Conversation
Fixes envoyproxy#23580 Full diff: google/tcmalloc@5940033...e33c7bc Signed-off-by: Keith Smiley <[email protected]>
repokitteh-read-only
bot
added
the
deps
|
CC @envoyproxy/dependency-shepherds: Your approval is needed for changes made to |
|
/lgtm deps However, CI pipelines need surgery |
repokitteh-read-only
bot
removed
the
deps
|
/retest |
|
Retrying Azure Pipelines: |
|
CI failures here appear to be regressions of //test/integration:stats_*, ptal /wait |
Signed-off-by: Keith Smiley <[email protected]>
| 100, true, [this](absl::string_view name) { store_.counterFromString(std::string(name)); }); | ||
| EXPECT_MEMORY_EQ(memory_test.consumedBytes(), 688080); // July 2, 2020 | ||
| EXPECT_MEMORY_LE(memory_test.consumedBytes(), 0.75 * million_); | ||
| EXPECT_MEMORY_LE(memory_test.consumedBytes(), 0.85 * million_); |
There was a problem hiding this comment.
I do not know the implications of bumping these
There was a problem hiding this comment.
I believe this just indicates the predictable pattern isn't so predictable. But there's a second question of whether this particular sort of tests belongs in envoyproxy or should be part of the testing of the underlying cpplib/absl implementations.
There was a problem hiding this comment.
It's just as possible that the change occurred in our own TestUtil::MemoryTest logic?
There was a problem hiding this comment.
it pretty reliably fails with this change, so I feel like there must be something in tcmalloc affecting it, but I agree that it could be better as an upstream test if that's all we're really testing. I didn't check how on the cusp of exceeding the previous limits we were before
Signed-off-by: Keith Smiley <[email protected]>
Signed-off-by: Keith Smiley <[email protected]>
This reverts commit 1c68be2. Signed-off-by: Alyssa Wilk <[email protected]>
This reverts commit 1c68be2. For #23891 Signed-off-by: Alyssa Wilk <[email protected]>
Fixes envoyproxy#23580 Full diff: google/tcmalloc@5940033...e33c7bc Signed-off-by: Keith Smiley <[email protected]> Signed-off-by: Ryan Northey <[email protected]>
3 participants
Fixes #23580
Full diff: google/tcmalloc@5940033...e33c7bc
Signed-off-by: Keith Smiley [email protected]