Before the release of Linux kernel 7.0, some interesting changes were planned in the Linux MultiMediaCard (MMC) subsystem, such as NXP IW61x device IDs for Wi-Fi chips via SDIO, support for production dates after 2025, optimized secure erase/TRIM for some Kingston eMMC cards, code cleanup in DW_MMC, support for Mediatek MT8189 in mtk-sd, and various SHDCI driver updates.
However, Linus Torvalds reviewed the proposed merge request and rejected it, commenting:
No.
These changes are complete nonsense and they don't even compile. Apparently, they were never in linux-next and were not tested during the build.
If we build this file core.o with CONFIG_MULTIPLEXER=m:
obj-$(CONFIG_MULTIPLEXER) += mux-core.o
but in include/linux/mux/consumer.h you have
#ifdef CONFIG_MULTIPLEXER
which will not be true (since CONFIG_MULTIPLEXER_MODULE is defined), so we will get a long stream of something like
drivers/mux/core.c:312:14: error: redefinition of ‘mux_control_states’
because in the header file mux/consumer.h there will be a dummy wrapper function defined.
In other words, commit ad314348ceb4 (‘mux: Add helper functions for getting optional and selected mux-state’) is pure, unadulterated, untested garbage.
I don't want to see from you a 'fixed' merge request. This is absolutely unacceptable, and I will not accept anything from you during the upcoming merge window. Stop sending me untested junk that hasn't been included in linux-next and doesn't even pass the most superficial problem check.
You can try again for version 7.1, but only after it has been included in linux-next and properly tested.
Thus, changes in the Linux MultiMediaCard will have to be postponed until the beginning of the Linux 7.1 merge window in mid-April, after the stable release of Linux 7.0.
Source: linux.org.ru
