On 2016/01/25 19:17, Xie, Huawei wrote:
> On 1/21/2016 7:09 PM, Tetsuya Mukawa wrote:
>> +static void
>> +qtest_handle_one_message(struct qtest_session *s, char *buf)
>> +{
>> + int ret;
>> +
>> + if (strncmp(buf, interrupt_message, strlen(interrupt_message)) == 0) {
>> + if (rte_atomic16_read(&s->enable_intr) == 0)
>> + return;
>> +
>> + /* relay interrupt to pipe */
>> + ret = write(s->irqfds.writefd, "1", 1);
> How about the interrupt latency? Seems it is quite long.
Yes, I agree with it.
Probably using evetfd or removing this read/write mechanism to handle
interrupts will be nice.
Let me check it more.
Tetsuya
