Conflict with optimization padding cond: "RuntimeError: Sizes of tensors must match except in dimension 1"

[SLAPaper]

Generating image of 800*1280, batch size 1, and getting this error:

    Traceback (most recent call last):
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\call_queue.py", line 58, in f
        res = list(func(*args, **kwargs))
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\call_queue.py", line 37, in f
        res = func(*args, **kwargs)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\txt2img.py", line 62, in txt2img
        processed = processing.process_images(p)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\processing.py", line 677, in process_images
        res = process_images_inner(p)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\extensions\sd-webui-controlnet\scripts\batch_hijack.py", line 42, in processing_process_images_hijack
        return getattr(processing, '__controlnet_original_process_images_inner')(p, *args, **kwargs)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\processing.py", line 794, in process_images_inner
        samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\processing.py", line 1054, in sample
        samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x))
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\sd_samplers_kdiffusion.py", line 498, in sample
        samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args={
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\sd_samplers_kdiffusion.py", line 337, in launch_sampling
        return func()
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\sd_samplers_kdiffusion.py", line 498, in <lambda>
        samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args={
      File "C:\Users\SLAPaper\scoop\apps\miniconda3\current\envs\sd_webui\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
        return func(*args, **kwargs)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\repositories\k-diffusion\k_diffusion\sampling.py", line 128, in sample_euler
        denoised = model(x, sigma_hat * s_in, **extra_args)
      File "C:\Users\SLAPaper\scoop\apps\miniconda3\current\envs\sd_webui\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
        return forward_call(*args, **kwargs)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\sd_samplers_kdiffusion.py", line 205, in forward
        uncond = pad_cond(uncond, num_repeats, empty)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\sd_samplers_kdiffusion.py", line 108, in pad_cond
        tensor['crossattn'] = pad_cond(tensor['crossattn'], repeats, empty)
      File "C:\Users\SLAPaper\workspace\stable-diffusion-webui\modules\sd_samplers_kdiffusion.py", line 106, in pad_cond
        return torch.cat([tensor, empty.repeat((tensor.shape[0], repeats, 1))], axis=1)
    RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 1280 but got size 2048 for tensor number 1 in the list.

Update:
It seems that conflicting with optimization padding cond:

[wcde]

Padding cond has nothing to do with this, cross attn should always be same size. I could not reproduce this error, something else changes its size. Perhaps this is because of ControlNet?

\extensions\sd-webui-controlnet\scripts\batch_hijack.py", line 42, in processing_process_images_hijack

[SLAPaper]

Padding cond has nothing to do with this, cross attn should always be same size. I could not reproduce this error, something else changes its size. Perhaps this is because of ControlNet?

\extensions\sd-webui-controlnet\scripts\batch_hijack.py", line 42, in processing_process_images_hijack

but I didn't using ControlNet since there is no XL CN available now
I'll try disable CN plugin for testing

[wcde]

It looks like disabled ControlNet is trying to restore original state of model and this breaks everything.