ghostscript: honor SOURCE_DATE_EPOCH=0 for reproducible build by bryan-minimal · Pull Request #272 · gominimal/pkgs

bryan-minimal · 2026-06-18T17:01:36Z

What

Makes the ghostscript package reproducible by fixing a SOURCE_DATE_EPOCH=0 handling bug in ghostscript's own build tools.

Why

Two from-scratch builds of ghostscript differed in exactly one place: an 8-byte value in usr/bin/gs — the gs_romfs_buildtime global (confirmed via the symbol table), a Unix build timestamp baked into the embedded ROM filesystem.

ghostscript's mkromfs and pack_ps tools try to honor SOURCE_DATE_EPOCH, but the logic is buggy:

if ((env_source_date_epoch = getenv("SOURCE_DATE_EPOCH"))) {
    buildtime = strtoul(env_source_date_epoch, NULL, 10);   // SDE="0" -> buildtime = 0
}
if (!buildtime)                 // 0 is falsy -> treats epoch 0 as "unset"
    buildtime = time(NULL);     // ...and falls back to wall-clock time

The build sandbox sets SOURCE_DATE_EPOCH=0 — which is the canonical reproducible-builds value, and exactly the value ghostscript mishandles: it reads 0, then if (!buildtime) treats it as unset and stamps wall-clock time() into the binary. (base/mkromfs.c:2614, base/pack_ps.c:344.)

Fix

Fall back to time() only when SOURCE_DATE_EPOCH is genuinely unset (!env_source_date_epoch), not when it parses to 0:

sed -i 's/if (!buildtime)/if (!env_source_date_epoch)/' base/mkromfs.c base/pack_ps.c

This is upstream-worthy — the bug bites any builder that sets SOURCE_DATE_EPOCH=0.

Verification

Build-twice-and-diff on aarch64 (--rebuild --no-fetch): with the fix, two from-scratch builds are byte-for-byte identical (repro-check diff, 196/196 files). Before the fix, usr/bin/gs diverged on the embedded gs_romfs_buildtime.

🤖 Generated with Claude Code

Summary by CodeRabbit

Chores
- Improved build reproducibility by refining how build timestamps are handled when using the SOURCE_DATE_EPOCH environment variable.
- Added safeguards to ensure the expected reproducibility-related logic is present, failing the build if it changes.

mkromfs/pack_ps read SOURCE_DATE_EPOCH but then do `if (!buildtime) buildtime = time(NULL)`, treating the sandbox's SOURCE_DATE_EPOCH=0 (epoch 0 is falsy) as unset and falling back to wall-clock time. That stamped a non-deterministic gs_romfs_buildtime into usr/bin/gs. Fall back to time() only when SOURCE_DATE_EPOCH is genuinely unset (base/mkromfs.c:2614, base/pack_ps.c:344). Verified on aarch64: two from-scratch builds are byte-identical (repro-check diff, 196/196 files). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

coderabbitai · 2026-06-18T17:01:50Z

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 175cb223-498d-414c-b8a4-2d5b0f84b1ef

📥 Commits

Reviewing files that changed from the base of the PR and between d515926 and ed17a06.

📒 Files selected for processing (1)

packages/ghostscript/build.sh

🚧 Files skipped from review as they are similar to previous changes (1)

packages/ghostscript/build.sh

📝 Walkthrough

Walkthrough

The Ghostscript build script gains a sed-based inline patch applied to base/mkromfs.c and base/pack_ps.c. It replaces the if (!buildtime) fallback condition with a check against env_source_date_epoch, so that a SOURCE_DATE_EPOCH value of 0 (epoch) no longer causes a nondeterministic wall-clock timestamp to be embedded.

Changes

Ghostscript Build Reproducibility Fix

Layer / File(s)	Summary
sed patch for buildtime guard `packages/ghostscript/build.sh`	Adds a comment and a `sed -i` invocation that rewrites the `if (!buildtime)` conditional in `base/mkromfs.c` and `base/pack_ps.c` to use `env_source_date_epoch`, fixing the zero-epoch edge case.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Suggested reviewers

twitchyliquid64

Poem

A rabbit checked the clock one day,
"Epoch zero? That's still okay!"
With a sed and a patch so neat,
The build timestamp stays concrete.
🐇 No more random ticks — hooray!

