Newbie questions

Newbie questions

[Ethy H. Brito]

Hi All.

I've been doing some home work reading the docs and some doubts have raised.
For reference, my environment is 
	Ubuntu 20.04
	kernel 5.4.0-66 
	tc utility, iproute2-ss200127.

1) https://xdp-project.net/areas/cpumap.html#cpumap--Create-script-MQ-HTB-silo-setup says that: 
	"XPS (Transmit Packet Steering) will take precedence over any changes to 
	skb->queue_mapping. You need to disable *XDP* via mask=00 in files 
	/sys/class/net/DEV/queues/tx-*/xps_cpus"

Shouldn't it say I need to disable *XPS* (not XDP) using mask=00??

2) Taking tc_mq_htb_setup_example.sh as reference, how to enable XDP?
Since I have to disable XPS, I assume I have to enable something in replacement, right?
How to set that CPU-0 will deal with mq queue 7FFF:1, CPU-1 will deal with 7FFF:2, and so on?

3) Is XDP available in kernel 5.4.0 ?

4) Is XPS a suitable "substitute" for XDP ?

Thank you for your time.

Regards Ethy

Re: Newbie questions

[Jesper Dangaard Brouer]

On Fri, 18 Jun 2021 13:31:06 -0300
"Ethy H. Brito" <ethy.brito@inexo.com.br> wrote:

> Hi All.
> 
> I've been doing some home work reading the docs and some doubts have raised.
> For reference, my environment is 
> 	Ubuntu 20.04
> 	kernel 5.4.0-66 
> 	tc utility, iproute2-ss200127.
> 
> 1) https://xdp-project.net/areas/cpumap.html#cpumap--Create-script-MQ-HTB-silo-setup says that: 
> 	"XPS (Transmit Packet Steering) will take precedence over any changes to 
> 	skb->queue_mapping. You need to disable *XDP* via mask=00 in files 
> 	/sys/class/net/DEV/queues/tx-*/xps_cpus"
> 
> Shouldn't it say I need to disable *XPS* (not XDP) using mask=00??

You are absolutely right it is a typo. Can I ask you to fix that and
send a GitHub PR?

The file you need to change is:
 https://github.com/xdp-project/xdp-project/blob/master/areas/cpumap.org

 
> 2) Taking tc_mq_htb_setup_example.sh as reference, how to enable XDP?

XDP gets enabled when you load an XDP program.

> Since I have to disable XPS, I assume I have to enable something in
> replacement, right?

No, XPS is just a kernel feature you need to disable, because it
conflicts with the TC-BPF program usage of skb->queue_mapping.


> How to set that CPU-0 will deal with mq queue 7FFF:1, CPU-1 will deal
> with 7FFF:2, and so on?

That is the role of the XDP program that redirect into a cpumap, and
the key in the cpumap is the CPU number.

 
> 3) Is XDP available in kernel 5.4.0 ?

I think so.

The real question is what NIC card drivers are you using?


> 4) Is XPS a suitable "substitute" for XDP ?

XPS and XDP are completely unrelated kernel features.

-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer

Re: Newbie questions

[Ethy H. Brito]

On Fri, 18 Jun 2021 19:40:17 +0200
Jesper Dangaard Brouer <brouer@redhat.com> wrote:

> On Fri, 18 Jun 2021 13:31:06 -0300
> "Ethy H. Brito" <ethy.brito@inexo.com.br> wrote:
> 
> > Hi All.
> > 
> > I've been doing some home work reading the docs and some doubts have raised.
> > For reference, my environment is 
> > 	Ubuntu 20.04
> > 	kernel 5.4.0-66 
> > 	tc utility, iproute2-ss200127.
> > 
> > 1) https://xdp-project.net/areas/cpumap.html#cpumap--Create-script-MQ-HTB-silo-setup says that: 
> > 	"XPS (Transmit Packet Steering) will take precedence over any changes to 
> > 	skb->queue_mapping. You need to disable *XDP* via mask=00 in files 
> > 	/sys/class/net/DEV/queues/tx-*/xps_cpus"
> > 
> > Shouldn't it say I need to disable *XPS* (not XDP) using mask=00??  
> 
> You are absolutely right it is a typo. Can I ask you to fix that and
> send a GitHub PR?
> 
> The file you need to change is:
>  https://github.com/xdp-project/xdp-project/blob/master/areas/cpumap.org
> 

I'll have to figure it out how to do that. 
Never edited a GitHub before.

>  
> > 2) Taking tc_mq_htb_setup_example.sh as reference, how to enable XDP?  
> 
> XDP gets enabled when you load an XDP program.

Ok.

> 
> > Since I have to disable XPS, I assume I have to enable something in
> > replacement, right?  
> 
> No, XPS is just a kernel feature you need to disable, because it
> conflicts with the TC-BPF program usage of skb->queue_mapping.

Pretty advanced topic. But I'll get there.

> 
> 
> > How to set that CPU-0 will deal with mq queue 7FFF:1, CPU-1 will deal
> > with 7FFF:2, and so on?  
> 
> That is the role of the XDP program that redirect into a cpumap, and
> the key in the cpumap is the CPU number.

You mean I have to master this tutorial:
https://github.com/xdp-project/xdp-tutorial/blob/master/setup_dependencies.org

right?

> 
>  
> > 3) Is XDP available in kernel 5.4.0 ?  
> 
> I think so.
> 
> The real question is what NIC card drivers are you using?

driver=i40e driverversion=2.8.20-k duplex=full firmware=8.15

Is it ok for xdp use?

> 
> 
> > 4) Is XPS a suitable "substitute" for XDP ?  
> 
> XPS and XDP are completely unrelated kernel features.

I thought their both primary goal was to redirect packets to be handled by a specific CPU.

The difference would be that XDP does this upstream compared with XPS.
At least, I understand this way reading
https://developers.redhat.com/blog/2021/05/13/receive-side-scaling-rss-with-ebpf-and-cpumap#
 at "Faster software receive steering with XDP" paragraph
Is my understanding incorrect?

Regards

Ethy


> 
> -- 
> Best regards,
>   Jesper Dangaard Brouer
>   MSc.CS, Principal Kernel Engineer at Red Hat
>   LinkedIn: http://www.linkedin.com/in/brouer
> 


-- 

Ethy H. Brito         /"\
InterNexo Ltda.       \ /  CAMPANHA DA FITA ASCII - CONTRA MAIL HTML
+55 (12) 3797-6860     X   ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
S.J.Campos - Brasil   / \ 
 
PGP key: http://www.inexo.com.br/~ethy/0xC3F222A0.asc

Re: Newbie questions

[Ethy H. Brito]

On Fri, 18 Jun 2021 17:37:17 -0300
"Ethy H. Brito" <ethy.brito@inexo.com.br> wrote:

> On Fri, 18 Jun 2021 19:40:17 +0200
> Jesper Dangaard Brouer <brouer@redhat.com> wrote:
> 
> > On Fri, 18 Jun 2021 13:31:06 -0300
> > "Ethy H. Brito" <ethy.brito@inexo.com.br> wrote:
> >   
> > > Hi All.
> > > 
> > > I've been doing some home work reading the docs and some doubts have raised.
> > > For reference, my environment is 
> > > 	Ubuntu 20.04
> > > 	kernel 5.4.0-66 
> > > 	tc utility, iproute2-ss200127.
> > > 
> > > 1) https://xdp-project.net/areas/cpumap.html#cpumap--Create-script-MQ-HTB-silo-setup says that: 
> > > 	"XPS (Transmit Packet Steering) will take precedence over any changes to 
> > > 	skb->queue_mapping. You need to disable *XDP* via mask=00 in files 
> > > 	/sys/class/net/DEV/queues/tx-*/xps_cpus"
> > > 
> > > Shouldn't it say I need to disable *XPS* (not XDP) using mask=00??    
> > 
> > You are absolutely right it is a typo. Can I ask you to fix that and
> > send a GitHub PR?
> > 
> > The file you need to change is:
> >  https://github.com/xdp-project/xdp-project/blob/master/areas/cpumap.org

File edited. PR sent.

> 
> > 
> >   
> > > How to set that CPU-0 will deal with mq queue 7FFF:1, CPU-1 will deal
> > > with 7FFF:2, and so on?    
> > 
> > That is the role of the XDP program that redirect into a cpumap, and
> > the key in the cpumap is the CPU number.  

OK. I see that in source code.
But I am still pretty in the dark here to start using XDP.
More newbie questions are necessary.

My goal is simple: to control the bandwidth of a few (or a lot) thousands users using an of-the-shelf (almost) box. Two 10Gbps ether interface. One internal, one external.
What come in thru eth0 goes out to eth0 or eth1 and what comes in thru eth1 comes out to eth0. 

Is there a road map about what to execute and in what order to achieve this task using xdp-cpumap-tc?

I have cloned xdp-cpumap-tc to try figuring it out reading the source code. 
But things did not get together.

For instance, tc_classify_kern.c (as tc_queue_mapping_kern.c)  "talks" about a "manuel" (sic)
setup:

	tc qdisc  add dev ixgbe2 clsact
	tc filter add dev ixgbe2 egress bpf da obj tc_classify_kern.o sec tc_classify

At what point these commands are to be executed?
They are not mentioned anywhere else. (tc_mq_htb_setup_example.sh forgot these perhaps?)

Which one is be to loaded tc_classify_kern or tc_queue_mapping_kern?
Or both? None? After and before what?

In the file tc_classify_kern.c, map_ifindex_type is defined differently from xdp_iphash_to_cpu_kern.c. 
	".size_value = sizeof(struct txq_config)" in the former
and  
	".size_value = sizeof(__u32)" int the later.
Is this a "Cut and paste" typo? Are they really meant to be two different maps?

Anyway, a step by step guide would be appreciated.
Maybe it is time to start populating that BNG-router repo I was told about.
How can I start helping with that? Worth doing it?

