summaryrefslogtreecommitdiff
path: root/TODO
diff options
context:
space:
mode:
authorRob Landley <rob@landley.net>2005-11-04 20:45:54 +0000
committerRob Landley <rob@landley.net>2005-11-04 20:45:54 +0000
commit7b7c99ca333b832c0d017292436f42efd71589cd (patch)
tree27373e0accdf8c7a291fc1d4a4356590737bee2a /TODO
parent3fc4ad1478e1c170d5a939104d5278ac90ca9e6f (diff)
downloadbusybox-w32-7b7c99ca333b832c0d017292436f42efd71589cd.tar.gz
busybox-w32-7b7c99ca333b832c0d017292436f42efd71589cd.tar.bz2
busybox-w32-7b7c99ca333b832c0d017292436f42efd71589cd.zip
Size I zapped the bb_close() error, might as well make a TODO item.
Diffstat (limited to 'TODO')
-rw-r--r--TODO14
1 files changed, 13 insertions, 1 deletions
diff --git a/TODO b/TODO
index 727fbbadf..92fd403a9 100644
--- a/TODO
+++ b/TODO
@@ -34,10 +34,22 @@ man
34--- 34---
35bzip2 35bzip2
36 Compression-side support. 36 Compression-side support.
37 37---
38init
39 General cleanup.
38 40
39Architectural issues: 41Architectural issues:
40 42
43bb_close() with fsync()
44 We should have a bb_close() in place of normal close, with a CONFIG_ option
45 to not just check the return value of close() for an error, but fsync().
46 Close can't reliably report anything useful because if write() accepted the
47 data then it either went out or it's in cache or a pipe buffer. Either way,
48 there's no guarantee it'll make it to its final destination before close()
49 gets called, so there's no guarantee that any error will be reported.
50 You need to call fsync() if you care about errors that occur after write(),
51 but that can have a big performance impact. So make it a config option.
52---
41Do a SUSv3 audit 53Do a SUSv3 audit
42 Look at the full Single Unix Specification version 3 (available online at 54 Look at the full Single Unix Specification version 3 (available online at
43 "http://www.opengroup.org/onlinepubs/009695399/nfindex.html") and 55 "http://www.opengroup.org/onlinepubs/009695399/nfindex.html") and