios of pairs of divisors of hw
+    closest_ratio = min(ratios, key=lambda x: abs(x - aspect_ratio)) # closest ratio to aspect_ratio
+    closest_pair = pairs[ratios.index(closest_ratio)] # closest pair of divisors to aspect_ratio
+    return closest_pair
+
+
+@cache
+def find_hw_candidates(hw:int, aspect_ratio:float) -> tuple[int, int]:
+    """
+    Finds h and w such that h*w = hw and h/w = aspect_ratio
+    """
+    h, w = round(math.sqrt(hw * aspect_ratio)), round(math.sqrt(hw / aspect_ratio))
+    # find h and w such that h*w = hw and h/w = aspect_ratio
+    if h * w != hw:
+        w_candidate = hw / h
+        # check if w is an integer
+        if not w_candidate.is_integer():
+            h_candidate = hw / w
+            # check if h is an integer
+            if not h_candidate.is_integer():
+                return iterative_closest_divisors(hw, aspect_ratio)
+            else:
+                h = int(h_candidate)
+        else:
+            w = int(w_candidate)
+    return h, w
+
+
+def self_attn_forward(params: HypertileParams, scale_depth=True) -> Callable:
+
+    @wraps(params.forward)
+    def wrapper(*args, **kwargs):
+        if not params.enabled:
+            return params.forward(*args, **kwargs)
+
+        latent_tile_size = max(128, params.tile_size) // 8
+        x = args[0]
+
+        # VAE
+        if x.ndim == 4:
+            b, c, h, w = x.shape
+
+            nh = random_divisor(h, latent_tile_size, params.swap_size)
+            nw = random_divisor(w, latent_tile_size, params.swap_size)
+
+            if nh * nw > 1:
+                x = rearrange(x, "b c (nh h) (nw w) -> (b nh nw) c h w", nh=nh, nw=nw)  # split into nh * nw tiles
+
+            out = params.forward(x, *args[1:], **kwargs)
+
+            if nh * nw > 1:
+                out = rearrange(out, "(b nh nw) c h w -> b c (nh h) (nw w)", nh=nh, nw=nw)
+
+        # U-Net
+        else:
+            hw: int = x.size(1)
+            h, w = find_hw_candidates(hw, params.aspect_ratio)
+            assert h * w == hw, f"Invalid aspect ratio {params.aspect_ratio} for input of shape {x.shape}, hw={hw}, h={h}, w={w}"
+
+            factor = 2 ** params.depth if scale_depth else 1
+            nh = random_divisor(h, latent_tile_size * factor, params.swap_size)
+            nw = random_divisor(w, latent_tile_size * factor, params.swap_size)
+
+            if nh * nw > 1:
+                x = rearrange(x, "b (nh h nw w) c -> (b nh nw) (h w) c", h=h // nh, w=w // nw, nh=nh, nw=nw)
+
+            out = params.forward(x, *args[1:], **kwargs)
+
+            if nh * nw > 1:
+                out = rearrange(out, "(b nh nw) hw c -> b nh nw hw c", nh=nh, nw=nw)
+                out = rearrange(out, "b nh nw (h w) c -> b (nh h nw w) c", h=h // nh, w=w // nw)
+
+        return out
+
+    return wrapper
+
+
+def hypertile_hook_model(model: nn.Module, width, height, *, enable=False, tile_size_max=128, swap_size=1, max_depth=3, is_sdxl=False):
+    hypertile_layers = getattr(model, "__webui_hypertile_layers", None)
+    if hypertile_layers is None:
+        if not enable:
+            return
+
+        hypertile_layers = {}
+        layers = DEPTH_LAYERS_XL if is_sdxl else DEPTH_LAYERS
+
+        for depth in range(4):
+            for layer_name, module in model.named_modules():
+                if any(layer_name.endswith(try_name) for try_name in layers[depth]):
+                    params = HypertileParams()
+                    module.__webui_hypertile_params = params
+                    params.forward = module.forward
+                    params.depth = depth
+                    params.layer_name = layer_name
+                    module.forward = self_attn_forward(params)
+
+                    hypertile_layers[layer_name] = 1
+
+        model.__webui_hypertile_layers = hypertile_layers
+
+    aspect_ratio = width / height
+    tile_size = min(largest_tile_size_available(width, height), tile_size_max)
+
+    for layer_name, module in model.named_modules():
+        if layer_name in hypertile_layers:
+            params = module.__webui_hypertile_params
+
+            params.tile_size = tile_size
+            params.swap_size = swap_size
+            params.aspect_ratio = aspect_ratio
+            params.enabled = enable and params.depth <= max_depth

extensions-builtin/hypertile/scripts/hypertile_script.py

@@ -0,0 +1,73 @@
+import hypertile
+from modules import scripts, script_callbacks, shared
+
+
+class ScriptHypertile(scripts.Script):
+    name = "Hypertile"
+
+    def title(self):
+        return self.name
+
+    def show(self, is_img2img):
+        return scripts.AlwaysVisible
+
+    def process(self, p, *args):
+        hypertile.set_hypertile_seed(p.all_seeds[0])
+
+        configure_hypertile(p.width, p.height, enable_unet=shared.opts.hypertile_enable_unet)
+
+    def before_hr(self, p, *args):
+        configure_hypertile(p.hr_upscale_to_x, p.hr_upscale_to_y, enable_unet=shared.opts.hypertile_enable_unet_secondpass or shared.opts.hypertile_enable_unet)
+
+
+def configure_hypertile(width, height, enable_unet=True):
+    hypertile.hypertile_hook_model(
+        shared.sd_model.first_stage_model,
+        width,
+        height,
+        swap_size=shared.opts.hypertile_swap_size_vae,
+        max_depth=shared.opts.hypertile_max_depth_vae,
+        tile_size_max=shared.opts.hypertile_max_tile_vae,
+        enable=shared.opts.hypertile_enable_vae,
+    )
+
+    hypertile.hypertile_hook_model(
+        shared.sd_model.model,
+        width,
+        height,
+        swap_size=shared.opts.hypertile_swap_size_unet,
+        max_depth=shared.opts.hypertile_max_depth_unet,
+        tile_size_max=shared.opts.hypertile_max_tile_unet,
+        enable=enable_unet,
+        is_sdxl=shared.sd_model.is_sdxl
+    )
+
+
+def on_ui_settings():
+    import gradio as gr
+
+    options = {
+        "hypertile_explanation": shared.OptionHTML("""
+    <a href='https://github.com/tfernd/HyperTile'>Hypertile</a> optimizes the self-attention layer within U-Net and VAE models,
+    resulting in a reduction in computation time ranging from 1 to 4 times. The larger the generated image is, the greater the
+    benefit.
+    """),
+
+        "hypertile_enable_unet": shared.OptionInfo(False, "Enable Hypertile U-Net").info("noticeable change in details of the generated picture; if enabled, overrides the setting below"),
+        "hypertile_enable_unet_secondpass": shared.OptionInfo(False, "Enable Hypertile U-Net for hires fix second pass"),
+        "hypertile_max_depth_unet": shared.OptionInfo(3, "Hypertile U-Net max depth", gr.Slider, {"minimum": 0, "maximum": 3, "step": 1}),
+        "hypertile_max_tile_unet": shared.OptionInfo(256, "Hypertile U-net max tile size", gr.Slider, {"minimum": 0, "maximum": 512, "step": 16}),
+        "hypertile_swap_size_unet": shared.OptionInfo(3, "Hypertile U-net swap size", gr.Slider, {"minimum": 0, "maximum": 6, "step": 1}),
+
+        "hypertile_enable_vae": shared.OptionInfo(False, "Enable Hypertile VAE").info("minimal change in the generated picture"),
+        "hypertile_max_depth_vae": shared.OptionInfo(3, "Hypertile VAE max depth", gr.Slider, {"minimum": 0, "maximum": 3, "step": 1}),
+        "hypertile_max_tile_vae": shared.OptionInfo(128, "Hypertile VAE max tile size", gr.Slider, {"minimum": 0, "maximum": 512, "step": 16}),
+        "hypertile_swap_size_vae": shared.OptionInfo(3, "Hypertile VAE swap size ", gr.Slider, {"minimum": 0, "maximum": 6, "step": 1}),
+    }
+
+    for name, opt in options.items():
+        opt.section = ('hypertile', "Hypertile")
+        shared.opts.add_option(name, opt)
+
+
+script_callbacks.on_ui_settings(on_ui_settings)

modules/errors.py

@@ -6,6 +6,21 @@ import traceback
 exception_records = []
 
