From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mailhub1.si.c-s.fr (2.236.17.93.rev.sfr.net [93.17.236.2]) by lists.ozlabs.org (Postfix) with ESMTP id E2C271A0018 for ; Thu, 8 Oct 2015 12:03:38 +1100 (AEDT) Subject: Re: [PATCH v2 22/25] powerpc32: move xxxxx_dcache_range() functions inline To: Scott Wood References: <20150929002935.GK6161@home.buserror.net> Cc: Benjamin Herrenschmidt , Paul Mackerras , Michael Ellerman , linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org From: Christophe Leroy Message-ID: <561514D4.9090406@c-s.fr> Date: Wed, 7 Oct 2015 14:49:24 +0200 MIME-Version: 1.0 In-Reply-To: <20150929002935.GK6161@home.buserror.net> Content-Type: text/plain; charset=windows-1252; format=flowed List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Le 29/09/2015 02:29, Scott Wood a écrit : > On Tue, Sep 22, 2015 at 06:51:13PM +0200, Christophe Leroy wrote: >> 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 >> >> They are good candidate for inlining, as __dma_sync() does >> almost nothing but calling them >> >> Signed-off-by: Christophe Leroy >> --- >> New in v2 >> >> arch/powerpc/include/asm/cacheflush.h | 55 +++++++++++++++++++++++++++-- >> arch/powerpc/kernel/misc_32.S | 65 ----------------------------------- >> arch/powerpc/kernel/ppc_ksyms.c | 2 ++ >> 3 files changed, 54 insertions(+), 68 deletions(-) >> >> diff --git a/arch/powerpc/include/asm/cacheflush.h 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 __flush_dcache_icache_phys(unsigned long physaddr) >> } >> #endif >> >> -extern void flush_dcache_range(unsigned long start, unsigned long stop); >> #ifdef CONFIG_PPC32 >> -extern void clean_dcache_range(unsigned long start, unsigned long stop); >> -extern void invalidate_dcache_range(unsigned long start, unsigned long stop); >> +/* >> + * Write any modified data cache blocks out to memory and invalidate them. >> + * Does not invalidate the corresponding instruction cache blocks. >> + */ >> +static inline void flush_dcache_range(unsigned long start, unsigned long stop) >> +{ >> + void *addr = (void *)(start & ~(L1_CACHE_BYTES - 1)); >> + unsigned int size = stop - (unsigned long)addr + (L1_CACHE_BYTES - 1); >> + unsigned int i; >> + >> + for (i = 0; i < size >> L1_CACHE_SHIFT; i++, addr += L1_CACHE_BYTES) >> + dcbf(addr); >> + if (i) >> + mb(); /* sync */ >> +} > I know this is 32-bit-specific code, but it's still bad practice to use > "unsigned int" for addresses or sizes thereof. > > Ok, I can fix size, but what about start and stop ? If I change that, it means I also have to fix all caller. Do you expect me to do that ? And it is very unlykely, but what if for some reason someone wants to invalidate the entire user address space which is 3Gbytes size ? A signed size would be negative here. Christophe