[haiku-commits] Re: r35156 - in haiku/trunk/src/libs/compat/freebsd_network: . compat/sys
- From: "Axel Dörfler" <axeld@xxxxxxxxxxxxxxxx>
- To: haiku-commits@xxxxxxxxxxxxx
- Date: Tue, 19 Jan 2010 16:04:51 +0100
Colin Günther <coling@xxxxxx> wrote:
> Axel Dörfler schrieb:
> > The _user_*() functions are syscalls, they are not supposed to be
> > called
> > within the kernel.
> > You probably want to use vm_map_file() instead.
> Mmh, where is the vm_map_file() counterpart?
> - I looked into munmap()'s callchain, which uses unmap_address_range
> in
> the end. Unfortunately unmap_address_range is declared static and
> thus
> not accessible by firmware_put().
You get an area, and you can delete those with delete_area() as usual.
It's even part of the public kernel interface as documented in the
BeBook.
A mapped file is just like any other area.
> - I looked into delete_area, quickly finding a comment in
> _user_delete_area about deleting areas is for userland created areas
> only. Mmh.
As I said, _user_*() functions are not supposed to be called from within
the kernel.
> - I looked in the code for close() in the hope that closing the file
> might trigger some unmapping. It doesn't look like there is taking
> place
> any unmapping along the call chain down to file_close() in vfs.cpp.
> (The
> curiousity let me look in the bfs_close(), too, pretty empty in there
> :)
The file descriptor is not used besides getting the file's cache - IOW
(as in mmap()) it's still your responsibility to close the file
descriptor, and that will not harm the area.
> Now I'm asking myself whether file mapping for firmware loading is
> the
> right approach at all. I mean instead of using the previous approach
> of
> reading the whole file in by using read(). Because then I could just
> use
> free() when firmware_put() is called.
Depending on the locking flags, mapping has the advantage that the data
is mapped in when needed, and unmapped if not (and that you save copying
the data to a buffer). Otherwise it doesn't really matter much.
In any case, you can still just call delete_area() when firmware_put()
is called.
Still confused? :-)
Bye,
Axel.
Other related posts:
