Skip to content

[ALMIOPEN-1951] [miopen] Fix install RPATH for MIOpenDriver and CK backend libs - #7485

Merged
SreecharanGundaboluAMD merged 1 commit into
developfrom
users/SreecharanGundaboluAMD/fix_install_rpath
May 26, 2026
Merged

SreecharanGundaboluAMD merged 1 commit into
developfrom
users/SreecharanGundaboluAMD/fix_install_rpath

Conversation

@SreecharanGundaboluAMD

@SreecharanGundaboluAMD SreecharanGundaboluAMD commented May 14, 2026 •

Copy link
Copy Markdown
Contributor

Motivation

MIOpenDriver and the CK backend shared libraries (MIOpen_CK_*) failed to locate libMIOpen.so at runtime after a standard install unless the install lib/ directory was already present in LD_LIBRARY_PATH. Neither target had an INSTALL_RPATH set in CMake, so the dynamic linker had no embedded hint about where to find its dependencies. A freshly installed package should be self-contained and runnable without any environment variable configuration.

Technical Details

Two one-line fixes, each adding a set_target_properties(... INSTALL_RPATH ...) call immediately before the corresponding install() invocation:

  • projects/miopen/driver/CMakeLists.txt — sets INSTALL_RPATH "$ORIGIN/../lib" on MIOpenDriver. The binary lands in <prefix>/bin/, so $ORIGIN/../lib resolves to <prefix>/lib/ where libMIOpen.so lives.

  • projects/miopen/src/ck_impl/CMakeLists.txt — sets INSTALL_RPATH "$ORIGIN" on each MIOpen_CK_* shared library (inside the per-GPU-target foreach loop). These libraries are installed into <prefix>/lib/ alongside libMIOpen.so, so $ORIGIN is sufficient.

Both RPATHs use $ORIGIN-relative notation, keeping the install tree fully relocatable regardless of the absolute install prefix.

Testing

Manually verified that MIOpenDriver and the CK backend libraries correctly resolve libMIOpen.so via their embedded RPATH after installation, with no LD_LIBRARY_PATH set. Existing CI test suite is unaffected (RPATH is stripped from build-tree binaries and does not influence in-tree test runs).

Risk Assessment

Low risk : This is a targeted bug fix for a missing CMake property with no code logic changes, validated by CI as well as Jason on his perf runs, and carries essentially no regression risk.

Submission Checklist

@JonathanLichtnerAMD JonathanLichtnerAMD left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, but let's add a "Risk Assessment" category in the PR description

@jdcampbe jdcampbe left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tested on my branches. Fixes the build issues for me.

@SreecharanGundaboluAMD
SreecharanGundaboluAMD merged commit ef8a4cf into develop May 26, 2026
37 checks passed
@SreecharanGundaboluAMD
SreecharanGundaboluAMD deleted the users/SreecharanGundaboluAMD/fix_install_rpath branch May 26, 2026 15:16
JonathanLichtnerAMD added a commit that referenced this pull request Jul 15, 2026
JIRA ID: ALMIOPEN-2312

## Motivation

`projects/miopen/CHANGELOG.md`'s "MIOpen 3.5.2 for ROCm 7.14.0" section
only had 4 entries, but 68 commits touched `projects/miopen` between the
`therock-7.13` tag and `release/therock-7.14`. Auditing that commit
range (via `git log --oneline
therock-7.13..remotes/origin/release/therock-7.14 -- projects/miopen`)
found 18 additional user-facing changes — new gfx950/gfx1250
architecture support, SystemDB refreshes, AI-heuristic improvements,
several correctness/crash bug fixes, and OpenCL solver removals — that
were never documented. This PR backfills those entries so the 7.14.0
changelog accurately reflects what shipped.

## Technical Details

Added 24 lines to the existing "MIOpen 3.5.2 for ROCm 7.14.0" section of
`projects/miopen/CHANGELOG.md`, merging new bullets into the existing
`Added`/`Changed`/`Resolved Issues` sub-sections (no new entries were
duplicated or reworded) and adding a new `Removed` sub-section for
solver removals. No existing content was altered.

