On Fri, 2024-02-23 at 17:51 +0200, Jarkko Sakkinen wrote: > On Tue Feb 13, 2024 at 7:13 PM EET, James Bottomley wrote: [...] > > --- a/drivers/char/tpm/tpm.h > > +++ b/drivers/char/tpm/tpm.h > > @@ -321,4 +321,14 @@ void tpm_bios_log_setup(struct tpm_chip > > *chip); > > void tpm_bios_log_teardown(struct tpm_chip *chip); > > int tpm_dev_common_init(void); > > void tpm_dev_common_exit(void); > > + > > +#ifdef CONFIG_TCG_TPM2_HMAC > > +int tpm2_sessions_init(struct tpm_chip *chip); > > I'm sorry but "sessions" and "init" are the worst possible > terminology I could every pick when trying to make a function that > self-documents itself :-)
About 95% of the kernel that run initialization routines are suffixed _init. The TCG documents that describe how to effect HMAC and encryption of commands call the entities used to do this "sessions". You can argue they should have chosen a better name, but given it's the name they chose, we have to use it to avoid confusion. > I'd like to see here well-scoped name so that it is easy to connect > this to its purpose when reviewing some other patches. It's purpose is to initialize the sessions handling code in the tpm2- sessions.c file; I'm not sure I could come up with anything as concise for self description. James
