Re: [PATCH v5] powerpc/irq: inline call_do_irq() and call_do_softirq() on PPC32

2020-06-01 Thread Christophe Leroy
Hi Michael, Le 07/12/2019 à 18:20, Christophe Leroy a écrit : call_do_irq() and call_do_softirq() are simple enough to be worth inlining. Inlining them avoids an mflr/mtlr pair plus a save/reload on stack. It also allows GCC to keep the saved ksp_limit in an nonvolatile reg. This is inspired f

Re: [PATCH v5] powerpc/irq: inline call_do_irq() and call_do_softirq() on PPC32

2019-12-12 Thread Christophe Leroy
Le 12/12/2019 à 13:52, Christoph Hellwig a écrit : On Sat, Dec 07, 2019 at 05:20:04PM +, Christophe Leroy wrote: call_do_irq() and call_do_softirq() are simple enough to be worth inlining. Inlining them avoids an mflr/mtlr pair plus a save/reload on stack. It also allows GCC to keep the

Re: [PATCH v5] powerpc/irq: inline call_do_irq() and call_do_softirq() on PPC32

2019-12-12 Thread Christoph Hellwig
On Sat, Dec 07, 2019 at 05:20:04PM +, Christophe Leroy wrote: > call_do_irq() and call_do_softirq() are simple enough to be > worth inlining. > > Inlining them avoids an mflr/mtlr pair plus a save/reload on stack. > It also allows GCC to keep the saved ksp_limit in an nonvolatile reg. > > Thi

[PATCH v5] powerpc/irq: inline call_do_irq() and call_do_softirq() on PPC32

2019-12-07 Thread Christophe Leroy
call_do_irq() and call_do_softirq() are simple enough to be worth inlining. Inlining them avoids an mflr/mtlr pair plus a save/reload on stack. It also allows GCC to keep the saved ksp_limit in an nonvolatile reg. This is inspired from S390 arch. Several other arches do more or less the same. The