Upstream the script thats used to generate patch for AST2700 by chander-nexthop · Pull Request #568 · sonic-net/sonic-linux-kernel

chander-nexthop · 2026-05-07T08:30:17Z

As discussed in the SONiC-BMC work group meeting yesterday I am submitting the script that I use to generate the patch file to add support for AST 2700 to the SONiC Linux Kernel. The script by default will create the patch based on the current top-of-tree in aspeed-master-v6.12. This can be overridded by running the script with the env var ASPEED_TAG, ex "ASPEED_TAG=v00.07.02 ./scripts/generate-aspeed-patch.sh"

… 2700 2. Script to generate the patch that adds support for Aspeed AST2700 SOC. The script will generate the patch in WORK_DIR/aspeed-ast2700-support-new.patch This needs to be moved to patches-sonic/aspeed-ast2700-support.patch. Signed-off-by: Chandrasekaran Swaminathan <chander@nexthop.ai>

mssonicbld · 2026-05-07T08:30:25Z

/azp run

azure-pipelines · 2026-05-07T08:30:36Z

Azure Pipelines successfully started running 1 pipeline(s).

paulmenzel · 2026-05-07T08:33:48Z

Thank you for sharing the script. Please make this two merge/pull requests.

Signed-off-by: Chandrasekaran Swaminathan <chander@nexthop.ai>

mssonicbld · 2026-05-07T08:38:45Z

/azp run

azure-pipelines · 2026-05-07T08:39:00Z

Azure Pipelines successfully started running 1 pipeline(s).

chander-nexthop · 2026-05-07T08:39:24Z

Thank you for sharing the script. Please make this two merge/pull requests.

Done.

saiarcot895 · 2026-05-12T02:39:35Z

+    else
+        echo "Warning: Patch file not found: $line"
+    fi
+done < <(head -228 "$KERNEL_DIR/patches-sonic/series")


Why the first 228 lines specifically?

Thank you @saiarcot895 for pointing it out . It a hardcoded cutoff for the pre aspeed section of the series file . it is replaced in the latest commit with an aspeed section skip so it adapts as patches are added or reordered.

saiarcot895 · 2026-05-12T02:40:14Z

+cat > "$WORK_DIR/smart_merge.py" << 'PYTHON_EOF'
+#!/usr/bin/env python3
+"""
+Smart merge for Kconfig/Makefile files.
+Uses a common ancestor approach: extract Aspeed additions from diff,
+add them to SONiC base at correct locations while preserving order.
+"""
+import sys
+import re
+import difflib
+
+def smart_merge(sonic_file, aspeed_file, target_file):
+    # Aspeed-related patterns
+    aspeed_patterns = [
+        re.compile(r'aspeed', re.IGNORECASE),
+        re.compile(r'ast2[567]00', re.IGNORECASE),
+        re.compile(r'ast1[78]00', re.IGNORECASE),
+        re.compile(r'AST2700_IRQ'),
+        re.compile(r'ARCH_ASPEED'),
+    ]
+
+    def is_aspeed_related(text):
+        return any(pattern.search(text) for pattern in aspeed_patterns)
+
+    # Read files
+    with open(sonic_file, 'r') as f:
+        sonic_lines = f.readlines()
+
+    with open(aspeed_file, 'r') as f:
+        aspeed_lines = f.readlines()
+
+    # Use difflib to get a proper sequence matcher
+    matcher = difflib.SequenceMatcher(None, sonic_lines, aspeed_lines)
+
+    result = []
+
+    for tag, i1, i2, j1, j2 in matcher.get_opcodes():
+        if tag == 'equal':
+            # Common lines - keep them
+            result.extend(sonic_lines[i1:i2])
+        elif tag == 'delete':
+            # Lines deleted in aspeed - KEEP THEM (preserve SONiC content)
+            result.extend(sonic_lines[i1:i2])
+        elif tag == 'insert':
+            # Lines added in aspeed - check if Aspeed-related
+            new_lines = aspeed_lines[j1:j2]
+            aspeed_related = any(is_aspeed_related(line) for line in new_lines)
+            if aspeed_related:
+                # Add these lines
+                result.extend(new_lines)
+            # Otherwise skip (non-Aspeed additions)
+        elif tag == 'replace':
+            # Lines changed - keep SONiC version, then add Aspeed additions if any
+            result.extend(sonic_lines[i1:i2])
+            new_lines = aspeed_lines[j1:j2]
+            aspeed_related = any(is_aspeed_related(line) for line in new_lines)
+            if aspeed_related:
+                # Also add the Aspeed replacements
+                result.extend(new_lines)
+
+    # Write result
+    with open(target_file, 'w') as f:
+        f.writelines(result)
+
+if __name__ == '__main__':
+    smart_merge(sys.argv[1], sys.argv[2], sys.argv[3])
+PYTHON_EOF


