summaryrefslogtreecommitdiff
path: root/open_issues/default_pager.mdwn
diff options
context:
space:
mode:
Diffstat (limited to 'open_issues/default_pager.mdwn')
-rw-r--r--open_issues/default_pager.mdwn44
1 files changed, 44 insertions, 0 deletions
diff --git a/open_issues/default_pager.mdwn b/open_issues/default_pager.mdwn
new file mode 100644
index 00000000..38c9a2be
--- /dev/null
+++ b/open_issues/default_pager.mdwn
@@ -0,0 +1,44 @@
+[[!meta copyright="Copyright © 2011, 2012, 2013, 2014 Free Software Foundation,
+Inc."]]
+
+[[!meta license="""[[!toggle id="license" text="GFDL 1.2+"]][[!toggleable
+id="license" text="Permission is granted to copy, distribute and/or modify this
+document under the terms of the GNU Free Documentation License, Version 1.2 or
+any later version published by the Free Software Foundation; with no Invariant
+Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license
+is included in the section entitled [[GNU Free Documentation
+License|/fdl]]."]]"""]]
+
+[[!tag open_issue_gnumach]]
+
+[[!toc]]
+
+
+# IRC, freenode, #hurd, 2011-08-31
+
+ <antrik> braunr: do you have any idea what could cause the paging errors
+ long before swap is exhausted?
+ <braunr> antrik: not really, but i know every project based on the mach vm
+ have rewritten their swap pager
+ <antrik> (and also I/O performance steadily dropping before that point is
+ reached?)
+
+[[performance/degradation]], [[ext2fs_page_cache_swapping_leak]].
+
+ <antrik> hm
+ <braunr> there could too many things
+ <antrik> perhaps we could "borrow" from one of them? :-)
+ <braunr> map entry fragmentation for example
+ <braunr> the freebsd one is the only possible candidate
+ <braunr> uvm is too different
+ <braunr> dragonflybsd maybe, but it's very close to freebsd
+ <braunr> i didn't look at darwin/xnu
+
+
+# [[trust_the_behavior_of_translators]]
+
+
+# IRC, freenode, #hurd, 2013-10-30
+
+ <braunr> it also seems that the kernel has trouble resuming processes that
+ have been swapped out