[PATCH 02 of 13] sparse-read: discard gap below 65K only

Boris Feld boris.feld at octobus.net
Tue Jul 10 09:27:11 EDT 2018


# HG changeset patch
# User Boris Feld <boris.feld at octobus.net>
# Date 1531182823 -7200
#      Tue Jul 10 02:33:43 2018 +0200
# Node ID 14c3643ceaf460b75dcd57866c5e5eca18e842d1
# Parent  c24db8e9c40d26bd106954c7186d1ff30dad6c19
# EXP-Topic write-for-sparse-read
# Available At https://bitbucket.org/octobus/mercurial-devel/
#              hg pull https://bitbucket.org/octobus/mercurial-devel/ -r 14c3643ceaf4
sparse-read: discard gap below 65K only

Testing on actual data shows that 65K is more efficient in both time and memory
than 256K.

diff --git a/mercurial/configitems.py b/mercurial/configitems.py
--- a/mercurial/configitems.py
+++ b/mercurial/configitems.py
@@ -597,7 +597,7 @@ coreconfigitem('experimental', 'sparse-r
     default=0.50,
 )
 coreconfigitem('experimental', 'sparse-read.min-gap-size',
-    default='256K',
+    default='65K',
 )
 coreconfigitem('experimental', 'treemanifest',
     default=False,


More information about the Mercurial-devel mailing list