-
Notifications
You must be signed in to change notification settings - Fork 493
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add ragged paged attention #8659
base: master
Are you sure you want to change the base?
Conversation
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
torch_xla/experimental/pallas_kernels/ragged_paged_attention_kernel.py
Outdated
Show resolved
Hide resolved
How is 65536 calculated? |
I found a ticket and someone uses it. I remember the number is the vmem limit on a TPU generation. |
ad2f87c
to
9e4b227
Compare
…_extreme_one_tokens_per_sequence.
…added runtime check.
9e4b227
to
7fe5071
Compare
|
cc onduty @lsy323 to assist with the CI test failure before we merge @vanbasten23 |
@@ -37,7 +37,8 @@ run_xla_hlo_debug python3 "$TEST_CDIR/scan/test_scan_debug.py" | |||
python3 "$TEST_CDIR/test_pallas.py" -v | |||
python3 "$TEST_CDIR/test_pallas_spmd.py" | |||
XLA_DISABLE_FUNCTIONALIZATION=1 python3 "$TEST_CDIR/test_pallas_spmd.py" | |||
python3 "$TEST_CDIR/test_tpu_paged_attention_kernel.py" | |||
python3 "$TEST_CDIR/test_multi_queries_paged_attention_kernel.py" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Isn't this line renamed? do you need it?
Test plan:
cc: @miladm