 
+def format_traceback(tb):
+    return [[f"{x.filename}, line {x.lineno}, {x.name}", x.line] for x in traceback.extract_tb(tb)]
+
+
+def format_exception(e, tb):
+    return {"exception": str(e), "traceback": format_traceback(tb)}
+
+
+def get_exceptions():
+    try:
+        return list(reversed(exception_records))
+    except Exception as e:
+        return str(e)
+
+
 def record_exception():
     _, e, tb = sys.exc_info()
     if e is None:
@@ -14,8 +29,7 @@ def record_exception():
     if exception_records and exception_records[-1] == e:
         return
 
-    from modules import sysinfo
+    exception_records.append(format_exception(e, tb))
-    exception_records.append(sysinfo.format_exception(e, tb))
 
     if len(exception_records) > 5:
         exception_records.pop(0)

modules/extensions.py

@@ -1,11 +1,14 @@
+from __future__ import annotations
+
+import configparser
 import os
 import threading
+import re
 
 from modules import shared, errors, cache, scripts
 from modules.gitpython_hack import Repo
 from modules.paths_internal import extensions_dir, extensions_builtin_dir, script_path  # noqa: F401
 
-extensions = []
 
 os.makedirs(extensions_dir, exist_ok=True)
 
@@ -19,11 +22,55 @@ def active():
         return [x for x in extensions if x.enabled]
 
 
+class ExtensionMetadata:
+    filename = "metadata.ini"
+    config: configparser.ConfigParser
+    canonical_name: str
+    requires: list
+
+    def __init__(self, path, canonical_name):
+        self.config = configparser.ConfigParser()
+
+        filepath = os.path.join(path, self.filename)
+        if os.path.isfile(filepath):
+            try:
+                self.config.read(filepath)
+            except Exception:
+                errors.report(f"Error reading {self.filename} for extension {canonical_name}.", exc_info=True)
+
+        self.canonical_name = self.config.get("Extension", "Name", fallback=canonical_name)
+        self.canonical_name = canonical_name.lower().strip()
+
+        self.requires = self.get_script_requirements("Requires", "Extension")
+
+    def get_script_requirements(self, field, section, extra_section=None):
+        """reads a list of requirements from the config; field is the name of the field in the ini file,
+        like Requires or Before, and section is the name of the [section] in the ini file; additionally,
+        reads more requirements from [extra_section] if specified."""
+
+        x = self.config.get(section, field, fallback='')
+
+        if extra_section:
+            x = x + ', ' + self.config.get(extra_section, field, fallback='')
+
+        return self.parse_list(x.lower())
+
+    def parse_list(self, text):
+        """converts a line from config ("ext1 ext2, ext3  ") into a python list (["ext1", "ext2", "ext3"])"""
+
+        if not text:
+            return []
+
+        # both "," and " " are accepted as separator
+        return [x for x in re.split(r"[,\s]+", text.strip()) if x]
+
+
 class Extension:
     lock = threading.Lock()
     cached_fields = ['remote', 'commit_date', 'branch', 'commit_hash', 'version']
+    metadata: ExtensionMetadata
 
-    def __init__(self, name, path, enabled=True, is_builtin=False):
+    def __init__(self, name, path, enabled=True, is_builtin=False, metadata=None):
         self.name = name
         self.path = path
         self.enabled = enabled
@@ -36,6 +83,8 @@ class Extension:
         self.branch = None
         self.remote = None
         self.have_info_from_repo = False
+        self.metadata = metadata if metadata else ExtensionMetadata(self.path, name.lower())
+        self.canonical_name = metadata.canonical_name
 
     def to_dict(self):
         return {x: getattr(self, x) for x in self.cached_fields}
@@ -56,6 +105,7 @@ class Extension:
                 self.do_read_info_from_repo()
 
                 return self.to_dict()
+
         try:
             d = cache.cached_data_for_file('extensions-git', self.name, os.path.join(self.path, ".git"), read_from_repo)
             self.from_dict(d)
@@ -136,9 +186,6 @@ class Extension:
 def list_extensions():
     extensions.clear()
 
