kernel/linux.git/tools/testing/cxl, branch v7.2-rc1

Merge branch 'for-7.2/cxl-type2-attach-region' into cxl-for-next

2026-06-12T20:47:53+00:00

cxl: Add dummy function for cxl_memdev_attach_region for !CONFIG_CXL_REGION cxl/region: Introduce devm_cxl_probe_mem() cxl/memdev: Introduce cxl_class_memdev_type cxl/memdev: Pin parents for entire memdev lifetime cxl/region: Resolve region deletion races cxl/region: Block region delete during region creation

cxl/region: Introduce devm_cxl_probe_mem()

2026-06-12T20:47:30+00:00

To date, platform firmware maps accelerator memory and accelerator drivers simply want an address range that they can map themselves. This typically results in a single region being auto-assembled upon registration of a memory device. Use the @attach mechanism of devm_cxl_add_memdev() parameter to retrieve that region while also adhering to CXL subsystem locking and lifetime rules. As part of adhering to current object lifetime rules, if the region or the CXL port topology is invalidated, the CXL core arranges for the accelertor driver to be detached as well. The locking and lifetime rules were validated with Dave's work-in-progress cxl-type-2 support for cxl_test. devm_cxl_add_classdev() supports the general memory expansion flow where region assembly is optional, dynamic, and user controlled. Cc: Alejandro Lucero Signed-off-by: Dan Williams Reviewed-by: Alejandro Lucero Tested-by: Alejandro Lucero Reviewed-by: Dave Jiang Link: https://patch.msgid.link/20260519210158.1499795-6-djbw@kernel.org Signed-off-by: Dave Jiang

tools/testing/cxl: Resolve auto-region decoder targets like real HW

2026-06-12T15:14:46+00:00

The mock auto-region created at module load wrote switch and host-bridge decoder target[] directly, in addition to target_map[]. Real HW programs only target_map[] and resolves target[] as dports enumerate, via update_decoder_targets(). Region replay already follows that ordering, the initial auto-region did not. Drop the direct target[] writes and call cxl_port_update_decoder_targets() so target[] is resolved the same way as real HW and region replay, exercising more of the auto-region driver path. This is inspired by the discussion [1] below: [1]: https://lore.kernel.org/all/20260521084806.28232-1-icheng@nvidia.com/ Suggested-by: Alison Schofield Signed-off-by: Richard Cheng Reviewed-by: Alison Schofield Reviewed-by: Dave Jiang Tested-by: Dave Jiang Link: https://patch.msgid.link/20260612011227.4220-1-icheng@nvidia.com Signed-off-by: Dave Jiang

cxl/test: Add check after kzalloc() memory in alloc_mock_res()

2026-06-12T01:08:34+00:00

alloc_mock_res() calls kzalloc() without checking the return value. Add scope based resource management to deal with the allocated memory cleanly. Reported-by: sashiko-bot Fixes: 67dcdd4d3b83 ("tools/testing/cxl: Introduce a mocked-up CXL port hierarchy") Reviewed-by: Alison Schofield Link: https://patch.msgid.link/20260611230305.197390-1-dave.jiang@intel.com Signed-off-by: Dave Jiang

cxl/test: Unregister cxl_acpi in cxl_test_init() error path

2026-06-12T01:08:02+00:00

In cxl_test_init(), Once cxl_mock_platform_device_add() succeeds, all error paths after needs to call platform_device_unregister() instead of platform_device_put() to clean up. Fixes: 67dcdd4d3b83 ("tools/testing/cxl: Introduce a mocked-up CXL port hierarchy") Reported-by: sashiko-bot Reviewed-by: Alison Schofield Link: https://patch.msgid.link/20260611230355.198912-1-dave.jiang@intel.com Signed-off-by: Dave Jiang

cxl/test: Zero out LSA backing memory to avoid leaking to user

2026-06-10T18:20:04+00:00

Memory through vmalloc() is not zeroed out. When this memory is copied into output payload, it leaks memory content to user. Use vzalloc() instead to zero out the memory. Suggested-by: sashiko-bot Link: https://lore.kernel.org/linux-cxl/20260605173146.2B9A31F00893@smtp.kernel.org/ Fixes: 7d3eb23c4ccf ("tools/testing/cxl: Introduce a mock memory device + driver") Reviewed-by: Alison Schofield Link: https://patch.msgid.link/20260605184426.4070913-1-dave.jiang@intel.com Signed-off-by: Dave Jiang