Each new bullet was traced to a specific commit + PR:

| Entry | Commit | PR |
|---|---|---|
| Added gfx950 (MI350X/MI355X) SystemDB | `60d8e1e5175` | #8081 |
| Added initial gfx1250 architecture support | `d3f057b2f22` | #7587 |
| Added JSON-formatted performance logs (`MIOPEN_PERFORMANCE_LOGS`) |
`ffbde871260` | #5736 |
| Refreshed gfx942 SystemDB | `faf754a9ef6` | #8174 |
| Refined 2D AI heuristics (Tunanet/KTN) for gfx942 & gfx950 |
`c24ad415884` | #7109 |
| Fixed thread-safety issue in AI-heuristic model caches | `cd957402309`
| #9149 (cherry-pick of #9131) |
| Fixed CK grouped-conv solvers not always registered in host |
`c971d54d7ae` | #9105 |
| Fixed legacy CK buffer resource descriptor (gfx1250) | `1dee011b8b0` |
#8758 |
| Removed stale gfx942 SystemDB naive-solver entries | `f42668f6316` |
#7028 |
| Fixed softmax NaN when `beta=0` | `fb8157c8512` | #8426 |
| Fixed BN backward spatial variant 1 vectorized memory access bug |
`1351a7e8b10` | #6765 |
| Fixed use-after-free of `kernel_id` in `InitInvokerFactoryNCHW` |
`394ebffb50d` | #7922 |
| Fixed SoftMarginLoss forward reduce-kernel kernarg corruption |
`1e65a31570f` | #8065, #8135 |
| Fixed build with `MIOPEN_ENABLE_AI_KERNEL_TUNING=OFF` | `e6288c3ca61`
| #6800 |
| Fixed install RPATH for MIOpenDriver / CK backend libs | `ef8a4cf2f93`
| #7485 |
| Fixed dangling lambda causing log-level 6 segfault | `190f9791737` |
#6953 |
| Enabled Windows long-path support for `addkernels` | `1f9d69724cf` |
#7387 |
| Removed OCL solver `ConvOclDirectFwd1x1` | `126c720cadd` | #6491 |
| Removed OCL solver `conv_ocl_dir2Dfwdgen` | `2d499ddf709` | #6633 |
| Removed OCL solvers `ConvOclBwdWrW1x1`, `ConvOclBwdWrW2` |
`b5d43a33dc6` | #6859 |
| Removed OCL solvers `OpTensorLeadingOnes`, `Op5dTensorGeneric` |
`8642c83b7de` | #7347 |

The remaining ~40 commits in the 68-commit range were deliberately
excluded (tests/CI/build/docs/internal refactors/1:1 OCL→HIP swaps with
no behavior change, per each commit's own stated scope) and are not
represented here.

## Risk Assessment

**Risk Level: 🟢 Low**

This is a documentation-only change to a single Markdown file. No code,
build configuration, tests, or public API surface is touched.

### Impacted Components

- `projects/miopen/CHANGELOG.md` only.

### Potential Side Effects

None expected. This does not change any runtime behavior; it only
documents changes that already shipped on `release/therock-7.14`.

### Mitigation Steps

None required beyond normal review — a reviewer familiar with the 7.14
release can spot-check a few entries against the linked commit hashes/PR
numbers for accuracy.

## Test Plan

Not applicable (no code changes). Verification was done by
cross-referencing every added bullet against a specific commit hash and
PR number pulled from `git log`, and by diffing the edited section
against `git show
remotes/origin/release/therock-7.14:projects/miopen/CHANGELOG.md` to
confirm no existing entry was duplicated or altered.

## Test Result

All 21 new bullets map 1:1 to a real commit/PR in the
`therock-7.13..release/therock-7.14` range for `projects/miopen`. No
pre-existing changelog text was modified — `git diff` shows a pure
addition (24 insertions, 0 deletions). Two borderline commits
(`a67834413e8` "Standalone build", `7c7cd5774a9` "enable warnings about
deprecated API/features") were deliberately left out pending author
judgment on whether they warrant an entry.
pschang-phy pushed a commit that referenced this pull request Jul 17, 2026
JIRA ID: ALMIOPEN-2312

## Motivation

`projects/miopen/CHANGELOG.md`'s "MIOpen 3.5.2 for ROCm 7.14.0" section
only had 4 entries, but 68 commits touched `projects/miopen` between the
`therock-7.13` tag and `release/therock-7.14`. Auditing that commit
range (via `git log --oneline
therock-7.13..remotes/origin/release/therock-7.14 -- projects/miopen`)
found 18 additional user-facing changes — new gfx950/gfx1250
architecture support, SystemDB refreshes, AI-heuristic improvements,
several correctness/crash bug fixes, and OpenCL solver removals — that
were never documented. This PR backfills those entries so the 7.14.0
changelog accurately reflects what shipped.

