[AMD] AMD CI - temporarily change to mi325 (#26443)

This commit is contained in:
YC Yen-Ching Tseng
2026-05-27 13:55:03 +08:00
committed by GitHub
parent 21d0e74aff
commit f32ca1e0e5
+22 -22
View File
@@ -63,7 +63,7 @@ on:
description: 'AMD runner pool to dispatch GPU jobs to'
required: false
type: choice
default: mi300
default: mi325
options:
- mi300
- mi325
@@ -180,7 +180,7 @@ jobs:
# =============================================== sgl-kernel ====================================================
sgl-kernel-unit-test-amd:
name: ${{ format('sgl-kernel-unit-test-amd (linux-{0}-1gpu-sglang)', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
name: ${{ format('sgl-kernel-unit-test-amd (linux-{0}-1gpu-sglang)', inputs.runner_arch || 'mi325') }}
needs: [check-changes, call-gate]
if: |
always() && !cancelled() &&
@@ -192,7 +192,7 @@ jobs:
needs.check-changes.outputs.sgl_kernel == 'true'
)
)
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -238,7 +238,7 @@ jobs:
exit $failures
sgl-kernel-unit-test-2-gpu-amd:
name: ${{ format('sgl-kernel-unit-test-2-gpu-amd (linux-{0}-2gpu-sglang)', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
name: ${{ format('sgl-kernel-unit-test-2-gpu-amd (linux-{0}-2gpu-sglang)', inputs.runner_arch || 'mi325') }}
needs: [check-changes, call-gate]
if: |
always() && !cancelled() &&
@@ -250,7 +250,7 @@ jobs:
needs.check-changes.outputs.sgl_kernel == 'true'
)
)
runs-on: ${{ format('linux-{0}-2gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-2gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -289,7 +289,7 @@ jobs:
# =============================================== primary ====================================================
stage-a-test-1-gpu-small-amd:
name: ${{ format('stage-a-test-1-gpu-small-amd (linux-{0}-1gpu-sglang)', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
name: ${{ format('stage-a-test-1-gpu-small-amd (linux-{0}-1gpu-sglang)', inputs.runner_arch || 'mi325') }}
needs: [check-changes, call-gate]
if: |
always() && !cancelled() &&
@@ -301,7 +301,7 @@ jobs:
((needs.check-changes.outputs.main_package == 'true') || (needs.check-changes.outputs.sgl_kernel == 'true'))
)
)
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -326,7 +326,7 @@ jobs:
bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-a-test-1-gpu-small-amd ${{ needs.check-changes.outputs.continue_on_error == 'true' && '--continue-on-error' || '' }}
jit-kernel-unit-test-amd:
name: ${{ format('jit-kernel-unit-test-amd (linux-{0}-1gpu-sglang)', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
name: ${{ format('jit-kernel-unit-test-amd (linux-{0}-1gpu-sglang)', inputs.runner_arch || 'mi325') }}
needs: [check-changes, call-gate]
if: |
always() && !cancelled() &&
@@ -338,7 +338,7 @@ jobs:
needs.check-changes.outputs.jit_kernel == 'true'
)
)
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -389,7 +389,7 @@ jobs:
max-wait-minutes: '240'
stage-b-test-1-gpu-small-amd:
name: ${{ format('stage-b-test-1-gpu-small-amd (linux-{0}-1gpu-sglang, {1})', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325'), matrix.part) }}
name: ${{ format('stage-b-test-1-gpu-small-amd (linux-{0}-1gpu-sglang, {1})', inputs.runner_arch || 'mi325', matrix.part) }}
needs: [check-changes, wait-for-stage-a-amd]
if: |
always() &&
@@ -405,7 +405,7 @@ jobs:
fail-fast: false
matrix:
part: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -429,7 +429,7 @@ jobs:
bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-b-test-1-gpu-small-amd --auto-partition-id ${{ matrix.part }} --auto-partition-size 14 --timeout-per-file 2400 ${{ needs.check-changes.outputs.continue_on_error == 'true' && '--continue-on-error' || '' }}
stage-b-test-1-gpu-small-amd-nondeterministic:
