New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fails to build with GCC-6 #56
Comments
maybe someone could propose this patch to dcoffin: https://gist.github.com/conchurnavid/ac19c8e882f1835f0310 |
Thank you for the patch. To be included in next Libraw 0.18 update |
Unfortunately, this patch will broke 4-color raw processing. To be reverted. |
Then you should re-open the issue, since it's not fixed yet. |
This is different issue. Patch (discussed above) fixes gcc-6 problems. These problems are fixed in current github branch, so issue is closed. |
Could you please point me to the patch fixing the GCC-6 FTBFS? It's considered a RC bug in Debian and LibRaw would be removed from testing/stretch suite due to this. |
This patch is already in LibRaw-master at github (this repo). |
Perfect. Thanks a lot. Il mer 27 lug 2016, 22:33 LibRaw LLC notifications@github.com ha scritto:
Matteo F. Vescovi |
I think I am the 0.01% who noticed that :) (http://www.libraw.org/node/2214) Has the bug been reported on debian ? |
we'll issue new public patch that fixes gcc6 incompatibility (with working vng) today. And, sure, we'll contact debian after that |
Wonderful :) |
Commited: 8a8723b |
@mfvescovi please re-patch debian copy of libraw 0.17.2 , there are users already affected by old broken patch. |
d690068 removes one extra warning (not significant :) |
0.17 branch: c3bf298 |
@LibRaw With the new patch LibRaw is FTBFS on all little-endian, as you can see at https://buildd.debian.org/libraw |
OK, let's replace type conversion with explicit value: 94ee55e Tested on both 3-color and 4-color images, but only on intel CPU |
@LibRaw in fact this new change fails again, see http://debomatic-armel.debian.net/distribution#unstable/libraw/0.17.2-6/buildlog as an example. Same story for armhf, powerpc and ppc64el, which I've tested before replying here. |
Sorry, this is different error. I thought this is the same vng_interpolate issue, but line 2491 is in kodak_radc_load_raw() and resulted from: Is char is unsigned by default in arm/etc architectures? |
As stated in https://wiki.debian.org/ArchitectureSpecificsMemo, all failing architectures have unsigned char. |
this patch should help: 4075372 |
@LibRaw much much better now: https://buildd.debian.org/libraw |
Hello. OK, as soon as the -6 is in testing I'll test it. |
I can confirm that the bug is fixed. Thank you. |
More info can be found at https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=811744
The text was updated successfully, but these errors were encountered: