Gvfs trace2 checkout and reset experiment #133
Conversation
jeffhostetler
requested review from
derrickstolee,
dscho,
jrbriggs and
kewillford
|
@kewillford These changes are fairly harmless by themselves, so I'm not worried about a regression here. But if we want to try a perf run on your stock configuration, then we could look and see if there are any voids in my data -- or areas where I should add some additional tracing. So, yeah. If you have time to do one that would be great! Thanks! |
|
@jeffhostetler there were some build errors. |
|
@kewillford sorry about that. i just pushed a fixup to take care of the cast error on the mac side. I've never seen the tag error before on the windows build. do you know what that is? |
|
There needs to be a tag in the history for the build to know what version number to stamp. Usually something like |
|
@kewillford How should we proceed? Do we want to create a temporary tag for now and reference that? |
|
@jeffhostetler yeah that will work for now. |
Signed-off-by: Jeff Hostetler <[email protected]>
Signed-off-by: Jeff Hostetler <[email protected]>
Signed-off-by: Jeff Hostetler <[email protected]>
Signed-off-by: Jeff Hostetler <[email protected]>
jeffhostetler
force-pushed
the
gvfs-trace2-checkout-experiment
branch
from
b9451f7 to
cab7dad
Compare
|
@kewillford I merged this PR and it looks like a tag was automatically created. |
…iment Gvfs trace2 checkout and reset experiment
…iment Gvfs trace2 checkout and reset experiment
…iment Gvfs trace2 checkout and reset experiment
…iment Gvfs trace2 checkout and reset experiment
Updates to v2.22.0.vfs.1 as discussed in microsoft/git#140. Includes these changes in microsoft/git: * microsoft/git#133 * microsoft/git#139 * microsoft/git#141 * microsoft/git#143 Also, the `post-indexchanged` hook was renamed to `post-index-change` as it was upstreamed.
Add some temporary telemetry around checkout and reset to help identify performance problems.
These are marked with the "exp" category.