kernel/linux.git/drivers/soc/litex, branch v6.12.80

soc: Switch back to struct platform_driver::remove()

2025-12-18T12:54:46+00:00

[ Upstream commit 511c06e3903563dba4472430e1b586745b6ae238 ] After commit 0edb555a65d1 ("platform: Make platform_driver::remove() return void") .remove() is (again) the right callback to implement for platform drivers. Convert all platform drivers below drivers/soc to use .remove(), with the eventual goal to drop struct platform_driver::remove_new(). As .remove() and .remove_new() have the same prototypes, conversion is done by just changing the structure member name in the driver initializer. On the way do a few whitespace changes to make indention consistent. Signed-off-by: Uwe Kleine-König Acked-by: Heiko Stuebner Acked-by: Herve Codina # for fsl/qe/{qmc,tsa}.c Acked-by: Bjorn Andersson # qcom parts Acked-by: Gabriel Somlo Acked-by: Andrew Jeffery # aspeed Link: https://lore.kernel.org/r/20241029074859.509587-2-u.kleine-koenig@baylibre.com Signed-off-by: Arnd Bergmann Stable-dep-of: 2286e18e3937 ("soc: qcom: gsbi: fix double disable caused by devm") Signed-off-by: Sasha Levin

drivers/soc/litex: drop obsolete dependency on COMPILE_TEST

2024-06-27T14:33:39+00:00

Since commit 0166dc11be91 ("of: make CONFIG_OF user selectable"), it is possible to test-build any driver which depends on OF on any architecture by explicitly selecting OF. Therefore depending on COMPILE_TEST as an alternative is no longer needed. It is actually better to always build such drivers with OF enabled, so that the test builds are closer to how each driver will actually be built on its intended target. Building them without OF may not test much as the compiler will optimize out potentially large parts of the code. In the worst case, this could even pop false positive warnings. Dropping COMPILE_TEST here improves the quality of our testing and avoids wasting time on non-existent issues. As a minor optimization, this also lets us drop of_match_ptr() and ifdef-guarding, as we now know what they will resolve to, we might as well save cpp some work. Signed-off-by: Jean Delvare Reviewed-by: Gabriel Somlo Cc: Karol Gugala Cc: Mateusz Holenko Cc: Joel Stanley Link: https://lore.kernel.org/r/20240617133004.59887629@endymion.delvare Signed-off-by: Arnd Bergmann

soc/litex: litex_soc_ctrl: Convert to platform remove callback returning void

2023-10-14T21:27:17+00:00

The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new() which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Acked-by: Gabriel Somlo Link: https://lore.kernel.org/r/20230925095532.1984344-13-u.kleine-koenig@pengutronix.de Signed-off-by: Uwe Kleine-König

drivers/soc/litex: remove 8-bit subregister option

2021-06-10T19:35:40+00:00

Since upstream LiteX recommends that Linux support be limited to designs configured with 32-bit CSR subregisters (see commit a2b71fde in upstream LiteX, https://github.com/enjoy-digital/litex), remove the option to select 8-bit subregisters, significantly reducing the complexity of LiteX CSR (MMIO register) accessor methods. NOTE: for details on the underlying mechanics of LiteX CSR registers, see https://github.com/enjoy-digital/litex/wiki/CSR-Bus or the original LiteX accessors (litex/soc/software/include/hw/common.h in the upstream repository). Signed-off-by: Gabriel Somlo Cc: Stafford Horne Cc: Florent Kermarrec Cc: Mateusz Holenko Cc: Joel Stanley Reviewed-by: Joel Stanley Signed-off-by: Stafford Horne

soc: litex: Remove duplicated header file inclusion

2021-04-03T20:46:46+00:00

The header file is already included above and can be removed here. Signed-off-by: Zhen Lei Signed-off-by: Mateusz Holenko Signed-off-by: Stafford Horne

Merge tag 'for-linus' of git://github.com/openrisc/linux

2021-02-26T22:16:06+00:00

Pull OpenRISC updates from Stafford Horne: - Update for Litex SoC controller to support wider width registers as well as reset. - Refactor SMP code to use device tree to define possible cpus. - Update build including generating vmlinux.bin * tag 'for-linus' of git://github.com/openrisc/linux: openrisc: Use devicetree to determine present cpus drivers/soc/litex: Add restart handler openrisc: add arch/openrisc/Kbuild drivers/soc/litex: make 'litex_[set|get]_reg()' methods private drivers/soc/litex: support 32-bit subregisters, 64-bit CPUs drivers/soc/litex: s/LITEX_REG_SIZE/LITEX_SUBREG_ALIGN/g drivers/soc/litex: separate MMIO from subregister offset calculation drivers/soc/litex: move generic accessors to litex.h openrisc: restart: Call common handlers before hanging openrisc: Add vmlinux.bin target

soc: litex: Properly depend on HAS_IOMEM

2021-01-29T21:36:10+00:00

The LiteX SOC controller driver makes use of IOMEM functions like devm_platform_ioremap_resource(), which are only available if CONFIG_HAS_IOMEM is defined. This causes the driver to be enable under make ARCH=um allyesconfig, even though it won't build. By adding a dependency on HAS_IOMEM, the driver will not be enabled on architectures which don't support it. Fixes: 22447a99c97e ("drivers/soc/litex: add LiteX SoC Controller driver") Signed-off-by: David Gow [shorne@gmail.com: Fix typo in commit message pointed out in review] Signed-off-by: Stafford Horne

Merge remote-tracking branch 'openrisc/or1k-5.11-fixes' into or1k-5.12-updates

2021-01-25T02:16:06+00:00

Pickup fixes that went upstream already in order to resolve conflicts in litex_soc_ctrl.c between 5.11 fixes that went upstream and the or1k-5.12-updates linux-next queue. Signed-off-by: Stafford Horne

drivers/soc/litex: Add restart handler

2021-01-21T12:40:05+00:00

Let the LiteX SoC Controller register a restart handler, which resets the LiteX SoC by writing 1 to CSR_CTRL_RESET_ADDR. Signed-off-by: Geert Uytterhoeven Reviewed-by: Joel Stanley Signed-off-by: Stafford Horne

drivers/soc/litex: make 'litex_[set|get]_reg()' methods private

2021-01-14T00:52:54+00:00

The 'litex_[set|get]_reg()' methods use the 'reg_size' parameter to specify the width of the LiteX CSR (MMIO) register being accessed. Since 'u64' is the widest data being supported, the value of 'reg_size' MUST be between 1 and sizeof(u64), which SHOULD be checked at runtime if these methods are publicly available for use by other LiteX device drivers. At the same time, none of the existing (or foreseeable) LiteX device drivers have a need to access registers whose size is unknown during compilation. As such, all LiteX device drivers should use fixed-width accessor methods such as 'litex_[write|read][8|16|32|64]()'. This patch renames 'litex_[set|get]_reg()' to '_litex_[set|get]_reg()', indicating that they should NOT be directly called from outside of the 'include/linux/litex.h' header file. Signed-off-by: Gabriel Somlo Signed-off-by: Stafford Horne