[QNN] Add SM8635 (Snapdragon 8s Gen 3) and SM8735 (Snapdragon 8s Gen 4) chipset support#19216
Draft
[QNN] Add SM8635 (Snapdragon 8s Gen 3) and SM8735 (Snapdragon 8s Gen 4) chipset support#19216
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/19216
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This PR needs a
|
…pset support Agent-Logs-Url: https://github.com/pytorch/executorch/sessions/3e9acd79-6c85-4c6c-860f-e364207f8e23 Co-authored-by: kirklandsign <107070759+kirklandsign@users.noreply.github.com>
Copilot
AI
changed the title
[WIP] Add support for SM8735 chip in QNN SDK
[QNN] Add SM8635 (Snapdragon 8s Gen 3) and SM8735 (Snapdragon 8s Gen 4) chipset support
Apr 30, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The Qualcomm 's' series chipsets (SM8635, SM8735) were missing from the QNN backend despite having compatible Hexagon HTP hardware supported by the QNN SDK.
Changes
qc_compiler_spec.fbs— AddSM8635 = 68andSM8735 = 85toQcomChipsetenumqc_schema.py— Add both chips toQcomChipsetIntEnum and_soc_info_table:SM8635: SoC Model 68, V73 arch, 8MB VTCM (same arch as SM8550)SM8735: SoC Model 85, V79 arch, 8MB VTCM (same arch as SM8750)utils.py— Updateget_soc_to_htp_arch_map(),get_soc_to_chipset_map(), and the compiler spec docstringREADME.md— Replace the incorrect note "chipset IDs are not accessible to community users" with a pointer to the QNN SDK's "Supported Snapdragon Devices" table where SoC model integers are documentedSoC model IDs verified from
QNN_SOC_MODEL_SM8635/QNN_SOC_MODEL_SM8735constants in the QNN SDK'sQnnTypes.h.Test plan
Validated that
QcomChipset.SM8635andQcomChipset.SM8735resolve correctly with the expected HTP arch and VTCM values, and that_soc_info_tablelookups succeed:Warning
Firewall rules blocked me from connecting to one or more addresses (expand for details)
I tried to connect to the following addresses, but was blocked by firewall rules:
aihub.qualcomm.com/home/REDACTED/work/_temp/ghcca-node/node/bin/node /home/REDACTED/work/_temp/ghcca-node/node/bin/node --enable-source-maps /home/REDACTED/work/_temp/copilot-developer-action-main/dist/index.js(dns block)android.googlesource.com/home/REDACTED/work/_temp/ghcca-node/node/bin/node /home/REDACTED/work/_temp/ghcca-node/node/bin/node --enable-source-maps /home/REDACTED/work/_temp/copilot-developer-action-main/dist/index.js(dns block)developer.qualcomm.com/home/REDACTED/work/_temp/ghcca-node/node/bin/node /home/REDACTED/work/_temp/ghcca-node/node/bin/node --enable-source-maps /home/REDACTED/work/_temp/copilot-developer-action-main/dist/index.js(dns block)discuss.pytorch.org/home/REDACTED/work/_temp/ghcca-node/node/bin/node /home/REDACTED/work/_temp/ghcca-node/node/bin/node --enable-source-maps /home/REDACTED/work/_temp/copilot-developer-action-main/dist/index.js(dns block)docs.qualcomm.com/home/REDACTED/work/_temp/ghcca-node/node/bin/node /home/REDACTED/work/_temp/ghcca-node/node/bin/node --enable-source-maps /home/REDACTED/work/_temp/copilot-developer-action-main/dist/index.js(dns block)https://api.github.com/repos/quic/ai-hub-models/contents/import urllib.request
Check quic/ai-hub-models
try:
url = 'REDACTED'
r = urllib.request.urlopen(url, timeout=10)
content = r.read().decode()
for line i e/backends/ir/IrBackend.h e/ba�� e/backends/QnnProfiler.h e/backends/QnnContextCommon.h e/backends/htp/HtpDevicePlatformInfoConfig.h e/backends/htp/Hgrep e/backends/htp/H-n e/backends/htp/HSM8|Snapdragon|soc_model e/backends/htp/H/home/REDACTED/.local/lib/python3.12/site-packages/qai_hub/client.py e/ba�� e/backends/htp/HtpGraph.h e/backends/htp/HtpContext.h stable-x86_64-REDACTED-linux-gnu/bin/rustc committer.email` (http block)
Let's look at QNN examples in the AI Engine Direct SDK for newer chip refs
import urllib.request
try:
r = urllib.request.urlopen('REDACTED` (dns block)
Look for newer versions of QAIRT documentation
for version in ['2.26', '2.27', '2.28', '2.29', '2.30', '2.31', '2.32', '2.33', '2.34', '2.35']:
url = f'REDACTED` (dns block)