From mboxrd@z Thu Jan 1 00:00:00 1970 From: Russell King Date: Wed, 01 Feb 2006 19:19:57 +0000 Subject: Re: [PATCH 1/12] generic *_bit() Message-Id: <20060201191957.GG3072@flint.arm.linux.org.uk> List-Id: References: <20060201180237.GA18464@infradead.org> <200602011807.k11I7ag15563@unix-os.sc.intel.com> In-Reply-To: <200602011807.k11I7ag15563@unix-os.sc.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: "Chen, Kenneth W" Cc: 'Christoph Hellwig' , 'Akinobu Mita' , Grant Grundler , Linux Kernel Development , linux-ia64@vger.kernel.org On Wed, Feb 01, 2006 at 10:07:28AM -0800, Chen, Kenneth W wrote: > Christoph Hellwig wrote on Wednesday, February 01, 2006 10:03 AM > > > Akinobu Mita wrote on Wednesday, January 25, 2006 7:29 PM > > > > This patch introduces the C-language equivalents of the functions below: > > > > > > > > - atomic operation: > > > > void set_bit(int nr, volatile unsigned long *addr); > > > > void clear_bit(int nr, volatile unsigned long *addr); > > > > void change_bit(int nr, volatile unsigned long *addr); > > > > int test_and_set_bit(int nr, volatile unsigned long *addr); > > > > int test_and_clear_bit(int nr, volatile unsigned long *addr); > > > > int test_and_change_bit(int nr, volatile unsigned long *addr); > > > > > > I wonder why you did not make these functions take volatile > > > unsigned int * address argument? > > > > Because they are defined to operate on arrays of unsigned long > > I think these should be defined to operate on arrays of unsigned int. > Bit is a bit, no matter how many byte you load (8/16/32/64), you can > only operate on just one bit. Invalid assumption, from the point of view of endianness across different architectures. Consider where bit 0 is for a LE and BE unsigned long * vs a LE and BE unsigned char *. -- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: 2.6 Serial core