Latest upstream release: 4.60 Current version/release in rawhide: 4.59-1.fc25 URL: http://www.gnu.org/software/glpk/ Please consult the package updates policy before you issue an update to a stable branch: https://fedoraproject.org/wiki/Updates_Policy More information about the service that created this bug can be found at: https://fedoraproject.org/wiki/Upstream_release_monitoring Please keep in mind that with any upstream change, there may also be packaging changes that need to be made. Specifically, please remember that it is your responsibility to review the new version to ensure that the licensing is still correct and that no non-free or legally problematic items have been added upstream. Based on the information from anitya: https://release-monitoring.org/project/1183/
Patching or scratch build for glpk and version 4.59 FAILED. See for details
Created attachment 1142777 [details] Rebase-helper rebase-helper-debug.log log file. See for details and report the eventual error to rebase-helper https://github.com/phracek/rebase-helper/issues.
Patches were not touched. All were applied properly
Latest upstream release: 4.61 Current version/release in rawhide: 4.59-1.fc25 URL: http://www.gnu.org/software/glpk/ Please consult the package updates policy before you issue an update to a stable branch: https://fedoraproject.org/wiki/Updates_Policy More information about the service that created this bug can be found at: https://fedoraproject.org/wiki/Upstream_release_monitoring Please keep in mind that with any upstream change, there may also be packaging changes that need to be made. Specifically, please remember that it is your responsibility to review the new version to ensure that the licensing is still correct and that no non-free or legally problematic items have been added upstream. Based on the information from anitya: https://release-monitoring.org/project/1183/
Patching or scratch build for glpk-4.59 failed.
Created attachment 1243484 [details] Rebase-helper rebase-helper-debug.log log file. See for details and report the eventual error to rebase-helper https://github.com/phracek/rebase-helper/issues.
Following patches has been deleted: ['glpk-4.59-unbundle-zlib.diff']
Following patches has been modified: ['/var/tmp/thn-rhuNsgEG/rebase-helper-results/glpk-4.59-unbundle-zlib.diff']
We've encountered a bug in 4.59 that has been resolved in 4.61. Is there a timeline for when 4.61 packages might become available? I'm happy to assist with the packaging if that's helpful too but I'd rather use something that's been accepted into Fedora vs. my own custom build.
I have a patch which updates glpk-4.59-unbundle-zlib.diff to apply cleanly to 4.61. I'm able to successfully run `fedpkg local` on a local centos test installation. I'd love to share this work with Fedora to get 4.61 support merged. Please let me know how to proceed.
Here's a gist containing the updated patch: https://gist.github.com/bfallik/4fb2f7e988f9940eb30891c087a70fb5
here's a copr test build: https://copr.fedorainfracloud.org/coprs/bfallik/glpk/build/534673/
It looks like glpk also bundles minisat version 1 and some version of suitesparse. We have had minisat version 2 in Fedora for the better part of a decade now, so I guess if version 1 is really needed the bundling is unavoidable. But in that case, the spec file should include: Provides: bundled(minisat) = 1.14.1 I wonder if the suitesparse bundling is really necessary. If it is, then a similar Provides should be added; if it isn't, then suitesparse should be unbundled as well. Also, past updates to glpk have sometimes broken python-cvxopt builds. Brian, have you checked if glpk consumers rebuild successfully with glpk 4.61?
