Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
syscall return values
1 view
Skip to first unread message
Adrian Kubala
May 8, 2005, 12:52:22 PM5/8/05
to
Can we always return 0 from asynchronous syscalls? Any call may have to
be suspended if the directory block is not yet loaded, so I don't see
what else we can do.
be suspended if the directory block is not yet loaded, so I don't see
what else we can do.
Mohnish Rao
May 8, 2005, 5:33:20 PM5/8/05
to
The scheduler is activated only after the first 5 "major" blocks have been
read in... so no syscalls will be executed before that.
read in... so no syscalls will be executed before that.
"Adrian Kubala" <adria...@sixfingeredman.net> wrote in message
news:slrnd7sgu6.4q...@sixfingeredman.net...
Adrian Kubala
May 8, 2005, 6:37:33 PM5/8/05
to
Mohnish Rao <mohni...@hotmail.com> schrieb:
> The scheduler is activated only after the first 5 "major" blocks have been
> read in... so no syscalls will be executed before that.
> The scheduler is activated only after the first 5 "major" blocks have been
> read in... so no syscalls will be executed before that.
No, despite the message in the code which seems to indicate that, I
don't think that's actually true. When I asked about it in another
thread, Sit says we should make processes wait on these blocks to be
loaded just like they would any other blocks (so processes can be
spawning other processes while the VFS is still starting up).
Mohnish Rao
May 9, 2005, 8:11:36 PM5/9/05
to
Are you sure about this? It seems to me that the first five blocks are
loaded before anything happens. If you look at OS's interrupt method, under
ST_LAST_INITIALIZE, it print's out the "Activiating scheduler" bit. This is
called only after the reading is done.
loaded before anything happens. If you look at OS's interrupt method, under
ST_LAST_INITIALIZE, it print's out the "Activiating scheduler" bit. This is
called only after the reading is done.
"Adrian Kubala" <adria...@sixfingeredman.net> wrote in message
news:slrnd7t55d.63...@sixfingeredman.net...
Adrian Kubala
May 9, 2005, 10:09:30 PM5/9/05
to
Mohnish Rao <mohni...@hotmail.com> schrieb:
> Are you sure about this? It seems to me that the first five blocks are
> loaded before anything happens. If you look at OS's interrupt method, under
> ST_LAST_INITIALIZE, it print's out the "Activiating scheduler" bit. This is
> called only after the reading is done.
> Are you sure about this? It seems to me that the first five blocks are
> loaded before anything happens. If you look at OS's interrupt method, under
> ST_LAST_INITIALIZE, it print's out the "Activiating scheduler" bit. This is
> called only after the reading is done.
You're right -- see Message-Id: <slrnd7vqdf.fo...@sixfingeredman.net>
0 new messages