Open
Description
What's wrong?
When running certain Pytorch programs the following panic may happen:
time=2025-05-29T21:26:08.153+02:00 level=DEBUG msg="submitting traces after batch is full" component=gpuevent.Tracer len=100
time=2025-05-29T21:26:08.153+02:00 level=INFO msg="GPU Kernel Launch" component=gpuevent.Tracer event="{Flags:1 Pad:[0 0 0] PidInfo:{HostPid:31028 UserPid:31028 Ns:4026531836} KernFuncOff:130653966699584 GridX:9216 GridY:1 GridZ:1 BlockX:128 BlockY:1 BlockZ:1 Stream:0 Args:[2359296 2473901162499 0 0 0 1784695698202641508 532789473 1093924747349361480 130654924177800 130654924296008 0 510601728 0 4709711173051059528 130654924176192 25769803777] UstackSz:5 Ustack:[130655445276672 130653966649216 140726006302952 130653964778608 9223372050679694477 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]}"
time=2025-05-29T21:26:08.154+02:00 level=DEBUG msg="storing new metric label set" component=prom.Expirer labelValues=[generic]
time=2025-05-29T21:26:08.154+02:00 level=DEBUG msg="storing new metric label set" component=prom.Expirer labelValues="[void at::native::elementwise_kernel<128, 2, __nv_hdl_wrapper_t<false, false, false, __nv_dl_tag<void (*)(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&), &(at::native::gpu_kernel_impl_nocast<at::native::CUDAFunctor_add<float> >), 1u>, void (int), OffsetCalculator<3, unsigned int, false>, std::array<char*, 3ul>, at::native::CUDAFunctor_add<float> const> >(int, __nv_hdl_wrapper_t<false, false, false, __nv_dl_tag<void (*)(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&), &(at::native::gpu_kernel_impl_nocast<at::native::CUDAFunctor_add<float> >), 1u>, void (int), OffsetCalculator<3, unsigned int, false>, std::array<char*, 3ul>, at::native::CUDAFunctor_add<float> const>) grouille-2.nancy.grid5000.fr:0 ]"
time=2025-05-29T21:26:08.154+02:00 level=DEBUG msg="storing new metric label set" component=prom.Expirer labelValues=[]
panic: inconsistent label cardinality: expected 5 label values but got 0 in []string(nil)
goroutine 228 [running]:
github.com/grafana/beyla/v2/pkg/export/prom.(*Expirer[...]).WithLabelValues.func1()
/home/runner/work/beyla/beyla/pkg/export/prom/expirer.go:47 +0x1ea
github.com/grafana/beyla/v2/pkg/export/expire.(*ExpiryMap[...]).GetOrCreate(0x2ac02e0, {0x5c01380, 0x0, 0x0}, 0xc003f9fd60)
/home/runner/work/beyla/beyla/pkg/export/expire/expiry_map.go:57 +0x1b9
github.com/grafana/beyla/v2/pkg/export/prom.(*Expirer[...]).WithLabelValues(0xc006532000?, {0x5c01380?, 0x0?, 0xfffffc26ec029554?})
/home/runner/work/beyla/beyla/pkg/export/prom/expirer.go:41 +0x5d
github.com/grafana/beyla/v2/pkg/export/prom.(*metricsReporter).observe(0xc000814908, 0xc006532000)
/home/runner/work/beyla/beyla/pkg/export/prom/prom.go:844 +0xafb
github.com/grafana/beyla/v2/pkg/export/prom.(*metricsReporter).collectMetrics(0xc000814908, {0x0?, 0x0?})
/home/runner/work/beyla/beyla/pkg/export/prom/prom.go:757 +0xe7
github.com/grafana/beyla/v2/pkg/export/prom.(*metricsReporter).reportMetrics(0xc000814908, {0x2a9e238, 0xc00092a1e0})
/home/runner/work/beyla/beyla/pkg/export/prom/prom.go:746 +0x99
github.com/grafana/beyla/v2/pkg/pipe/swarm.(*Runner).Start.func2()
/home/runner/work/beyla/beyla/pkg/pipe/swarm/runner.go:61 +0x42
created by github.com/grafana/beyla/v2/pkg/pipe/swarm.(*Runner).Start in goroutine 171
/home/runner/work/beyla/beyla/pkg/pipe/swarm/runner.go:60 +0x11a
Steps to reproduce
Sample:
import argparse
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
from torch.optim.lr_scheduler import StepLR
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(1, 32, 3, 1)
self.conv2 = nn.Conv2d(32, 64, 3, 1)
self.dropout1 = nn.Dropout(0.25)
self.dropout2 = nn.Dropout(0.5)
self.fc1 = nn.Linear(9216, 128)
self.fc2 = nn.Linear(128, 10)
def forward(self, x):
x = self.conv1(x)
x = F.relu(x)
x = self.conv2(x)