Suggestion for the future: split this into a separate file just for editing/review purposes.

it is taken care @saiarcot895 . extracted into scripts/smart_merge.py now. Thank you.

mssonicbld · 2026-05-12T15:21:33Z

/azp run

azure-pipelines · 2026-05-12T15:21:45Z

Azure Pipelines successfully started running 1 pipeline(s).

paulmenzel

700 lines is quite much. Why can’t nexthop not publish a git repository with the all the SONiC patches applied, and the Aspeed patches on top? (SONiC should also publish it’s own Linux kernel git archive with the patches applied.)

paulmenzel · 2026-05-12T17:22:34Z

+# Example, run with ASPEED_TAG=v00.07.02 ./scripts/generate-aspeed-patch.sh
+ASPEED_BRANCH="${ASPEED_TAG:-aspeed-master-v6.12}"
+ASPEED_REPO="https://github.com/AspeedTech-BMC/linux.git"
+WORK_DIR="/tmp/aspeed-patch-gen"


Please consider $TMPDIR.

Thank you @paulmenzel for the feedback. review comment incorporated .

paulmenzel · 2026-05-12T17:28:11Z

+OUTPUT_PATCH="$WORK_DIR/aspeed-ast2700-support-new.patch"
+
+# Read kernel version from Makefile
+KERNEL_VERSION=$(grep "^KERNEL_VERSION" "$KERNEL_DIR/Makefile" | cut -d= -f2 | tr -d ' ')


sed -n 's/^KERNEL_VERSION\s*=\s*//p' "$KERNEL_DIR/Makefile"

KERNEL_VERSION ?= 6.12.41

A literal = regex doesnt match ?=, so the variable would expand to empty and break the rest of the script. I generalized the pattern. Thank you for feedback .

paulmenzel · 2026-05-12T17:29:32Z

+fi
+mkdir -p "$SONIC_SRC"
+cd "$SONIC_SRC"
+wget -q "$SONIC_KERNEL_URL"


Why redownload it each time?

Tarball is now cached . Thank you.

paulmenzel · 2026-05-12T17:30:52Z

+wget -q "$SONIC_KERNEL_URL"
+tar -xf "linux_${KERNEL_VERSION}.orig.tar.xz" --strip-components=1
+rm "linux_${KERNEL_VERSION}.orig.tar.xz"
+echo "SONiC kernel source ready: $(du -sh $SONIC_SRC | cut -f1)"


Explicitly mention the size?

Explicitly mentioned the size now. Thank you.

- Updated patch series file to move ASPEED patches at the end - Move inline smart_merge Python out of the bash heredoc into scripts/smart_merge.py; bash invokes it via $SMART_MERGE. - Replace the hard-coded "head -228" cap with a break at the "###-> aspeed" marker so the patch-application loop stays correct as patches-sonic/series evolves. - Warn patches listed after "###-> aspeed-end" with a note that they may need regeneration if they touch files also modified by the regenerated aspeed patch. Signed-off-by: Chinmoy Dey <chinmoy@nexthop.ai>

mssonicbld · 2026-05-19T09:16:34Z

/azp run

azure-pipelines · 2026-05-19T09:16:41Z

Azure Pipelines will not run the associated pipelines, because the pull request was updated after the run command was issued. Review the pull request again and issue a new run command.

- WORK_DIR now honors ${TMPDIR:-/tmp}. - KERNEL_VERSION read via sed (handles ?= / := / += / =). - SONiC kernel tarball is cached across runs; only the extracted $SONIC_SRC tree is wiped and re-extracted each time. - progress lines now label the "size on disk" figure explicitly. Merge remote-tracking branch 'upstream/master' into chander-ast2700-patch-script Signed-off-by: Chinmoy Dey <chinmoy@nexthop.ai>

mssonicbld · 2026-05-19T11:37:59Z

/azp run

azure-pipelines · 2026-05-19T11:38:09Z

Azure Pipelines successfully started running 1 pipeline(s).

chinmoy-nexthop · 2026-05-27T11:31:45Z

All review comments have been incorporated.

chinmoy-nexthop · 2026-06-01T05:29:52Z

@saiarcot895 @paulmenzel , please let me know if you have any additional feedback on this PR. Otherwise, we can proceed with the next steps here.

saiarcot895 · 2026-06-01T06:22:40Z

+
+############################################################
+#
+# Let us keep Aspeed patches at the end


Is there a reason for keeping Aspeed patches at the end?

The Aspeed generated patch block is intentionally kept at the end to ensure that the base patch set is applied successfully before any Aspeed specific patches are processed. This avoids patch ordering issues and keeps the script behaviour deterministic as patches-sonic/series changes over time. Any new patch added to the patch stack should be validated to ensure it does not introduce conflicts with the Aspeed generated patches.

saiarcot895 · 2026-06-01T06:23:10Z

 0003-neighbor-Add-NTF_EXT_VALIDATED-flag-for-externally-v.patch

+###-> nvidia_aspeed_bmc-start
+###-> nvidia_aspeed_bmc-end


Nvidia's aspeed patch marker ordering has been swapped, which may cause impact to them.

This is just a marker at the moment, and there are no patches in this section yet. If its current placement is a concern, I can certainly move it to the end as well.

The main intent is to provide a clear location for any future Aspeed generated patches. Whenever a new patch is introduced that touches the same files as the Aspeed generated patches, we should ensure that any conflicts or interactions are reviewed and resolved appropriately. That is the primary expectation behind keeping this marker at the end.

paulmenzel · 2026-06-01T09:31:02Z

700 lines is quite much. Why can’t nexthop not publish a git repository with the all the SONiC patches applied, and the Aspeed patches on top? (SONiC should also publish it’s own Linux kernel git archive with the patches applied.)

I haven’t seen a reply to this comment.

chinmoy-nexthop · 2026-06-05T09:30:13Z

700 lines is quite much. Why can’t nexthop not publish a git repository with the all the SONiC patches applied, and the Aspeed patches on top? (SONiC should also publish it’s own Linux kernel git archive with the patches applied.)

I haven’t seen a reply to this comment.

Thats a fair question. The main reason for upstreaming the script is reproducibility. Rather than maintaining a separate rebased kernel tree, the script regenerates the AST2700 patch from the current SONiC kernel source and the selected Aspeed baseline. The recent changes were primarily focused on making the workflow more maintainable and less ad hoc. The goal of this PR is simply to upstream the generation method so that others can reliably reproduce the same patch set.

chander-nexthop added 2 commits May 7, 2026 08:19

Merge branch 'master' into chander-ast2700-patch-script

be17d98

chander-nexthop requested a review from a team as a code owner May 7, 2026 08:30

Revert the change to the options. Will raise a separate PR

28506e6

Signed-off-by: Chandrasekaran Swaminathan <chander@nexthop.ai>

chander-nexthop changed the title ~~Fix an unwanted option and upstream the script thats used to generate patch for AST2700~~ Upstream the script thats used to generate patch for AST2700 May 7, 2026

saiarcot895 reviewed May 12, 2026

View reviewed changes

paulmenzel suggested changes May 12, 2026

View reviewed changes

chinmoy-nexthop force-pushed the chander-ast2700-patch-script branch from df0c469 to 5e82787 Compare May 19, 2026 09:16

saiarcot895 reviewed Jun 1, 2026

View reviewed changes

Conversation

chander-nexthop commented May 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mssonicbld commented May 7, 2026

Uh oh!

azure-pipelines Bot commented May 7, 2026

Uh oh!

paulmenzel commented May 7, 2026

Uh oh!

mssonicbld commented May 7, 2026

Uh oh!

azure-pipelines Bot commented May 7, 2026

Uh oh!

chander-nexthop commented May 7, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mssonicbld commented May 12, 2026

Uh oh!

azure-pipelines Bot commented May 12, 2026

Uh oh!

paulmenzel left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

chinmoy-nexthop May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

chinmoy-nexthop May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mssonicbld commented May 19, 2026

Uh oh!

azure-pipelines Bot commented May 19, 2026

Uh oh!

mssonicbld commented May 19, 2026

Uh oh!

azure-pipelines Bot commented May 19, 2026

Uh oh!

chinmoy-nexthop commented May 27, 2026

Uh oh!

chinmoy-nexthop commented Jun 1, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

paulmenzel commented Jun 1, 2026

Uh oh!

chinmoy-nexthop commented Jun 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

chander-nexthop commented May 7, 2026 •

edited

Loading

chinmoy-nexthop May 19, 2026 •

edited

Loading

chinmoy-nexthop May 19, 2026 •

edited

Loading