On Tue, 2025-07-22 at 07:37 -0700, Jeff Johnson wrote: > > This predates me becoming a maintainer, and I don't see any comment from Kalle > to the patch, however he did have the following comment in a separate thread > where the issue was being discussed: > https://lore.kernel.org/all/87cz8v2xb2.fsf@xxxxxxxxxx/ > > > Though I am happy to take your MODULE_SOFTDEP() patch, just wondering if > > there is a better way to solve this. For example net/mac80211 (the > > 802.11 stack) has a lot of crypto dependencies: > > > > select CRYPTO > > select CRYPTO_LIB_ARC4 > > select CRYPTO_AES > > select CRYPTO_CCM > > select CRYPTO_GCM > > select CRYPTO_CMAC > > select CRC32 > > > > And it's not using MODULE_SOFTDEP() at all. > > So I'm guessing he considered this to be an issue that should be solved > external to the individual drivers. I just happened to see this and never saw the other discussion from Dec 2022, but it's somewhat interesting ... for some of those modules mac80211 has a hard dependency (such as lib_arc3, crc32) but for the other crypto modules it instantiates the algorithms via the string descriptions (such as "gcm(aes)") so it has no hard dependency. Maybe crypto modules are treated specially by installers and the like? johannes