## Technical Details

Added 24 lines to the existing "MIOpen 3.5.2 for ROCm 7.14.0" section of
`projects/miopen/CHANGELOG.md`, merging new bullets into the existing
`Added`/`Changed`/`Resolved Issues` sub-sections (no new entries were
duplicated or reworded) and adding a new `Removed` sub-section for
solver removals. No existing content was altered.

Each new bullet was traced to a specific commit + PR:

| Entry | Commit | PR |
|---|---|---|
| Added gfx950 (MI350X/MI355X) SystemDB | `60d8e1e5175` | #8081 |
| Added initial gfx1250 architecture support | `d3f057b2f22` | #7587 |
| Added JSON-formatted performance logs (`MIOPEN_PERFORMANCE_LOGS`) |
`ffbde871260` | #5736 |
| Refreshed gfx942 SystemDB | `faf754a9ef6` | #8174 |
| Refined 2D AI heuristics (Tunanet/KTN) for gfx942 & gfx950 |
`c24ad415884` | #7109 |
| Fixed thread-safety issue in AI-heuristic model caches | `cd957402309`
| #9149 (cherry-pick of #9131) |
| Fixed CK grouped-conv solvers not always registered in host |
`c971d54d7ae` | #9105 |
| Fixed legacy CK buffer resource descriptor (gfx1250) | `1dee011b8b0` |
#8758 |
| Removed stale gfx942 SystemDB naive-solver entries | `f42668f6316` |
#7028 |
| Fixed softmax NaN when `beta=0` | `fb8157c8512` | #8426 |
| Fixed BN backward spatial variant 1 vectorized memory access bug |
`1351a7e8b10` | #6765 |
| Fixed use-after-free of `kernel_id` in `InitInvokerFactoryNCHW` |
`394ebffb50d` | #7922 |
| Fixed SoftMarginLoss forward reduce-kernel kernarg corruption |
`1e65a31570f` | #8065, #8135 |
| Fixed build with `MIOPEN_ENABLE_AI_KERNEL_TUNING=OFF` | `e6288c3ca61`
| #6800 |
| Fixed install RPATH for MIOpenDriver / CK backend libs | `ef8a4cf2f93`
| #7485 |
| Fixed dangling lambda causing log-level 6 segfault | `190f9791737` |
#6953 |
| Enabled Windows long-path support for `addkernels` | `1f9d69724cf` |
#7387 |
| Removed OCL solver `ConvOclDirectFwd1x1` | `126c720cadd` | #6491 |
| Removed OCL solver `conv_ocl_dir2Dfwdgen` | `2d499ddf709` | #6633 |
| Removed OCL solvers `ConvOclBwdWrW1x1`, `ConvOclBwdWrW2` |
`b5d43a33dc6` | #6859 |
| Removed OCL solvers `OpTensorLeadingOnes`, `Op5dTensorGeneric` |
`8642c83b7de` | #7347 |

The remaining ~40 commits in the 68-commit range were deliberately
excluded (tests/CI/build/docs/internal refactors/1:1 OCL→HIP swaps with
no behavior change, per each commit's own stated scope) and are not
represented here.

## Risk Assessment

**Risk Level: 🟢 Low**

This is a documentation-only change to a single Markdown file. No code,
build configuration, tests, or public API surface is touched.

### Impacted Components

- `projects/miopen/CHANGELOG.md` only.

### Potential Side Effects

None expected. This does not change any runtime behavior; it only
documents changes that already shipped on `release/therock-7.14`.

### Mitigation Steps

None required beyond normal review — a reviewer familiar with the 7.14
release can spot-check a few entries against the linked commit hashes/PR
numbers for accuracy.

## Test Plan

Not applicable (no code changes). Verification was done by
cross-referencing every added bullet against a specific commit hash and
PR number pulled from `git log`, and by diffing the edited section
against `git show
remotes/origin/release/therock-7.14:projects/miopen/CHANGELOG.md` to
confirm no existing entry was duplicated or altered.

## Test Result

All 21 new bullets map 1:1 to a real commit/PR in the
`therock-7.13..release/therock-7.14` range for `projects/miopen`. No
pre-existing changelog text was modified — `git diff` shows a pure
addition (24 insertions, 0 deletions). Two borderline commits
(`a67834413e8` "Standalone build", `7c7cd5774a9` "enable warnings about
deprecated API/features") were deliberately left out pending author
judgment on whether they warrant an entry.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants