On 10/26/2015 06:54 PM, Eric Blake wrote:
> On 10/26/2015 04:06 PM, John Snow wrote:
>> Signed-off-by: John Snow <js...@redhat.com>
>> ---
>> qemu-io-cmds.c | 58
>> +++++++++++++++++++++++++++++++++++++---------------------
>> 1 file changed, 37 insertions(+), 21 deletions(-)
>>
>> diff --git a/qemu-io-cmds.c b/qemu-io-cmds.c
>> index e2477fc..92c6b87 100644
>> --- a/qemu-io-cmds.c
>> +++ b/qemu-io-cmds.c
>> @@ -146,6 +146,21 @@ static int64_t cvtnum(const char *s)
>> return ret;
>> }
>>
>> +static void print_cvtnum_err(int64_t rc, const char *arg)
>> +{
>> + switch (rc) {
>> + case -EINVAL:
>> + printf("Parsing error: non-numeric argument,"
>> + " or extraneous/unrecognized suffix -- %s\n", arg);
>
> s/ --/:/
>
>> + break;
>> + case -ERANGE:
>> + printf("Parsing error: argument too large -- %s\n", arg);
>> + break;
>> + default:
>> + printf("Parsing error -- %s\n", arg);
>
> Twice more.
>
> With that change,
> Reviewed-by: Eric Blake <ebl...@redhat.com>
>
>> @@ -2199,10 +2214,11 @@ static int sigraise_f(BlockBackend *blk, int argc,
>> char **argv)
>> {
>> int64_t sig = cvtnum(argv[1]);
>> if (sig < 0) {
>> - printf("non-numeric signal number argument -- %s\n", argv[1]);
>> + print_cvtnum_err(sig, argv[1]);
>> return 0;
>> - } else if (sig > INT_MAX) {
>> - printf("signal argument '%s' is too large\n", argv[1]);
>> + } else if (sig > NSIG) {
>> + printf("signal argument '%s' is too large to be a valid signal\n",
>> + argv[1]);
>
> Should the comparison against NSIG rather than INT_MAX be squashed into
> patch 1?
>
Yes.
