kernel.git/drivers/crypto/ccree/cc_driver.c, branch linux-5.10.y

crypto: ccree - Remove debugfs when platform_driver_register failed

2023-01-14T09:15:49Z

[ Upstream commit 4f1c596df706c9aca662b6c214fad84047ae2a97 ] When platform_driver_register failed, we need to remove debugfs, which will caused a resource leak, fix it. Failed logs as follows: [ 32.606488] debugfs: Directory 'ccree' with parent '/' already present! Fixes: 4c3f97276e15 ("crypto: ccree - introduce CryptoCell driver") Signed-off-by: Gaosheng Cui Signed-off-by: Herbert Xu Signed-off-by: Sasha Levin

crypto: ccree - Simplify with dev_err_probe()

2020-09-04T07:57:16Z

Common pattern of handling deferred probe can be simplified with dev_err_probe(). Less code and also it prints the error value. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Herbert Xu

crypto: ccree - use existing dev helper in init_cc_resources()

2020-02-22T01:25:46Z

Use the existing dev helper variable instead of plat_dev->dev. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu

crypto: ccree - spelling s/Crytpcell/Cryptocell/

2020-02-22T01:25:46Z

Fix a typo in a comment. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu

crypto: ccree - remove cc_pm_is_dev_suspended() wrapper

2020-02-22T01:25:44Z

If CONFIG_PM=y, cc_pm_is_dev_suspended() is just a wrapper around pm_runtime_suspended(). If CONFIG_PM=n, cc_pm_is_dev_suspended() a dummy that behaves exactly the same as the dummy for pm_runtime_suspended(). Hence remove cc_pm_is_dev_suspended(), and call pm_runtime_suspended() directly. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu

crypto: ccree - use of_device_get_match_data()

2020-02-22T01:25:44Z

If the driver is probed, it means a match was found in arm_ccree_dev_of_match[]. Hence we can just use the of_device_get_match_data() helper. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu

crypto: ccree - simplify Runtime PM handling

2020-02-22T01:25:44Z

Currently, a large part of the probe function runs before Runtime PM is enabled. As the driver manages the device's clock manually, this may work fine on some systems, but may break on platforms with a more complex power hierarchy. Fix this by moving the initialization of Runtime PM before the first register access (in cc_wait_for_reset_completion()), and putting the device to sleep only after the last access (in cc_set_ree_fips_status()). This allows to remove the pm_on flag, which was used to track manually if Runtime PM had been enabled or not. Remove the cc_pm_{init,go,fini}() wrappers, as they are called only once, and obscure operation. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu

crypto: ccree - clean up clock handling

2020-02-22T01:25:43Z

Use devm_clk_get_optional() instead of devm_clk_get() and explicit optional clock handling. As clk_prepare_enable() and clk_disable_unprepare() handle optional clocks fine, the cc_clk_on() and cc_clk_off() wrappers can be removed. While at it, use the new "%pe" format specifier to print error codes. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu

crypto: ccree - remove empty cc_sram_mgr_fini()

2020-02-22T01:25:43Z

cc_sram_mgr_fini() doesn't do anything, so it can just be removed. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu

crypto: ccree - drop duplicated error message on SRAM exhaustion

2020-02-22T01:25:43Z

When no SRAM can be allocated, cc_sram_alloc() already prints an error message. Hence there is no need to duplicate this in all callers. Signed-off-by: Geert Uytterhoeven Signed-off-by: Herbert Xu