kernel.git/drivers/base/regmap/regmap-kunit.c, branch linux-6.10.y

regmap: kunit: Fix memory leaks in gen_regmap() and gen_raw_regmap()

2024-08-14T13:34:05Z

[ Upstream commit c3820641da87442251e0c00b6874ef1022da8f58 ] - Use kunit_kcalloc() to allocate the defaults table so that it will be freed when the test case ends. - kfree() the buf and *data buffers on the error paths. - Use kunit_add_action_or_reset() instead of kunit_add_action() so that if it fails it will call regmap_exit(). Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240411103724.54063-1-rf@opensource.cirrus.com Signed-off-by: Mark Brown Signed-off-by: Sasha Levin

regmap: kunit: Fix array overflow in stride() test

2024-05-17T17:23:12Z

Force the max_register of the test regmap to be one register longer than the number of test registers, to prevent an array overflow in the test loop. The test defines num_reg_defaults = 6. With 6 registers and stride == 2 the valid register addresses would be 0, 2, 4, 6, 8, 10. However the loop checks attempting to access the odd address, so on the final register it accesses address 11, and it writes entry [11] of the read/written arrays. Originally this worked because the max_register of the regmap was hardcoded to be BLOCK_TEST_SIZE (== 12). commit 710915743d53 ("regmap: kunit: Run sparse cache tests at non-zero register addresses") introduced the ability to start the test address range from any address, which means adjusting the max_register. If max_register was not forced, it was calculated either from num_reg_defaults or BLOCK_TEST_SIZE. This correctly calculated that with num_reg_defaults == 6 and stride == 2 the final valid address is 10. So the read/written arrays are allocated to contain entries [0..10]. When stride attempted to access [11] it was overflowing the array. Signed-off-by: Richard Fitzgerald Fixes: 710915743d53 ("regmap: kunit: Run sparse cache tests at non-zero register addresses") Tested-by: Guenter Roeck Link: https://msgid.link/r/20240517144703.1200995-1-rf@opensource.cirrus.com Signed-off-by: Mark Brown

regmap: kunit: Fix an NULL vs IS_ERR() check

2024-04-15T10:54:57Z

The kunit_device_register() function returns error pointers, not NULL. Passing an error pointer to get_device() will lead to an Oops. Also get_device() returns the same device you passed to it. Fix it! ;) Fixes: 7b7982f14315 ("regmap: kunit: Create a struct device for the regmap") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/b32e80cf-b385-40cd-b8ec-77ec73e07530@moroto.mountain Signed-off-by: Mark Brown

regmap: kunit: Add test cases for regmap_read_bypassed()

2024-04-08T15:38:15Z

This adds test cases to prove that regmap_read_bypassed() reads the hardware value while the regmap is in cache-only. Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240408144600.230848-12-rf@opensource.cirrus.com Signed-off-by: Mark Brown

regmap: kunit: Add cache-drop test with multiple cache blocks

2024-04-08T15:38:14Z

Add a test case for dropping only some cache blocks and leaving others unchanged. The regmap is divided into 8 register ranges, and only 4 of these are written with values. This creates 4 non-contiguous ranges of registers with cached values. One whole range is then dropped, and part of another range. A cache sync is then performed to check that the correct registers were written, and the correct values were written to these registers. Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240408144600.230848-11-rf@opensource.cirrus.com Signed-off-by: Mark Brown

regmap: kunit: Replace a kmalloc/kfree() pair with KUnit-managed alloc

2024-04-08T15:38:13Z

Replace the kmalloc() and kfree() in raw_read_defaults() with a kunit_kmalloc() so that KUnit will free it automatically. Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240408144600.230848-10-rf@opensource.cirrus.com Signed-off-by: Mark Brown

regmap: kunit: Use a KUnit action to call regmap_exit()

2024-04-08T15:38:12Z

Registert a KUnit action handler to call regmap_exit() when a test terminates. This ensures that regmap_exit() will be called if a test function returns early or aborts. Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240408144600.230848-9-rf@opensource.cirrus.com Signed-off-by: Mark Brown

regmap: kunit: Add more cache-sync tests

2024-04-08T15:38:11Z

Extend the testing of cache-sync. - cache_sync() renamed cache_sync_marked_dirty() for clarity of what conditions it is testing. - cache_sync_defaults() renamed cache_sync_defaults_marked_dirty() for clarity. Added code to write the register back to its default value to check that a dirty sync doesn't write out the default value. - Added cache_sync_after_cache_only(). Tests syncing the cache without calling regcache_mark_dirty(). A register written while in cache-only should be written out by regcache_sync(). - Added cache_sync_default_after_cache_only. This is similar to cache_sync_after_cache_only(), but the register is changed to its default value while in cache-only. Because regcache_mark_dirty() was NOT called, regacache_sync() should write out the register. Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240408144600.230848-8-rf@opensource.cirrus.com Signed-off-by: Mark Brown

regmap: kunit: Add more cache-drop tests

2024-04-08T15:38:10Z

Extend the testing of cache-drop. - Added cache_drop_all_and_sync_marked_dirty(). If all registers are dropped from the cache a regcache_mark_dirty() followed by regcache_sync() should not write anything because the cache is empty. - Added cache_drop_all_and_sync_no_defaults(). This is similar to cache_drop_all_and_sync_marked_dirty() except that regcache_mark_dirty() is NOT called. All registers were dropped so regcache_sync() should not write anything. - Added cache_drop_all_and_sync_has_defaults(). This is the same as cache_drop_all_and_sync_no_defaults() except that the regmap has a table of default values. Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240408144600.230848-7-rf@opensource.cirrus.com Signed-off-by: Mark Brown

regmap: kunit: Run non-sparse cache tests at non-zero register addresses

2024-04-08T15:38:09Z

Change the tests parameterized by real_cache_types_list[] to test at some register addresses that are not 0. The cache_range_window_reg() test has hardcoded address assumptions that are not present in any other tests using real_cache_types_list[] table. So it has been given a separate parameter table, real_cache_types_only_list[], that preserves the original parameterization. Signed-off-by: Richard Fitzgerald Link: https://msgid.link/r/20240408144600.230848-6-rf@opensource.cirrus.com Signed-off-by: Mark Brown