Regards 

Ethy

Re: Newbie questions

[Jesper Dangaard Brouer]

On Mon, 21 Jun 2021 22:28:09 -0300
"Ethy H. Brito" <ethy.brito@inexo.com.br> wrote:

> On Fri, 18 Jun 2021 17:37:17 -0300
> "Ethy H. Brito" <ethy.brito@inexo.com.br> wrote:
> 
> > On Fri, 18 Jun 2021 19:40:17 +0200
> > Jesper Dangaard Brouer <brouer@redhat.com> wrote:
> >   
> > > On Fri, 18 Jun 2021 13:31:06 -0300
> > > "Ethy H. Brito" <ethy.brito@inexo.com.br> wrote:
> > >     
> > > > Hi All.
> > > > 
> > > > I've been doing some home work reading the docs and some doubts have raised.
> > > > For reference, my environment is 
> > > > 	Ubuntu 20.04
> > > > 	kernel 5.4.0-66 
> > > > 	tc utility, iproute2-ss200127.
> > > > 
> > > > 1) https://xdp-project.net/areas/cpumap.html#cpumap--Create-script-MQ-HTB-silo-setup says that: 
> > > > 	"XPS (Transmit Packet Steering) will take precedence over any changes to 
> > > > 	skb->queue_mapping. You need to disable *XDP* via mask=00 in files 
> > > > 	/sys/class/net/DEV/queues/tx-*/xps_cpus"
> > > > 
> > > > Shouldn't it say I need to disable *XPS* (not XDP) using mask=00??      
> > > 
> > > You are absolutely right it is a typo. Can I ask you to fix that and
> > > send a GitHub PR?
> > > 
> > > The file you need to change is:
> > >  https://github.com/xdp-project/xdp-project/blob/master/areas/cpumap.org  
> 
> File edited. PR sent.

Thanks merged it.

> > >     
> > > > How to set that CPU-0 will deal with mq queue 7FFF:1, CPU-1 will deal
> > > > with 7FFF:2, and so on?      
> > > 
> > > That is the role of the XDP program that redirect into a cpumap, and
> > > the key in the cpumap is the CPU number.    
> 
> OK. I see that in source code.

Yes, see the explanation in the source code.
Also read: https://github.com/xdp-project/xdp-cpumap-tc/blob/master/src/howto_debug.org

The "tc_queue_mapping_kern.c" program[1] is the simplest solution,
which only does the skb->queue_mapping, and you have to configure Linux
to set the correct TC minor:major number on a per packet basis (e.g.
via iptables see comment in code).

The "tc_classify_kern.c" program[2] is more advanced and have
implemented a IP-lookup map that have this[3] config per entry:

  struct ip_hash_info {
	/* lookup key: __u32 IPv4-address */
	__u32 cpu;
	__u32 tc_handle; /* TC handle MAJOR:MINOR combined in __u32 */
  };

 [1] https://github.com/xdp-project/xdp-cpumap-tc/blob/master/src/tc_queue_mapping_kern.c#L40-L76
 [2] https://github.com/xdp-project/xdp-cpumap-tc/blob/master/src/tc_classify_kern.c#L277
 [3] https://github.com/xdp-project/xdp-cpumap-tc/blob/master/src/common_kern_user.h#L29-L33

> But I am still pretty in the dark here to start using XDP.

Okay, then let me explain some basic concepts for xdp-cpumap-tc.

 1. XDP need to run on physical NIC with driver that supports (native) XDP.

 2. XDP is a layer before network stack, before the "SKB" is created.

 3. XDP redirect the raw frame to another CPU via XDP_REDIRECT'ing into a cpumap.

 4. The cpumap (kthread) running on remote CPU will create the SKB and
    call normal network stack on this CPU.

 5. The TC-BPF program running on remote CPU update skb->queue_mapping
    (and possibly skb->priority) to map packet into the TC-queue of
    your choosing.

Notice for you scenario there are 4 BPF-progs running, two XDP and two
TC-BPF.  See what is running via cmdline: "bpftool net"

 # bpftool net
 xdp:
 eno49(4) driver id 22
 eno50(5) driver id 26

 tc:
 eno49(4) clsact/egress tc_classify_kern.o:[tc_classify] id 42
 eno50(5) clsact/egress tc_classify_kern.o:[tc_classify] id 43

All the BPF-programs share BPF-maps to have the same config.

Maps pinned:
 # ls -1 /sys/fs/bpf/tc/globals/
 map_ifindex_type
 map_ip_hash
 map_txq_config


> More newbie questions are necessary.
> 
> My goal is simple: to control the bandwidth of a few (or a lot)
> thousands users using an of-the-shelf (almost) box. Two 10Gbps ether
> interface. One internal, one external.

I have access to a production system, that have 2x 25Gbit/s NIC (plus
VLANs for each apartment building), let me check how many customers
they have added.  They are using[2] "tc_classify_kern.c" and their
IP-map contains 6086 entries (more than I expected actually).

> What come in thru eth0 goes out to eth0 or eth1 and what comes in
> thru eth1 comes out to eth0. 
> 
> Is there a road map about what to execute and in what order to
> achieve this task using xdp-cpumap-tc?

This is already available today, and running in production at an ISP.
Sorry for the lack of documentation on how to use it, but it is done.

> I have cloned xdp-cpumap-tc to try figuring it out reading the source code. 
> But things did not get together.
> 
> For instance, tc_classify_kern.c (as tc_queue_mapping_kern.c)  "talks" about a "manuel" (sic)
> setup:
> 
> 	tc qdisc  add dev ixgbe2 clsact
> 	tc filter add dev ixgbe2 egress bpf da obj tc_classify_kern.o sec tc_classify
> 
> At what point these commands are to be executed?
> They are not mentioned anywhere else. (tc_mq_htb_setup_example.sh forgot these perhaps?)

This is handled by: tc_classify_user
 https://github.com/xdp-project/xdp-cpumap-tc/blob/master/src/tc_classify_user.c

The TC commands are called from C-code in this file:
 https://github.com/xdp-project/xdp-cpumap-tc/blob/master/src/common_user.c

The roadmap is to convert this to use the new libbpf TC API instead, as
it is a mess to have a dependency on the right iproute2 version.

 
> Which one is be to loaded tc_classify_kern or tc_queue_mapping_kern?
> Or both? None? After and before what?

Actually due to limitation in iproute2 loader, you should load
XDP-programs first (as it will create maps with BTF info).

You cannot load tc_classify_kern and tc_queue_mapping_kern simultaneously.

> 
> In the file tc_classify_kern.c, map_ifindex_type is defined
> differently from xdp_iphash_to_cpu_kern.c.
>
> 	".size_value = sizeof(struct txq_config)" in the former
> and  
> 	".size_value = sizeof(__u32)" int the later.
>
> Is this a "Cut and paste" typo? Are they really meant to be two
> different maps?

Hmm... this looks like a copy-paste error.  The tc_classify_kern.c
map_ifindex_type should have size_value = sizeof(__u32). It happens to
work because sizeof(struct txq_config) is also 4 bytes.


> Anyway, a step by step guide would be appreciated.

I'm hoping you will create/document that once you learn howto use these
programs ;-)

> Maybe it is time to start populating that BNG-router repo I was told about.
> How can I start helping with that? Worth doing it?

I think we need to convince other ISP's to join in...
... let me CC those guys again.

-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer

Re: newbie questions

[Johannes Berg]

On Mon, 2015-08-31 at 09:26 -0500, Pierre-Louis Bossart wrote:
> 
> What I was really looking for is the --integrate option that does 
> merge the backport into an existing tree.
> 

Ah. I'd suggest you not actually start with that though - it's more
complex and I'd think it's much easier to start getting all of this off
the ground as modules. Assuming the drivers you care about can be built
as modules, of course.

johannes
--
To unsubscribe from this list: send the line "unsubscribe backports" in

Re: newbie questions

[Pierre-Louis Bossart]

[-- Attachment #1: Type: text/plain, Size: 1393 bytes --]

On 08/31/2015 08:33 AM, Johannes Berg wrote:
> On Mon, 2015-08-31 at 08:21 -0500, Pierre-Louis Bossart wrote:
>>>
>>> The *to* version is never relevant. A given backport will compile
>>> against many different *to* versions.
>>
>> I guess I completely missed the concept. I was thinking that the 
>> gentree.py command would only port and adjust the delta between linux
>> -next and the version I wanted. Looks like the backport is really an 
>> add-on that will apply to multiple versions. Not sure I understand 
>> how successive changes in the tree are handled if there is a single 
>> backport.
> 
> Well, there's a single *from* version, as you say that was currently
> "next-20150731" (or that was the one you used). The backport git
> repository is maintained in lockstep with the *from* version (although
> there's usually quite a bit of wiggle room)
> 
> The result, the output of backports, will/should compile against any
> kernel starting from the earliest supported, all the way up to the
> *from* version, right now I think that's 3.0 until 4.1 or so. 

What I was really looking for is the --integrate option that does merge the backport into an existing tree.
This option is broken btw on newer kernels due to a change in the kernel Makefile in October 2014, the attached patch provides a correction (not sure how to use the right patch for the right kernel though).
-Pierre


[-- Attachment #2: 0001-enable-backports-built-in.patch --]
[-- Type: text/x-patch, Size: 1628 bytes --]

Allow backports to be integrated into vmlinux with newer Makefiles

diff --git a/Makefile b/Makefile
index f5c8983..31c7cbf 100644
--- a/Makefile
+++ b/Makefile
@@ -555,6 +555,7 @@ scripts: scripts_basic include/config/auto.conf include/config/tristate.conf \
 	$(Q)$(MAKE) $(build)=$(@)
 
 # Objects we will link into vmlinux / subdirs we need to visit
+backports-y 	:= backports/
 init-y		:= init/
 drivers-y	:= drivers/ sound/ firmware/
 net-y		:= net/
@@ -887,11 +888,13 @@ core-y		+= kernel/ mm/ fs/ ipc/ security/ crypto/ block/
 
 vmlinux-dirs	:= $(patsubst %/,%,$(filter %/, $(init-y) $(init-m) \
 		     $(core-y) $(core-m) $(drivers-y) $(drivers-m) \
+		     $(backports-y) $(backports-m) \
 		     $(net-y) $(net-m) $(libs-y) $(libs-m)))
 
 vmlinux-alldirs	:= $(sort $(vmlinux-dirs) $(patsubst %/,%,$(filter %/, \
-		     $(init-) $(core-) $(drivers-) $(net-) $(libs-))))
+		     $(init-) $(core-) $(drivers-) $(backports-) $(net-) $(libs-))))
 
+backports-y	:= $(patsubst %/, %/built-in.o, $(backports-y))
 init-y		:= $(patsubst %/, %/built-in.o, $(init-y))
 core-y		:= $(patsubst %/, %/built-in.o, $(core-y))
 drivers-y	:= $(patsubst %/, %/built-in.o, $(drivers-y))
@@ -902,7 +905,7 @@ libs-y		:= $(libs-y1) $(libs-y2)
 
 # Externally visible symbols (used by link-vmlinux.sh)
 export KBUILD_VMLINUX_INIT := $(head-y) $(init-y)
-export KBUILD_VMLINUX_MAIN := $(core-y) $(libs-y) $(drivers-y) $(net-y)
+export KBUILD_VMLINUX_MAIN := $(core-y) $(libs-y) $(drivers-y) $(net-y) $(backports-y)
 export KBUILD_LDS          := arch/$(SRCARCH)/kernel/vmlinux.lds
 export LDFLAGS_vmlinux
 # used by scripts/pacmage/Makefile

Re: newbie questions

[Johannes Berg]

On Mon, 2015-08-31 at 08:21 -0500, Pierre-Louis Bossart wrote:
> > 
> > The *to* version is never relevant. A given backport will compile
> > against many different *to* versions.
> 
> I guess I completely missed the concept. I was thinking that the 
> gentree.py command would only port and adjust the delta between linux
> -next and the version I wanted. Looks like the backport is really an 
> add-on that will apply to multiple versions. Not sure I understand 
> how successive changes in the tree are handled if there is a single 
> backport.

