Re: [Qemu-devel] [PATCH] scsi: Don't deference in_buf if NULL

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1265 bytes --]

On 01/03/2018 08:42 PM, Fam Zheng wrote:
> scsi_disk_emulate_command passes in_buf=NULL and in_len=0 in the
> REQUEST_SENSE branch. Inline the fixed_in evaluation and put it after
> the in_len test.
> 
> Signed-off-by: Fam Zheng <famz@redhat.com>
> ---
>  scsi/utils.c | 4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/scsi/utils.c b/scsi/utils.c
> index ddae650a99..9a0a925ef9 100644
> --- a/scsi/utils.c
> +++ b/scsi/utils.c
> @@ -320,10 +320,8 @@ int scsi_convert_sense(uint8_t *in_buf, int in_len,
>                         uint8_t *buf, int len, bool fixed)
>  {
>      SCSISense sense;
> -    bool fixed_in;
>  
> -    fixed_in = (in_buf[0] & 2) == 0;
> -    if (in_len && fixed == fixed_in) {
> +    if (in_len && !!fixed == ((in_buf[0] & 2) == 0)) {

Huh?  'fixed' is already a bool, so '!!fixed' is just extra typing that
makes things harder to read.  Did you mean:

if (in_len && fixed == !(in_buf[0] & 2))

as something that is slightly more legible (the LHS is already bool, the
RHS uses a single ! to convert a bitwise test into a bool with the
correct sense)?

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 619 bytes --]