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title directly and concisely describes the main change: fixing ghostscript to properly handle SOURCE_DATE_EPOCH=0 for reproducible builds.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

Create stacked PR
Commit on current branch

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch bryan/ghostscript-reproducible

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

coderabbitai

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@packages/ghostscript/build.sh`:
- Around line 16-22: The reproducibility fix depends on SOURCE_DATE_EPOCH=0
being set by the caller, but if it is not explicitly set in the script, the
fallback to wall-clock time in mkromfs and pack_ps will cause non-deterministic
behavior. Add explicit export statements at the beginning of the build.sh script
to set both SOURCE_DATE_EPOCH=0 and PYTHONHASHSEED=0 before the sed command that
modifies base/mkromfs.c and base/pack_ps.c. This ensures deterministic behavior
is guaranteed regardless of the caller's environment.
- Line 21: The sed substitution on line 21 can silently fail to apply if the
Ghostscript source changes upstream, causing reproducibility to regress without
a build error. Add a guard check before the sed command to verify that the
pattern "if (!buildtime)" exists in both base/mkromfs.c and base/pack_ps.c files
using grep, and add a verification check after the sed command to confirm the
substitution was successful by checking that "if (!env_source_date_epoch)" now
exists in both files. If either guard check fails, the script should exit with
an error to prevent silent reproducibility failures.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

Push a commit to this branch (recommended)
Create a new PR with the fixes

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: d69ac19d-3d0f-430d-8ffd-5fc796b600ff

📥 Commits

Reviewing files that changed from the base of the PR and between 087287e and d515926.

📒 Files selected for processing (1)

packages/ghostscript/build.sh

coderabbitai · 2026-06-18T17:04:39Z

+# Reproducibility: mkromfs/pack_ps read SOURCE_DATE_EPOCH but then do
+# `if (!buildtime) buildtime = time(NULL)`, which treats the sandbox's
+# SOURCE_DATE_EPOCH=0 (epoch 0 is falsy) as "unset" and falls back to wall-clock
+# time -> non-deterministic gs_romfs_buildtime baked into the gs binary. Only
+# fall back to time() when SOURCE_DATE_EPOCH is genuinely unset.
+sed -i 's/if (!buildtime)/if (!env_source_date_epoch)/' base/mkromfs.c base/pack_ps.c
+


⚠️ Potential issue | 🟠 Major | ⚡ Quick win

Export reproducibility env vars in-script to guarantee deterministic behavior.

This fix depends on SOURCE_DATE_EPOCH=0; if it is not set by the caller, the code still falls back to wall-clock time. Set reproducibility env vars explicitly in this script.

Suggested addition

export LDFLAGS="-Wl,--build-id=none" export CXXFLAGS="${CFLAGS}" +export SOURCE_DATE_EPOCH=0 +export PYTHONHASHSEED=0 # Reproducibility: mkromfs/pack_ps read SOURCE_DATE_EPOCH but then do

As per coding guidelines, for embedded timestamps and Python-based build steps in packages/*/build.sh, export SOURCE_DATE_EPOCH=0 and PYTHONHASHSEED=0.

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

# Reproducibility: mkromfs/pack_ps read SOURCE_DATE_EPOCH but then do

# `if (!buildtime) buildtime = time(NULL)`, which treats the sandbox's

# SOURCE_DATE_EPOCH=0 (epoch 0 is falsy) as "unset" and falls back to wall-clock

# time -> non-deterministic gs_romfs_buildtime baked into the gs binary. Only

# fall back to time() when SOURCE_DATE_EPOCH is genuinely unset.

sed -i 's/if (!buildtime)/if (!env_source_date_epoch)/' base/mkromfs.c base/pack_ps.c

export LDFLAGS="-Wl,--build-id=none"

export CXXFLAGS="${CFLAGS}"

export SOURCE_DATE_EPOCH=0

export PYTHONHASHSEED=0

# Reproducibility: mkromfs/pack_ps read SOURCE_DATE_EPOCH but then do

# `if (!buildtime) buildtime = time(NULL)`, which treats the sandbox's

# SOURCE_DATE_EPOCH=0 (epoch 0 is falsy) as "unset" and falls back to wall-clock

# time -> non-deterministic gs_romfs_buildtime baked into the gs binary. Only

# fall back to time() when SOURCE_DATE_EPOCH is genuinely unset.

sed -i 's/if (!buildtime)/if (!env_source_date_epoch)/' base/mkromfs.c base/pack_ps.c

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@packages/ghostscript/build.sh` around lines 16 - 22, The reproducibility fix depends on SOURCE_DATE_EPOCH=0 being set by the caller, but if it is not explicitly set in the script, the fallback to wall-clock time in mkromfs and pack_ps will cause non-deterministic behavior. Add explicit export statements at the beginning of the build.sh script to set both SOURCE_DATE_EPOCH=0 and PYTHONHASHSEED=0 before the sed command that modifies base/mkromfs.c and base/pack_ps.c. This ensures deterministic behavior is guaranteed regardless of the caller's environment.

