[GC] Preserve workstation GC diagnostic data in heap dumps

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.

[max-charlamb]

Summary

• Append the workstation card_table pointer to GcDacVars at GC interface minor version 9.
• Use the versioned GC-DAC variable when returning workstation heap details, retaining the VM-global fallback when the field is unavailable.
• Explicitly enumerate the workstation card-table pointer storage.
• Preserve the complete workstation interesting_data_per_heap, compact_reasons_per_heap, expand_mechanisms_per_heap, and interesting_mechanism_bits_per_heap arrays.

Before this change, CDB /mw heap dumps did not preserve the workstation card-table pointer storage and preserved only the first element of each GC diagnostics array. After the card-table read was fixed, legacy DAC !dumpgcdata could load the dump and print global GC mechanisms, but failed when requesting per-heap interesting information:

[info per heap]
Error requesting interesting GC info

cDAC's GC contract also eagerly reads these complete arrays while constructing GCHeapData, so it encountered the same missing-memory problem. Explicitly preserving the card-table pointer and complete diagnostic-array ranges makes this data available to both the legacy DAC API used by !dumpgcdata and the cDAC GC heap reader.

Testing

• ./build.cmd clr -rc checked
• ./build.cmd clr+libs -rc release
• ./build.cmd clr+libs+host -rc release
• src/tests/build.cmd -GenerateLayoutOnly x64 Release
• Generated a CDB /mw dump from a workstation GC process after GC initialization.
• Verified the pre-change dump reports Error requesting interesting GC info from !dumpgcdata.
• Verified the post-change dump prints all 9 interesting-data points, 11 compacting reasons, 6 expansion mechanisms, and 2 mechanism bits through legacy DAC !dumpgcdata.

Note

This pull request description was generated with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

Tagging subscribers to this area: @anicka-net, @dotnet/gc
See info in area-owners.md if you want to be subscribed.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Updates CoreCLR GC/DAC plumbing to surface the workstation GC card-table pointer via versioned GcDacVars, enabling DAC consumers to read/enumerate it when available while preserving fallback behavior for older standalone GCs.

Changes:

• Bump GC interface minor version to 9.
• Add a new versioned GcDacVars entry for the WKS card_table pointer and populate it when supported.
• Update DAC heap-detail retrieval and WKS global memory enumeration to use the versioned card_table when available (fallback to g_card_table otherwise).

File summaries

File	Description
src/coreclr/gc/gcinterface.h	Bumps GC_INTERFACE_MINOR_VERSION to 9.
src/coreclr/gc/gcinterface.dacvars.def	Appends card_table to the versioned DAC vars definition (v5.9).
src/coreclr/gc/gc.cpp	Extends PopulateDacVars to populate the new v9 card_table DAC var in WKS builds.
src/coreclr/debug/daccess/request.cpp	Uses the versioned card_table when GC DAC minor >= 9 and enumerates its storage for WKS dumps.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟡 Changes recommended

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

src/coreclr/gc/gc.cpp:4531

• PopulateDacVars uses the incoming gcDacVars->minor_version_number to gate which fields are populated (v2/v4/v6/v8/v9), but then unconditionally overwrites minor_version_number with 9. In a standalone-GC mismatch (new GC + older runtime), this can report v9 even when v9 fields weren’t populated, which can lead the DAC to read/enumerate fields that aren’t actually present/populated.

 gcDacVars->major_version_number = 2;
gcDacVars->minor_version_number = 9;
if (v2)

• Files reviewed: 4/4 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🔵 Needs a closer look

Review details

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite

[steveisok]

I generated a dump w/ cdb and the runtime with this PR - all good.