oh well! Hopefully it's still useful info for somebody.

If you wish to take a look at optimizing it, you can look at the the phoronix test suite for fftw. Which runs fftw-source/tests/bench --time-repeat 100 -opatient libc64 There are many other ranges you can throw it it like ibc4096x4096 to libc32. You won't be able to use the phoronix test at it compiles it's own fftw and doesn't test the system version.

Will need to be rebased against master, apologies for inconvenience.

I have no idea how phab figured there was a differential for this.

Pushing manually to bootstrap properly.

Update to 9.2.0. Address niggles.

We can try backporting the proposed fix https://github.com/OpenPrinting/cups-filters/pull/128. Should be noted that it is not finalized yet.

For clarification, while I appreciate you putting the time in to make the patch it's just basically double the cost to maintain two versions of a package, especially, when few things support the new version at this point.

I only have 8GB so that explains it I guess! Workstation upgrades are well past needed at this point. I'll try your solution when building for clang (good work figuring that one out btw).

Thanks for testing pytorch. I actually can't build it due to ld.bfd sucking all my memory, and, can't build it with clang + lld due to --copy-needed-dt-entries being inherited.

You need to add abi_symbols and abi_used_libs files to the diff. For info libguile-2.0.so.22 has been bumped to libguile-2.2.so.1, all rev deps of this will need to be rebuilt. The pkgconfig name has also changed from pkgconfig(guile-2.0) to pkgconfig(guile-2.2). Already I know that weechat and autogen do not support the newest guile. GCC tools like gdb also do not support it. Until they do, I doubt we will create two guile packages unless we are forced to.

Resolved by R2900:4093704fc214

Waiting for GCC 9.2 tarballs and will update Monday/Tuesday. If tarballs are delayed it'll have to wait another sync until next Saturday.

Updated mlt to 6.16.0 and shotcut to 19.08beta, it builds and seems to work. Haven't tested the non-beta version yet but it could be updated regardless.

Deprecating shotcut!? Know how many people use that. Pretty sure mlt and shotcut just need updating.

Can confirm your bug with the ?? command not working as expected within rstudio with R 3.6.1. I've updated rstudio to the latest version (1.2.1335) locally and the issue still persists with R 3.6.1. However, downgrading R to 3.5.1 the ?? command works as expected again. It should be noted however, the ?? command works as expected when running from the terminal with both R 3.5.1 and R 3.6.1.

gfortran rebuilds are done as requested by DataDrake. There was a bit of trouble with scipy via numpy and plplot needed an update, but the rest were smooth sailing. I couldn't find anything in the repo that directly links against the address sanitizers so that deleted symbol shouldn't be a problem (GCC is supposed to be abi forward compatible anyway). I'm in the mind that i'll let some dust settle on this and see if any issues crop up locally, then push it early next sync cycle barring any misfortune.

isl rebuilds are done and tested for GCC AVR and GCC ARM. Compiled some basic programs for ARM and AVR targets. Though I don't have any hardware to actually test them on. Also updated GCC ARM and GCC AVR to version GCC 9 for additional testing. Successfully compiled some basic examples for ARM and AVR targets once again. So it'll be greatly appreciated if the original submitter @chax could look at updating GCC AVR and ARM and look for any bug fixes need backporting from the GCC 9 branch when this get merged.

Also ran some quick benchmarks via graphicsmagick, imagemagick and openjpeg via libpng to check if there weren't any obvious performance regressions in libgcc. Maybe not the most ideal but these are what came to mind as i've been testing libpng decode perf recently.

@Girtablulu For manaplus chroot in and look at the config.log. There was a similar error i encountered when updating R with missing curl, but it was just additional missing pkgconfigs.

@DataDrake do you want this added to the monolithic llvm build or as a seperate repo like lldb? Points to consider, this is not officially part of llvm, i've also been testing a PGO'd LLVM again and we may not want to make that build longer if it get merged.

Reverse dependencies will need to support it first. Version 4 is a fairly big ABI and API break afaik. Started doing the rebuilds and already eviacam gmic pytorch don't build against it. pytorch is unmaintained by us and will need some lovin'. eviacam seemed to have stopped development so we'll need to manually patch it. Other reverse deps should support it once their respective upstreams add support if they haven't already.

Pushing to finish up poppler rebuilds