[Bug 284165] biology/mmseqs2: update 16 -747c6 → 17-b804f

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=284165

            Bug ID: 284165
           Summary: biology/mmseqs2: update 16-747c6 → 17-b804f
           Product: Ports & Packages
           Version: Latest
          Hardware: Any
               URL: https://github.com/soedinglab/MMseqs2/releases
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: Individual Port(s)
          Assignee: jwb@freebsd.org
          Reporter: alster@vinterdalen.se
                CC: jwb@freebsd.org
             Flags: maintainer-feedback?(jwb@freebsd.org)
          Assignee: jwb@freebsd.org
 Attachment #256808 maintainer-approval?(jwb@freebsd.org)
             Flags:
                CC: jwb@freebsd.org

Created attachment 256808
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=256808&action=edit
[PATCH] biology/mmseqs2: update 16-747c6 → 17-b804f

MMseqs2 Release 17-b804f Latest

MMseqs2 Release 17 is mostly a bug fix release. Highlights include usability
improvements in MMseqs2-GPU and fix for a common crash in the prefilter that
was affecting many clustering runs.
New Features and Enhancements

    result2profile can print frequencies in TSV format (c2c3ad9)
    add a new masking mode --mask-n-repeat (c2c3ad9)
    Improve GPU clients in server mode to wait for databasess to be loaded
(e095774)
    GPU server now takes CUDA_VISIBLE_DEVICES into account. (b804fbe)
    Reduced glibc requirements for precompiled MMseqs2-GPU binaries to 2.17
(i.e. CentOS 7; db8ad2d)

Bug Fixes

    Segmentation fault in easy-cluster starting in #916 (dc7f8ad)
    GPU version generated corrupted sequence outputs #912 (e3b16fa)
    Sequences starting with * could break Sequence mapping #927 (492297b)
    Indexes without k-mer index are masked now (4766f92)
    Invalid taxids check in majoritylca does not abort the whole process
(8d17137)
    Merged taxID larger than any taxid in nodes.dmp could corrupt memory #931
(fd37b37)

Developer Notes

    Export NATIVE_ARCH in cmake (17cd5c0)

-- 
You are receiving this mail because:
You are the assignee for the bug.