name: ${{ format('stage-b-test-1-gpu-small-amd-nondeterministic (linux-{0}-1gpu-sglang)', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
name: ${{ format('stage-b-test-1-gpu-small-amd-nondeterministic (linux-{0}-1gpu-sglang)', inputs.runner_arch || 'mi325') }}
needs: [check-changes, wait-for-stage-a-amd]
if: |
always() &&
@@ -441,7 +441,7 @@ jobs:
((needs.check-changes.outputs.main_package == 'true') || (needs.check-changes.outputs.sgl_kernel == 'true'))
)
)
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -504,7 +504,7 @@ jobs:
bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-b-test-1-gpu-small-amd-mi35x ${{ needs.check-changes.outputs.continue_on_error == 'true' && '--continue-on-error' || '' }}
stage-b-test-1-gpu-large-amd:
name: ${{ format('stage-b-test-1-gpu-large-amd (linux-{0}-1gpu-sglang, {1})', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325'), matrix.part) }}
name: ${{ format('stage-b-test-1-gpu-large-amd (linux-{0}-1gpu-sglang, {1})', inputs.runner_arch || 'mi325', matrix.part) }}
needs: [check-changes, wait-for-stage-a-amd]
if: |
always() &&
@@ -520,7 +520,7 @@ jobs:
fail-fast: false
matrix:
part: [0, 1, 2]
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-1gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -544,7 +544,7 @@ jobs:
bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-b-test-1-gpu-large-amd --auto-partition-id ${{ matrix.part }} --auto-partition-size 3 --timeout-per-file 2700 ${{ needs.check-changes.outputs.continue_on_error == 'true' && '--continue-on-error' || '' }}
stage-b-test-2-gpu-large-amd:
name: ${{ format('stage-b-test-2-gpu-large-amd (linux-{0}-2gpu-sglang, {1})', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325'), matrix.part) }}
name: ${{ format('stage-b-test-2-gpu-large-amd (linux-{0}-2gpu-sglang, {1})', inputs.runner_arch || 'mi325', matrix.part) }}
needs: [check-changes, wait-for-stage-a-amd]
if: |
always() &&
@@ -560,7 +560,7 @@ jobs:
fail-fast: false
matrix:
part: [0, 1]
runs-on: ${{ format('linux-{0}-2gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-2gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -889,7 +889,7 @@ jobs:
max-wait-minutes: '480'
stage-c-test-4-gpu-amd:
name: ${{ format('stage-c-test-4-gpu-amd (linux-{0}-4gpu-sglang, {1})', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325'), matrix.part) }}
name: ${{ format('stage-c-test-4-gpu-amd (linux-{0}-4gpu-sglang, {1})', inputs.runner_arch || 'mi325', matrix.part) }}
needs: [check-changes, call-gate, wait-for-stage-b-amd]
if: |
always() &&
@@ -905,7 +905,7 @@ jobs:
fail-fast: false
matrix:
part: [0]
runs-on: ${{ format('linux-{0}-4gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-4gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4
@@ -945,7 +945,7 @@ jobs:
${{ needs.check-changes.outputs.continue_on_error == 'true' && '--continue-on-error' || '' }}
stage-c-test-large-8-gpu-amd:
name: ${{ format('stage-c-test-large-8-gpu-amd (linux-{0}-8gpu-sglang, {1})', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325'), matrix.part) }}
name: ${{ format('stage-c-test-large-8-gpu-amd (linux-{0}-8gpu-sglang, {1})', inputs.runner_arch || 'mi325', matrix.part) }}
needs: [check-changes, call-gate, wait-for-stage-b-amd]
if: |
always() &&
@@ -958,12 +958,12 @@ jobs:
)
)
env:
RUNNER_LABELS: ${{ format('linux-{0}-8gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
RUNNER_LABELS: ${{ format('linux-{0}-8gpu-sglang', inputs.runner_arch || 'mi325') }}
strategy:
fail-fast: false
matrix:
part: [0, 1, 2]
runs-on: ${{ format('linux-{0}-8gpu-sglang', inputs.runner_arch || (github.event_name == 'pull_request' && 'mi300' || 'mi325')) }}
runs-on: ${{ format('linux-{0}-8gpu-sglang', inputs.runner_arch || 'mi325') }}
steps:
- name: Checkout code
uses: actions/checkout@v4