GeezerVisonThe World Wide Web, restored to 800 pixels of glory. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Home|About|Gaming cabinet|Archives|How it works|Starter sites | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GeezerVison is showing therock-hud.amd.com — TheRock HUD Weekly digestTrends last 14 daysWhat shipped today —Loading today's release status…
Key failures (last 7d) —Loading…
Top recently merged PRs —Loading…
Top recently closed issues —Loading…
Recent activity —Loading…
Loading insights…
Failures over timeDaily job-level failures across all monitored repos, stacked by category. Click a category in the legend to toggle it. Loading… —
Filter hotspots:
Release Nightly —Loading…
CI Nightly —Loading…
Bump PR —Loading…
Other —Loading…
Most affected architecturesLoading…
Most affected workflowsLoading…
Slowest jobs —Top 15 (workflow, job pattern) pairs by median wall time across the selected window. Useful for identifying CI cost-optimization targets. Loading…
Never been green —Jobs that have been failing for 3+ consecutive days with zero successful runs in the last 14 days. These are the silent CI-cost burners — fix them, exclude them from the matrix, or accept them as known-bad and add to a quarantine list. Loading chronically-broken jobs…
How to read this table.
Each row is one nightly
multi_arch_release.yml run plus its dispatched downstreams
(test_artifacts, PyTorch wheels). Columns follow pipeline execution order across
12 lanes (ROCm Linux Builds → Linux Packaging → Publish → Tests → PyTorch Builds → PyTorch Tests → same for Windows).
Click ▸ on a column header to expand Math Libs / Test Artifacts / PyTorch Builds / PyTorch Tests into per-arch dots.
Click any row to open the full pipeline tree, or the ⬇ Downloads
button for install commands pinned to that exact build.
Date range:
From
→
To
ROCm version:
Release type:
Legend
PASS All jobs in this column succeeded
FAIL One or more jobs failed
RUNNING Stage still in progress
N/A No jobs for this stage on this version yet
Lanes: ROCm Linux Builds · ROCm Linux Packaging · Linux Publish · Linux Tests · Linux PyTorch Builds · Linux PyTorch Tests · Linux JAX Builds · Linux JAX Tests · ROCm Windows Builds · ROCm Windows Packaging · Windows Publish · Windows Tests · Windows PyTorch Builds · Windows PyTorch Tests — each lane is shaded differently (cool blues/greens for Linux, warmer ambers/reds for Windows) and starts with a thicker primary-tinted border.
LINUX / WINDOWS tags above each column header indicate the OS.
Cell badges show passed / total jobs for that column on that ROCm version (multiple workflow runs for the same nightly are summed together).
Build jobs are routed by job-name path (
Failed jobs per day in the Multi-Arch Release nightly pipeline.
One chart per OS; each plots failures per UTC day, bucketed the same way the
Multi-Arch Release
matrix routes jobs (Build = compile + packaging; Test = Test Artifacts). Two lines per chart:
■ Build failures ·
■ Test failures
(timed-out jobs count as failed; Publish + PyTorch-wheels phases are not shown).
Click a point to open that day's nightly in the Multi-Arch Release tab.
ROCm Linux — failures/day
—
ROCm Windows — failures/day
—
Status:
Author:
Search:
Showing 0 of 0 commits
Legend
✓ Passed
✗ Failed
⏳ Running
○ Skipped/N/A
Tip: Click on a commit row to expand and see detailed job status. Click any job status to go to GitHub.
Pre-merge (presubmit) CI quality for open pull requests. Companion to CI HUD, which shows post-merge CI on the default branch.
Status:
Author:
Hide drafts
Search:
Showing 0 of 0 PRs
Legend
✓ Passed
✗ Failed
⏳ Running
○ Skipped/N/A
Tip: Click a PR row to expand per-job presubmit status. The relation panel above pairs each recently merged PR's presubmit result with its post-merge result.
Legend
✓ All required checks passed
✗ Any required check failed
⏳ Required checks still running
— No data available
open PR not yet merged — Postsubmit shows "—" since there's no merge commit to run postsubmit CI on
Each required-check column: top icon = Presubmit, bottom icon = Postsubmit (hover an icon for details, click a column header to expand into separate Pre/Post sub-columns)
n/a Postsubmit not applicable — that check only runs on pull requests (e.g. Security Scan)
optional under a column name: shown as a useful signal but not required by the repo's branch protection (e.g. Multi-Arch CI on TheRock)
ASAN Build & Test Pipeline
TSAN Build & Test Pipeline
Legend
PASS Build/Test Successful
FAIL Build/Test Failed
SKIPPED Skipped
IN PROGRESS Currently Running
CANCELLED Cancelled
PENDING Waiting to Start
Time format: Run time +Queue time
One card per ROCm Nightly Tests run (newest first) — filter by Branch above.
Each card mirrors the workflow's run-time job summary: a row per GPU platform with
Tests Run /
Pass /
Fail /
Error
and Pass %; platforms not in that run show as Skipped — platform not available.
Runs are dispatched from
ROCm/rockrel's multi_arch_release.yml — click
Multi-Arch pipeline for that nightly's build tree, or source for the GitHub run.
Loading nightly test results…
Per-GPU-architecture trends across the ROCm Nightly Tests runs.
One line per arch: failing tests
(fail + error) and pass rate
per nightly. One point per ROCm build tested — a nightly that gets re-dispatched
after an infra failure collapses into a single point (newest attempt wins; the tooltip names the
superseded run). Gaps mean "no nightly", not zero failures.
An arch can cover several CI targets (
gfx942 → MI300A + MI308X); the per-arch numbers
sum them and the table below breaks them back out.
Click a point to open that run in the
Nightly Tests tab.
Failing tests per nightly
—
Pass rate per nightly
—
Loading nightly trends…
One card per ROCm Weekly Tests run — the long-running suites tagged
weekly in ROCm/rocm-tests, which are too slow for the nightly cadence.
Same layout as Nightly Tests:
a row per GPU platform with Tests Run /
Pass /
Fail /
Error, and a ▸ caret to
drill into per-test results. Because these run roughly weekly, the window defaults to 90 days.
Loading weekly test results…
Per-GPU-architecture trends across the ROCm Weekly Tests runs.
One line per arch: failing tests
(fail + error) and pass rate.
One point per ROCm build tested — a re-dispatched run collapses into a single point
(newest attempt wins; the tooltip names the superseded run). Gaps mean "no run", not zero failures.
Click a point to open that run in the
Weekly Tests tab.
Failing tests per run
—
Pass rate per run
—
Loading weekly trends…
Date range (scopes Bump Triage + History below):
From
→
To
Preset:
—
StatisticsIn-Progress Bump PRsOpen bump PRs awaiting merge — updating submodule versions in TheRock. Click ▸ on a tracked-submodule row to see what upstream PRs that bump pulls in.
Bump Triage bump-PR → issue → fix tracking (AI-assisted)Each bump-pr issue mapped to the bump PR that surfaced it, with an AI-synthesized root cause, commenter quotes, and the applied fix. Rows that share a Bump PR group together so you can see all issues caused by one bump. Scoped by the Date range below (kept in sync with the bar at the top of this tab).
Date range:
From
→
To
Preset:
—
Class:
Component:
Issue:
PR:
Reason coverage:
Search:
—
Bump History — HistoricalHistorical bump PRs (merged + closed + open) for tracked submodules, most recent first. Excludes
Date range:
From
→
To
Preset:
—
Status
Submodule Version StatusCurrent submodule versions from .gitmodules
Legend
OPEN PR is open
REVIEW Awaiting review
DRAFT Draft PR
CHANGES Changes requested
APPROVED Approved
MERGED Merged
10+ Significantly behind upstream
1-9 Slightly behind
0 Up to date
Status:
Labels:
Match ALL selected
Assignee:
Date Range (applies to everything):
From
→
To
Search:
Preset:
—
Issues Report
—
0
Total Issues
0
Open
0
Closed
0%
Close Rate (snapshot)
🏷️ Top Labels
Loading…
👥 Top Contributors
Loading…
Repository IssuesIssues from ROCm/TheRock repository
Showing 0-0 of 0 issues
Per page:
🧮 Issues Pivot — Open issues grouped by GPU architecture × framework version
Open
All
Platform:
Click any number to drill into those issues
IssuesLegend
OPEN Issue is open
CLOSED Issue is closed
Note: Use the multi-select filter to find issues by label. Data updates in real-time via webhooks.
Failure ClustersRecent failures grouped by job name + workflow + arch. Larger clusters mean a stage is broken across many runs — start here when triaging.
Flake DetectionJobs where a re-run produced a different result than the original attempt — the test flapped pass↔fail without code changes.
Legend
✓ Attempt passed
✗ Attempt failed
○ Cancelled
Tip: Click any "Logs" link to jump to the failing job on GitHub. A flake of
fail → pass usually means a transient issue (resource, network); pass → fail means a regression slipped past the first run.
Generate notes between two commits
Paste two commit SHAs (7+ hex chars). The lambda will look up their timestamps and pull every PR merged in that window across the selected repos. Tip: copy SHAs from the CI HUD tab. Share with
From SHA (older):
To SHA (newer):
Repos:
TheRock
rocm-systems
rocm-libraries
Release Notes
Compare on GitHub
Preview
Markdown
How PRs are categorized Rules apply in order — first match wins · 5 rules
1
GitHub label match
bug → Fixes
feature → Features
breaking → Breaking
2
Conventional commit prefix
feat: → Features
fix: → Fixes
chore: → Internal
3
Bracket prefix
[CI] → CI/Build
[docs] → Documentation
[bump] → Dependencies
4
Verb-start heuristic
Add … → Features
Fix … → Fixes
Refactor … → Refactor
5
Fallback
No rule matched → Other
14
Repo:
Merge throughputOpen PR backlog by ageTop contributorsAuthor × Category
Recently merged —Notes
Time-to-first-review is omitted: our ingestion lambda doesn't subscribe to the
pull_request_review webhook event. Adding it requires (1) updating the GitHub webhook subscription, (2) extending the lambda + ClickHouse schema, (3) backfilling. Worth doing if review responsiveness becomes a focus area.
