13:05 @dcoutts we can add a GC_GLOBAL_SYNC event later :-)
and later:
16:32 @dcoutts mikolaj: one alternative I was thinking about btw, was a separate generic event to indicate a GC sync
16:32 @dcoutts that is, indicating that we're doing a stop-the-world GC and all other HECs should be between a GC_START/END at that moment
16:32 @dcoutts mikolaj: so that we don't have to use GC_STATS_GHC for that purpose
16:32 @dcoutts which I find a bit unpleasant
16:33 @dcoutts ideally GC_STATS_GHC ought only to be extra ghc-specific info
16:33 @dcoutts not something we have to rely on to be able to match up GC across HECs
16:34 @dcoutts mikolaj: if we're relying on GC_START/END and GC_SYNC then yes, we should not need GC_STATS_GHC for GC vs MUT time
13:05 @dcoutts we can add a GC_GLOBAL_SYNC event later :-)
and later:
16:32 @dcoutts mikolaj: one alternative I was thinking about btw, was a separate generic event to indicate a GC sync 16:32 @dcoutts that is, indicating that we're doing a stop-the-world GC and all other HECs should be between a GC_START/END at that moment 16:32 @dcoutts mikolaj: so that we don't have to use GC_STATS_GHC for that purpose 16:32 @dcoutts which I find a bit unpleasant 16:33 @dcoutts ideally GC_STATS_GHC ought only to be extra ghc-specific info 16:33 @dcoutts not something we have to rely on to be able to match up GC across HECs 16:34 @dcoutts mikolaj: if we're relying on GC_START/END and GC_SYNC then yes, we should not need GC_STATS_GHC for GC vs MUT time