From eb8c1c4b7b414b73e81c8be1cbd2e5b616a8c443 Mon Sep 17 00:00:00 2001 From: Pino Toscano Date: Sun, 31 Mar 2013 09:52:26 +0200 Subject: misc changes for libfuse and time --- hurd/libfuse.mdwn | 7 ++++--- open_issues/time.mdwn | 15 ++++++++++++++- 2 files changed, 18 insertions(+), 4 deletions(-) diff --git a/hurd/libfuse.mdwn b/hurd/libfuse.mdwn index bc6a9d4a..45ff97ec 100644 --- a/hurd/libfuse.mdwn +++ b/hurd/libfuse.mdwn @@ -15,7 +15,7 @@ initially written by Stefan Siegl. The implementation takes advantage of the [[translators|translator]] facilities of Hurd: this means that applications that implement a FUSE filesystem, when -compiled against libfuse-hurd, become translators to be set with usual `settrans` +compiled against libfuse-hurd, become translators to be set with usual [[settrans]] etc. @@ -23,9 +23,10 @@ etc. * Only part of the API is implemented * lowlevel API not implemented - * Options handling not implemented + * Options handling (`fuse_parse_cmdline` and `fuse_opt_*`) not implemented * CUSE lowlevel not supported (compatibility level 29) -* Supports the compatibility level 25 (while current libfuse 2.9.x provides 26) +* Supports up to the compatibility level 25 (while current libfuse 2.9.x provides 26) +* File I/O is quite slow. # Source diff --git a/open_issues/time.mdwn b/open_issues/time.mdwn index ab239aef..13cb0a41 100644 --- a/open_issues/time.mdwn +++ b/open_issues/time.mdwn @@ -51,7 +51,7 @@ GNU time's *elapsed* value is off by some factor. user 0m0.000s sys 0m0.010s -As above; also here all the running time should be attriuted to *user* time. +As above; also here all the running time should be attributed to *user* time. This is probably a [[!taglink open_issue_gnumach]]. @@ -67,3 +67,16 @@ While testing some [[performance/IPC_virtual_copy]] performance issues: And I can confirm that with dd if=/dev/zero of=/dev/null bs=4k running, a parallel sleep 10 takes about 20 s (on strauss). + +# 2013-03-30/31 + +Investigating time's `configure`, a difference of the output between Linux and +Hurd shows: + + -checking for wait3 that fills in rusage... yes + +checking for wait3 that fills in rusage... no + +This causes a different code path in `resuse.c` to be used; such code path does +not get a define for `HZ`, which is then defined with a fallback value of 60. + +[[!debbug 704283]] has been filed with a fix for this no-wait3 case. -- cgit v1.2.3