Well, there's a single *from* version, as you say that was currently
"next-20150731" (or that was the one you used). The backport git
repository is maintained in lockstep with the *from* version (although
there's usually quite a bit of wiggle room)

The result, the output of backports, will/should compile against any
kernel starting from the earliest supported, all the way up to the
*from* version, right now I think that's 3.0 until 4.1 or so.

johannes
--
To unsubscribe from this list: send the line "unsubscribe backports" in

Re: newbie questions

[Pierre-Louis Bossart]

>> humm... What I wanted is backport next-20150731 to v4.1.6. linux-next 
>> is checked out to next-20150731.
>> which of the two does git-revision refer two? the linux-next origin 
>> or the destination?
> 
> The *to* version is never relevant. A given backport will compile
> against many different *to* versions.

I guess I completely missed the concept. I was thinking that the gentree.py command would only port and adjust the delta between linux-next and the version I wanted. Looks like the backport is really an add-on that will apply to multiple versions. Not sure I understand how successive changes in the tree are handled if there is a single backport.

--
To unsubscribe from this list: send the line "unsubscribe backports" in

Re: newbie questions

[Johannes Berg]

On Mon, 2015-08-31 at 07:50 -0500, Pierre-Louis Bossart wrote:
> 
> > In general, this is possible. Check out the copy-list file. You can
> > edit this, and it'll copy only what's listed there.
> 
> ok. I was thinking of a more automated way that doesn't require me to 
> maintain a branch but that'll do for now.

maintaining a branch of the copy-list is simple though.

> > In order to add sound though, you'll also have to at least change
> > backport/Makefile.kernel and backport/Kconfig.sources to include 
> > the
> > sound directory in the kconfig and build.
> 
> yes that part works fine. I still have to figure out how Kconfig 
> works in the backports, I didn't get the explanations.

You can submit this part upstream btw - it'll just be disabled in the
default copy-list... unless you also submit that upstream.

> humm... What I wanted is backport next-20150731 to v4.1.6. linux-next 
> is checked out to next-20150731.
> which of the two does git-revision refer two? the linux-next origin 
> or the destination?

The *to* version is never relevant. A given backport will compile
against many different *to* versions.

johannes
--
To unsubscribe from this list: send the line "unsubscribe backports" in

Re: newbie questions

[Pierre-Louis Bossart]

Thanks for the quick answers Johannes, more below.

On 08/31/2015 02:43 AM, Johannes Berg wrote:
> Hi,
> 
>> question1: is there a way the list of packages can be selected? I'd 
>> like to use the backports infrastructure only for what I care about 
>> (media, usb, audio). It'd be really nice if I could limit the updates 
>> to the parts I care about (media, usb, sound) and remove all the 
>> networking part I don't really have a use for. It's fine if the 
>> releases include everything, I only want to limit the changes in my 
>> own workspace.
> 
> In general, this is possible. Check out the copy-list file. You can
> edit this, and it'll copy only what's listed there.

ok. I was thinking of a more automated way that doesn't require me to maintain a branch but that'll do for now.

> In order to add sound though, you'll also have to at least change
> backport/Makefile.kernel and backport/Kconfig.sources to include the
> sound directory in the kconfig and build.

yes that part works fine. I still have to figure out how Kconfig works in the backports, I didn't get the explanations.

> The existing semantic patches will be applied to it automatically, but
> it seems fairly unlikely that it'll all compile cleanly without more
> effort.

indeed.

>> question2: I am not sure I understand how the code is generated. 
>> Nothing seems to work/compile even without my changes... If I use 
>> linux-next with the next-20150731 tag and the latest backport code, I 
>> get:
>>
>> ~/backports$ ./gentree.py --clean --verbose --gitdebug --git-revision
>>   v4.1.6  ../linux-next ../linux-next-backport
> 
> You're not using next-20150731 tag? You're using v4.1.6 tag. If you
> checked out linux-next in ../linux-next, then you can simply remove the
> "--git-revision ..." argument and make the script a bit faster even.

humm... What I wanted is backport next-20150731 to v4.1.6. linux-next is checked out to next-20150731.
which of the two does git-revision refer two? the linux-next origin or the destination?

> 
>> Exception: Patch failed
> 
> This is because you're not using the right tag.
> 
>> Now if I use branch origin/linux-4.1.y, the tree is generated but 
>> usually doesn't compile (random options selected)
>>
>> Can anyone help with a known config that works?
>>
>>
>> question 3: how exactly does ckmake work and what am I missing?
>> :~/backports$ ./devel/ckmake 
>> [snip] backports/Makefile does not exist
>>
> 
> Don't worry about it for now. It's just a test tool to compile against
> a lot of trees - you need to run it on the result of backport.
> 
> johannes
> --
> To unsubscribe from this list: send the line "unsubscribe backports" in
> 

--
To unsubscribe from this list: send the line "unsubscribe backports" in

Re: newbie questions

[Johannes Berg]

Hi,

> question1: is there a way the list of packages can be selected? I'd 
> like to use the backports infrastructure only for what I care about 
> (media, usb, audio). It'd be really nice if I could limit the updates 
> to the parts I care about (media, usb, sound) and remove all the 
> networking part I don't really have a use for. It's fine if the 
> releases include everything, I only want to limit the changes in my 
> own workspace.

In general, this is possible. Check out the copy-list file. You can
edit this, and it'll copy only what's listed there.

In order to add sound though, you'll also have to at least change
backport/Makefile.kernel and backport/Kconfig.sources to include the
sound directory in the kconfig and build.

The existing semantic patches will be applied to it automatically, but
it seems fairly unlikely that it'll all compile cleanly without more
effort.

> question2: I am not sure I understand how the code is generated. 
> Nothing seems to work/compile even without my changes... If I use 
> linux-next with the next-20150731 tag and the latest backport code, I 
> get:
> 
> ~/backports$ ./gentree.py --clean --verbose --gitdebug --git-revision
>   v4.1.6  ../linux-next ../linux-next-backport

You're not using next-20150731 tag? You're using v4.1.6 tag. If you
checked out linux-next in ../linux-next, then you can simply remove the
"--git-revision ..." argument and make the script a bit faster even.

> Exception: Patch failed

This is because you're not using the right tag.

> Now if I use branch origin/linux-4.1.y, the tree is generated but 
> usually doesn't compile (random options selected)
> 
> Can anyone help with a known config that works?
> 
> 
> question 3: how exactly does ckmake work and what am I missing?
> :~/backports$ ./devel/ckmake 
> [snip] backports/Makefile does not exist
> 

Don't worry about it for now. It's just a test tool to compile against
a lot of trees - you need to run it on the result of backport.

johannes
--
To unsubscribe from this list: send the line "unsubscribe backports" in

newbie questions

[Pierre-Louis Bossart]

Hi,
I am trying to figure out how complicated it'd be to add the sound subsystem to the list of supported packages - there are many cases where I only want the latest and greatest sound drivers and nothing else beyond the minimal dependencies. Bear with me if I am asking silly questions, I am not familiar at all with the backports environment and only tried to progress from the wiki and code.

question1: is there a way the list of packages can be selected? I'd like to use the backports infrastructure only for what I care about (media, usb, audio). It'd be really nice if I could limit the updates to the parts I care about (media, usb, sound) and remove all the networking part I don't really have a use for. It's fine if the releases include everything, I only want to limit the changes in my own workspace.

question2: I am not sure I understand how the code is generated. Nothing seems to work/compile even without my changes... If I use linux-next with the next-20150731 tag and the latest backport code, I get:

