Messages in this thread | | | Date | Tue, 05 May 2015 21:15:54 +0800 | From | Hanjun Guo <> | Subject | Re: [PATCH v2 3/7] ACPI / processor: Introduce invalid_logical_cpuid() |
| |
On 2015年05月05日 20:04, Rafael J. Wysocki wrote: > On Tuesday, May 05, 2015 12:15:13 PM Sudeep Holla wrote: >> >> On 05/05/15 03:46, Hanjun Guo wrote: >>> In ACPI processor drivers, we use direct comparisons of cpu logical >>> id with -1 which are error prone in case logical cpuid is accidentally >>> assinged an error code and prevents us from returning an error-encoding >>> cpuid directly in some cases. >>> >>> So introduce invalid_logical_cpuid() to identify cpu with invalid >>> logical cpu num, then it will be used to replace the direct comparisons >>> with -1. >>> >> >> Ah, OK I see that this fixes the issue I raised in PATCH 1/7, so I think >> you need to reorder this and 1/7 patch IMO. > > Well, comparing an unsigned int with -1 is not technically invalid (although it > involves an implicit type conversion), but yes, Hanjun, please reorder the > patches.
Sure, I will.
Thanks Hanjun
| |