kernel/linux.git/drivers/base/regmap/regmap-mmio.c, branch v6.1.181

regmap: mmio: replace return 0 with break in switch statement

2022-09-22T11:33:18+00:00

Variable min_stride is assigned a value that is never read, fix this by replacing the return 0 with a break statement. This also makes the case statement consistent with the other cases in the switch statement. Signed-off-by: Colin Ian King Link: https://lore.kernel.org/r/20220922080445.818020-1-colin.i.king@gmail.com Signed-off-by: Mark Brown

regmap: mmio: Use swabXX_array() helpers

2022-09-07T11:42:26+00:00

Since we have a few helpers to swab elements of a given size in an array use them instead of open coded variants. Signed-off-by: Andy Shevchenko Reviewed-by: Linus Walleij Link: https://lore.kernel.org/r/20220831212744.56435-2-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown

regmap: mmio: Fix rebase error

2022-09-01T10:56:26+00:00

A dangling pointless "ret 0" was left in and some unneeded whitespace can go too. Fixes: 81c0386c1376 ("regmap: mmio: Support accelerared noinc operations") Reported-by: Andy Shevchenko Signed-off-by: Linus Walleij Link: https://lore.kernel.org/r/20220831141303.501548-1-linus.walleij@linaro.org Signed-off-by: Mark Brown

regmap: mmio: Support accelerared noinc operations

2022-08-17T12:00:32+00:00

Use the newly added callback for accelerated noinc MMIO to provide writesb, writesw, writesl, writesq, readsb, readsw, readsl and readsq. A special quirk is needed to deal with big endian regmaps: there are no accelerated operations defined for big endian, so fall back to calling the big endian operations itereatively for this case. The Hexagon architecture turns out to have an incomplete : writesb() is not implemented. Fix this by doing what other architectures do: include into the file. Cc: Brian Cain Cc: linux-hexagon@vger.kernel.org Cc: Arnd Bergmann Signed-off-by: Linus Walleij Link: https://lore.kernel.org/r/20220816204832.265837-2-linus.walleij@linaro.org Signed-off-by: Mark Brown

regmap: mmio: Fix MMIO accessors to avoid talking to IO port

2022-08-15T00:20:11+00:00

Currently regmap MMIO is inconsistent with IO accessors. I.e. the Big Endian counterparts are using ioreadXXbe() / iowriteXXbe() which are not clean implementations of readXXbe(). That said, reimplement current Big Endian MMIO accessors by replacing ioread()/iowrite() with respective read()/write() and swab() calls. Note, there are no current in-kernel users that may utilize the functionality of the IO ports on Big Endian hardware. All drivers that use regmap MMIO either Little Endian, or they don't map IO ports in a way that ioreadXX()/iowriteXX() may be utilized. Signed-off-by: Andy Shevchenko Acked-by: William Breathitt Gray Link: https://lore.kernel.org/r/20220808203401.35153-5-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown

regmap: mmio: Introduce IO accessors that can talk to IO port

2022-08-15T00:20:10+00:00

Some users may use regmap MMIO for IO ports, and this can be done by assigning ioreadXX()/iowriteXX() and their Big Endian counterparts to the regmap context. Add IO port support with a corresponding flag added. While doing that, make sure that user won't select relaxed MMIO access along with IO port because the latter have no relaxed variants. Signed-off-by: Andy Shevchenko Acked-by: William Breathitt Gray Link: https://lore.kernel.org/r/20220808203401.35153-4-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown

regmap: mmio: Get rid of broken 64-bit IO

2022-08-15T00:20:09+00:00

The current implementation, besides having no active users, is broken by design of regmap. For 64-bit IO we need to supply 64-bit value, otherwise there is no way to handle upper 32 bits in 64-bit register. Hence, remove the broken IO accessors for good and wait for real user that can fix entire regmap API for that. Signed-off-by: Andy Shevchenko Acked-by: William Breathitt Gray Link: https://lore.kernel.org/r/20220808203401.35153-3-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown

regmap: mmio: Remove mmio_relaxed member from context

2022-08-15T00:20:08+00:00

There is no need to keep mmio_relaxed member in the context, it's onetime used during generation of the context. Remove it. Signed-off-by: Andy Shevchenko Acked-by: William Breathitt Gray Link: https://lore.kernel.org/r/20220808203401.35153-2-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown

regmap: Prefer unsigned int to bare use of unsigned

2021-07-11T22:48:43+00:00

Fix checkpatch warnings: WARNING: Prefer 'unsigned int' to bare use of 'unsigned' Signed-off-by: Jinchao Wang Link: https://lore.kernel.org/r/20210628171907.63646-1-wjc@cdjrlc.com Signed-off-by: Mark Brown

regmap: mmio: add config option to allow relaxed MMIO accesses

2020-10-26T19:54:59+00:00

On some platforms (eg armv7 due to the CONFIG_ARM_DMA_MEM_BUFFERABLE) MMIO R/W operations always add memory barriers which can increase load, decrease battery life or in general reduce performance unnecessarily on devices which access a lot of configuration registers and where ordering does not matter (eg. media accelerators like the Verisilicon / Hantro video decoders). Drivers used to call the relaxed MMIO variants directly but since they are now accessing the MMIO registers via regmaps (to compensate for different VPU HW reg layouts via regmap fields), there is a need for a relaxed API / config to preserve existing behaviour. Cc: Mark Brown Signed-off-by: Adrian Ratiu Link: https://lore.kernel.org/r/20201014203024.954369-1-adrian.ratiu@collabora.com Signed-off-by: Mark Brown