Jim Phillips

This issue only seems to affect rasters loaded from a URL. Loading from an array buffer looks fine.

I think this is fixed by https://github.com/jcphill/georaster-layer-for-leaflet/commit/2996308f1c7ed66cf0dc2fc37ac23d0d01a8baf8

> I think this is fixed by [jcphill@2996308](https://github.com/jcphill/georaster-layer-for-leaflet/commit/2996308f1c7ed66cf0dc2fc37ac23d0d01a8baf8) That's only a partial fix. Most of the edge shifting is due to how COG sampling interacts with axis rotation in projections.

Thank you for your reply. The logic is correct in the pull request at https://github.com/geotiffjs/geotiff.js/blob/dd72ff1a106e7113b6626a9e54f00987893357b8/src/source/blockedsource.js#L190 I'm glad to see that there is an allowFullFile option in the pull request to...

I'd suggest just getting colvars running on the GPU in whatever way is possible with the fewest code changes using whatever bleeding-edge features might help, even if it only runs...

*Original date: 2016-04-14 14:26:08* --- This appears to be the node queue receive function in src/arch/util/machine-common-core.c: ``` c++ void *CmiGetNonLocalNodeQ(void) { CmiState cs = CmiGetState(); char *result = 0; CmiIdleLock_checkMessage(&cs->idle);...

*Original date: 2016-05-13 16:33:21* --- I tried the above variant, as well as adding a lock to restrict node queue access to a single consumer at a time. Neither showed...

This is not an issue if ++n is used instead of +p, so the rule should be to require ++ppn whenever ++mpiexec-no-n is used with +p on an smp layer.

*Original date: 2017-09-14 02:33:31* --- 6.8.0 from Sept 5 (v6.8.0-0-ga36028e-namd-charm-6.8.0-build-2017-Sep-05-28093). No observed performance difference on Bridges.

*Original date: 2017-09-14 02:59:42* --- I see the exact same performance for v6.7.0-574-g7d61794-namd-charm-6.8.0-build-2017-Jan-23-80737 and v6.7.0-0-g46f867c-namd-charm-6.7.0-build-2015-Dec-21-45876. Definitely not a recent change.