open_issues/performance/io_system/read-ahead: New.

2 files changed, 53 insertions, 5 deletions

diff --git a/open_issues/performance/io_system.mdwn b/open_issues/performance/io_system.mdwn
index 0d41d3c7..dbf7012a 100644
--- a/open_issues/performance/io_system.mdwn
+++ b/open_issues/performance/io_system.mdwn
@@ -1,4 +1,4 @@
-[[!meta copyright="Copyright © 2008, 2009, 2010 Free Software Foundation,
+[[!meta copyright="Copyright © 2008, 2009, 2010, 2011 Free Software Foundation,
 Inc."]]
 
 [[!meta license="""[[!toggle id="license" text="GFDL 1.2+"]][[!toggleable
@@ -6,8 +6,8 @@ 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]]."]]"""]]
+is included in the section entitled [[GNU Free Documentation
+License|/fdl]]."]]"""]]
 
 [[!meta title="I/O System"]]
 
@@ -20,7 +20,7 @@ slow hard disk access.
 The reason for this slowness is lack and/or bad implementation of common
 optimization techniques, like scheduling reads and writes to minimize head
 movement; effective block caching; effective reads/writes to partial blocks;
-reading/writing multiple blocks at once; and read-ahead.  The
+reading/writing multiple blocks at once; and [[read-ahead]].  The
 [[ext2_filesystem_server|hurd/translator/ext2fs]] might also need some
 optimizations at a higher logical level.
 
@@ -30,7 +30,7 @@ requires understanding the data flow through the various layers involved in
 disk access on the Hurd ([[filesystem|hurd/virtual_file_system]],
 [[pager|hurd/libpager]], driver), and general experience with
 optimizing complex systems.  That said, the killing feature we are definitely
-missing is the read-ahead, and even a very simple implementation would bring
+missing is the [[read-ahead]], and even a very simple implementation would bring
 very big performance speedups.
 
 Here are some real testcases:
diff --git a/open_issues/performance/io_system/read-ahead.mdwn b/open_issues/performance/io_system/read-ahead.mdwn
new file mode 100644
index 00000000..b3b139c7
--- /dev/null
+++ b/open_issues/performance/io_system/read-ahead.mdwn
@@ -0,0 +1,48 @@
+[[!meta copyright="Copyright © 2011 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 open_issue_hurd]]
+
+IRC, #hurd, freenode, 2011-02-13:
+
+    <etenil> youpi: Would libdiskfs/diskfs.h be in the right place to make
+      readahead functions?
+    <youpi> etenil: no, it'd rather be at the memory management layer,
+      i.e. mach, unfortunately
+    <youpi> because that's where you see the page faults
+    <etenil> youpi: Linux also provides a readahead() function for higher level
+      applications. I'll probably have to add the same thing in a place that's
+      higher level than mach
+    <youpi> well, that should just be hooked to the same common implementation
+    <etenil> the man page for readahead() also states that portable
+      applications should avoid it, but it could be benefic to have it for
+      portability
+    <youpi> it's not in posix indeed
+
+IRC, #hurd, freenode, 2011-02-14:
+
+    <etenil> youpi: I've investigated prefetching (readahead) techniques. One
+      called DiskSeen seems really efficient. I can't tell yet if it's patented
+      etc. but I'll keep you informed
+    <youpi> don't bother with complicated techniques, even the most simple ones
+      will be plenty :)
+    <etenil> it's not complicated really
+    <youpi> the matter is more about how to plug it into mach
+    <etenil> ok
+    <youpi> then don't bother with potential pattents
+    <antrik> etenil: please take a look at the work KAM did for last year's
+      GSoC
+    <youpi> just use a trivial technique :)
+    <etenil> ok, i'll just go the easy way then
+
+    <braunr> antrik: what was etenil referring to when talking about
+      prefetching ?
+    <braunr> oh, madvise() stuff
+    <braunr> i could help him with that