Bruce M Simpson <[EMAIL PROTECTED]> writes:
> On Sat, Sep 25, 2004 at 01:35:07PM +0200, Dag-Erling Smørgrav wrote:
> > > If we were to implement POSIX spawn(), we'd need something like this.
> > Uh, no. posix_spawn(3) can be implemented entirely in userland and
> > does not require any special ker
On Sat, Sep 25, 2004 at 01:35:07PM +0200, Dag-Erling Smørgrav wrote:
> > If we were to implement POSIX spawn(), we'd need something like this.
>
> Uh, no. posix_spawn(3) can be implemented entirely in userland and
> does not require any special kernel support.
Discussions with peter@ and others
Bruce M Simpson <[EMAIL PROTECTED]> writes:
> On Fri, Sep 24, 2004 at 06:21:25PM +0200, Dag-Erling Smørgrav wrote:
> > You could, of course, write a kernel API for creating processes from
> > scratch. They'd still need a parent, but you can use init(8) (pid 1)
> > for that.
> If we were to impleme
On Fri, Sep 24, 2004 at 06:21:25PM +0200, Dag-Erling Smørgrav wrote:
> You could, of course, write a kernel API for creating processes from
> scratch. They'd still need a parent, but you can use init(8) (pid 1)
> for that.
If we were to implement POSIX spawn(), we'd need something like this.
So t
Robert Watson <[EMAIL PROTECTED]> writes:
> Well, we have kproc/kthread APIs, but none of that is semantically
> compatible with the notion of execve(), which is a very user-centric
> concept ("replace the address space with a mapping of binary "). You
> could fudge together a related notion, thou
On Fri, 24 Sep 2004, Dag-Erling Smørgrav wrote:
> execve(2) assumes you already have a process. You get a process by
> forking another process. The only process we ever create from scratch
> is init(8), and that takes a s**tload of work (see kern/init_main.c).
> This is why we have stuff like
Mike Meyer <[EMAIL PROTECTED]> writes:
> Gordon David <[EMAIL PROTECTED]> writes:
> > Kqueue is a good method to notify the user. But I want the code in
> > the kernel directly calls a user program.
> How about starting with the code in kern/kern_exec.c?
execve(2) assumes you already have a proces
In <[EMAIL PROTECTED]>, Gordon David <[EMAIL PROTECTED]> typed:
>
> >From: "Bjoern A. Zeeb" <[EMAIL PROTECTED]>
> >To: Gordon David <[EMAIL PROTECTED]>
> >CC: [EMAIL PROTECTED]
> >Subject: Re: execute a user process in the kernel
>
Gordon David wrote:
Hello,
I have a question. Anyone would like to tell me how to execute a user
process or shell script in the kernel?
As we know, the kernel forks a process named initproc and executes
/sbin/init, etc. If I want to execute a user level process, such as a
simple printf("Hello wor
On Thu, Sep 23, 2004 at 08:18:14AM +, Gordon David wrote:
> Kqueue is a good method to notify the user. But I want the code in the
> kernel directly calls a user program.
This is Very, Very Hard indeed, because it's not something supported by
the system, but it should be possible. Look at cre
From: "Bjoern A. Zeeb" <[EMAIL PROTECTED]>
To: Gordon David <[EMAIL PROTECTED]>
CC: [EMAIL PROTECTED]
Subject: Re: execute a user process in the kernel
Date: Thu, 23 Sep 2004 08:02:18 + (UTC)
On Thu, 23 Sep 2004, Gordon David wrote:
> That's the point. I do no
On Thu, 23 Sep 2004, Gordon David wrote:
> That's the point. I do not want the userland program to check /dev/fooctl
> from time to time. I want the kernel to notify the userland program
> instead. So how shall I do it? Maybe linker_load_file is a better way.
man 2 kqueue ?
--
Bjoern A. Zeeb
> On Wed, Sep 22, 2004 at 08:49:11AM +, Gordon David wrote:
> > Hello,
> >
> > I have a question. Anyone would like to tell me how to execute a
> > user process or shell script in the kernel?
>
> You probably really don't want to do this and if you do, there is
> likely a better approach.
Readi
On Wed, 22 Sep 2004 18:46:49 -0400
Allan Fields <[EMAIL PROTECTED]> wrote:
> On Wed, Sep 22, 2004 at 08:49:11AM +, Gordon David wrote:
> > Hello,
> >
> > I have a question. Anyone would like to tell me how to execute a
> > user process or shell script in the kernel?
>
> You probably really d
On Wed, Sep 22, 2004 at 08:49:11AM +, Gordon David wrote:
> Hello,
>
> I have a question. Anyone would like to tell me how to execute a user
> process or shell script in the kernel?
You probably really don't want to do this and if you do, there is
likely a better approach.
> As we know, the
Hello,
I have a question. Anyone would like to tell me how to execute a user
process or shell script in the kernel?
As we know, the kernel forks a process named initproc and executes
/sbin/init, etc. If I want to execute a user level process, such as a
simple printf("Hello world") in a driver,
16 matches
Mail list logo
