From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: [PATCH v2 26/31] arm64: Miscellaneous library functions Date: Wed, 15 Aug 2012 15:21:14 +0000 Message-ID: <201208151521.14663.arnd@arndb.de> References: <1344966752-16102-1-git-send-email-catalin.marinas@arm.com> <1344966752-16102-27-git-send-email-catalin.marinas@arm.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset="windows-1252" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1344966752-16102-27-git-send-email-catalin.marinas@arm.com> Sender: linux-kernel-owner@vger.kernel.org To: Catalin Marinas Cc: linux-arch@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, Marc Zyngier , Will Deacon List-Id: linux-arch.vger.kernel.org On Tuesday 14 August 2012, Catalin Marinas wrote: > + > +/* > + * Use compiler builtins for simple inline operations. > + */ > +static inline unsigned long __ffs(unsigned long word) > +{ > + return __builtin_ffsl(word) - 1; > +} > + > +static inline int ffs(int x) > +{ > + return __builtin_ffs(x); > +} > + > +static inline unsigned long __fls(unsigned long word) > +{ > + return BITS_PER_LONG - 1 - __builtin_clzl(word); > +} > + > +static inline int fls(int x) > +{ > + return x ? sizeof(x) * BITS_PER_BYTE - __builtin_clz(x) : 0; > +} These are all great, but I think whether to use them or not should depend on the compiler version rather than the architecture in general. Do we know a minimum gcc version that supports all of the above? Then we could put that code into the generic files. If that's not possible, we could still make the implementation available for other architectures by moving it to asm-generic/bitops/builtin-__ffs.h asm-generic/bitops/builtin-ffs.h asm-generic/bitops/builtin-__fls.h asm-generic/bitops/builtin-fls.h > --- /dev/null > +++ b/arch/arm64/lib/bitops.c > @@ -0,0 +1,25 @@ > +/* > + * Copyright (C) 2012 ARM Limited > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License version 2 as > + * published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + * You should have received a copy of the GNU General Public License > + * along with this program. If not, see . > + */ > + > +#include > +#include > +#include > + > +#ifdef CONFIG_SMP > +arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned = { > + [0 ... (ATOMIC_HASH_SIZE-1)] = __ARCH_SPIN_LOCK_UNLOCKED > +}; > +#endif What? I suppose this is a leftover from an earlier version using the generic bitops, right? Arnd From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from moutng.kundenserver.de ([212.227.17.10]:53117 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755170Ab2HOPVX (ORCPT ); Wed, 15 Aug 2012 11:21:23 -0400 From: Arnd Bergmann Subject: Re: [PATCH v2 26/31] arm64: Miscellaneous library functions Date: Wed, 15 Aug 2012 15:21:14 +0000 References: <1344966752-16102-1-git-send-email-catalin.marinas@arm.com> <1344966752-16102-27-git-send-email-catalin.marinas@arm.com> In-Reply-To: <1344966752-16102-27-git-send-email-catalin.marinas@arm.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="windows-1252" Content-Transfer-Encoding: 7bit Message-ID: <201208151521.14663.arnd@arndb.de> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Catalin Marinas Cc: linux-arch@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, Marc Zyngier , Will Deacon Message-ID: <20120815152114.ebre8H2uuhCWcLLA9gDCETY2QhbLY_a6D7wHXxrWnPw@z> On Tuesday 14 August 2012, Catalin Marinas wrote: > + > +/* > + * Use compiler builtins for simple inline operations. > + */ > +static inline unsigned long __ffs(unsigned long word) > +{ > + return __builtin_ffsl(word) - 1; > +} > + > +static inline int ffs(int x) > +{ > + return __builtin_ffs(x); > +} > + > +static inline unsigned long __fls(unsigned long word) > +{ > + return BITS_PER_LONG - 1 - __builtin_clzl(word); > +} > + > +static inline int fls(int x) > +{ > + return x ? sizeof(x) * BITS_PER_BYTE - __builtin_clz(x) : 0; > +} These are all great, but I think whether to use them or not should depend on the compiler version rather than the architecture in general. Do we know a minimum gcc version that supports all of the above? Then we could put that code into the generic files. If that's not possible, we could still make the implementation available for other architectures by moving it to asm-generic/bitops/builtin-__ffs.h asm-generic/bitops/builtin-ffs.h asm-generic/bitops/builtin-__fls.h asm-generic/bitops/builtin-fls.h > --- /dev/null > +++ b/arch/arm64/lib/bitops.c > @@ -0,0 +1,25 @@ > +/* > + * Copyright (C) 2012 ARM Limited > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License version 2 as > + * published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + * You should have received a copy of the GNU General Public License > + * along with this program. If not, see . > + */ > + > +#include > +#include > +#include > + > +#ifdef CONFIG_SMP > +arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned = { > + [0 ... (ATOMIC_HASH_SIZE-1)] = __ARCH_SPIN_LOCK_UNLOCKED > +}; > +#endif What? I suppose this is a leftover from an earlier version using the generic bitops, right? Arnd