2009/8/13 Blue Swirl : > On Thu, Aug 13, 2009 at 6:44 PM, Artyom > Tarasenko wrote: >> 2009/8/12 Igor Kovalenko : >>> PUT_PSR is asking to be made inline function with arch-specific parts, >>> the sparc64 counterpart is too similar. >> Ok, here we go: > > Sorry, it does not apply. Ups, copy&paste error, sorry. This time I send it as an attachment. >> -#if !defined(CONFIG_USER_ONLY) >> -                            cpu_check_irqs(env); >> -#endif > > This should be in a separate patch, preferably one with the check > moved to helper_wrpsr. But that's the purpose of the patch. The check is moved to PUT_PSR, not to helper_wrpsr, because helper_wrpsr is not the only function calling PUT_PSR. From the code I can't tell that the other calls can not trigger an irq as well. >> +/* sum4m.c, sun4u.c */ >> +void cpu_check_irqs(CPUSPARCState *env); > > The prototypes should remain in header files. The two lines are moved from the line number 481 to the line number 588 within the same file. Btw there's a typo, I guess it should be su_n_4m. :) Fixed this too.