Thanks for you input, Jerry. (In reply to Jerry James from comment #13) > It looks like glpk also bundles minisat version 1 and some version of > suitesparse. We have had minisat version 2 in Fedora for the better part of > a decade now, so I guess if version 1 is really needed the bundling is > unavoidable. But in that case, the spec file should include: > > Provides: bundled(minisat) = 1.14.1 Yes, I see ministat and I can retest with the Provides line. I don't see suiteparse though. Is this the library http://faculty.cse.tamu.edu/davis/suitesparse.html? If so I don't see any reference to it in the source tarball. > > I wonder if the suitesparse bundling is really necessary. If it is, then a > similar Provides should be added; if it isn't, then suitesparse should be > unbundled as well. > > Also, past updates to glpk have sometimes broken python-cvxopt builds. > Brian, have you checked if glpk consumers rebuild successfully with glpk > 4.61? I have not checked. What's the process? I can try to rebuild the python-cvxopt package on a system with 4.61 but I wonder if that's enough to catch breaking changes within the source code. Any tips here would be appreciated. brian
(In reply to Brian from comment #14) > I don't see suiteparse though. Is this the library > http://faculty.cse.tamu.edu/davis/suitesparse.html? If so I don't see any > reference to it in the source tarball. The amd and colamd directories are lifted from the suitesparse distribution. This spec file needs to be cleaned up and modernized in several ways. I'll attach a suggestion for what I think should be done. > I have not checked. What's the process? I can try to rebuild the > python-cvxopt package on a system with 4.61 but I wonder if that's enough to > catch breaking changes within the source code. Any tips here would be > appreciated. Personally, I like to use libabigail for this kind of thing. It requires building the new version (I like to use mock for that), and downloading the previous build from koji. The easiest way to do that is to visit http://bugz.fedoraproject.org/glpk/ and follow the "Koji Builds" link on the lower left. Then you run the libabigail tool like this: $ abipkgdiff --d1 glpk-debuginfo-4.59-2.fc26.x86_64.rpm --d2 glpk-debuginfo-4.61-1.fc27.x86_64.rpm glpk-4.59-2.fc26.x86_64.rpm glpk-4.61-1.fc27.x86_64.rpm ================ changes of 'libglpk.so.40.0.0'=============== Functions changes summary: 0 Removed, 1 Changed (157 filtered out), 1 Added functions Variables changes summary: 0 Removed, 0 Changed, 0 Added variable 1 Added function: 'function const char* glp_config(const char*)' {glp_config} 1 function with some indirect sub-type change: [C]'function int glp_add_cols(glp_prob*, int)' at prob1.c:354:1 has some indirect sub-type changes: parameter 1 of type 'glp_prob*' has sub-type changes: in pointed to type 'typedef glp_prob' at glpk.h:39:1: underlying type 'struct glp_prob' at prob.h:43:1 changed: type size changed from 1536 to 1472 bits 1 data member deletion: 'unsigned int glp_prob::magic', at offset 0 (in bits) at prob.h:41:1 27 data member changes: 'DMP* glp_prob::pool' offset changed from 64 to 0 (in bits) (by -64 bits) type of 'glp_tree* glp_prob::tree' changed: in pointed to type 'typedef glp_tree' at glpk.h:154:1: underlying type 'struct glp_tree' at glpios.h:39:1 changed: type size hasn't changed 1 data member changes (1 filtered): type of 'const glp_iocp* glp_tree::parm' changed: in pointed to type 'const glp_iocp': in unqualified underlying type 'typedef glp_iocp' at glpk.h:204:1: underlying type 'struct __anonymous_struct__' at glpk.h:157:1 changed: type size changed from 2688 to 2624 bits 1 data member insertion: 'int __anonymous_struct__::flip', at offset 1120 (in bits) at glpk.h:201:1 1 data member changes (1 filtered): type of 'double __anonymous_struct__::foo_bar[24]' changed: type name changed from 'double[24]' to 'double[23]' array type size changed from 1536 to 1472 bits: array type subrange 1 changed length from 24 to 23 and offset changed from 128 to 64 (in bits) (by -64 bits) 'char* glp_prob::name' offset changed from 192 to 128 (in bits) (by -64 bits) 'char* glp_prob::obj' offset changed from 256 to 192 (in bits) (by -64 bits) 'int glp_prob::dir' offset changed from 320 to 256 (in bits) (by -64 bits) 'double glp_prob::c0' offset changed from 384 to 320 (in bits) (by -64 bits) 'int glp_prob::m_max' offset changed from 448 to 384 (in bits) (by -64 bits) 'int glp_prob::n_max' offset changed from 480 to 416 (in bits) (by -64 bits) 'int glp_prob::m' offset changed from 512 to 448 (in bits) (by -64 bits) 'int glp_prob::n' offset changed from 544 to 480 (in bits) (by -64 bits) 'int glp_prob::nnz' offset changed from 576 to 512 (in bits) (by -64 bits) 'GLPROW** glp_prob::row' offset changed from 640 to 576 (in bits) (by -64 bits) 'GLPCOL** glp_prob::col' offset changed from 704 to 640 (in bits) (by -64 bits) 'AVL* glp_prob::r_tree' offset changed from 768 to 704 (in bits) (by -64 bits) 'AVL* glp_prob::c_tree' offset changed from 832 to 768 (in bits) (by -64 bits) 'int glp_prob::valid' offset changed from 896 to 832 (in bits) (by -64 bits) 'int* glp_prob::head' offset changed from 960 to 896 (in bits) (by -64 bits) 'BFD* glp_prob::bfd' offset changed from 1024 to 960 (in bits) (by -64 bits) 'int glp_prob::pbs_stat' offset changed from 1088 to 1024 (in bits) (by -64 bits) 'int glp_prob::dbs_stat' offset changed from 1120 to 1056 (in bits) (by -64 bits) 'double glp_prob::obj_val' offset changed from 1152 to 1088 (in bits) (by -64 bits) 'int glp_prob::it_cnt' offset changed from 1216 to 1152 (in bits) (by -64 bits) 'int glp_prob::some' offset changed from 1248 to 1184 (in bits) (by -64 bits) 'int glp_prob::ipt_stat' offset changed from 1280 to 1216 (in bits) (by -64 bits) 'double glp_prob::ipt_obj' offset changed from 1344 to 1280 (in bits) (by -64 bits) 'int glp_prob::mip_stat' offset changed from 1408 to 1344 (in bits) (by -64 bits) 'double glp_prob::mip_obj' offset changed from 1472 to 1408 (in bits) (by -64 bits) ================ end of changes of 'libglpk.so.40.0.0'=============== That means that any consumers that call glp_add_cols() or use the glp_prob type should probably be rebuilt; other consumers probably don't need to be rebuilt. Here are the packages currently in Fedora that consume glpk: $ dnf repoquery --disablerepo=\* --enablerepo=rawhide-source --arch=src --whatrequires glpk-devel Fedora - Rawhide - Source 2.8 MB/s | 6.7 MB 00:02 Macaulay2-0:1.6-32.fc26.src cbmc-0:5.6-2.fc26.src coin-or-Alps-0:1.5.5-2.fc26.src coin-or-Bcp-0:1.4.3-2.fc26.src coin-or-Bcps-0:0.94.4-2.fc26.src coin-or-Blis-0:0.94.4-2.fc26.src coin-or-Bonmin-0:1.8.4-4.fc26.src coin-or-Cbc-0:2.9.8-2.fc26.src coin-or-Cgl-0:0.59.9-2.fc26.src coin-or-Clp-0:1.16.10-2.fc26.src coin-or-CoinMP-0:1.8.3-2.fc26.src coin-or-Couenne-0:0.5.6-4.fc26.src coin-or-Dip-0:0.92.2-2.fc26.src coin-or-DyLP-0:1.10.3-2.fc26.src coin-or-FlopC++-0:1.2.4-2.fc26.src coin-or-Ipopt-0:3.12.7-2.fc27.src coin-or-OS-0:2.10.1-3.fc26.src coin-or-Osi-0:0.107.8-2.fc26.src coin-or-SYMPHONY-0:5.6.14-2.fc26.src coin-or-Vol-0:1.5.3-3.fc26.src coin-or-lemon-0:1.3.1-9.fc26.src latte-integrale-0:1.7.3b-2.fc26.src octave-6:4.2.1-1.fc26.src openms-0:2.1.0-6.20170131gitbde813.fc26.src ppl-0:1.2-2.fc27.src python-cvxopt-0:1.1.9-3.fc26.src sagemath-0:7.4-4.fc27.src shogun-0:5.0.0-0.5.fc26.src I am currently working towards a rebuild of Macaulay2 and sagemath for other reasons. If you need provenpackager help, I don't mind rebuilding glpk and whichever of these packages need rebuilding as a consequence while I'm at it. Do you have time to take a look at these packages and see which would need a rebuild? That would be very helpful.
Created attachment 1268754 [details] Proposed spec file
Created attachment 1268755 [details] Proposed patch to unbundle zlib
Created attachment 1268756 [details] Proposed patch to unbundle suitesparse
Created attachment 1268757 [details] Proposed patch to fix aliasing problems in minisat 1
(In reply to Jerry James from comment #15) > (In reply to Brian from comment #14) > > I don't see suiteparse though. Is this the library > > http://faculty.cse.tamu.edu/davis/suitesparse.html? If so I don't see any > > reference to it in the source tarball. > > The amd and colamd directories are lifted from the suitesparse distribution. > This spec file needs to be cleaned up and modernized in several ways. I'll > attach a suggestion for what I think should be done. > > > I have not checked. What's the process? I can try to rebuild the > > python-cvxopt package on a system with 4.61 but I wonder if that's enough to > > catch breaking changes within the source code. Any tips here would be > > appreciated. > > Personally, I like to use libabigail for this kind of thing. It requires > building the new version (I like to use mock for that), and downloading the > previous build from koji. The easiest way to do that is to visit > http://bugz.fedoraproject.org/glpk/ and follow the "Koji Builds" link on the > lower left. Then you run the libabigail tool like this: > > $ abipkgdiff --d1 glpk-debuginfo-4.59-2.fc26.x86_64.rpm --d2 > glpk-debuginfo-4.61-1.fc27.x86_64.rpm glpk-4.59-2.fc26.x86_64.rpm > glpk-4.61-1.fc27.x86_64.rpm > ================ changes of 'libglpk.so.40.0.0'=============== > Functions changes summary: 0 Removed, 1 Changed (157 filtered out), 1 > Added functions > Variables changes summary: 0 Removed, 0 Changed, 0 Added variable > > 1 Added function: > > 'function const char* glp_config(const char*)' {glp_config} > > 1 function with some indirect sub-type change: > > [C]'function int glp_add_cols(glp_prob*, int)' at prob1.c:354:1 has some > indirect sub-type changes: > parameter 1 of type 'glp_prob*' has sub-type changes: > in pointed to type 'typedef glp_prob' at glpk.h:39:1: > underlying type 'struct glp_prob' at prob.h:43:1 changed: > type size changed from 1536 to 1472 bits > 1 data member deletion: > 'unsigned int glp_prob::magic', at offset 0 (in bits) at > prob.h:41:1 > > 27 data member changes: > 'DMP* glp_prob::pool' offset changed from 64 to 0 (in bits) (by > -64 bits) > type of 'glp_tree* glp_prob::tree' changed: > in pointed to type 'typedef glp_tree' at glpk.h:154:1: > underlying type 'struct glp_tree' at glpios.h:39:1 changed: > type size hasn't changed > 1 data member changes (1 filtered): > type of 'const glp_iocp* glp_tree::parm' changed: > in pointed to type 'const glp_iocp': > in unqualified underlying type 'typedef glp_iocp' at > glpk.h:204:1: > underlying type 'struct __anonymous_struct__' at > glpk.h:157:1 changed: > type size changed from 2688 to 2624 bits > 1 data member insertion: > 'int __anonymous_struct__::flip', at offset > 1120 (in bits) at glpk.h:201:1 > 1 data member changes (1 filtered): > type of 'double > __anonymous_struct__::foo_bar[24]' changed: > type name changed from 'double[24]' to > 'double[23]' > array type size changed from 1536 to 1472 > bits: > array type subrange 1 changed length from 24 > to 23 > > > and offset changed from 128 to 64 (in bits) (by -64 bits) > 'char* glp_prob::name' offset changed from 192 to 128 (in bits) > (by -64 bits) > 'char* glp_prob::obj' offset changed from 256 to 192 (in bits) > (by -64 bits) > 'int glp_prob::dir' offset changed from 320 to 256 (in bits) > (by -64 bits) > 'double glp_prob::c0' offset changed from 384 to 320 (in bits) > (by -64 bits) > 'int glp_prob::m_max' offset changed from 448 to 384 (in bits) > (by -64 bits) > 'int glp_prob::n_max' offset changed from 480 to 416 (in bits) > (by -64 bits) > 'int glp_prob::m' offset changed from 512 to 448 (in bits) (by > -64 bits) > 'int glp_prob::n' offset changed from 544 to 480 (in bits) (by > -64 bits) > 'int glp_prob::nnz' offset changed from 576 to 512 (in bits) > (by -64 bits) > 'GLPROW** glp_prob::row' offset changed from 640 to 576 (in > bits) (by -64 bits) > 'GLPCOL** glp_prob::col' offset changed from 704 to 640 (in > bits) (by -64 bits) > 'AVL* glp_prob::r_tree' offset changed from 768 to 704 (in > bits) (by -64 bits) > 'AVL* glp_prob::c_tree' offset changed from 832 to 768 (in > bits) (by -64 bits) > 'int glp_prob::valid' offset changed from 896 to 832 (in bits) > (by -64 bits) > 'int* glp_prob::head' offset changed from 960 to 896 (in bits) > (by -64 bits) > 'BFD* glp_prob::bfd' offset changed from 1024 to 960 (in bits) > (by -64 bits) > 'int glp_prob::pbs_stat' offset changed from 1088 to 1024 (in > bits) (by -64 bits) > 'int glp_prob::dbs_stat' offset changed from 1120 to 1056 (in > bits) (by -64 bits) > 'double glp_prob::obj_val' offset changed from 1152 to 1088 (in > bits) (by -64 bits) > 'int glp_prob::it_cnt' offset changed from 1216 to 1152 (in > bits) (by -64 bits) > 'int glp_prob::some' offset changed from 1248 to 1184 (in bits) > (by -64 bits) > 'int glp_prob::ipt_stat' offset changed from 1280 to 1216 (in > bits) (by -64 bits) > 'double glp_prob::ipt_obj' offset changed from 1344 to 1280 (in > bits) (by -64 bits) > 'int glp_prob::mip_stat' offset changed from 1408 to 1344 (in > bits) (by -64 bits) > 'double glp_prob::mip_obj' offset changed from 1472 to 1408 (in > bits) (by -64 bits) > > > ================ end of changes of 'libglpk.so.40.0.0'=============== > > > That means that any consumers that call glp_add_cols() or use the glp_prob > type should probably be rebuilt; other consumers probably don't need to be > rebuilt. Here are the packages currently in Fedora that consume glpk: > > $ dnf repoquery --disablerepo=\* --enablerepo=rawhide-source --arch=src > --whatrequires glpk-devel > Fedora - Rawhide - Source 2.8 MB/s | 6.7 MB 00:02 > Macaulay2-0:1.6-32.fc26.src > cbmc-0:5.6-2.fc26.src > coin-or-Alps-0:1.5.5-2.fc26.src > coin-or-Bcp-0:1.4.3-2.fc26.src > coin-or-Bcps-0:0.94.4-2.fc26.src > coin-or-Blis-0:0.94.4-2.fc26.src > coin-or-Bonmin-0:1.8.4-4.fc26.src > coin-or-Cbc-0:2.9.8-2.fc26.src > coin-or-Cgl-0:0.59.9-2.fc26.src > coin-or-Clp-0:1.16.10-2.fc26.src > coin-or-CoinMP-0:1.8.3-2.fc26.src > coin-or-Couenne-0:0.5.6-4.fc26.src > coin-or-Dip-0:0.92.2-2.fc26.src > coin-or-DyLP-0:1.10.3-2.fc26.src > coin-or-FlopC++-0:1.2.4-2.fc26.src > coin-or-Ipopt-0:3.12.7-2.fc27.src > coin-or-OS-0:2.10.1-3.fc26.src > coin-or-Osi-0:0.107.8-2.fc26.src > coin-or-SYMPHONY-0:5.6.14-2.fc26.src > coin-or-Vol-0:1.5.3-3.fc26.src > coin-or-lemon-0:1.3.1-9.fc26.src > latte-integrale-0:1.7.3b-2.fc26.src > octave-6:4.2.1-1.fc26.src > openms-0:2.1.0-6.20170131gitbde813.fc26.src > ppl-0:1.2-2.fc27.src > python-cvxopt-0:1.1.9-3.fc26.src > sagemath-0:7.4-4.fc27.src > shogun-0:5.0.0-0.5.fc26.src > > I am currently working towards a rebuild of Macaulay2 and sagemath for other > reasons. If you need provenpackager help, I don't mind rebuilding glpk and > whichever of these packages need rebuilding as a consequence while I'm at > it. Do you have time to take a look at these packages and see which would > need a rebuild? That would be very helpful. Hi Jerry, Yes, I have some time to help with this but want to make sure I'm answer you correctly before I traverse that list. I ran `dnf download --source cbmc` and then `rpm -ivh cbmc-5.6-1.fc25.src.rpm`. I then used a simple grep to find those tokens: $ grep -r glp_add_cols cbmc-cbmc-5.6/ cbmc-cbmc-5.6/src/goto-instrument/wmm/instrumenter_strategies.cpp: glp_add_cols(lp, edges.size()); cbmc-cbmc-5.6/src/musketeer/fence_inserter.cpp: glp_add_cols(ilp.lp, unique*fence_options); $ grep -r glp_prob cbmc-cbmc-5.6/ cbmc-cbmc-5.6/src/goto-instrument/wmm/instrumenter_strategies.cpp: glp_prob* lp; cbmc-cbmc-5.6/src/musketeer/ilp.h: glp_prob* lp; which suggest, to me, that cbmc will need a rebuild. Did I get that right? One gap from your list is that dnf downloaded the f25 rpm instead of f26. I'm not sure if there's a simpler/better way or if I should continue through the list using this mechanism. Thanks, brian
(In reply to Brian from comment #20) > Did I get that right? One gap from your list is that dnf downloaded the f25 > rpm instead of f26. I'm not sure if there's a simpler/better way or if I > should continue through the list using this mechanism. Are you a packager? If so, then I would do it this way: $ fedpkg clone cbmc $ cd cbmc $ fedpkg sources If not, then you could also visit http://bugz.fedoraproject.org/cbmc/, select "Koji Builds" from the bottom left, choose the latest Rawhide or F26 build (should be near the top), then find the "RPMS" section and look for "src".
I'm not a packager but I suspect I can use fedpkg clone --anonymous for this test. I'll run through the list and post my results here.
Great! You don't need to bother with sagemath or Macaulay2, since I need to rebuild those packages anyway.
Oh, you can skip latte-integrale, too. I'm also rebuilding that one anyway.
jjames's glpk-4.61-1.fc27 completed http://koji.fedoraproject.org/koji/buildinfo?buildID=876339
here are my results so far: cbmc uses glp_add_cols, glp_prob coin-or-Alps uses neither coin-or-Bcp uses neither coin-or-Bcps uses neither coin-or-Blis uses neither coin-or-Bonmin uses neither coin-or-Cbc uses glp_prob coin-or-Cgl uses neither coin-or-CoinMP uses neither coin-or-Couenne uses neither coin-or-Dip uses neither
More results: coin-or-DyLP uses neither coin-or-FlopC++ uses neither coin-or-Ipopt uses neither coin-or-OS uses neither coin-or-Ose uses glp_prob coin-or-SYMPHONY uses glp_prob coin-or-Vol uses neither This is tedious!
last batch: coin-or-lemon uses glp_add_cols octave uses glp_prob openms uses both ppl uses glp_prob python-cvxopt uses both shogun uses both
Thank you, Brian. I appreciate you doing that tedious job. I'll start working on the ones that need rebuilds. I'm doing Rawhide builds today. Once those are done I will give everything a quick check and, if no problems appear, start doing F-26 builds.
Macaulay2-1.9.2-1.fc26, Singular-4.1.0p2-1.fc26, arb-2.10.0-1.fc26, cbmc-5.7-1.fc26, coin-or-Cbc-2.9.8-3.fc26, coin-or-Osi-0.107.8-3.fc26, coin-or-SYMPHONY-5.6.14-3.fc26, coin-or-lemon-1.3.1-10.fc26, eclib-20170330-1.fc26, fflas-ffpack-2.2.2-5.fc26, flint-2.5.2-15.fc26, frobby-0.9.0-10.fc26, gap-pkg-float-0.7.5-2.fc26, giac-1.2.3-3.25.fc26, givaro-4.0.2-5.fc26, glpk-4.61-1.fc26, latte-integrale-1.7.3b-3.fc26, libfplll-5.0.3-1.fc26, libgap-4.8.6-1.fc26, linbox-1.4.2-7.fc26, normaliz-3.1.4-1.fc26, ntl-10.3.0-1.fc26, octave-4.2.1-2.fc26, openms-2.1.0-7.20170131gitbde813.fc26, polymake-3.1-2.fc26, pynac-0.7.3-1.fc26, python-cvxopt-1.1.9-4.fc26, sagemath-7.5.1-1.fc26 has been pushed to the Fedora 26 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2017-4e8cf20cbb
Macaulay2-1.9.2-1.fc26, Singular-4.1.0p2-1.fc26, arb-2.10.0-1.fc26, cbmc-5.7-1.fc26, coin-or-Cbc-2.9.8-3.fc26, coin-or-Osi-0.107.8-3.fc26, coin-or-SYMPHONY-5.6.14-3.fc26, coin-or-lemon-1.3.1-10.fc26, eclib-20170330-1.fc26, fflas-ffpack-2.2.2-5.fc26, flint-2.5.2-15.fc26, frobby-0.9.0-10.fc26, gap-pkg-float-0.7.5-2.fc26, giac-1.2.3-3.25.fc26, givaro-4.0.2-5.fc26, glpk-4.61-1.fc26, latte-integrale-1.7.3b-3.fc26, libfplll-5.0.3-1.fc26, libgap-4.8.6-1.fc26, linbox-1.4.2-7.fc26, normaliz-3.1.4-1.fc26, ntl-10.3.0-1.fc26, octave-4.2.1-2.fc26, openms-2.1.0-7.20170131gitbde813.fc26, polymake-3.1-2.fc26, pynac-0.7.3-1.fc26, python-cvxopt-1.1.9-4.fc26, sagemath-7.5.1-1.fc26 has been pushed to the Fedora 26 stable repository. If problems still persist, please make note of it in this bug report.