, with the exception mtdchar.
Jason
--
Revert hiding of inline_map_* functions in linux/mtd.h, and mark
MTD_CHAR as incompatible with uml until a better solution can be found
there.
Signed-off-by: Jason Lunz l...@acm.org
---
drivers/mtd/Kconfig |1 +
include/linux/mtd/map.h | 18
warnings. That doesn't seem to me like an improvement.
Jason
--
Allow parts of drivers/mtd to compile on uml by pushing the HAS_IOMEM
dependencies down closer to the parts of mtd that actually need it.
This allows enough of mtd to build to let jffs2 be used on uml.
Signed-off-by: Jason Lunz l
On one hand you've got uml, which simply doesn't have mmio. On the other
there's mtd, which began as a method for accessing hardware devices that
are often accessed using mmio. But then the mtd subsystem developed
emulations of that hardware that are software based and thus don't
require mmio.
On Tue, Dec 14, 2010 at 06:49:02PM -0600, Rob Landley wrote:
The problem is that jffs2 is a filesystem, and thus something people would
really like to be able to loopback mount, but it's hardwired to assume it's
only ever stored on a certain type of hardware, and thus requies incestuous
On Tue, Dec 07, 2010 at 10:39:41AM +0100, richard -rw- weinberger wrote:
On Tue, Dec 7, 2010 at 8:29 AM, Jason Lunz l...@acm.org wrote:
Allow parts of drivers/mtd to compile on uml by pushing the HAS_IOMEM
dependencies down closer to the parts of mtd that actually need it.
This allows
Allow parts of drivers/mtd to compile on uml by pushing the HAS_IOMEM
dependencies down closer to the parts of mtd that actually need it.
This allows enough of mtd to build to let jffs2 be used on uml.
Signed-off-by: Jason Lunz l...@acm.org
---
arch/um/Kconfig.rest|4 +---
drivers
On Mon, Apr 02, 2007 at 05:44:34PM -0400, Jeff Dike wrote:
There are sites (http://uml.nagafix.co.uk/ being the best one I know
of) where, with two downloads, two uncompressions, and one command
line later, you have a booted UML.
The only way I know of to improve on this, aside from
. It passes my testing - I can still use
lvm within uml. I have not tested CONFIG_HIGHMEM, but here's an
implementation against 2.6.20.
Jeff, please drop my other patch and use this one.
Signed-off-by: Jason Lunz [EMAIL PROTECTED]
---
arch/um/kernel/mem.c |3 +--
1 file changed, 1 insertion
can
see, max_pfn and max_low_pfn don't get much use after the
bootmem-allocator stops being used anyway, except that they initialize
the block layer's blk_max_low_pfn/blk_max_pfn.
With this applied I can use lvm to create logical volumes without
crashing UML.
Signed-off-by: Jason Lunz [EMAIL
.
Signed-off-by: Jason Lunz [EMAIL PROTECTED]
---
arch/um/kernel/mem.c |3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
Index: linux-2.6.20-rc5-uml/arch/um/kernel/mem.c
===
--- linux-2.6.20-rc5-uml.orig/arch/um/kernel/mem.c
On Fri, Jan 26, 2007 at 04:05:05PM -0800, blah deblah wrote:
I've made myself a slackware 10.2 uml virtual machine using the
2.6.19.2 kernel. I've configured it to use slirp for networking. My
problem is that the upload speed from the virtual machine seems to be
capped at 46KB/s. For example
Between 2.6.19 and 2.6.20, an #else clause was added to the
#ifdef !__ASSEMBLY__ in arch/i386/desc.h. This results in the skas3
patch inserting a declaration into the assembly portion of the file.
This is fixed by moving the declaration back into the proper branch of
the #ifdef.
Jason
---
linux/config.h is now included automatically. #including it by hand is a
build error in 2.6.19; this removes the three instances in the latest
skas3-2.6 patch.
Jason
---
include/linux/proc_mm.h |1 -
mm/proc_mm-mod.c|1 -
mm/proc_mm.c|1 -
3 files changed, 3
On Thu, Nov 16, 2006 at 10:48:16AM +0100, Nicolas Boullis wrote:
Half of all MAC-48 addresses are locally administered: those with the
second-least-significant bit of the first byte set.
Uh?
Looking at the list of allocated OUI
(http://standards.ieee.org/regauth/oui/oui.txt), I see that
On Wed, Sep 27, 2006 at 01:57:47PM -0400, Jeff Dike wrote:
Now, if there is no MAC from the command line, one is generated. We
use the microseconds from gettimeofday (20 bits), plus the low 12
bits of the pid to seed the random number generator. random() is
called twice, with 16 bits of each
[EMAIL PROTECTED] said:
Hmmm, I'm not seeing it there either. The closest I see is
skas-2.6.17-rc5-v9-pre9. Is v9 now the stable line?
Yes, you can use it safely.
maybe it's time to finally call it 2.6.17-skas3-v9? and dispense with
the -pre stuff?
Jason
Using Tomcat but need to do more?
On Mon, Jun 05, 2006 at 12:31:48PM +0200, Blaisorblade wrote:
nope. thanks, though.
map : /proc/mm map failed, err = 9
map : /proc/mm map failed, err = 9
map : /proc/mm map failed, err = 9
the 'map failed' thing repeats forever.
