View previous topic :: View next topic |
Author |
Message |
thegreenhat n00b
Joined: 06 Jul 2018 Posts: 3
|
Posted: Fri Jul 06, 2018 6:49 pm Post subject: an obnoxious issue with emerging libunistring-0.9.7 |
|
|
Solved
Last edited by thegreenhat on Sat Jul 07, 2018 2:57 pm; edited 1 time in total |
|
Back to top |
|
|
John R. Graham Administrator
Joined: 08 Mar 2005 Posts: 10589 Location: Somewhere over Atlanta, Georgia
|
Posted: Fri Jul 06, 2018 7:13 pm Post subject: |
|
|
Builds here error free with the same toolchain. Sometime makefiles have unintended parallelism errors, though. Try the libunistring emerge again with: Code: | MAKEOPTS="-j1" emerge -1v libunistring | to see if that is the case here.
- John _________________ I can confirm that I have received between 0 and 499 National Security Letters. |
|
Back to top |
|
|
Hu Moderator
Joined: 06 Mar 2007 Posts: 21635
|
Posted: Sat Jul 07, 2018 12:19 am Post subject: |
|
|
That does not look like a parallelism bug to me. Rather, it looks like an upstream bug due to mismatched declarations. For some platforms, size_t might be a typedef alias for unsigned int, which would allow this to work despite being wrong. For others, it would alias some other type and this failure would occur. Preferably, the declarations should match exactly. Ideally, there should be exactly one declaration and exactly one definition, and those would match exactly. |
|
Back to top |
|
|
thegreenhat n00b
Joined: 06 Jul 2018 Posts: 3
|
Posted: Sat Jul 07, 2018 10:19 am Post subject: |
|
|
John R. Graham wrote: | Builds here error free with the same toolchain. Sometime makefiles have unintended parallelism errors, though. Try the libunistring emerge again with: Code: | MAKEOPTS="-j1" emerge -1v libunistring | to see if that is the case here.
- John |
Not the case, still does not work :/.
Hu wrote: | That does not look like a parallelism bug to me. Rather, it looks like an upstream bug due to mismatched declarations. For some platforms, size_t might be a typedef alias for unsigned int, which would allow this to work despite being wrong. For others, it would alias some other type and this failure would occur. Preferably, the declarations should match exactly. Ideally, there should be exactly one declaration and exactly one definition, and those would match exactly. |
What should I do though? Should I edit libunistring by myself?
*Notes*
To fix this issue just compile the 0.9.10 version |
|
Back to top |
|
|
Hu Moderator
Joined: 06 Mar 2007 Posts: 21635
|
Posted: Sat Jul 07, 2018 3:18 pm Post subject: |
|
|
Since it is fixed in a later version, encouraging stabilization of that version is probably the simplest solution.
If it were not fixed, I would suggest that you edit it yourself to verify that my explanation is correct and produces a working build. If it does, then bug upstream to fix it and supply your change as a demonstration of what is needed. You could also post the change here for our review/critique before going upstream if you are concerned about whether you did the right change. |
|
Back to top |
|
|
thegreenhat n00b
Joined: 06 Jul 2018 Posts: 3
|
Posted: Sat Jul 07, 2018 6:00 pm Post subject: |
|
|
Hu wrote: | Since it is fixed in a later version, encouraging stabilization of that version is probably the simplest solution.
If it were not fixed, I would suggest that you edit it yourself to verify that my explanation is correct and produces a working build. If it does, then bug upstream to fix it and supply your change as a demonstration of what is needed. You could also post the change here for our review/critique before going upstream if you are concerned about whether you did the right change. |
It might sounds a little agonizing, however, I do believe the issue is due to some miscellaneous components in my system.
Such assortment is causing many unique errors.
In my opinion, stabilization the latest version is a valid solution |
|
Back to top |
|
|
Ant P. Watchman
Joined: 18 Apr 2009 Posts: 6920
|
Posted: Wed Sep 05, 2018 6:48 pm Post subject: Re: an obnoxious issue with emerging libunistring-0.9.7 |
|
|
thegreenhat wrote: | Solved :) |
Please don't delete questions; others may be searching for an answer to whatever problem you had. |
|
Back to top |
|
|
|