Port patches and fixes over from node #124
Conversation
piscisaureus
changed the title
|
I'm not a fan of the flaky test stuff in f6433a9 and da0ef90 and would rather see those be tackled properly |
|
@rvagg The list with flaky tests in particular? |
|
just the fact that we need the list in the first place; you are probably more familiar with these tests than me and why they are being marked as flaky, do you think it's justified and the best way forward with this? |
|
Marking test-net-GH-5504 as flaky is reasonable, it hits a bug with some (old) Linux kernels. The others seem rather suspect, though. That said, it's a test runner switch and we don't have to use it. |
|
I'll land both patches, but I'll make sure only GH-5504 ends up on the flaky test list. |
piscisaureus
force-pushed
the
merge
branch
2 times, most recently
from
7e44f01 to
8b56072
Compare
piscisaureus
changed the title
|
Merge conflicts made many of these not apply cleanly, I cut up the async listener removal into two parts; one where Async Listener itself is removed, and another where the tracing module is removed. If desired I can squash those two commits again, but I kind of like the way it ended up. |
|
@bnoordhuis or @rvagg, would you mind taking another look at the unreviewed commits (the ones with that cute little yellow puppet in front of them). |
|
@piscisaureus Reviewed. |
On Windows a long integer is always 32-bits, even when the target architecture uses 64-bit pointers. PR-URL: nodejs#124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
uv_thread_t is a HANDLE (void pointer) on Windows, which means that on 64-bit windows it cannot be stored with CRYPTO_THREADID_set_numeric without potential data loss. PR-URL: nodejs#124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: nodejs#124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
That test can trigger a bug on some older Linux kernels. PR-URL: nodejs#124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: nodejs#124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
On Windows a long integer is always 32-bits, even when the target architecture uses 64-bit pointers. PR-URL: #124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
uv_thread_t is a HANDLE (void pointer) on Windows, which means that on 64-bit windows it cannot be stored with CRYPTO_THREADID_set_numeric without potential data loss. PR-URL: #124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: #124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
That test can trigger a bug on some older Linux kernels. PR-URL: #124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: #124 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
No description provided.