~/backports$ ./gentree.py --clean --verbose --gitdebug --git-revision  v4.1.6  ../linux-next ../linux-next-backport
Get original source files from git ...
Applying patches from patches to ../linux-next-backport ...
Applying patch 0000-upstream-backport-changes/0005-uapi-compromise/mwifiex.patch
> patching file drivers/net/wireless/mwifiex/decl.h
> Hunk #1 succeeded at 26 with fuzz 1.
> patching file drivers/net/wireless/mwifiex/sta_rx.c
Applying patch backport-adjustments/crypto-ccm.patch
> patching file compat/crypto-ccm.c
> Hunk #2 succeeded at 872 (offset -60 lines).
> Hunk #3 succeeded at 898 (offset -60 lines).
Applying patch backport-adjustments/devcoredump.patch
> patching file compat/drivers-base-devcoredump.c
> patching file include/linux/backport-devcoredump.h
Not applying backport-adjustments/flow_dissector.patch, not needed
Applying patch collateral-evolutions/media/0001-dma_mmap_coherent-revert.patch
> patching file drivers/media/v4l2-core/videobuf2-dma-contig.c
Applying patch collateral-evolutions/media/0002-no_dmabuf/v4l2.patch
> patching file drivers/media/v4l2-core/Kconfig
> Hunk #1 succeeded at 68 (offset -11 lines).
> Hunk #2 succeeded at 79 (offset -11 lines).
> patching file drivers/media/v4l2-core/v4l2-mem2mem.c
> Hunk #1 succeeded at 440 (offset -19 lines).
> Hunk #2 succeeded at 454 (offset -19 lines).
> Hunk #3 succeeded at 805 with fuzz 2 (offset -36 lines).
> Hunk #4 succeeded at 814 (offset -36 lines).
> patching file drivers/media/v4l2-core/videobuf2-core.c
> Hunk #1 succeeded at 252 (offset -1 lines).
> Hunk #2 succeeded at 281 (offset -1 lines).
> Hunk #3 succeeded at 425 (offset -1 lines).
> Hunk #4 succeeded at 787 (offset -1 lines).
> Hunk #5 succeeded at 801 (offset -1 lines).
> Hunk #6 succeeded at 835 (offset -1 lines).
> Hunk #7 succeeded at 1502 (offset -5 lines).
> Hunk #8 succeeded at 1621 (offset -5 lines).
> Hunk #9 succeeded at 1684 (offset -5 lines).
> Hunk #10 succeeded at 2074 (offset -10 lines).
> Hunk #11 succeeded at 2085 (offset -10 lines).
> Hunk #12 succeeded at 2094 (offset -10 lines).
> Hunk #13 succeeded at 2401 (offset -14 lines).
> Hunk #14 succeeded at 2479 (offset -14 lines).
> Hunk #15 succeeded at 3417 (offset -22 lines).
> Hunk #16 succeeded at 3427 (offset -22 lines).
> patching file drivers/media/v4l2-core/videobuf2-dma-contig.c
> patching file drivers/media/v4l2-core/videobuf2-vmalloc.c
> patching file include/media/v4l2-mem2mem.h
> Hunk #1 succeeded at 119 (offset -2 lines).
> Hunk #2 succeeded at 244 (offset -2 lines).
> patching file include/media/videobuf2-core.h
> Hunk #3 succeeded at 463 (offset -4 lines).
> Hunk #4 succeeded at 630 (offset -13 lines).
> patching file drivers/media/usb/uvc/uvc_queue.c
> Hunk #1 FAILED at 270.
> Hunk #2 succeeded at 269 (offset -12 lines).
> 1 out of 2 hunks FAILED -- saving rejects to file drivers/media/usb/uvc/uvc_queue.c.rej
> patching file drivers/media/v4l2-core/videobuf2-dma-sg.c
> Hunk #1 succeeded at 373 (offset -4 lines).
Traceback (most recent call last):
  File "./gentree.py", line 1091, in <module>
    ret = _main()
  File "./gentree.py", line 724, in _main
    logwrite=logwrite)
  File "./gentree.py", line 906, in process
    apply_patches(args, "backport", source_dir, 'patches', bpid.target_dir, logwrite)
  File "./gentree.py", line 552, in apply_patches
    raise Exception('Patch failed')
Exception: Patch failed

Now if I use branch origin/linux-4.1.y, the tree is generated but usually doesn't compile (random options selected)

Can anyone help with a known config that works?


question 3: how exactly does ckmake work and what am I missing?
:~/backports$ ./devel/ckmake 
[snip] backports/Makefile does not exist

Thanks,
-Pierre
--
To unsubscribe from this list: send the line "unsubscribe backports" in

Re: Newbie questions

[Mark Kampe]

In your original question, I assumed you were talking about the
partitioning of a single cluster.  Now you are talking about
Geographic Disaster Recovery: the replication of data across
multiple (relatively) independent clusters.  This is not yet
supported, but it is most definitely on the road-map.

On 10/6/2012 5:08 PM, Adam Nielsen wrote:
>> The problem you are describing is called split-brain.  Ceph has an odd
>> number of monitors and quorum is required before objects can be served.
>> The partition with the smaller number of monitors will wait harmlessly
>> until connectivity is reestablished .
>
> Ah right, that makes sense.  Is this set in stone or can it be
> configured? I'm just thinking that in this scenario it could be
> beneficial to allow read-only access from the partition with the smaller
> number of monitors, if there are also clients that can only see those
> hosts.  (For example, a business with two sites, and the link between
> them goes down, so client PCs can only see their site-local servers.)

Re: Newbie questions

[Adam Nielsen]

> The problem you are describing is called split-brain.  Ceph has an odd
> number of monitors and quorum is required before objects can be served.
> The partition with the smaller number of monitors will wait harmlessly
> until connectivity is reestablished .

Ah right, that makes sense.  Is this set in stone or can it be configured? 
I'm just thinking that in this scenario it could be beneficial to allow 
read-only access from the partition with the smaller number of monitors, if 
there are also clients that can only see those hosts.  (For example, a 
business with two sites, and the link between them goes down, so client PCs 
can only see their site-local servers.)

Thanks,
Adam.

Re: Newbie questions

[Mark Kampe]

The problem you are describing is called split-brain.  Ceph has an odd number of monitors and quorum is required before objects can be served.  The partition with the smaller number of monitors will wait harmlessly until connectivity is reestablished .

Adam Nielsen <a.nielsen@shikadi.net> wrote:

>Thanks both for your answers - very informative.  I think I will set up a test 
>Ceph system on my home servers to try it out.
>
>I have one more question:
>
>Ceph seems to handle failed nodes well enough, but what about failed network 
>links?  Say you have a few systems in two locations, connected by a single 
>link.  If the link fails, you will have two isolated networks, each of which 
>will think the other has failed and presumably will try to go on as best it 
>can.  What happens when the link comes back up again?  What if the same file 
>was modified by both isolated clusters when the link was down?  What version 
>will end up back in the cluster?
>
>Thanks again,
>Adam.
>
>
>--
>To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Newbie questions

[Adam Nielsen]

Thanks both for your answers - very informative.  I think I will set up a test 
Ceph system on my home servers to try it out.

I have one more question:

Ceph seems to handle failed nodes well enough, but what about failed network 
links?  Say you have a few systems in two locations, connected by a single 
link.  If the link fails, you will have two isolated networks, each of which 
will think the other has failed and presumably will try to go on as best it 
can.  What happens when the link comes back up again?  What if the same file 
was modified by both isolated clusters when the link was down?  What version 
will end up back in the cluster?

Thanks again,
Adam.

Re: Newbie questions

[Sage Weil]

On Mon, 1 Oct 2012, Adam Nielsen wrote:
> Hi all,
> 
> I've been investigating cluster filesystems for a while now, and I have a few
> questions about Ceph I hope you don't mind me asking here.  This is in the
> context of using Ceph as a POSIX filesystem and alternative to something like
> NFS.
> 
>   1. Is Ceph stable enough for "real" use yet?  I read that upgrading to v0.48
> required a reformat, which I imagine would be a bit of an issue in a
> production system.  Is this how upgrades are normally done?  Is anyone running
> Ceph in a production environment with real data yet?

The upgrade to v0.48 required a conversion the first time each ceph-osd 
daemon was started with new code.  It was transparent to the user except 
for the fact that the conversion was slow.  It is also atypical.

The goal is for all future Ceph upgrades to be transparent and rolling 
(i.e., upgrade one daemon/machine at a time while the cluster remains 
completely available).  This is an absolute requirement for 
upgrades/updates within a stable series (e.g., v0.48 -> v0.48.2), and a 
high priority and probability between versions (argonaut -> bobtail, v0.51 
-> v0.52, etc.).

>   3. If I have multiple disks in a machine that I can dedicate to Ceph, is it
> better to RAID them and present Ceph with a single filesystem, or do you get
> better results by giving Ceph a filesystem on each disk and letting it look
> after the striping and any faulty disks?

There are arguments to be made for both configurations.  Currently we are 
using (and generally recommending) one ceph-osd per disk.  I would 
consider going the RAID route if you have limited RAM on the node or have 
a high-end RAID array that you want to take full advantage of.

sage

Re: Newbie questions

[Joao Eduardo Luis]

Hello Adam,

On 10/01/2012 01:30 PM, Adam Nielsen wrote:
> Hi all,
> 
> I've been investigating cluster filesystems for a while now, and I have
> a few questions about Ceph I hope you don't mind me asking here.  This
> is in the context of using Ceph as a POSIX filesystem and alternative to
> something like NFS.
> 
>   1. Is Ceph stable enough for "real" use yet?  I read that upgrading to
> v0.48 required a reformat, which I imagine would be a bit of an issue in
> a production system.  Is this how upgrades are normally done?  Is anyone
> running Ceph in a production environment with real data yet?

The rbd and radosgw are considered stable. CephFS on the other hand is
not advised for production.

With 0.48 (aka argonaut) came big changes, and that made the upgrade a
one-way street: upgrading to 0.48 was possible, but going back was not.
Argonaut's release notes made that bit clear
(http://ceph.com/releases/v0-48-argonaut-released/)

And yes, there are production systems backed up by Ceph. Probably the
one with most visibility is Dreamhost's DreamObjects
(http://dreamhost.com/cloud/dreamobjects/) which is now on an open
public beta.

>   2. Why does the wiki say that you can run one or three monitor
> daemons, but running two is worse than one?  Wouldn't running two be
> less work than running three?

First of all, a better source for update documentation would be
http://ceph.com/docs/master/

The monitors must reach a quorum, thus it is advised to have an
odd-number of monitors. Having an even-number of monitors may result in
some confusion if both halves disagree on something such as "what's the
most recent version of a given map". Hence, running an odd-number of
monitors is advised.

>   3. If I have multiple disks in a machine that I can dedicate to Ceph,
> is it better to RAID them and present Ceph with a single filesystem, or
> do you get better results by giving Ceph a filesystem on each disk and
> letting it look after the striping and any faulty disks?

I'm sure someone else will be able to address this better than I am!

> 
>   4. How resilient is the system?  I can find a lot of information
> saying one node can go away without any data loss, but does that mean
> losing a second node will take everything down?  Can you configure it
> such that every node has a complete copy of the cluster, so as long as
> any one node survives, all the data is available?

Depending on the replication level, how the data is placed across the
cluster and how many nodes you have in place, I would say it is fairly
certain that you wouldn't lose any data. I, for one, haven't got to
configure and deal with such a system (although other have), but ceph is
all about avoiding such data loss.

See, data is replicated across the cluster osds. As long as the
surviving nodes also have the data that was lost on the other two,
three, ..., servers, your cluster should be up and running.

You can configure how data is placed across the cluster by configuring
the crush map (http://ceph.com/docs/master/cluster-ops/crush-map/). I
don't know if having a complete copy of the cluster in each node is the
best idea ever, unless you are aiming for either a small-ish aggregated
storage capacity or massive servers.

>   5. Given that the cluster filesystem contains files, which are then
> stored as other files in a different filesystem, does this affect
> performance much? I'm thinking of something like a git repository which
> accesses file metadata a lot, and seems to suffer a bit if it's not
> running off a local disk.