Source: Coding guidelines

coderabbitai · 2026-06-18T17:04:39Z

+# SOURCE_DATE_EPOCH=0 (epoch 0 is falsy) as "unset" and falls back to wall-clock
+# time -> non-deterministic gs_romfs_buildtime baked into the gs binary. Only
+# fall back to time() when SOURCE_DATE_EPOCH is genuinely unset.
+sed -i 's/if (!buildtime)/if (!env_source_date_epoch)/' base/mkromfs.c base/pack_ps.c


⚠️ Potential issue | 🟠 Major | ⚡ Quick win

Make the inline sed patch fail-fast on upstream drift.

Line 21 can silently no-op if Ghostscript source changes, and reproducibility regresses without a build failure. Add guard checks before/after substitution.

Suggested hardening

+# Fail fast if upstream changes and the patch no longer applies. +grep -q 'if (!buildtime)' base/mkromfs.c +grep -q 'if (!buildtime)' base/pack_ps.c sed -i 's/if (!buildtime)/if (!env_source_date_epoch)/' base/mkromfs.c base/pack_ps.c +! grep -q 'if (!buildtime)' base/mkromfs.c +! grep -q 'if (!buildtime)' base/pack_ps.c +grep -q 'if (!env_source_date_epoch)' base/mkromfs.c +grep -q 'if (!env_source_date_epoch)' base/pack_ps.c

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@packages/ghostscript/build.sh` at line 21, The sed substitution on line 21 can silently fail to apply if the Ghostscript source changes upstream, causing reproducibility to regress without a build error. Add a guard check before the sed command to verify that the pattern "if (!buildtime)" exists in both base/mkromfs.c and base/pack_ps.c files using grep, and add a verification check after the sed command to confirm the substitution was successful by checking that "if (!env_source_date_epoch)" now exists in both files. If either guard check fails, the script should exit with an error to prevent silent reproducibility failures.

…hanges) Adopts the hardening idea from the Edge Delta bot's #274 (cleanly — no whole-file reindentation, executable bit preserved): verify the 'if (!buildtime)' pattern exists in both files before sed-ing, so a future ghostscript that renames it fails the build instead of silently shipping a non-reproducible binary. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

coderabbitai Bot reviewed Jun 18, 2026

View reviewed changes

edge-delta Bot mentioned this pull request Jun 19, 2026

build hardening: filter ldd copies, guard sed patches, validate strtoll #274

Closed

6 tasks

mitodrummer approved these changes Jun 22, 2026

View reviewed changes

bryan-minimal added this pull request to the merge queue Jun 22, 2026

Merged via the queue into main with commit 5c8b5b7 Jun 22, 2026
4 checks passed

bryan-minimal deleted the bryan/ghostscript-reproducible branch June 22, 2026 17:00

Navigation Menu

Uh oh!

ghostscript: honor SOURCE_DATE_EPOCH=0 for reproducible build - #272

ghostscript: honor SOURCE_DATE_EPOCH=0 for reproducible build#272
bryan-minimal merged 2 commits into
mainfrom
bryan/ghostscript-reproducible

bryan-minimal commented Jun 18, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

coderabbitai Bot commented Jun 18, 2026 •

edited

Loading

Walkthrough

Changes

Estimated code review effort

Suggested reviewers

Poem

Uh oh!

coderabbitai Bot left a comment

Uh oh!

coderabbitai Bot Jun 18, 2026

Uh oh!

coderabbitai Bot Jun 18, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

bryan-minimal commented Jun 18, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What

Why

Fix

Verification

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Jun 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Estimated code review effort

Suggested reviewers

Poem

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

coderabbitai Bot Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

coderabbitai Bot Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

bryan-minimal commented Jun 18, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Jun 18, 2026 •

edited

Loading