[swift-users] Is this really a race condition?

[Edward Connell]

Hi Guillaume,
There is still a race condition report, but I think maybe we are almost
there. I had to modify your example because it didn't quite build, and a
loop to do it twice since it randomly complains if you only do it once.
Btw, this is on Linux.
The warning is triggered when deinitialize is called. It wasn't in your
example, but it seems that it is required since the Elements are not a
trivial type like Int.

Thanks, Ed

for _ in 0..<2 {
let count = 1000
let items = UnsafeMutablePointer<[UInt8]?>.allocate(capacity: count)
items.initialize(to: nil, count: count)

DispatchQueue.concurrentPerform(iterations: count) {
items[$0] = [UInt8](repeating: 7, count: 10)
}

items.deinitialize(count: count)
items.deallocate(capacity: count)
}

==================
WARNING: ThreadSanitizer: data race (pid=24076)
  Write of size 8 at 0x7d0c00005fa0 by main thread:
    #0 free <null> (libtsan.so.0+0x000000025819)
    #1 _TwXxOs31_ClosedRangeIndexRepresentation <null>
(libswiftCore.so+0x00000027e071)

  Previous write of size 8 at 0x7d0c00005fa0 by thread T20:
    #0 malloc <null> (libtsan.so.0+0x0000000254a3)
    #1 swift_slowAlloc <null> (libswiftCore.so+0x0000002ee3e5)

  Thread T20 (tid=24288, running) created by main thread at:
    #0 pthread_create <null> (libtsan.so.0+0x000000027577)
    #1 manager_workqueue_additem
/home/buildnode/disk2/workspace/oss-swift-3.0-package-linux-ubuntu-16_04/swift-corelibs-libdispatch/libpwq/src/posix/manager.c:815
(libdispatch.so+0x00000007c6b1)

SUMMARY: ThreadSanitizer: data race ??:0 __interceptor_free
==================


ThreadSanitizer: reported 1 warnings



On Wed, Mar 1, 2017 at 6:48 PM, Edward Connell <ewconnell at gmail.com> wrote:

> Ahh! thank you. That makes sense.
>
> On Wed, Mar 1, 2017 at 3:29 PM, Guillaume Lessard <
> glessard at tffenterprises.com> wrote:
>
>>
>> > On Mar 1, 2017, at 3:21 PM, Edward Connell via swift-users <
>> swift-users at swift.org> wrote:
>> >
>> > The thread sanitizer on Linux is reporting that I have race conditions
>> in libswiftcore. I eliminated enough code down to this trivial example. Is
>> there really a race condition here or are these bogus errors?
>> >
>> >       let count = 1000
>> >       var items = [[UInt8]?](repeating: nil, count: count)
>> >
>> >       DispatchQueue.concurrentPerform(iterations: count) {
>> >               items[$0] = [UInt8](repeating: 7, count: 10)
>> >       }
>> >
>> > My real scenario is retrieving data asynchronously, so I just threw in
>> a buffer assignment.
>>
>> The assignments to array elements are where the race lies.
>>
>> I don’t know about the libswiftcore part, but: assigning to a shared
>> Array concurrently from multiple threads won't work, because of Array's
>> copy-on-write behaviour. You could do
>>
>> let items = UnsafeMutablePointer<[UInt8]?>.allocate(capacity: 1)
>> items.initialize(to: nil, count: count)
>>
>> DispatchQueue.concurrentPerform(iterations: count) {
>>   items[$0].initialize([UInt8](repeating: 7, count: 10))
>> }
>>
>> // you’ll be able to see here that they’re all initialized
>>
>> items.deallocate(capacity: count)
>>
>> Cheers,
>> Guillaume Lessard
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.swift.org/pipermail/swift-users/attachments/20170302/4707baac/attachment.html>