Re: [RFC 0/2] tls: add macros for coroutine-safe TLS variables

[Thomas Huth <thuth@redhat.com>]

On 26/10/2021 19.10, Richard Henderson wrote:
> On 10/26/21 9:34 AM, Stefan Hajnoczi wrote:
>> On Tue, Oct 26, 2021 at 08:10:16AM -0700, Richard Henderson wrote:
>>> On 10/26/21 6:22 AM, Stefan Hajnoczi wrote:
>>>> If "safe" TLS variables are opt-in then we'll likely have obscure bugs
>>>> when code changes to access a TLS variable that was previously never
>>>> accessed from a coroutine. There is no compiler error and no way to
>>>> detect this. When it happens debugging it is painful.
>>>
>>> Co-routines are never used in user-only builds.
>>
>> If developers have the choice of using __thread then bugs can slip
>> through.
> 
> Huh?  How.  No, really.
> 
>> Are you concerned about performance, the awkwardness of calling
>> getters/setters, or something else for qemu-user?
> 
> Awkwardness first, performance second.
> 
> I'll also note that coroutines never run on vcpu threads, only io threads.  
> So I'll resist any use of these interfaces in TCG as well.

Would it maybe make sense to tweak check_patch.pl to forbid __thread in 
certain folders only, e.g. block/ and util/ (i.e. where we know that there 
might be code that the iothreads are using)?

  Thomas