summaryrefslogtreecommitdiff
path: root/apt-pkg/contrib/srvrec.cc
diff options
context:
space:
mode:
authorDavid Kalnischkies <david@kalnischkies.de>2016-08-12 10:02:28 +0200
committerDavid Kalnischkies <david@kalnischkies.de>2016-08-12 11:12:10 +0200
commit379a36f43d3f4db4afa5ad4fb6f79b89824c999c (patch)
tree03820e0718fe0b888581c1738a61c2c6eba229c2 /apt-pkg/contrib/srvrec.cc
parent1cb047079aa2c26a8159d100348b7e69a49bc117 (diff)
ensure a good clock() value for usage and tests
We use clock() as a very cheap way of getting a "random" value, but the manpage warns that this could return -1, so we should be dealing with this. Additionally, e.g. on hurd-i386 the value increases only slowly – to slow for our fast running tests for randomness hence producing the same range in both samples, so we introduce a simple busy-wait loop (as clock is counting processor time used by the program) in the test which delays the second sample just enough making our randomness a bit more predictable.
Diffstat (limited to 'apt-pkg/contrib/srvrec.cc')
-rw-r--r--apt-pkg/contrib/srvrec.cc2
1 files changed, 1 insertions, 1 deletions
diff --git a/apt-pkg/contrib/srvrec.cc b/apt-pkg/contrib/srvrec.cc
index be159bad9..327e59937 100644
--- a/apt-pkg/contrib/srvrec.cc
+++ b/apt-pkg/contrib/srvrec.cc
@@ -185,7 +185,7 @@ SrvRec PopFromSrvRecs(std::vector<SrvRec> &Recs)
[&I](SrvRec const &J) { return I->priority != J.priority; });
// clock seems random enough.
- I += clock() % std::distance(I, J);
+ I += std::max(static_cast<clock_t>(0), clock()) % std::distance(I, J);
SrvRec const selected = std::move(*I);
Recs.erase(I);