Click contributor or label bars to drill into the Issues tab filtered by author or label.
Issues on this tab are scoped to the
Date range (applies to everything):
From
→
To
Preset:
—
0
Merged Bump PRs
0
Reverted Bumps
0
Issues (bump-pr)
0
Code Regressions
0
Infra Issues
0
Miscellaneous
Issues by ComponentClick a row to filter the issue list below.
Issues by CategoryLabel-driven buckets: Code Regression, Test Infra, Test Flaky, Infra Machine, Infra Timeout, CICD, Test Timeout, and Miscellaneous (no category label). Click a row to filter the issue list.
Reverts by Component bump-pr-revert · selected repo
Loading…
Issues — actionable backlog — all bump-pr issues.
Date range:
From
→
To
Preset:
—
Component:
Category:
Status:
Search:
Or click a row in the breakdown tables above.
Submodule Bump Trends TheRock only rocm-systems, rocm-libraries, and Extras (all other submodules)Each panel shows total merges + average per week, a scatter chart, a weekly breakdown table, and the full per-period PR list.
Date range:
From
→
To
Preset:
—
Submodule:
Three panels: rocm-systems, rocm-libraries, and Extras (the other 8 .gitmodules submodules). Pick one to focus.
Loading submodule trends…
Reverts — list bump-pr-revert · selected repo
Every PR labeled
Bump Merge Latency TheRock only time from PR open → merge · overlay of GPU runner events
Each dot is one bump PR — x-axis is when it was merged, y-axis is how many hours it sat open before merging.
Vertical dashed lines mark OSSCI GPU runner events from
Date range (synced):
From
→
To
Preset:
—
Legend & Methodology
Components are derived from the issue's
project: / component: / test: labels only (see _classifyIssueComponent in lambdas/query-proxy/handlers-misc.js). An issue whose label isn't in the component map is bucketed as Other.
Category is driven entirely by the issue's GitHub labels — no AI, no title/body regex. First match wins: code-regression → test-infra → test-flaky → infra-flaky → infra-machine → infra-timeout → cicd → test-timeout. Issues carrying none of these fall back to miscellaneous (see ISSUE_CATEGORY_LABEL_MAP in lambdas/query-proxy/handlers-misc.js).
Scope: Issues breakdowns and Reverted Bumps follow the Repo dropdown. Issues are gated by the bump-pr label (nightly triage bugs have their own Multi-Arch Nightly Issues tab). Merged Bump PRs are resolved to one of TheRock's 10 .gitmodules submodules (component bumps roll up to their parent submodule; third-party / sysdep version bumps and automation PRs are excluded) and always source from ROCm/TheRock — as do Submodule Bump Trends and Bump Merge Latency, which ignore the Repo dropdown. Reverted Bumps counts every PR carrying the bump-pr-revert label in the selected repo; the Reverts by Component breakdown groups them by ROCm component. All are narrowed by the selected period.
Scoped to GitHub issues labelled
Date range (applies to everything):
From
→
To
Preset:
—
Component:
Category:
GPU Arch:
Platform:
Priority:
Status:
Search:
Or click a row in the breakdown tables below.
0
Nightly Issues
0
Open Now (all time)
0
Closed in Period
0
Code Regressions
—
Top GPU Arch
—
Top Component
Issues by Component
Resolved from the issue's
Issues by CategoryLabel-driven buckets. Most nightly bugs carry no category label yet, so Miscellaneous dominates until triagers add one. Click a row to filter the issue list.
Issues by GPU Arch
Read from the issue's
Issues by Platform
From the
Issues by Priority
From GitHub's native Issue Fields
Issues — actionable backlog — all nightly triage issues in the selected period.
Legend & Methodology
Scope: issues labelled
release-nightly, in the monitored repos plus ROCm/rockrel. An issue is in the period if it was created or updated inside it.
Component and Category use the same classifiers as the Bump PR Insights tab (_classifyIssueComponent / _classifyIssueType in lambdas/query-proxy/handlers-misc.js). Category is label-only, so an issue with no category label is miscellaneous.
GPU Arch and Platform are multi-valued: an issue reading "Linux+Windows gfx900/gfx906" is counted once in each bucket, so those tables' totals are issue-instances and exceed the issue count. Both fall back to Unspecified when nothing can be resolved.
Open now ignores the selected period — it's the current all-time backlog for that bucket, so a long-open bug with no recent activity still shows up.
Loading failure context…
Trend—
—
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| You are visitor 01402503 — GeezerVison does not store the pages you look up. Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
