On Fri, Oct 15, 2021 at 9:16 AM Jonathan Cameron Jonathan.Cameron@huawei.com wrote:
On Sat, 9 Oct 2021 09:44:13 -0700 Dan Williams dan.j.williams@intel.com wrote:
cxl_pci_map_regblock() may return an ERR_PTR(), but cxl_pci_setup_regs() is only prepared for NULL as the error case.
What's the logic behind doing this rather than adjusting the call site to check for an error pointer?
Minimize the fix for the stable backport. In the later patches the cxl_pci_map_regblock() => cxl_map_regblock() conversion goes from returning a pointer to an error code.
Either approach is fine as far as I'm concerned though so this is really just a request for a bit more info in this patch description.
I can include that note above to clarify.
FWIW
Reviewed-by: Jonathan Cameron Jonathan.Cameron@huawei.com
Thanks.