x = F.relu(x)
x = F.max_pool2d(x, 2)
x = self.dropout1(x)
x = torch.flatten(x, 1)
x = self.fc1(x)
x = F.relu(x)
x = self.dropout2(x)
x = self.fc2(x)
output = F.log_softmax(x, dim=1)
return output
def train(args, model, device, train_loader, optimizer, epoch):
model.train()
for batch_idx, (data, target) in enumerate(train_loader):
data, target = data.to(device), target.to(device)
optimizer.zero_grad()
output = model(data)
loss = F.nll_loss(output, target)
loss.backward()
optimizer.step()
if batch_idx % args.log_interval == 0:
print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
epoch, batch_idx * len(data), len(train_loader.dataset),
100. * batch_idx / len(train_loader), loss.item()))
if args.dry_run:
break
def test(model, device, test_loader):
model.eval()
test_loss = 0
correct = 0
with torch.no_grad():
for data, target in test_loader:
data, target = data.to(device), target.to(device)
output = model(data)
test_loss += F.nll_loss(output, target, reduction='sum').item() # sum up batch loss
pred = output.argmax(dim=1, keepdim=True) # get the index of the max log-probability
correct += pred.eq(target.view_as(pred)).sum().item()
test_loss /= len(test_loader.dataset)
print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
test_loss, correct, len(test_loader.dataset),
100. * correct / len(test_loader.dataset)))
def main():
# Training settings
parser = argparse.ArgumentParser(description='PyTorch MNIST Example')
parser.add_argument('--batch-size', type=int, default=64, metavar='N',
help='input batch size for training (default: 64)')
parser.add_argument('--test-batch-size', type=int, default=1000, metavar='N',
help='input batch size for testing (default: 1000)')
parser.add_argument('--epochs', type=int, default=14, metavar='N',
help='number of epochs to train (default: 14)')
parser.add_argument('--lr', type=float, default=1.0, metavar='LR',
help='learning rate (default: 1.0)')
parser.add_argument('--gamma', type=float, default=0.7, metavar='M',
help='Learning rate step gamma (default: 0.7)')
parser.add_argument('--no-accel', action='store_true',
help='disables accelerator')
parser.add_argument('--dry-run', action='store_true',
help='quickly check a single pass')
parser.add_argument('--seed', type=int, default=1, metavar='S',
help='random seed (default: 1)')
parser.add_argument('--log-interval', type=int, default=10, metavar='N',
help='how many batches to wait before logging training status')
parser.add_argument('--save-model', action='store_true',
help='For Saving the current Model')
args = parser.parse_args()
use_accel = not args.no_accel and torch.accelerator.is_available()
torch.manual_seed(args.seed)
if use_accel:
device = torch.accelerator.current_accelerator()
else:
device = torch.device("cpu")
train_kwargs = {'batch_size': args.batch_size}
test_kwargs = {'batch_size': args.test_batch_size}
if use_accel:
accel_kwargs = {'num_workers': 1,
'pin_memory': True,
'shuffle': True}
train_kwargs.update(accel_kwargs)
test_kwargs.update(accel_kwargs)
transform=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
])
dataset1 = datasets.MNIST('../data', train=True, download=True,
transform=transform)
dataset2 = datasets.MNIST('../data', train=False,
transform=transform)
train_loader = torch.utils.data.DataLoader(dataset1,**train_kwargs)
test_loader = torch.utils.data.DataLoader(dataset2, **test_kwargs)
model = Net().to(device)
optimizer = optim.Adadelta(model.parameters(), lr=args.lr)
scheduler = StepLR(optimizer, step_size=1, gamma=args.gamma)
for epoch in range(1, args.epochs + 1):
train(args, model, device, train_loader, optimizer, epoch)
test(model, device, test_loader)
scheduler.step()
if args.save_model:
torch.save(model.state_dict(), "mnist_cnn.pt")
if __name__ == '__main__':
main()
System information
No response
Software version
v2.3.4-alloy-pre
Configuration
Logs