RE: [PATCH] Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-10-13 Thread Travis Vitek
: Liviu Nicoara Sent: Thursday, October 11, 2012 5:28 AM To: dev@stdcxx.apache.org Subject: Re: [PATCH] Re: STDCXX-1072 SPARC V8 mutex alignment requirements I applied the patch on 4.2.x. If someone with access to a SPARC machine could give it a runall and post the results here it would

Re: [PATCH] Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-10-13 Thread Liviu Nicoara
] 140 % I also tested with POSIX mutexes and saw the same behavior. Travis From: Liviu Nicoara Sent: Thursday, October 11, 2012 5:28 AM To: dev@stdcxx.apache.org Subject: Re: [PATCH] Re: STDCXX-1072 SPARC V8 mutex alignment requirements I applied the patch on 4.2.x. If someone with access

Re: [PATCH] Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-10-11 Thread Liviu Nicoara
I applied the patch on 4.2.x. If someone with access to a SPARC machine could give it a runall and post the results here it would be awesome. I will postpone closing the incident until then. Thanks! Liviu On 10/06/12 16:56, Liviu Nicoara wrote: On 09/29/12 15:33, Liviu Nicoara wrote: On

[PATCH] Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-29 Thread Liviu Nicoara
On 9/28/12 11:32 AM, Travis Vitek wrote: -Original Message- From: Liviu Nicoara Sent: Friday, September 28, 2012 5:29 AM [...] The patch assumes the type `long double' exists on every platform. While I do believe that it is available everywhere, we have lots of conditional code

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Liviu Nicoara
On 09/28/12 08:29, Liviu Nicoara wrote: I have created the above and linked it to the closed STDCXX-1066. [...] IMO, the patch I attached does not break binary compatibility. Scratch this, I haven't thought it through. Thanks, Liviu

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Liviu Nicoara
On 09/28/12 08:45, Liviu Nicoara wrote: On 09/28/12 08:29, Liviu Nicoara wrote: I have created the above and linked it to the closed STDCXX-1066. [...] IMO, the patch I attached does not break binary compatibility. Scratch this, I haven't thought it through. Actually, after more thought, I

RE: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Travis Vitek
-Original Message- From: Liviu Nicoara Sent: Friday, September 28, 2012 5:29 AM In short, my reading about this issue is that the kernel patch changed the alignment of the userland mutex objects from a machine word to a double-word boundary. No changes are required of the users

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Martin Sebor
The patch looks reasonable to me, except for the missing guard for _RWSTD_NO_LONG_DOUBLE. For C++ 11 compilers, we might want to replace the union with the alignas features. Of course, that will require another configuration test and macro, and most likely won't help the current Sun Studio

RE: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Travis Vitek
Liviu, Sorry I didn't look until just now, but it appears that I could have re-opened STDCXX-1066. I only see the 'Reopen Issue' button for STDCXX issues, but it is most definitely there. Perhaps there is some sort of permission issue for you? Also, STDCXX-1066 appears to have been a duplicate

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Martin Sebor
On 09/28/2012 09:32 AM, Travis Vitek wrote: -Original Message- From: Liviu Nicoara Sent: Friday, September 28, 2012 5:29 AM In short, my reading about this issue is that the kernel patch changed the alignment of the userland mutex objects from a machine word to a double-word

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Liviu Nicoara
On 09/28/12 11:45, Travis Vitek wrote: Liviu, Sorry I didn't look until just now, but it appears that I could have re-opened STDCXX-1066. I only see the 'Reopen Issue' button for STDCXX issues, but it is most definitely there. Perhaps there is some sort of permission issue for you? It's ok,

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Liviu Nicoara
On 09/28/12 11:32, Travis Vitek wrote: -Original Message- From: Liviu Nicoara Sent: Friday, September 28, 2012 5:29 AM In short, my reading about this issue is that the kernel patch changed the alignment of the userland mutex objects from a machine word to a double-word boundary.

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Martin Sebor
On 09/28/2012 11:27 AM, Liviu Nicoara wrote: On 09/28/12 11:45, Travis Vitek wrote: Liviu, Sorry I didn't look until just now, but it appears that I could have re-opened STDCXX-1066. I only see the 'Reopen Issue' button for STDCXX issues, but it is most definitely there. Perhaps there is some

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Liviu Nicoara
On 09/28/12 13:51, Martin Sebor wrote: [...] One other comment: I would suggest choosing subjects for bug reports that reflect the problem rather than a fix for it or a rationale for it. For STDCXX-1066 I think something like Library mutex objects misaligned on SPARCV8 would better capture the

Re: STDCXX-1072 SPARC V8 mutex alignment requirements

2012-09-28 Thread Martin Sebor
On 09/28/2012 11:55 AM, Liviu Nicoara wrote: On 09/28/12 13:51, Martin Sebor wrote: [...] One other comment: I would suggest choosing subjects for bug reports that reflect the problem rather than a fix for it or a rationale for it. For STDCXX-1066 I think something like Library mutex objects