diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2023-06-28 13:26:19 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2023-06-28 13:26:19 -0700 |
commit | 4171a9aa235988fc5cb19d84d493496cb73e6988 (patch) | |
tree | 9daf77086005adcb9c08465c057285db1960060b /drivers/spi/spidev.c | |
parent | 1b2c92a1cb2469d8c0079dbf496ab86e22e1cb7c (diff) | |
parent | d0c99ffe212679b338d12fe283964e6e43ce1501 (diff) | |
download | lwn-4171a9aa235988fc5cb19d84d493496cb73e6988.tar.gz lwn-4171a9aa235988fc5cb19d84d493496cb73e6988.zip |
Merge tag 'regmap-v6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap
Pull regmap updates from Mark Brown:
"Another busy release for regmap with the second half of the maple tree
register cache implementation, there's some smaller optimisations that
could be done but this should now be able to replace the rbtree cache
for most devices.
We also had a followup from Aidan MacDonald's refactoring of some of
the regmap-irq interfaces, the conversion is complete so the old
interfaces are removed. This means that even with the new features for
the maple tree cache we'd have a nice negative diffstat were it not
for the addition of a bunch more KUnit coverage.
There's one GPIO patch in here, it was a dependency for a cleanup of
an API in the regmap-irq code for which the gpio-104-dio-48e driver
was the only user.
Highlights:
- The maple tree cache can now load in default values more
efficiently, and is capabale of syncing multiple registers
in a single write during cache sync
- More KUnit coverage, including some coverage for raw I/O
and a dummy RAM backed cache to support it
- Removal of several old interfaces in regmap-irq now all
users have been modernised"
* tag 'regmap-v6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap: (23 commits)
regmap: Allow reads from write only registers with the flat cache
regmap: Drop early readability check
regmap: Check for register readability before checking cache during read
regmap: Add test to make sure we don't sync to read only registers
regmap: Add a test case for write only registers
regmap: Add test that writes to write only registers are prevented
regmap: Add debugfs file for forcing field writes
regmap: Don't check for changes in regcache_set_val()
regmap: maple: Implement block sync for the maple tree cache
regmap: Provide basic KUnit coverage for the raw register I/O
regmap: Provide a ram backed regmap with raw support
regmap: Add missing cache_only checks
regmap: regmap-irq: Move handle_post_irq to before pm_runtime_put
regmap: Load register defaults in blocks rather than register by register
regmap: mmio: Allow passing an empty config->reg_stride
regmap-irq: Drop backward compatibility for inverted mask/unmask
regmap-irq: Minor adjustments to .handle_mask_sync()
regmap-irq: Remove support for not_fixed_stride
regmap-irq: Remove type registers
regmap-irq: Remove virtual registers
...
Diffstat (limited to 'drivers/spi/spidev.c')
0 files changed, 0 insertions, 0 deletions