Re: [PATCH] net/9p: fix endian issues

[Harvey Harrison <harvey.harrison@gmail.com>]

On Tue, 2009-02-03 at 09:51 -0600, ericvh@gmail.com wrote:
> From: Eric Van Hensbergen <ericvh@gmail.com>
> 
> When the changes were done to the protocol last release, some endian
> bugs crept in.  This patch fixes those endian problems and has been
> verified to run on 32/64 bit and x86/ppc architectures.
> 
> Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
> ---
>  net/9p/protocol.c |   12 ++++++------
>  1 files changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/net/9p/protocol.c b/net/9p/protocol.c
> index dcd7666..19ad5d2 100644
> --- a/net/9p/protocol.c
> +++ b/net/9p/protocol.c
> @@ -164,7 +164,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int optional, const char *fmt, va_list ap)
>  					errcode = -EFAULT;
>  					break;
>  				}
> -				*val = cpu_to_le16(*val);
> +				*val = le16_to_cpu(*val);
>  			}
>  			break;
>  		case 'd':{
> @@ -173,7 +173,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int optional, const char *fmt, va_list ap)
>  					errcode = -EFAULT;
>  					break;
>  				}
> -				*val = cpu_to_le32(*val);
> +				*val = le32_to_cpu(*val);
>  			}
>  			break;
>  		case 'q':{
> @@ -182,7 +182,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int optional, const char *fmt, va_list ap)
>  					errcode = -EFAULT;
>  					break;
>  				}
> -				*val = cpu_to_le64(*val);
> +				*val = le64_to_cpu(*val);
>  			}

Umm, the above three have absolutely no functional effect you know.  And you're just
trading one set of sparse warnings for another.

>  			break;
>  		case 's':{
> @@ -362,19 +362,19 @@ p9pdu_vwritef(struct p9_fcall *pdu, int optional, const char *fmt, va_list ap)
>  			}
>  			break;
>  		case 'w':{
> -				int16_t val = va_arg(ap, int);
> +				int16_t val = cpu_to_le16(va_arg(ap, int));
>  				if (pdu_write(pdu, &val, sizeof(val)))
>  					errcode = -EFAULT;
>  			}
>  			break;
>  		case 'd':{
> -				int32_t val = va_arg(ap, int32_t);
> +				int32_t val = cpu_to_le32(va_arg(ap, int32_t));
>  				if (pdu_write(pdu, &val, sizeof(val)))
>  					errcode = -EFAULT;
>  			}
>  			break;
>  		case 'q':{
> -				int64_t val = va_arg(ap, int64_t);
> +				int64_t val = cpu_to_le64(va_arg(ap, int64_t));
>  				if (pdu_write(pdu, &val, sizeof(val)))
>  					errcode = -EFAULT;
>  			}

And here even if you are fixing endian bugs, you're actively _introducing_
sparse warnings, it would be nice if you could annotate the variable types
properly too.

Cheers,

Harvey