Re: [PATCH v2 22/25] powerpc32: move xxxxx_dcache_range() functions inline

On Tue, 2015-09-22 at 13:58 -0500, Scott Wood wrote:

On Tue, 2015-09-22 at 18:12 +0000, Joakim Tjernlund wrote:

quoted

On Tue, 2015-09-22 at 18:51 +0200, Christophe Leroy wrote:

quoted

flush/clean/invalidate _dcache_range() functions are all very
similar and are quite short. They are mainly used in __dma_sync()
perf_event locate them in the top 3 consumming functions during
heavy ethernet activity
=20
They are good candidate for inlining, as __dma_sync() does
almost nothing but calling them
=20
Signed-off-by: Christophe Leroy <redacted>
---
New in v2
=20
 arch/powerpc/include/asm/cacheflush.h | 55 +++++++++++++++++++++++++=

++--

quoted

quoted

 arch/powerpc/kernel/misc_32.S         | 65 -------------------------=

-----

quoted

quoted

-----
 arch/powerpc/kernel/ppc_ksyms.c       |  2 ++
 3 files changed, 54 insertions(+), 68 deletions(-)
=20

diff --git a/arch/powerpc/include/asm/cacheflush.h=20

b/arch/powerpc/include/asm/cacheflush.h
index 6229e6b..6169604 100644

--- a/arch/powerpc/include/asm/cacheflush.h
+++ b/arch/powerpc/include/asm/cacheflush.h

@@ -47,12 +47,61 @@ static inline void=20

__flush_dcache_icache_phys(unsigned long physaddr)
 }
 #endif
=20
-extern void flush_dcache_range(unsigned long start, unsigned long st=

op);

quoted

quoted

 #ifdef CONFIG_PPC32
-extern void clean_dcache_range(unsigned long start, unsigned long st=

op);

quoted

quoted

-extern void invalidate_dcache_range(unsigned long start, unsigned lo=

ng=20

quoted

quoted

stop);
+/*
+ * Write any modified data cache blocks out to memory and invalidate=

=20

quoted

quoted

them.
+ * Does not invalidate the corresponding instruction cache blocks.
+ */
+static inline void flush_dcache_range(unsigned long start, unsigned =

long=20

quoted

quoted

stop)
+{
+   void *addr =3D (void *)(start & ~(L1_CACHE_BYTES - 1));
+   unsigned int size =3D stop - (unsigned long)addr + (L1_CACHE_BYTE=

S - 1);

quoted

quoted

+   unsigned int i;
+
+   for (i =3D 0; i < size >> L1_CACHE_SHIFT; i++, addr +=3D L1_CACHE=

_BYTES)

quoted

quoted

+           dcbf(addr);
+   if (i)
+           mb();   /* sync */
+}

=20
This feels optimized for the uncommon case when there is no invalidatio=

n.

=20
If you mean the "if (i)", yes, that looks odd.

Yes.

=20

quoted

I THINK it would be better to bail early=20

=20
Bail under what conditions?

test for "i =3D 0" and return.=20

=20

quoted

and use do { .. } while(--i); instead.

=20
GCC knows how to optimize loops.  Please don't make them less readable.

Been a while since I checked but it used to be bad att transforming post in=
c to pre inc/dec
I remain unconvinced until I have seen it.

 Jocke