cd6c352512
- update to 8.6.2
* Several compiler panics observed in GHC 8.6.1 have been fixed
* An integer overflow, resulting in some encodeFloat uses returning
incorrect results, has been fixed
* A long-standing bug in the LLVM code generator, resulting in incorrect
floating point evaluation, has been fixed.
* A long-standing bug exposed in GHC 8.6.1, has been fixed. This issue
resulted in undefined runtime behavior with some uses of the dataToTag#
primop. Note that this issue, while less likely to manifest, has existed
in some form in all GHC releases prior to this release and may result
in silent, incorrect evaluation.
- Make use of memory-constraints to avoid OOM errors
- Remove forced 'make -j2' for aarch64 and %arm
OBS-URL: https://build.opensuse.org/request/show/646895
OBS-URL: https://build.opensuse.org/package/show/devel:languages:haskell/ghc?expand=0&rev=285