Well, yes. Adding an extra layer of abstraction is bound to result in
performance loss. But on the other hand, using those native file systems
simplify ceph's task, design and also allows ceph to leverage certain
capabilities those file systems offer and that would otherwise need to
be supported by ceph itself (btrfs snapshots come to mind). So, I
suppose there's a trade-off here.

How much it affects performance? Well, there are a couple of recent
threads in the mailing list regarding such issues (feel free to skim
over them), and how they are being address.

> 
> Hopefully I'm not asking questions which are already covered in the
> documentation - if so please point me in the right direction.

http://ceph.com/docs/master/

Cheers,
  -Joao

> 
> Many thanks,
> Adam.

Newbie questions

[Adam Nielsen]

Hi all,

I've been investigating cluster filesystems for a while now, and I have a few 
questions about Ceph I hope you don't mind me asking here.  This is in the 
context of using Ceph as a POSIX filesystem and alternative to something like NFS.

   1. Is Ceph stable enough for "real" use yet?  I read that upgrading to 
v0.48 required a reformat, which I imagine would be a bit of an issue in a 
production system.  Is this how upgrades are normally done?  Is anyone running 
Ceph in a production environment with real data yet?

   2. Why does the wiki say that you can run one or three monitor daemons, but 
running two is worse than one?  Wouldn't running two be less work than running 
three?

   3. If I have multiple disks in a machine that I can dedicate to Ceph, is it 
better to RAID them and present Ceph with a single filesystem, or do you get 
better results by giving Ceph a filesystem on each disk and letting it look 
after the striping and any faulty disks?

   4. How resilient is the system?  I can find a lot of information saying one 
node can go away without any data loss, but does that mean losing a second 
node will take everything down?  Can you configure it such that every node has 
a complete copy of the cluster, so as long as any one node survives, all the 
data is available?

   5. Given that the cluster filesystem contains files, which are then stored 
as other files in a different filesystem, does this affect performance much? 
I'm thinking of something like a git repository which accesses file metadata a 
lot, and seems to suffer a bit if it's not running off a local disk.

Hopefully I'm not asking questions which are already covered in the 
documentation - if so please point me in the right direction.

Many thanks,
Adam.

Re: Newbie questions

[Ralph Alvy]

Alain wrote:
> Ralph Alvy escreveu:
>> To run my business, I use Slackware 10.2, and run the following DOS
>> programs simultaneously in a dosemu 1.2.1 MS-DOS 6.22 box:
> 
> That dosemu is too old and really bad. I coudn't use anything on it.
> 
>> WordPerfect Shell 3.1 (TSR)
>> ZipKey (TSR)
>> DPMouse (TSR)
>> DPPrint (TSR)
> 
> What are these?

WP Shell 3.1 is a menu/shell program, from which other programs are
launched. It stays in residence when the other programs are open.

ZipKey is a TSR that allows the user to, say, enter a Zip Code in a City
field, and then hit F11 (this is how I have it configured) to get ZipKey to
fill in the correct city in the City field, the correct state in the State
field, and copy the entered zip code in the Zip Code field.
 
DPMouse is a TSR that allows the use of a mouse in a DataPerfect
application. Otherwise, native DataPerfect applications don't recognize the
mouse. It also adds other capabilities to a DataPerfect application.

DPPrint is a TSR that adds certain print functions to a DataPerfect
application, including a print spooler. This print spooler doesn't work
well with dosemu 1.2.2 and dosemu 1.3.2, but works fine with dosemu 1.2.1.

>> DataPerfect (fully relational database manager)
> 
> Is this a normal "well behaviored" program?

Yes. And without the other TSRs I mentioned running at the same time,
DataPerfect applications run fine under freedos.

>> Never tried later versions of freedos after that. I'll take a look at the
>> latest official release of freedos later.
> 
> See my latest message: try "said" unstable.

Yeah, I know you mentioned "unstable", but the applications I'm talking
about are manipulating mission critical data for my business. Not sure if
the unstable freedos versions out there are okay or not.

> PS: this is list stuf, send it to the list, I wiil get it too...

When I got your message about this, I got it in private email and didn't see
it in the list, so thought you must have decided to do this privately for
some reason. Now I see the list got your message some time later. I must
have responded shortly after you posted it.

Ralph

Re: Newbie questions

[Alain]

Ralph Alvy escreveu:
> To run my business, I use Slackware 10.2, and run the following DOS programs 
> simultaneously in a dosemu 1.2.1 MS-DOS 6.22 box:

That dosemu is too old and really bad. I coudn't use anything on it.

> WordPerfect Shell 3.1 (TSR)
> ZipKey (TSR)
> DPMouse (TSR)
> DPPrint (TSR)

What are these?

> DataPerfect (fully relational database manager)

Is this a normal "well behaviored" program?

> They run perfectly in a dosemu 1.2.1 MS-DOS 6.22 box. However, in a dosemu 
> 1.2.2 MS-DOS 6.22 box, printing doesn't work properly. It also fails to work 
> properly in a dosemu 1.3.2 MS-DOS 6.22 box.

Only wit CVS-dosemu that ZI couls run all that I wanted

> In a dosemu 1.2.1 freedos box, I can't get all 5 programs to run together. 

?? You could explain better

> Never tried later versions of freedos after that. I'll take a look at the 
> latest official release of freedos later.

See my latest message: try "said" unstable.

Alain
PS: this is list stuf, send it to the list, I wiil get it too...

Re: Newbie questions

[Ralph Alvy]

Larry Alkoff wrote:
>> 
>> I imagine IBM-DOS will work, but I haven't tried it. If freedos allowed
>> me to run the DOS apps I use on a daily basis, I'd use that instead of
>> MS-DOS 6.22, but it doesn't.

> Ralph what DOS apps do not run in IBM-DOS?
> Do you know what's lacking in IBM-DOS so they can't run?
> Larry
> 
I've never tried IBM-DOS. In terms of freedos, I can't get WordPerfect Shell
3.1 to work with two TSRs I use: ZipKey and DPPrint. The former allows me
to insert City, State, Zip into a database by just typing in the Zip. The
latter allows me to certain printing things with DataPerfect relational
database applications. These work fine with MS-DOS 6.22 under dosemu,
better under dosemu 1.2.1 than either 1.2.2 or 1.3.2.

Re: Newbie questions

[Larry Alkoff]

Ralph Alvy wrote:
> Larry Alkoff wrote:
> 
> 
>>John R. Sowden wrote:
>>
>>>On Wed November 2 2005 20:12, Larry Alkoff wrote:
>>>
>>>
>>>>Ralph Alvy wrote:
>>>>
>>>>
>>>>>Larry Alkoff wrote:
>>>>>
>>>>>
>>>>>>Can I use 4dos as a replacement for command.com?
>>>>>
>>>>>Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
>>>>>
>>>>>
>>>>>-
>>>>>To unsubscribe from this list: send the line "unsubscribe linux-msdos"
>>>>>in the body of a message to majordomo@vger.kernel.org
>>>>>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>>
>>>>Thanks Ralph.  That answers the 4dos question quite satisfactorily.
>>>>
>>>>Larry
>>>
>>>I use 4dos 7.5, the latest version and ms-dos 7.10, which, if I remember
>>>correctly is the dos from win98.  I do this so I can access my 32 bit
>>>partitions from dos.
>>
>>John I presently use
>>4DOS 7.50   (Win98) DOS 7.10
>>
>>However I'd like to use IBM dos which perhaps can be purchased to get
>>away from any software liability from using Windows 98 dos.
>>
>>Any idea which IBM dos version I need and where I could get it?
>>Otherwise I'll use freedos.
> 
> 
> I imagine IBM-DOS will work, but I haven't tried it. If freedos allowed me
> to run the DOS apps I use on a daily basis, I'd use that instead of MS-DOS
> 6.22, but it doesn't.
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-msdos" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
Ralph what DOS apps do not run in IBM-DOS?
Do you know what's lacking in IBM-DOS so they can't run?
Larry


-- 
Larry Alkoff N2LA - Austin TX
Using Thunderbird on Slackware Linux

Re: Newbie questions

[Ralph Alvy]

Larry Alkoff wrote:

> John R. Sowden wrote:
>> On Wed November 2 2005 20:12, Larry Alkoff wrote:
>> 
>>>Ralph Alvy wrote:
>>>
>>>>Larry Alkoff wrote:
>>>>
>>>>>Can I use 4dos as a replacement for command.com?
>>>>
>>>>Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
>>>>
>>>>
>>>>-
>>>>To unsubscribe from this list: send the line "unsubscribe linux-msdos"
>>>>in the body of a message to majordomo@vger.kernel.org
>>>>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>
>>>Thanks Ralph.  That answers the 4dos question quite satisfactorily.
>>>
>>>Larry
>> 
>> I use 4dos 7.5, the latest version and ms-dos 7.10, which, if I remember
>> correctly is the dos from win98.  I do this so I can access my 32 bit
>> partitions from dos.
> 
> John I presently use
> 4DOS 7.50   (Win98) DOS 7.10
> 
> However I'd like to use IBM dos which perhaps can be purchased to get
> away from any software liability from using Windows 98 dos.
> 
> Any idea which IBM dos version I need and where I could get it?
> Otherwise I'll use freedos.

I imagine IBM-DOS will work, but I haven't tried it. If freedos allowed me
to run the DOS apps I use on a daily basis, I'd use that instead of MS-DOS
6.22, but it doesn't.

Re: Newbie questions

[Justin Zygmont]

On Thu, 3 Nov 2005, John R. Sowden wrote:

> On Thu November 3 2005 11:02, Larry Alkoff wrote:
>> John R. Sowden wrote:
>>> On Wed November 2 2005 20:12, Larry Alkoff wrote:
>>>> Ralph Alvy wrote:
>>>>> Larry Alkoff wrote:
>>>>>> Can I use 4dos as a replacement for command.com?
>>>>>
>>>>> Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
>>>>>
>>>>>
>>>>> -
>>>>> To unsubscribe from this list: send the line "unsubscribe linux-msdos"
>>>>> in the body of a message to majordomo@vger.kernel.org
>>>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>>
>>>> Thanks Ralph.  That answers the 4dos question quite satisfactorily.
>>>>
>>>> Larry
>>>
>>> I use 4dos 7.5, the latest version and ms-dos 7.10, which, if I remember
>>> correctly is the dos from win98.  I do this so I can access my 32 bit
>>> partitions from dos.
>>
>> John I presently use
>> 4DOS 7.50   (Win98) DOS 7.10
>>
>> However I'd like to use IBM dos which perhaps can be purchased to get
>> away from any software liability from using Windows 98 dos.
>>
>> Any idea which IBM dos version I need and where I could get it?
>> Otherwise I'll use freedos.
>
> I don't know the differences, although I assume that freedos is the only one
> that is actually in development.  what we need is a comparison chart listing
> the different dos brands/versions and the features.  then let users send in
> the features they know of.   That would be a neat site.
>
> In the big scheme of things, I am not worried about microsoft giving me grief
> re: using their dos under linux.  I have a original cd and receipt for every
> copy that I use.  I know of someone who was selling used computers and
> putting winxx and office on them.  microsoft went after him, and now he puts
> freedos on them, but that was over the top.
>
> IBM has a site that has a link for pcdos, and they were still selling it about
> 6 months ago when I saw the site.