Try this one - it'll be shortly on the website, I've
[EMAIL PROTECTED] said:
I've being thinking to this and I'm wondering why we shouldn't do it. When we
have set no IP or 0.0.0.0, which is not a unique IP, and we bring it up, we
should choose a random MAC to use.
I agree this makes sense. Currently I'm forced to do it in a script.
It's
[EMAIL PROTECTED] said:
So what is the second bit? I only know about the broadcast/multicast bit, and
no one has bothered clueing me in on any other special bits :-)
the other one's the locally administered bit. It's a lot like rfc1918
address space in ipv4, only for ethernet. wikipedia's
[EMAIL PROTECTED] said:
Below is my take on this. Compiled, but not run, on 2.6.17-rc5.
nope. thanks, though.
Checking for the skas3 patch in the host...found
Checking for /proc/mm...found
Checking PROT_EXEC mmap in /tmp...OK
Checking for /dev/anon on the host...Not available (open failed with
[EMAIL PROTECTED] said:
Even in the other solution (PPP over SSH), or in whichever one you can think
of (I would use normal Ethernet subnets, but have many different ones, each
one with only two hosts on it), you get to use an UML as router...
Another nice tool for playing with these things
[EMAIL PROTECTED] said:
My udhcpd config allows for a small range of dhcp handled addresses
and sets the router as 192.168.1.1 and the DNS as the host's (the
laptop's) local DNS server (the laptop runs dnsmasq as a caching DNS
server).
If you configure dnsmasq properly, you can probably do
Are there remaining problems with running a skas3-patched amd64 kernels?
I'm considering doing that, and I know there were problems with it in
the past but I don't know whether they're still around.
thanks,
Jason
---
Using Tomcat but need to
[EMAIL PROTECTED] said:
I'm running a vanilla 2.6.16 UML (compiled without TT) on a vanilla
skas3-patched 2.6.16 host.
Works like a charm.
With a nonstandard memory split? Jeff said skas3 + no-TT + non-1/3 splet
should work, but I haven't tried it yet.
Jason
[EMAIL PROTECTED] said:
Now... all this works in terms of the plumbing but *all* the UMLs have
the same eth device HWAddr (mac address) so they all get the same IP.
I want them to have different IPs.
I thought, because they were all using different tun devices that they
would get
In 2.6.16, the kernel allows user-selectable address space splits. I
have 1.5G of memory, so choosing 2G/2G lets me turn off highmem. I also
have the new 2.6.16 skas3 v8.2 patch applied.
With this configuration, all umls die immediately, whether I try skas0,
skas3 or TT.
Is this something uml
On Thu, Mar 23, 2006 at 02:19:16PM -0800, David Lang wrote:
I don't know about other non-standard memory configs, but there is an
explicit config option when compiling a uml kernel to support the 2G/2G
memory split.
I see. Do you know whether a single uml binary can support arbitrary
host
[EMAIL PROTECTED] said:
we're evaluating different options for these virtual machines. for the
vmware option there is a claim that the host doesn't need to have an IP
address on a particular nic to allow a virtual machine to access things on
that nic. if the host doesn't try to process the
[EMAIL PROTECTED] said:
In the page I pointed you to there's an explaination about bridging, and on
my
homepage there's a links sections with various other articles. Between the
various stuff, you'll see that eth0 and tap0 are given no IP but only br0 is;
I'm not sure if this can be
[EMAIL PROTECTED] said:
does the uml pcap network backend still work? that's more or less what
you're asking for.
...or not. It doesn't let the guest transmit, does it? I've never tried
it.
It would be cool if a uml pcap interface could be attached directly to a
saved 'tcpdump -w' packet dump
On Wed, Mar 22, 2006 at 10:51:34AM -0800, David Lang wrote:
if your goal is for the uml to see *every* packet, bridging doesn't do
what you want. linux bridging acts as a switch, and it won't forward
packets through to the uml if it knows that the dst mac is on the
physical (ethN) side of the
[EMAIL PROTECTED] said:
My goal is to have one subnet on tap0 in which all the UMLs are too, so
you get one big virtual ethernet network.
That's possible, but you need to make UML use the daemon transport, and run
uml_switch beforehand, connecting it to a pre-setup tap0 device. See the main
[EMAIL PROTECTED] said:
I'm confused, why would you have executables in a shared library directory
instead of /usr/bin or some such? (These are runnable elf binaries, not
shared libraries, correct?)
The FHS says /usr/lib includes object files, libraries, and internal
binaries that are not
[EMAIL PROTECTED] said:
I'm confused, why would you have executables in a shared library directory
instead of /usr/bin or some such? (These are runnable elf binaries, not
shared libraries, correct?)
The FHS says /usr/lib includes object files, libraries, and internal
binaries that are not
[EMAIL PROTECTED] said:
If your distro, instead of doing:
ifconfig eth0 up
ifconfig eth0 address, does
ifconfig eth0 address, then UML will decide the Mac address based on the IP
address (will be FE:FD:ip address in hex, maybe reversed).
Has any thought been given to just choosing a random
35 matches
Mail list logo
