On Tue, Oct 24, 2017 at 12:52:08PM -0600, Jason Gunthorpe wrote: > On Mon, Oct 23, 2017 at 02:38:14PM +0200, Jarkko Sakkinen wrote: > > The reasoning is simple and obvious. Since every call site passes the > > value TPM_ANY_NUM (0xFFFF) the parameter does not have right to exist. > > Refined the documentation of the corresponding functions. > > I like this patch, but how about a slightly different take, make this > change instead: > > -struct tpm_chip *tpm_chip_find_get(int chip_num) > +struct tpm_chip *tpm_chip_find_get(struct tpm_chip *chip); > > Where chip == NULL means the default TPM. > > And then at all call sites swap TPM_ANY_NUM to NULL and instead of > flowing an 'int chip_num' to tpm_chip_find_get, flow the 'struct > tpm_chip *' directly. > > This gets us much closer to the desired API with about the same amount > of churn as this patch has. > > Jason I can do that. I will rework this as a patch set that includes this proposal. Gives us more clean position to continue. Thank you. /Jarkko
