RE: [PATCH v13 4/5] x86/sgx: Implement ENCLS[EUPDATESVN]

[Reshetova, Elena]

> -----Original Message-----
> From: Huang, Kai <kai.huang@xxxxxxxxx>
> Sent: Wednesday, August 13, 2025 2:08 PM
> To: Reshetova, Elena <elena.reshetova@xxxxxxxxx>; Hansen, Dave
> <dave.hansen@xxxxxxxxx>
> Cc: seanjc@xxxxxxxxxx; mingo@xxxxxxxxxx; Scarlata, Vincent R
> <vincent.r.scarlata@xxxxxxxxx>; x86@xxxxxxxxxx; jarkko@xxxxxxxxxx;
> Annapurve, Vishal <vannapurve@xxxxxxxxxx>; linux-kernel@xxxxxxxxxxxxxxx;
> Mallick, Asit K <asit.k.mallick@xxxxxxxxx>; Aktas, Erdem
> <erdemaktas@xxxxxxxxxx>; Cai, Chong <chongc@xxxxxxxxxx>; Bondarevska,
> Nataliia <bondarn@xxxxxxxxxx>; linux-sgx@xxxxxxxxxxxxxxx; Raynor, Scott
> <scott.raynor@xxxxxxxxx>
> Subject: Re: [PATCH v13 4/5] x86/sgx: Implement ENCLS[EUPDATESVN]
> 
> 
> >
> > +/* Counter to count the active SGX users */
> > +static int sgx_usage_count;
> >
> 
> [...]
> 
> > + * Return:
> > + * %0:			- Success or not supported
> > + * %-EAGAIN:	- Can be safely retried, failure is due to lack of
> > + *				entropy in RNG
> > + * %-EIO:		- Unexpected error, retries are not advisable
> > + */
> 
> This time I actually downloaded those patches and applied to my local, and
> I found the descriptions of the error codes are not vertically aligned.
> 
> Please fix (and it's sad we still need to fix this type of thing in v13).

Sorry about this, it seems like I need to change the editor, which keeps reverting
this on rebases ((

> 
> Nit: as said before, the k-doc comment doc says:
> 
>     .. in order to produce the desired line breaks, you need to use a ReST
>     list, e. g.:
> 
>     * Return:
>     * * %0            - OK to runtime suspend the device
>     * * %-EBUSY       - Device should not be runtime suspended
> 
> (hint: there's an additional '*' before the '%'.)
> 
> But I guess it's just a nit but not a blocker.

Yes, I did go and check the kdoc style, but missed the additional '*'.
Will fix. 

> 
> > +static int __maybe_unused sgx_update_svn(void)
> > +{
> > +	int ret;
> > +
> > +	/*
> > +	 * If EUPDATESVN is not available, it is ok to
> > +	 * silently skip it to comply with legacy behavior.
> > +	 */
> > +	if (!cpu_feature_enabled(X86_FEATURE_SGX_EUPDATESVN))
> > +		return 0;
> > +
> > +	/*
> > +	 * EPC is guaranteed to be empty when there are no users.
> > +	 * Ensure we are on our first user before proceeding further.
> > +	 */
> > +	WARN(sgx_usage_count != 1, "Elevated usage count when calling
> EUPDATESVN\n");
> 
> It seems you are obsessed to use "!= 1", rather than "!= 0".
> 
> IIUC, Dave suggested the latter [*]:
> 
>     	/* EPC is guaranteed to be empty when there are no users: */
> 	WARN(count, "Elevated usage count...");
> 
> .. which is my natural response too.
> 
> And the odd is I actually need to look at the next patch to see why "!= 1"
> is used.

I can change it to this version given that I change the code in the previous patch. 

Thank you very much for your prompt review!

Best Regards,
Elena.