Christophe Lombard <clomb...@linux.vnet.ibm.com> writes:
> This patches removes this restriction and all page faults, whatever the
> reason, will be handled. In this case, the interruption is always
> acknowledged.
This can also be done with adding call to cxl_ack_ae(ctx) at the end of
cxl_handle_fault().
> static bool cxl_is_page_fault(struct cxl_context *ctx, u64 dsisr)
> {
> - u64 crs; /* Translation Checkout Response Status */
> -
> if ((cxl_is_power8()) && (dsisr & CXL_PSL_DSISR_An_DM))
> return true;
>
> - if (cxl_is_power9()) {
> - crs = (dsisr & CXL_PSL9_DSISR_An_CO_MASK);
> - if ((crs == CXL_PSL9_DSISR_An_PF_SLR) ||
> - (crs == CXL_PSL9_DSISR_An_PF_RGC) ||
> - (crs == CXL_PSL9_DSISR_An_PF_RGP) ||
> - (crs == CXL_PSL9_DSISR_An_PF_HRH) ||
> - (crs == CXL_PSL9_DSISR_An_PF_STEG) ||
> - (crs == CXL_PSL9_DSISR_An_URTCH)) {
> - return true;
> - }
> - }
Removing this check will force call to copro_handle_mm_fault even for
nmmu reported errors that only need to be communicated to the AFU and
not handled by mm subsystem as such.
> + if (cxl_is_power9())
> + return true;
Optional: With this change the function has become a good candidate for
being inlined or even converting to a macro.
--
Vaibhav Jain <vaib...@linux.vnet.ibm.com>
Linux Technology Center, IBM India Pvt. Ltd.