PC-DOS 2000 (PS-DOS 8.0)  is available from their site for about $80 last 
time I checked.  You can still use DR-DOS 7.03, this site has it: 
telnet://solarflow.dyndns.org

Re: Newbie questions

[John R. Sowden]

On Thu November 3 2005 11:02, Larry Alkoff wrote:
> John R. Sowden wrote:
> > On Wed November 2 2005 20:12, Larry Alkoff wrote:
> >>Ralph Alvy wrote:
> >>>Larry Alkoff wrote:
> >>>>Can I use 4dos as a replacement for command.com?
> >>>
> >>>Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
> >>>
> >>>
> >>>-
> >>>To unsubscribe from this list: send the line "unsubscribe linux-msdos"
> >>> in the body of a message to majordomo@vger.kernel.org
> >>>More majordomo info at  http://vger.kernel.org/majordomo-info.html
> >>
> >>Thanks Ralph.  That answers the 4dos question quite satisfactorily.
> >>
> >>Larry
> >
> > I use 4dos 7.5, the latest version and ms-dos 7.10, which, if I remember
> > correctly is the dos from win98.  I do this so I can access my 32 bit
> > partitions from dos.
>
> John I presently use
> 4DOS 7.50   (Win98) DOS 7.10
>
> However I'd like to use IBM dos which perhaps can be purchased to get
> away from any software liability from using Windows 98 dos.
>
> Any idea which IBM dos version I need and where I could get it?
> Otherwise I'll use freedos.

I don't know the differences, although I assume that freedos is the only one 
that is actually in development.  what we need is a comparison chart listing 
the different dos brands/versions and the features.  then let users send in 
the features they know of.   That would be a neat site.

In the big scheme of things, I am not worried about microsoft giving me grief 
re: using their dos under linux.  I have a original cd and receipt for every 
copy that I use.  I know of someone who was selling used computers and 
putting winxx and office on them.  microsoft went after him, and now he puts 
freedos on them, but that was over the top.

IBM has a site that has a link for pcdos, and they were still selling it about 
6 months ago when I saw the site.
-- 
John R. Sowden
AMERICAN SENTRY SYSTEMS, INC.
Residential & Commercial Alarm Service
UL Listed Central Station
Serving the San Francisco Bay Area Since 1967
mail@americansentry.net
www.americansentry.net

Re: Newbie questions

[Larry Alkoff]

John R. Sowden wrote:
> On Wed November 2 2005 20:12, Larry Alkoff wrote:
> 
>>Ralph Alvy wrote:
>>
>>>Larry Alkoff wrote:
>>>
>>>>Can I use 4dos as a replacement for command.com?
>>>
>>>Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
>>>
>>>
>>>-
>>>To unsubscribe from this list: send the line "unsubscribe linux-msdos" in
>>>the body of a message to majordomo@vger.kernel.org
>>>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>>Thanks Ralph.  That answers the 4dos question quite satisfactorily.
>>
>>Larry
> 
> I use 4dos 7.5, the latest version and ms-dos 7.10, which, if I remember 
> correctly is the dos from win98.  I do this so I can access my 32 bit 
> partitions from dos.

John I presently use
4DOS 7.50   (Win98) DOS 7.10

However I'd like to use IBM dos which perhaps can be purchased to get 
away from any software liability from using Windows 98 dos.

Any idea which IBM dos version I need and where I could get it?
Otherwise I'll use freedos.


-- 
Larry Alkoff N2LA - Austin TX
Using Thunderbird on Slackware Linux

Re: Newbie questions

[John R. Sowden]

On Wed November 2 2005 20:12, Larry Alkoff wrote:
> Ralph Alvy wrote:
> > Larry Alkoff wrote:
> >>Can I use 4dos as a replacement for command.com?
> >
> > Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
> >
> >
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-msdos" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
> Thanks Ralph.  That answers the 4dos question quite satisfactorily.
>
> Larry
I use 4dos 7.5, the latest version and ms-dos 7.10, which, if I remember 
correctly is the dos from win98.  I do this so I can access my 32 bit 
partitions from dos.
-- 
John R. Sowden
AMERICAN SENTRY SYSTEMS, INC.
Residential & Commercial Alarm Service
UL Listed Central Station
Serving the San Francisco Bay Area Since 1967
mail@americansentry.net
www.americansentry.net

Re: Newbie questions

[Ralph Alvy]

Larry Alkoff wrote:

> Ralph Alvy wrote:
>> Larry Alkoff wrote:
>> 
>> 
>>>Can I use 4dos as a replacement for command.com?
>> 
>> 
>> Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
>> 
>> 
>> -
>> To unsubscribe from this list: send the line "unsubscribe linux-msdos" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> 
> Thanks Ralph.  That answers the 4dos question quite satisfactorily.
> 
> Larry
> 
In terms of stuff in your autoexec.bat, I even load TSRs there. However, I
find that I must use dosemu 1.2.1 instead of either 1.2.2 or 1.3.2.
Otherwise printing doesn't work well. If remove those TSRs, then I can
print fine under 1.2.2 and 1.3.2, but I need those TSRs for my DOS app that
runs my business.

You can find all you need to get printing to work in the dosemu.conf file.

Re: Newbie questions

[Larry Alkoff]

Ralph Alvy wrote:
> Larry Alkoff wrote:
> 
> 
>>Can I use 4dos as a replacement for command.com?
> 
> 
> Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.
> 
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-msdos" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
Thanks Ralph.  That answers the 4dos question quite satisfactorily.

Larry

-- 
Larry Alkoff N2LA - Austin TX
Using Thunderbird on Slackware Linux

Re: Newbie questions

[Ralph Alvy]

Larry Alkoff wrote:

> Can I use 4dos as a replacement for command.com?

Yes. I use MS-DOS 6.22 with 4DOS replacing command.com.

Re: Newbie questions

[Justin Zygmont]

On Tue, 1 Nov 2005, Larry Alkoff wrote:

> Some questions from a dosemu newbie that I could not find in documentation. 
> Using dosemu version 1.2.2.2.0 under Slackware 10.1.
>
> 0.  Where can I find more documentation?
>
> 1.  How do I print?  My system uses CUPS.
>
>    In particular I wish to print reports using the Supercalc version 4 
> spreadsheet program and a Deskjet 500 through the CUPS system. Supercalc is 
> working just fine except for printing.  Does anyone have any experience with 
> this program and dosemu?
>
> 2.  Can I use 4dos as a replacement for command.com?
>
> 3.  Where can I find guidance on the allow commands in config.sys and 
> autoexec.bat?

I think there is the best documentation available on the dosemu site: 
http://dosemu.org  Also, i'd strongly recomment to upgrade your dosemu to 
the latest version.  You should see the settings to print from the 
dosemu.conf.

Newbie questions

[Larry Alkoff]

Some questions from a dosemu newbie that I could not find in 
documentation.  Using dosemu version 1.2.2.2.0 under Slackware 10.1.

0.  Where can I find more documentation?

1.  How do I print?  My system uses CUPS.

     In particular I wish to print reports using the Supercalc version 4 
    spreadsheet program and a Deskjet 500 through the CUPS system. 
Supercalc is working just fine except for printing.  Does anyone have 
any experience with this program and dosemu?

2.  Can I use 4dos as a replacement for command.com?

3.  Where can I find guidance on the allow commands in config.sys and 
autoexec.bat?

-- 
Larry Alkoff N2LA - Austin TX
Using Thunderbird on Slackware Linux

Re: Newbie questions

[Luke Kenneth Casson Leighton]

On Thu, Oct 06, 2005 at 05:04:28PM -0400, Ivan Gyurdiev wrote:
> 
> >Q1. Is the idea here for SELinux to create specialized domains for all
> >possible desktop apps (on the server side it seems the plan is to
> >eventually confine all daemons)?
> I think that's what the plan is/was...whether this is actually possible 
> remains to be discovered. Desktop apps are very difficult to deal with, 
> since they tend to be highly complex, and interact in complicated ways.

 one of those complexities is well illustrated by KDE's
 favourite game: doing an optimisation job which should
 otherwise really be done by the kernel, the compiler or
 the linker.

 for example, prior to the existence of prelink, a tool for prelinking
 dynamically linked libraries and preallocating fixed locations right
 across the board of all .sos and their references in all programs that
 use them - yes it takes about an hour to run prelink on a desktop
 system - the kde team came up with "kdeinit".

 kdeinit is a program that preloads virtually every component of the kde
 system instead of as independent executables as .so libraries.

 then, instead of exec()ing new programs, a fork() is performed instead
 and the alternative .so called.  so, you run kdesktop - actually it's
 kdeinit loading the kde_kdesktop.so - and that runs the kde start menu
 program - and actually it's kdeinit loading the kde_kstartmenu.so.

 the fork()ed process has all of the previously called dynamically
 allocated library functions already sorted.

 
 a similar thing was done with the networking: one program, the kde io
 handler, will fork() and maintain a cache of outgoing connections,
 whereby you will be presented with access to those network connections
 via the kde io handler, which performs proxying on your behalf via unix
 domain sockets.

 the idea is to have the kde io handler "cache" connections
 such as authenticated IMAP, POP3, SSL connections, HTTP
 connections which required a password - that sort of thing.

 fortunately, the KDE team saw fit to do away with both these things if
 you set some environment variables, one of which is only active in KDE
 3.4 and above.

 export KDE_IS_PRELINKED="1" will stop the kdeinit silliness and

 export KDE_EXECS_SLAVES="1" will stop the io slave silliness.

 i pointed out to the kde development team that their cacheing
 of network connections presents quite a serious security risk,
 namely that any one program (e.g. irc) could be compromised
 and end up hacking in to _all_ outgoing network connections.

 they were quite good about it and did actually acknowledge that there
 was a potential problem.

 i don't know if anything came of my pointing it out, though.

 l.


--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to majordomo@tycho.nsa.gov with
the words "unsubscribe selinux" without quotes as the message.

Re: Newbie questions

[Ivan Gyurdiev]

> Q1. Is the idea here for SELinux to create specialized domains for all
> possible desktop apps (on the server side it seems the plan is to
> eventually confine all daemons)?
I think that's what the plan is/was...whether this is actually possible 
remains to be discovered. Desktop apps are very difficult to deal with, 
since they tend to be highly complex, and interact in complicated ways.
>   If not then any SELinux unaware app on
> exec will run within  Mozilla domain if called form within the browser
> (viewers/players/editors)?
>   
The app does not have to be SELinux aware - this is done automatically 
for the app, based on policy. Applications which do not have a 
transition defined in policy from mozilla to a different type continue 
to run as mozilla (and usually get lots of denials).
> What about invocation from shell...will it then run within user_t domain?
>   
That depends on whether or not transitions are defined from user_t to a 
different domain upon executing the application. If no policy has been 
written for an application, for example, it would run as user_t (and as 
$1_mozilla_t in the case above).


--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to majordomo@tycho.nsa.gov with
the words "unsubscribe selinux" without quotes as the message.

Newbie questions

[Gaurav Poothia]

Hello!
I have been looking at macros/program/mozilla_macros.te in file (ver 1.26)

Mozilla seems to run in its own derived domain(based on the user
running it). It then seems to create these domain transitions for many
execs such as:
-> To a derived domain on calling a java program
-> To specialized domains on exec of mplayer, lpr etc

Q1. Is the idea here for SELinux to create specialized domains for all
possible desktop apps (on the server side it seems the plan is to
eventually confine all daemons)?  If not then any SELinux unaware app on
exec will run within  Mozilla domain if called form within the browser
(viewers/players/editors)?
What about invocation from shell...will it then run within user_t domain?

Q2 When I ran an strace on mozilla opening a document using xpdf it seemed
to call a clone() which then exec to xpdf rather than call fork().
Are both fork and clone treated identically by the SELinux?
If yes would xpdf then run inside the prefix_mozilla_t domain (since
SELinux seems be xpdf unaware)

I am sorry I pbbly could answer some of these questions empirically but I
haven't installed the policy. Btw is there a play machine out there with
1.26
installed?

Thanks
-Gaurav


--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to majordomo@tycho.nsa.gov with
the words "unsubscribe selinux" without quotes as the message.

Newbie questions

[Gaurav Poothia]

Hello!
I have been looking at macros/program/mozilla_macros.te in file (ver 1.26)

Mozilla seems to run in its own derived domain(based on the user
running it). It then seems to create these domain transitions for many
execs such as:
-> To a derived domain on calling a java program
-> To specialized domains on exec of mplayer, lpr etc

Q1. Is the idea here for SELinux to create specialized domains for all
possible desktop apps (on the server side it seems the plan is to
eventually confine all daemons)?  If not then any SELinux unaware app on
exec will run within  Mozilla domain if called form within the browser
(viewers/players/editors)?
What about invocation from shell...will it then run within user_t domain?

Q2 When I ran an strace on mozilla opening a document using xpdf it seemed
to call a clone() which then exec to xpdf rather than call fork().
Are both fork and clone treated identically by the SELinux?
If yes would xpdf then run inside the prefix_mozilla_t domain (since
SELinux seems be xpdf unaware)

I am sorry I pbbly could answer some of these questions empirically but I
haven't installed the policy. Btw is there a play machine out there with 1.26
installed?

Thanks
-Gaurav

--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to majordomo@tycho.nsa.gov with
the words "unsubscribe selinux" without quotes as the message.

Re: Newbie questions

[Scott Miller]

I'm moving again.  Thank you!

Geert Uytterhoeven wrote:

> On Wed, 19 Jan 2005, Scott Miller wrote:

>>Is there a missing step in my configuration, or am I using the wrong tools?
> 
> 
> You must use a compiler targeted for mips(el)-linux.

Re: Newbie questions

[Geert Uytterhoeven]

On Wed, 19 Jan 2005, Scott Miller wrote:
> I'm trying to build a MIPS kernel (2.4) and I'm stuck.
> 
> I'm using the MIPS SDE-lite installed on a linux box, but I don't think I have
> it configured properly.  (I have to define things like -D_MIPS_SZLONG in
> CPPFLAGS to get it to compile.)
> 
> I've gotten it all to compile, but it fails to link at the end with undefined
> reference to handle_reserved, handle_watch, handle_adel, etc.
> I assume these are intrinsics in the MIPS libraries, but I can't find them.
> 
> Is there a missing step in my configuration, or am I using the wrong tools?

You must use a compiler targeted for mips(el)-linux.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Newbie questions

[Scott Miller]

Hi.

I'm trying to build a MIPS kernel (2.4) and I'm stuck.

I'm using the MIPS SDE-lite installed on a linux box, but I don't think 
I have it configured properly.  (I have to define things like 
-D_MIPS_SZLONG in CPPFLAGS to get it to compile.)

I've gotten it all to compile, but it fails to link at the end with 
undefined reference to handle_reserved, handle_watch, handle_adel, etc.
I assume these are intrinsics in the MIPS libraries, but I can't find them.

Is there a missing step in my configuration, or am I using the wrong tools?
-- 
"Outside of a dog, a book is a mans best friend. Inside of a dog, it's
  too dark to read." - Groucho Marx

Re: Newbie Questions

[Marco Gerards]

Joseph Swaminathan <jswamina@cisco.com> writes:

Hi Joseph,

>      Thanks very much for the reply. By FC HBA I
> meant Fiber Chanel Host Bus Adapter. Since you
> mentioned no networking support, I assume FC HBA
> is also ruled out.

Right.

If it is important for you, you could add support to GRUB Legacy.  It
has networking support.  The disadvantage of doing that is that new
features are not accepted for GRUB Legacy.  Another disadvantage is
that it will become obsolete someday.

If you are interested in working on networking support for GRUB 2,
that would be great...

--
Marco

Re: Newbie Questions

[Joseph Swaminathan]

Marco,

     Thanks very much for the reply. By FC HBA I
meant Fiber Chanel Host Bus Adapter. Since you
mentioned no networking support, I assume FC HBA
is also ruled out.

thanks
Joseph

Marco Gerards wrote:
> Joseph Swaminathan <jswamina@cisco.com> writes:
> 
> 
>>    Iam new to grub. I have few basic questions.
>>Please let me know if this not the right forum
>>for such questions.
> 
> 
> For GRUB 2, it is.
> 
> 
>>    1. Does Grub/Grub 2, support iSCSI / FC HBA,
>>       for loading from SAN.
> 
> 
> 
> For iSCSI: GRUB 2 does not have networking support yet.  I have no
> knowledge about FC HBA, I don't even know what it is, but it is save
> to assume it is not supported by GRUB 2.
> 
> 
>>    2. Is there a document on Grub 2.
> 
> 
> Only the wiki, documentation has to be written and it is an important
> item on the todo list.
> 
> 
>>    3. I couldnt access the GRUB wiki page,
>>       http://www.autistici.org/grub, given
>>       in the GNU Grub web site. It says
>>       connection refused. Is this site down,
>>       or I need some access permission to
>>       visit this site.
> 
> 
> Right, it seems to be down.  AFAIK you do not need special
> permission...
> 
> --
> Marco
> 
> 
> 
> _______________________________________________
> Grub-devel mailing list
> Grub-devel@gnu.org
> http://lists.gnu.org/mailman/listinfo/grub-devel
> 

Re: Newbie Questions

[Marco Gerards]

Joseph Swaminathan <jswamina@cisco.com> writes:

>     Iam new to grub. I have few basic questions.
> Please let me know if this not the right forum
> for such questions.

For GRUB 2, it is.

>     1. Does Grub/Grub 2, support iSCSI / FC HBA,
>        for loading from SAN.


For iSCSI: GRUB 2 does not have networking support yet.  I have no
knowledge about FC HBA, I don't even know what it is, but it is save
to assume it is not supported by GRUB 2.

>     2. Is there a document on Grub 2.

Only the wiki, documentation has to be written and it is an important
item on the todo list.

>     3. I couldnt access the GRUB wiki page,
>        http://www.autistici.org/grub, given
>        in the GNU Grub web site. It says
>        connection refused. Is this site down,
>        or I need some access permission to
>        visit this site.

Right, it seems to be down.  AFAIK you do not need special
permission...

--
Marco

Newbie Questions

[Joseph Swaminathan]

    Iam new to grub. I have few basic questions.
Please let me know if this not the right forum
for such questions.

    1. Does Grub/Grub 2, support iSCSI / FC HBA,
       for loading from SAN.

    2. Is there a document on Grub 2.

    3. I couldnt access the GRUB wiki page,
       http://www.autistici.org/grub, given
       in the GNU Grub web site. It says
       connection refused. Is this site down,
       or I need some access permission to
       visit this site.

thanks
Joseph

Re: Newbie questions

[Jan Rychter]

[-- Attachment #1: Type: text/plain, Size: 2458 bytes --]

>>>>> "Steven" == Steven Hand <Steven.Hand@cl.cam.ac.uk>:
 >> I've just started looking at Xen and I have some questions. These
 >> are all fairly basic, as there is preciously little overview
 >> documentation (most of it seems to be fairly detailed).
 >>
 >> First of all, I'm coming from an UML (User-Mode Linux) and VMware
 >> background. I'm using UML for a number of things, and I'm generally
 >> happy with it, except for the performance and lack of suspend/resume
 >> functionality.

[...many interesting explanations snipped...]

 >> -- if yes, is it reasonable to expect that a patched (customized for
 >>    a laptop) kernel that includes swsusp2 and encryption (among
 >>    other things) will be able to support Xen?

 Steven> So encryption should not be a problem; swsusp2 might be
 Steven> depending on how much power-management support it requires. To
 Steven> try this out you'd need to build a xenolinux kernel from
 Steven> scratch (i.e. get the source, build xen, build a xenolinux tree
 Steven> (see HOWTO), patch xenolinux tree, cross fingers, build,
 Steven> boot...).

 Steven> Note that the xenolinux kernel is based on 2.4.25 -- there is
 Steven> no support for 2.6 as yet.

 >> -- is Xen at all suitable for the following usage scenario: a laptop
 >> host, with frequently changing networking, software suspend (via
 >> swsusp2), Xenolinux without real-world networking
 >> (e.g. host<->Xenolinux communication only)?