cxl/test: Fix integer overflow in mock LSA bounds checks

2026-06-10T18:19:10+00:00

Pre-existing issue discovered by sashiko-bot. mock_get_lsa() and mock_set_lsa() validate the requested LSA range with "offset + length > LSA_SIZE". Both offset and length are u32 and, in mock_get_lsa(), both are taken directly from the user-supplied payload. The addition is evaluated modulo 2^32, so a large offset combined with a small length wraps around and passes the check. Rewrite the checks to first bound offset, then compare length against the remaining LSA size. Suggested-by: sashiko-bot Fixes: 7d3eb23c4ccf ("tools/testing/cxl: Introduce a mock memory device + driver") Link: https://lore.kernel.org/linux-cxl/20260605143748.235271F00893@smtp.kernel.org/ Assisted-by: Claude:claude-opus-4-8 Reviewed-by: Alison Schofield Signed-off-by: Dave Jiang

cxl/test: Verify cmd->size_in before accessing payload

2026-06-10T17:50:35+00:00

Several mock mailbox handlers access input payload fields before verifying that cmd->size_in is large enough for the corresponding structure. To ensure invalid commands are rejected before any payload data is consumed, add missing size checks and move existing checks ahead of the first payload field access. [dj: Updated commit log per Alison's comments. ] Fixes: 7d3eb23c4ccf ("tools/testing/cxl: Introduce a mock memory device + driver") Fixes: d1dca858f058 ("cxl/test: Add generic mock events") Fixes: f6448cb5f2f3 ("tools/testing/cxl: add firmware update emulation to CXL memdevs") Fixes: e77e9c107978 ("cxl/test: Add Get Feature support to cxl_test") Link: https://lore.kernel.org/linux-cxl/20260605143748.235271F00893@smtp.kernel.org/ Suggested-by: sashiko-bot Tested-by: Alison Schofield Reviewed-by: Alison Schofield Signed-off-by: Dave Jiang

cxl/test: Fix __fortify_panic

2026-06-04T15:44:00+00:00

Fix a runtime assertion in setup_xor_mapping(). Fortify complains that it is potentially overflowing the xormaps array per __counted_by(nr_maps). Quiet the false positive by initializing @nr_maps earlier. memcpy: detected buffer overflow: 32 byte write of buffer size 0 WARNING: lib/string_helpers.c:1036 at __fortify_report+0x4d/0xa0, CPU#8: modprobe/2728 Call Trace: __fortify_panic+0xd/0xf setup_xor_mapping+0x6c/0xa0 [cxl_translate] [ dj: Fixed up @nr_entries to @nr_maps in commit log. ] Fixes: 06377c54a133 ("cxl/test: Add cxl_translate module for address translation testing") Signed-off-by: Dan Williams Reviewed-by: Alison Schofield Reviewed-by: Dave Jiang Link: https://patch.msgid.link/20260519221204.1517773-3-djbw@kernel.org Signed-off-by: Dave Jiang

cxl/test: Enforce PMD alignment for volatile mock regions

2026-06-04T15:44:00+00:00

cxl_test allocates synthetic CFMWS HPA windows from a gen_pool with SZ_256M alignment. On arm64 with CONFIG_ARM64_64K_PAGES=y and CONFIG_PGTABLE_LEVELS=3, PMD_SIZE is 512M, so every CXL region carved from a volatile window inherits a non-PMD-aligned start, and cxl_dax_region_probe() -> alloc_dax_region() fails: """ cxl_dax_region dax_region1: probe with driver cxl_dax_region failed with error -12 """ Enforce that every volatile mock CFMWS is PMD-aligned in both start and size Reviewed-by: Dave Jiang Acked-by: Kai-Heng Feng Signed-off-by: Richard Cheng Reviewed-by: Alison Schofield Link: https://patch.msgid.link/20260527090332.30002-1-icheng@nvidia.com Signed-off-by: Dave Jiang