-    if not os.path.isdir(extensions_dir):
-        return
-
     if shared.cmd_opts.disable_all_extensions:
         print("*** \"--disable-all-extensions\" arg was used, will not load any extensions ***")
     elif shared.opts.disable_all_extensions == "all":
@@ -148,18 +195,43 @@ def list_extensions():
     elif shared.opts.disable_all_extensions == "extra":
         print("*** \"Disable all extensions\" option was set, will only load built-in extensions ***")
 
-    extension_paths = []
+    loaded_extensions = {}
-    for dirname in [extensions_dir, extensions_builtin_dir]:
+
+    # scan through extensions directory and load metadata
+    for dirname in [extensions_builtin_dir, extensions_dir]:
         if not os.path.isdir(dirname):
-            return
+            continue
 
         for extension_dirname in sorted(os.listdir(dirname)):
             path = os.path.join(dirname, extension_dirname)
             if not os.path.isdir(path):
                 continue
 
-            extension_paths.append((extension_dirname, path, dirname == extensions_builtin_dir))
+            canonical_name = extension_dirname
+            metadata = ExtensionMetadata(path, canonical_name)
 
-    for dirname, path, is_builtin in extension_paths:
+            # check for duplicated canonical names
-        extension = Extension(name=dirname, path=path, enabled=dirname not in shared.opts.disabled_extensions, is_builtin=is_builtin)
+            already_loaded_extension = loaded_extensions.get(metadata.canonical_name)
-        extensions.append(extension)
+            if already_loaded_extension is not None:
+                errors.report(f'Duplicate canonical name "{canonical_name}" found in extensions "{extension_dirname}" and "{already_loaded_extension.name}". Former will be discarded.', exc_info=False)
+                continue
+
+            is_builtin = dirname == extensions_builtin_dir
+            extension = Extension(name=extension_dirname, path=path, enabled=extension_dirname not in shared.opts.disabled_extensions, is_builtin=is_builtin, metadata=metadata)
+            extensions.append(extension)
+            loaded_extensions[canonical_name] = extension
+
+    # check for requirements
+    for extension in extensions:
+        for req in extension.metadata.requires:
+            required_extension = loaded_extensions.get(req)
+            if required_extension is None:
+                errors.report(f'Extension "{extension.name}" requires "{req}" which is not installed.', exc_info=False)
+                continue
+
+            if not extension.enabled:
+                errors.report(f'Extension "{extension.name}" requires "{required_extension.name}" which is disabled.', exc_info=False)
+                continue
+
+
+extensions: list[Extension] = []

modules/img2img.py

@@ -44,6 +44,8 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=Fal
     steps = p.steps
     override_settings = p.override_settings
     sd_model_checkpoint_override = get_closet_checkpoint_match(override_settings.get("sd_model_checkpoint", None))
+    batch_results = None
+    discard_further_results = False
     for i, image in enumerate(images):
         state.job = f"{i+1} out of {len(images)}"
         if state.skipped:
@@ -127,7 +129,21 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=Fal
 
         if proc is None:
             p.override_settings.pop('save_images_replace_action', None)
-            process_images(p)
+            proc = process_images(p)
+
+        if not discard_further_results and proc:
+            if batch_results:
+                batch_results.images.extend(proc.images)
+                batch_results.infotexts.extend(proc.infotexts)
+            else:
+                batch_results = proc
+
+            if 0 <= shared.opts.img2img_batch_show_results_limit < len(batch_results.images):
+                discard_further_results = True
+                batch_results.images = batch_results.images[:int(shared.opts.img2img_batch_show_results_limit)]
+                batch_results.infotexts = batch_results.infotexts[:int(shared.opts.img2img_batch_show_results_limit)]
+
+    return batch_results
 
 
 def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_styles, init_img, sketch, init_img_with_mask, inpaint_color_sketch, inpaint_color_sketch_orig, init_img_inpaint, init_mask_inpaint, steps: int, sampler_name: str, mask_blur: int, mask_alpha: float, inpainting_fill: int, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, denoising_strength: float, selected_scale_tab: int, height: int, width: int, scale_by: float, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, img2img_batch_use_png_info: bool, img2img_batch_png_info_props: list, img2img_batch_png_info_dir: str, request: gr.Request, *args):
@@ -212,10 +228,10 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s
     with closing(p):
         if is_batch:
             assert not shared.cmd_opts.hide_ui_dir_config, "Launched with --hide-ui-dir-config, batch img2img disabled"
+            processed = process_batch(p, img2img_batch_input_dir, img2img_batch_output_dir, img2img_batch_inpaint_mask_dir, args, to_scale=selected_scale_tab == 1, scale_by=scale_by, use_png_info=img2img_batch_use_png_info, png_info_props=img2img_batch_png_info_props, png_info_dir=img2img_batch_png_info_dir)
 
-            process_batch(p, img2img_batch_input_dir, img2img_batch_output_dir, img2img_batch_inpaint_mask_dir, args, to_scale=selected_scale_tab == 1, scale_by=scale_by, use_png_info=img2img_batch_use_png_info, png_info_props=img2img_batch_png_info_props, png_info_dir=img2img_batch_png_info_dir)
+            if processed is None:
-
+                processed = Processed(p, [], p.seed, "")
-            processed = Processed(p, [], p.seed, "")
         else:
             processed = modules.scripts.scripts_img2img.run(p, *args)
             if processed is None:

modules/launch_utils.py

@@ -441,7 +441,7 @@ def dump_sysinfo():
     import datetime
 
     text = sysinfo.get()
-    filename = f"sysinfo-{datetime.datetime.utcnow().strftime('%Y-%m-%d-%H-%M')}.txt"
+    filename = f"sysinfo-{datetime.datetime.utcnow().strftime('%Y-%m-%d-%H-%M')}.json"
 
     with open(filename, "w", encoding="utf8") as file:
         file.write(text)

modules/options.py

@@ -76,7 +76,7 @@ class Options:
 
     def __init__(self, data_labels: dict[str, OptionInfo], restricted_opts):
         self.data_labels = data_labels
-        self.data = {k: v.default for k, v in self.data_labels.items()}
+        self.data = {k: v.default for k, v in self.data_labels.items() if not v.do_not_save}
         self.restricted_opts = restricted_opts
 
     def __setattr__(self, key, value):
@@ -210,7 +210,7 @@ class Options:
 
     def add_option(self, key, info):
         self.data_labels[key] = info
-        if key not in self.data:
+        if key not in self.data and not info.do_not_save:
             self.data[key] = info.default
 
     def reorder(self):

modules/processing.py

@@ -799,7 +799,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
 
     infotexts = []
     output_images = []
-
     with torch.no_grad(), p.sd_model.ema_scope():
         with devices.autocast():
             p.init(p.all_prompts, p.all_seeds, p.all_subseeds)
@@ -873,7 +872,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
             else:
                 if opts.sd_vae_decode_method != 'Full':
                     p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method
-
                 x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True)
 
             x_samples_ddim = torch.stack(x_samples_ddim).float()
@@ -1147,6 +1145,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
 
         if not self.enable_hr:
             return samples
+        devices.torch_gc()
 
         if self.latent_scale_mode is None:
             decoded_samples = torch.stack(decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True)).to(dtype=torch.float32)
@@ -1156,8 +1155,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
         with sd_models.SkipWritingToConfig():
             sd_models.reload_model_weights(info=self.hr_checkpoint_info)
 
-        devices.torch_gc()
-
         return self.sample_hr_pass(samples, decoded_samples, seeds, subseeds, subseed_strength, prompts)
 
     def sample_hr_pass(self, samples, decoded_samples, seeds, subseeds, subseed_strength, prompts):
@@ -1165,7 +1162,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
             return samples
 
         self.is_hr_pass = True
-
         target_width = self.hr_upscale_to_x
         target_height = self.hr_upscale_to_y
 
@@ -1254,7 +1250,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
         decoded_samples = decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True)
 
         self.is_hr_pass = False
-
         return decoded_samples
 
     def close(self):

modules/scripts.py

@@ -311,20 +311,113 @@ scripts_data = []
 postprocessing_scripts_data = []
 ScriptClassData = namedtuple("ScriptClassData", ["script_class", "path", "basedir", "module"])
 
+def topological_sort(dependencies):
+    """Accepts a dictionary mapping name to its dependencies, returns a list of names ordered according to dependencies.
+    Ignores errors relating to missing dependeencies or circular dependencies
+    """
+
+    visited = {}
+    result = []
+
+    def inner(name):
+        visited[name] = True
+
+        for dep in dependencies.get(name, []):
+            if dep in dependencies and dep not in visited:
+                inner(dep)
+
+        result.append(name)
+
+    for depname in dependencies:
+        if depname not in visited:
+            inner(depname)
+
+    return result
+
+
+@dataclass
+class ScriptWithDependencies:
+    script_canonical_name: str
+    file: ScriptFile
+    requires: list
+    load_before: list
+    load_after: list
+
 
 def list_scripts(scriptdirname, extension, *, include_extensions=True):
-    scripts_list = []
+    scripts = {}
 
-    basedir = os.path.join(paths.script_path, scriptdirname)
+    loaded_extensions = {ext.canonical_name: ext for ext in extensions.active()}
-    if os.path.exists(basedir):
+    loaded_extensions_scripts = {ext.canonical_name: [] for ext in extensions.active()}
-        for filename in sorted(os.listdir(basedir)):
+
-            scripts_list.append(ScriptFile(paths.script_path, filename, os.path.join(basedir, filename)))
+    # build script dependency map
+    root_script_basedir = os.path.join(paths.script_path, scriptdirname)
+    if os.path.exists(root_script_basedir):
+        for filename in sorted(os.listdir(root_script_basedir)):
+            if not os.path.isfile(os.path.join(root_script_basedir, filename)):
+                continue
+
+            if os.path.splitext(filename)[1].lower() != extension:
+                continue
+
+            script_file = ScriptFile(paths.script_path, filename, os.path.join(root_script_basedir, filename))
+            scripts[filename] = ScriptWithDependencies(filename, script_file, [], [], [])
 
     if include_extensions:
         for ext in extensions.active():
-            scripts_list += ext.list_files(scriptdirname, extension)
+            extension_scripts_list = ext.list_files(scriptdirname, extension)
+            for extension_script in extension_scripts_list:
+                if not os.path.isfile(extension_script.path):
+                    continue
 
-    scripts_list = [x for x in scripts_list if os.path.splitext(x.path)[1].lower() == extension and os.path.isfile(x.path)]
+                script_canonical_name = ("builtin/" if ext.is_builtin else "") + ext.canonical_name + "/" + extension_script.filename
+                relative_path = scriptdirname + "/" + extension_script.filename
+
+                script = ScriptWithDependencies(
+                    script_canonical_name=script_canonical_name,
+                    file=extension_script,
+                    requires=ext.metadata.get_script_requirements("Requires", relative_path, scriptdirname),
+                    load_before=ext.metadata.get_script_requirements("Before", relative_path, scriptdirname),
+                    load_after=ext.metadata.get_script_requirements("After", relative_path, scriptdirname),
+                )
+
+                scripts[script_canonical_name] = script
+                loaded_extensions_scripts[ext.canonical_name].append(script)
+
+    for script_canonical_name, script in scripts.items():
+        # load before requires inverse dependency
+        # in this case, append the script name into the load_after list of the specified script
+        for load_before in script.load_before:
+            # if this requires an individual script to be loaded before
+            other_script = scripts.get(load_before)
+            if other_script:
+                other_script.load_after.append(script_canonical_name)
+
+            # if this requires an extension
+            other_extension_scripts = loaded_extensions_scripts.get(load_before)
+            if other_extension_scripts:
+                for other_script in other_extension_scripts:
+                    other_script.load_after.append(script_canonical_name)
+
+        # if After mentions an extension, remove it and instead add all of its scripts
+        for load_after in list(script.load_after):
+            if load_after not in scripts and load_after in loaded_extensions_scripts:
+                script.load_after.remove(load_after)
+
+                for other_script in loaded_extensions_scripts.get(load_after, []):
+                    script.load_after.append(other_script.script_canonical_name)
+
+    dependencies = {}
+
+    for script_canonical_name, script in scripts.items():
+        for required_script in script.requires:
+            if required_script not in scripts and required_script not in loaded_extensions:
+                errors.report(f'Script "{script_canonical_name}" requires "{required_script}" to be loaded, but it is not.', exc_info=False)
+
+        dependencies[script_canonical_name] = script.load_after
+
+    ordered_scripts = topological_sort(dependencies)
+    scripts_list = [scripts[script_canonical_name].file for script_canonical_name in ordered_scripts]
 
     return scripts_list
 
@@ -365,15 +458,9 @@ def load_scripts():
             elif issubclass(script_class, scripts_postprocessing.ScriptPostprocessing):
                 postprocessing_scripts_data.append(ScriptClassData(script_class, scriptfile.path, scriptfile.basedir, module))
 
-    def orderby(basedir):
+    # here the scripts_list is already ordered
-        # 1st webui, 2nd extensions-builtin, 3rd extensions
+    # processing_script is not considered though
-        priority = {os.path.join(paths.script_path, "extensions-builtin"):1, paths.script_path:0}
+    for scriptfile in scripts_list:
-        for key in priority:
-            if basedir.startswith(key):
-                return priority[key]
-        return 9999
-
-    for scriptfile in sorted(scripts_list, key=lambda x: [orderby(x.basedir), x]):
         try:
             if scriptfile.basedir != paths.script_path:
                 sys.path = [scriptfile.basedir] + sys.path

modules/sd_samplers_extra.py

@@ -60,7 +60,7 @@ def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=No
                 sigma_restart = get_sigmas_karras(restart_steps, sigmas[min_idx].item(), sigmas[max_idx].item(), device=sigmas.device)[:-1]
                 while restart_times > 0:
                     restart_times -= 1
-                    step_list.extend([(old_sigma, new_sigma) for (old_sigma, new_sigma) in zip(sigma_restart[:-1], sigma_restart[1:])])
+                    step_list.extend(zip(sigma_restart[:-1], sigma_restart[1:]))
 
     last_sigma = None
     for old_sigma, new_sigma in tqdm.tqdm(step_list, disable=disable):

modules/shared_options.py

@@ -189,6 +189,7 @@ options_templates.update(options_section(('img2img', "img2img"), {
     "img2img_inpaint_sketch_default_brush_color": OptionInfo("#ffffff", "Inpaint sketch initial brush color", ui_components.FormColorPicker, {}).info("default brush color of img2img inpaint sketch").needs_reload_ui(),
     "return_mask": OptionInfo(False, "For inpainting, include the greyscale mask in results for web"),
     "return_mask_composite": OptionInfo(False, "For inpainting, include masked composite in results for web"),
+    "img2img_batch_show_results_limit": OptionInfo(32, "Show the first N batch img2img results in UI", gr.Slider, {"minimum": -1, "maximum": 1000, "step": 1}).info('0: disable, -1: show all images. Too many images can cause lag'),
 }))
 
 options_templates.update(options_section(('optimizations', "Optimizations"), {

modules/sysinfo.py

@@ -1,7 +1,6 @@
 import json
 import os
 import sys
-import traceback
 
 import platform
 import hashlib
@@ -84,7 +83,7 @@ def get_dict():
         "Checksum": checksum_token,
         "Commandline": get_argv(),
         "Torch env info": get_torch_sysinfo(),
-        "Exceptions": get_exceptions(),
+        "Exceptions": errors.get_exceptions(),
         "CPU": {
             "model": platform.processor(),
             "count logical": psutil.cpu_count(logical=True),
@@ -104,21 +103,6 @@ def get_dict():
     return res
 
 
-def format_traceback(tb):
-    return [[f"{x.filename}, line {x.lineno}, {x.name}", x.line] for x in traceback.extract_tb(tb)]
-
-
-def format_exception(e, tb):
-    return {"exception": str(e), "traceback": format_traceback(tb)}
-
-
-def get_exceptions():
-    try:
-        return list(reversed(errors.exception_records))
-    except Exception as e:
-        return str(e)
-
-
 def get_environment():
     return {k: os.environ[k] for k in sorted(os.environ) if k in environment_whitelist}
 

modules/ui.py

@@ -635,12 +635,6 @@ def create_ui():
                                     scale_by.release(**on_change_args)
                                     button_update_resize_to.click(**on_change_args)
 
-                                    # the code below is meant to update the resolution label after the image in the image selection UI has changed.
-                                    # as it is now the event keeps firing continuously for inpaint edits, which ruins the page with constant requests.
-                                    # I assume this must be a gradio bug and for now we'll just do it for non-inpaint inputs.
-                                    for component in [init_img, sketch]:
-                                        component.change(fn=lambda: None, _js="updateImg2imgResizeToTextAfterChangingImage", inputs=[], outputs=[], show_progress=False)
-
                             tab_scale_to.select(fn=lambda: 0, inputs=[], outputs=[selected_scale_tab])
                             tab_scale_by.select(fn=lambda: 1, inputs=[], outputs=[selected_scale_tab])
 
@@ -701,6 +695,12 @@ def create_ui():
                     if category not in {"accordions"}:
                         scripts.scripts_img2img.setup_ui_for_section(category)
 
+            # the code below is meant to update the resolution label after the image in the image selection UI has changed.
+            # as it is now the event keeps firing continuously for inpaint edits, which ruins the page with constant requests.
+            # I assume this must be a gradio bug and for now we'll just do it for non-inpaint inputs.
+            for component in [init_img, sketch]:
+                component.change(fn=lambda: None, _js="updateImg2imgResizeToTextAfterChangingImage", inputs=[], outputs=[], show_progress=False)
+
             def select_img2img_tab(tab):
                 return gr.update(visible=tab in [2, 3, 4]), gr.update(visible=tab == 3),
 
@@ -1308,7 +1308,7 @@ def setup_ui_api(app):
         from fastapi.responses import PlainTextResponse
 
         text = sysinfo.get()
-        filename = f"sysinfo-{datetime.datetime.utcnow().strftime('%Y-%m-%d-%H-%M')}.txt"
+        filename = f"sysinfo-{datetime.datetime.utcnow().strftime('%Y-%m-%d-%H-%M')}.json"
 
         return PlainTextResponse(text, headers={'Content-Disposition': f'{"attachment" if attachment else "inline"}; filename="{filename}"'})
 

pyproject.toml

@@ -16,6 +16,7 @@ exclude = [
 
 ignore = [
 	"E501", # Line too long
+	"E721", # Do not compare types, use `isinstance`
 	"E731", # Do not assign a `lambda` expression, use a `def`
 	
 	"I001", # Import block is un-sorted or un-formatted

webui.sh

@@ -89,7 +89,7 @@ delimiter="################################################################"
 
 printf "\n%s\n" "${delimiter}"
 printf "\e[1m\e[32mInstall script for stable-diffusion + Web UI\n"
-printf "\e[1m\e[34mTested on Debian 11 (Bullseye)\e[0m"
+printf "\e[1m\e[34mTested on Debian 11 (Bullseye), Fedora 34+ and openSUSE Leap 15.4 or newer.\e[0m"
 printf "\n%s\n" "${delimiter}"
 
 # Do not run as root
@@ -223,7 +223,7 @@ fi
 # Try using TCMalloc on Linux
 prepare_tcmalloc() {
     if [[ "${OSTYPE}" == "linux"* ]] && [[ -z "${NO_TCMALLOC}" ]] && [[ -z "${LD_PRELOAD}" ]]; then
-        TCMALLOC="$(PATH=/usr/sbin:$PATH ldconfig -p | grep -Po "libtcmalloc(_minimal|)\.so\.\d" | head -n 1)"
+        TCMALLOC="$(PATH=/sbin:$PATH ldconfig -p | grep -Po "libtcmalloc(_minimal|)\.so\.\d" | head -n 1)"
         if [[ ! -z "${TCMALLOC}" ]]; then
             echo "Using TCMalloc: ${TCMALLOC}"
             export LD_PRELOAD="${TCMALLOC}"