livie
/
phoenix-firestorm
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

On Windows, make "very large message" test ridiculously small. 

This test must not be subject to spurious environmental failures, else some
kind soul will disable it entirely. We observe that APR specifies a hard-coded
buffer size of 64Kbytes for pipe creation -- use that and cross fingers.
master
Nat Goodspeed 2012-03-15 16:51:34 -04:00
parent 2a6f919cdb
commit be669d4a1f
1 changed files with 2 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
indra/llcommon/tests/llleap_test.cpp
View File

@ -56,14 +56,8 @@ const size_t BUFFERED_LENGTH = 1023*1024; // try wrangling just under a megabyte
// ones -- how?? Empirically this behavior is only observed when writing a
// "very large message". To be able to move forward at all, try to bypass this
// particular failure by adjusting the size of a "very large message" on
// Windows. When the test fails at BUFFERED_LENGTH, the test_or_split()
// function performs a binary search to find the largest size that will work.
// Running several times on a couple different Windows machines produces a
// range of "largest successful size" results... suggesting that it may be a
// matter of available OS buffer space? In any case, pick something small
// enough to be optimistic, while hopefully remaining comfortably larger than
// real messages we'll encounter in the wild.
const size_t BUFFERED_LENGTH = 256*1024;
// Windows.
const size_t BUFFERED_LENGTH = 65336;
#endif // LL_WINDOWS
void waitfor(const std::vector<LLLeap*>& instances, int timeout=60)