summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorSamuel Thibault <samuel.thibault@ens-lyon.org>2019-01-15 01:16:17 +0100
committerSamuel Thibault <samuel.thibault@ens-lyon.org>2019-01-15 01:16:17 +0100
commit0937ec41b2a52d346bd91365e1f6b54da1c21d51 (patch)
tree1403c84dcf836b5e5c17ae049314deef8ad87d96
parent1750f279a7919606cdee58e10f27df4ec4e78cf0 (diff)
Note the 32bit vs 64bit difference
-rw-r--r--faq/slow.mdwn5
1 files changed, 4 insertions, 1 deletions
diff --git a/faq/slow.mdwn b/faq/slow.mdwn
index 80c4ed00..ffa20fe3 100644
--- a/faq/slow.mdwn
+++ b/faq/slow.mdwn
@@ -18,7 +18,10 @@ completely usable.
Take care when running the Hurd in fully-virtualized machines: virtualization
software use ugly heuristics to make Linux run faster, which will not work on
the Hurd (or BSD, etc.) so comparisons in virtualized environments do not really
-hold.
+hold. Also take care of not comparing Hurd (which is currently 32bit) with a
+64bit Linux execution: 64bit has *much* more computational and optimization
+possibilities than 32bit. At least, use -m32 to build a 32bit Linux version of a
+test for comparison.
The main reason for slowness is *not* because of the overhead of RPCs. It's
mostly simply because less care has been done on implementing what makes Linux