[patch V2 14/44] x86/smp: Remove pointless duplicated interrupt code

2017-08-25 Thread Thomas Gleixner
Two NOP5 are really a good tradeoff vs. the unholy IDT switching mess, which duplicates code all over the place. The rescheduling interrupt gets optimized in a later step. Make the ordering of function call and statistics increment the same as in other places. stats first, then function call.

[patch V2 14/44] x86/smp: Remove pointless duplicated interrupt code

2017-08-25 Thread Thomas Gleixner
Two NOP5 are really a good tradeoff vs. the unholy IDT switching mess, which duplicates code all over the place. The rescheduling interrupt gets optimized in a later step. Make the ordering of function call and statistics increment the same as in other places. stats first, then function call.