Re: [RFC PATCH v3 2/4] x86/sgx: Implement support for MADV_WILLNEED

[Jarkko Sakkinen <jarkko@kernel.org>]

On Wed, Jan 25, 2023 at 01:40:39PM -0600, Haitao Huang wrote:
> On Tue, 15 Nov 2022 16:03:45 -0600, Jarkko Sakkinen <jarkko@kernel.org>
> wrote:
> 
> 
> > > +	if (!encl)
> > > +		return -EINVAL;
> > 
> > Why !encl check is needed?
> It was intended as sanity check. But I think encl should not be null at this
> point so will remove in next version.
> 
> > 
> > > +	if (!cpu_feature_enabled(X86_FEATURE_SGX2))
> > > +		return -EINVAL;
> > 
> > This should be done first before doing anything else in the function.
> 
> Will do
> 
> > 
> > > +
> > > +	if (offset + len < offset)
> > > +		return -EINVAL;
> > > +	if (encl->base + offset < encl->base)
> > > +		return -EINVAL;
> > > +	start  = offset + encl->base;
> > > +	end = start + len;
> > 
> > These could be just as well assigned in the declarations, which would
> > definitely be also less convoluted.
> > 
> Will do
> 
> ...
> > > diff --git a/arch/x86/kernel/cpu/sgx/encl.c
> > > b/arch/x86/kernel/cpu/sgx/encl.c
> > > index 1abc5e7f2660..c57e60d5a0aa 100644
> > > --- a/arch/x86/kernel/cpu/sgx/encl.c
> > > +++ b/arch/x86/kernel/cpu/sgx/encl.c
> > > @@ -305,11 +305,11 @@ struct sgx_encl_page
> > > *sgx_encl_load_page(struct sgx_encl *encl,
> > >   * on a SGX2 system then the EPC can be added dynamically via the SGX2
> > >   * ENCLS[EAUG] instruction.
> > >   *
> > > - * Returns: Appropriate vm_fault_t: VM_FAULT_NOPAGE when PTE was
> > > installed
> > > - * successfully, VM_FAULT_SIGBUS or VM_FAULT_OOM as error otherwise.
> > > + * Returns: 0 when PTE was installed successfully, -EBUSY for
> > > waiting on
> > > + * reclaimer to free EPC, -ENOMEM for out of RAM, -EFAULT as error
> > > otherwise.
> > >   */
> > > -vm_fault_t sgx_encl_eaug_page(struct vm_area_struct *vma,
> > > -			      struct sgx_encl *encl, unsigned long addr)
> > > +int sgx_encl_eaug_page(struct vm_area_struct *vma,
> > > +		       struct sgx_encl *encl, unsigned long addr)
> > >  {
> > >  	vm_fault_t vmret = VM_FAULT_SIGBUS;
> > >  	struct sgx_pageinfo pginfo = {0};
> > > @@ -318,10 +318,10 @@ vm_fault_t sgx_encl_eaug_page(struct
> > > vm_area_struct *vma,
> > >  	struct sgx_va_page *va_page;
> > >  	unsigned long phys_addr;
> > >  	u64 secinfo_flags;
> > > -	int ret;
> > > +	int ret = -EFAULT;
> > 
> > Why?
> Original code uses ret only to temporarily store return values from the
> called functions.
> Now I also use it as return of this function and assign -EFAULT as default
> in all cases that ret is not assigned explicitly below, e.g., -EBUSY cases.
> Basically -EFAULT cases are the same as VM_FAULT_SIGBUS cases in original
> code.
> I'll clarify in comments above for the return values.
> 
> Thanks
> Haitao

OK, will look forward to it.

BR, Jarkko