[
https://issues.apache.org/jira/browse/THRIFT-4552?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16437025#comment-16437025
]
Can Celasun commented on THRIFT-4552:
-------------------------------------
Let me try to give an example:
{code:go}
func main() {
wg := &sync.WaitGroup{}
go work(wg)
wg.Wait()
}
func work(wg *sync.WaitGroup) {
wg.Add(1)
// stuff
defer wg.Done()
}
{code}
Here, there is a race conditon because if {{wg.Wait()}} executes before
{{wg.Add(1)}}, Wait will panic as
[documented|https://godoc.org/sync#WaitGroup.Add]:
{quote}Note that calls with a positive delta that occur when the counter is
zero must happen before a Wait.{quote}
Still, I agree with you that this has a non-neglible performance impact. Let me
try a few things and post back there.
> why acquire a lock in TSimpleServer implementation for Go?
> ----------------------------------------------------------
>
> Key: THRIFT-4552
> URL: https://issues.apache.org/jira/browse/THRIFT-4552
> Project: Thrift
> Issue Type: Improvement
> Affects Versions: 0.11.0
> Reporter: gansteed
> Priority: Major
>
> I've sent a email to groups, but I think maybe here will be better?
>
> I'm using Thrift and I'm reading thrift implementation for Go, I found code
> in `TSimpleServer.AcceptLoop` like this:
>
> {code:go}
> func (p *TSimpleServer) AcceptLoop() error {
> for {
> client, err := p.serverTransport.Accept()
> p.mu.Lock()
> if atomic.LoadInt32(&p.closed) != 0 {
> return nil
> }
> if err != nil {
> return err
> }
> if client != nil {
> p.wg.Add(1)
> go func() {
> defer p.wg.Done()
> if err := p.processRequests(client); err !=
> nil {
> log.Println("error processing
> request:", err)
> }
> }()
> }
> p.mu.Unlock()
> }
> }
> {code}
>
> every time it accept a request,it:
>
> 1. read if protocol had been closed, this step is atomic, it does not need a
> lock.
> 2. p.wg.Add(1) to accumulate a goroutine? this step is atomic, too, it does
> not need a lock
> 3. after processor processed the request, it do p.wg.Done(), it's atomic,
> too, and it does not need a lock.
>
> by the way, it seems that `p.wg.Done()` do not need to put in defer? just put
> it after p.processRequests(client)?
>
> so is there any particular to do it in this way?if not, I would like to
> submit a PR to reduce unneccessary performance overhead in TSimpleServer
> implementation.
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
