[MIPS] Provide empty irq_enable_hazard definition for legacy and R1 cores.
authorRalf Baechle <ralf@linux-mips.org>
Mon, 3 Sep 2007 14:22:26 +0000 (16:22 +0200)
committerRalf Baechle <ralf@linux-mips.org>
Tue, 11 Sep 2007 18:03:25 +0000 (19:03 +0100)
Following a strict interpretation the empty definition of irq_enable_hazard
has always been a bug - but an intentional one because it didn't bite.
This has now changed, for uniprocessor kernels mm/slab.c:do_drain()

[...]
        on_each_cpu(do_drain, cachep, 1, 1);
        check_irq_on();
[...]

may be compiled into a mtc0 c0_status; mfc0 c0_status sequence resulting
in a back-to-back hazard.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
include/asm-mips/hazards.h

index 918a4894b587d10fe8450a96239ad42f0b7e63e5..6a5fa32f615bc24e63ed4278267ce3160185c5e5 100644 (file)
@@ -172,6 +172,7 @@ ASMMACRO(tlb_probe_hazard,
         nop; nop; nop
        )
 ASMMACRO(irq_enable_hazard,
+        _ssnop; _ssnop; _ssnop;
        )
 ASMMACRO(irq_disable_hazard,
        nop; nop; nop