 Steven> Laptop is fine in general (several of us run xen and xenolinux
 Steven> on our laptops), although some funky laptops may cause problems
 Steven> (we support only a subset of device drivers in the current
 Steven> model)

[...]

Thanks a lot for your (and Ian Pratt's) answers. They have indeed helped
me a lot in understanding whan Xen is (and what it is not).

I was really looking for a way to make my virtual world persistent
(suspendable and preferably also snapshottable) and independent of the
crash of the day being served to me by Linux. In particular, the USB
subsystem in Linux ranks highly among the crappiest things I have to
use.

From your explanations it seems that Xen is designed for virtual hosting
environment and isn't really the way to go for me. UML (user-mode linux)
suits my needs better. Except it doesn't have suspend/resume
functionality. Oh well, I guess I'll just have to wait for that.

many thanks for your time,
--J.

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

Re: Newbie questions

[Ian Pratt]

> I've just started looking at Xen and I have some questions. These are
> all fairly basic, as there is preciously little overview documentation
> (most of it seems to be fairly detailed).
> 
> First of all, I'm coming from an UML (User-Mode Linux) and VMware
> background. I'm using UML for a number of things, and I'm generally
> happy with it, except for the performance and lack of suspend/resume
> functionality.
> 
> My questions:
> 
>   -- as I understand, Xen *requires* running a custom kernel, it isn't
>      just a user-level application like UML or VMware?

Yes. It's more like VMware ESX server: rather than having a 'hostOS' 
there's a custom VMM.

>   -- if yes, is it reasonable to expect that a patched (customized for a
>      laptop) kernel that includes swsusp2 and encryption (among other
>      things) will be able to support Xen?

Given that the host is a custom VMM rather than linux, this
question doesn't make sense.

You can suspend/migrate guest OSes, but not the host VMM itself.

As regards encryption, you can compile it into the
guests. There's also some support for network encryption in the
Xen 1.3vnet tree. This will be rather more complete in the new IO
model in 1.4.

>   -- is Xen at all suitable for the following usage scenario: a laptop
>      host, with frequently changing networking, software suspend (via
>      swsusp2), Xenolinux without real-world networking
>      (e.g. host<->Xenolinux communication only)?

Xen is (currently) more targeted at clusters of servers rather
than laptops. However, I do have Xen installed on my Toshiba
R100.

Ian


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click

Re: Newbie questions

[Steven Hand]

>I've just started looking at Xen and I have some questions. These are
>all fairly basic, as there is preciously little overview documentation
>(most of it seems to be fairly detailed).
>
>First of all, I'm coming from an UML (User-Mode Linux) and VMware
>background. I'm using UML for a number of things, and I'm generally
>happy with it, except for the performance and lack of suspend/resume
>functionality.
>
>My questions:
>
>  -- as I understand, Xen *requires* running a custom kernel, it isn't
>     just a user-level application like UML or VMware?

So Xen = "the host" from your point of view; however it is /not/ an 
operating system itself (i.e. it has no notion of filesystem, network 
stack, process, user, etc), and so you cannot do anything with a system
without at least one guest OS. This is by contrast with UML and (most
versions of) VMWare. 

The basic setup then is Xen + Domain 0. This is the first guest OS 
booted, and is booted via grub along with Xen itself. Domain 0 by 
default runs a XenoLinux kernel which is a modified kernel built to 
understand that it is running on top of Xen rather than the bare 
hardware. The XenoLinux kernel in turn allows proceses, etc, as per
regular linux. In terms of priviliege levels: 

   - Xen runs in ring 0 and can access all physical memory and devices

   - XenoLinux kernel runs in ring 1 and can access a subset of physical
     memory as allocated by Xen, and can access devices via Xen  (this 
     final bit is changing currently but is not yet ready for general 
     release) 
  
   - XenoLinux processes run in ring 3 exactly as per normal. 

Hence once you log into domain 0, it looks pretty much exactly like 
a regular linux, with full application binary compatability, etc. In 
addition there's an extended interface which allows one to create 
additional guest operating systems (domains 1, 2, 3, ....), each with
their own allocation of physical memory, virtual devices, etc. 

>  -- if yes, is it reasonable to expect that a patched (customized for a
>     laptop) kernel that includes swsusp2 and encryption (among other
>     things) will be able to support Xen?

So encryption should not be a problem; swsusp2 might be depending on 
how much power-management support it requires. To try this out you'd 
need to build a xenolinux kernel from scratch (i.e. get the source, 
build xen, build a xenolinux tree (see HOWTO), patch xenolinux tree, 
cross fingers, build, boot...). 

Note that the xenolinux kernel is based on 2.4.25 -- there is no 
support for 2.6 as yet. 


>  -- is Xen at all suitable for the following usage scenario: a laptop
>     host, with frequently changing networking, software suspend (via
>     swsusp2), Xenolinux without real-world networking
>     (e.g. host<->Xenolinux communication only)?

Laptop is fine in general (several of us run xen and xenolinux on our 
laptops), although some funky laptops may cause problems (we support only
a subset of device drivers in the current model) 

Frequently changing networking: shouldn't be a problem as long as 
standard linux stuff is used. 

So w.r.t suspend, see above. 

"Xenolinux without real-world networking (e.g. host<->Xenolinux communication
only)" -- this doesn't make much sense in our world as the 'host' (Xen) is
not itself doing any networking save for basic device driver stuff. The
analagous setup for us would be domain 0 with 'real world' networking, and
all other domains with e.g. 169.254.x.y networking. This is a fairly 
standard setup for us (with NAT in domain 0 if you want ot allow other 
domains to access the real world, or no NAT if you don't want this). Details
in the HOWTOs and docs I belive. 

>I'd appreciate any advice.

HTH, 

cheers, 

S.



-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click

Newbie questions

[Jan Rychter]

[-- Attachment #1: Type: text/plain, Size: 1018 bytes --]

I've just started looking at Xen and I have some questions. These are
all fairly basic, as there is preciously little overview documentation
(most of it seems to be fairly detailed).

First of all, I'm coming from an UML (User-Mode Linux) and VMware
background. I'm using UML for a number of things, and I'm generally
happy with it, except for the performance and lack of suspend/resume
functionality.

My questions:

  -- as I understand, Xen *requires* running a custom kernel, it isn't
     just a user-level application like UML or VMware?

  -- if yes, is it reasonable to expect that a patched (customized for a
     laptop) kernel that includes swsusp2 and encryption (among other
     things) will be able to support Xen?

  -- is Xen at all suitable for the following usage scenario: a laptop
     host, with frequently changing networking, software suspend (via
     swsusp2), Xenolinux without real-world networking
     (e.g. host<->Xenolinux communication only)?

I'd appreciate any advice.

thanks,
--J.

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

Re: Newbie questions

[Axel Siebenwirth]

Hi Gustavo!

On Sat, 03 Aug 2002, Gustavo Sverzut Barbieri wrote:

> I'm trying to use ACPI withou success. The kernel
> 2.5.{29,30} don't even compile (error with: save_flags
> and restore_flags, listed below).

drivers/acpi/system.c lacks the inclusion of linux/interrupt.h.

So add
#include <linux/interrupt.h>
to drivers/acpi/system.c and it should build.

> I got acpid, but I got nothing more than mapping the
> buttons (power and lid) to run what I desire. I wanna
> know how I put it to sleep (suspend, and how to come
> back)

Don't really know about but Pavel Machek is working on software suspend for
2.5 kernels. Tried it every now and then but it's not 100% working yet since
it's still under development as the whole 2.5 kernel is.
Better is you stickt to 2.4 kernels.
To get acpid work with 2.5 kernels you need a patch for acpid to get it to 
work with latest acpi.

Regards,
Axel


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Newbie questions

[Gustavo Sverzut Barbieri]

Hello,

I'm a newbie to ACPI (just bought a
desknote(www.desknote.biz) laptop that ships w/ that).
I'm trying to use ACPI withou success. The kernel
2.5.{29,30} don't even compile (error with: save_flags
and restore_flags, listed below).

I'm using red hat 2.4.18-3 recompiled right now, that
looks too old (I tried 2.5.24 before, the /proc/acpi
subsys gives more info)

I got acpid, but I got nothing more than mapping the
buttons (power and lid) to run what I desire. I wanna
know how I put it to sleep (suspend, and how to come
back)

Best regards.

Gustavo

--------------
make bzImage errors:

drivers/built-in.o: In function `acpi_system_suspend':
drivers/built-in.o(.text+0x3220e): undefined reference
to `save_flags'
drivers/built-in.o(.text+0x3223b): undefined reference
to `restore_flags'


_______________________________________________________________________
Yahoo! PageBuilder
O super editor para criação de sites: é grátis, fácil e rápido.
http://br.geocities.yahoo.com/v/pb.html


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Newbie Questions

[Daniel]

1) if you can use cin.getline(), for parameters see in doc
2) to convert you can use atoi-i for int (const char *) or atol - l for
long,
    see man atoi, also you have to include stdlib (man tells you the needed
header also)
3) include <time.h> and use gmtime, it returns a pointer to a struct with
all information
    you will not need to convert the year at all, it is in this struct .(as
member)
4) I don't know what is wrong, but try it in this way
    int c;
    while((c=fgetc(pF))!=EOF){// it reads first from file, c is an int !!}
Ok , good luck

----- Original Message -----
From: "The Gyzmo" <gyzmobro@linuxfreemail.com>
Newsgroups: ka.lists.linux.c.programming
Sent: Sunday, April 07, 2002 3:57 AM
Subject: Newbie Questions


> Hello. I'm new to C and I'm working on an age program. I have some
> questions:
>
> 1) Gcc tells me that the gets() function is dangerous and should not be
> used - which function could I replace gets() with that does exactly the
> same thing?
>
> 2) In my program, I want to calculate the age of people from their birth
> date whithout having to make the user input the current year. Which time
> funciton should I use to get the year and if it outputs a string, how
> would I convert that string to an integer?
>
> 3) One of the goals of my program is to input birth dates from a
> database file, display them all, and then find the average. I have a
> 'while(!feof(fp))' loop that takes each line in the file and extracts
> the birth dates and counts how many entries there are in the file, but
> it loops one too many times - that is it also loops after it reaches the
> EOF. This results in 2 of the same entry being displayed and the counter
> for how many entries there are being 1 too high. Which function can I
> use to loop only before it reaches the EOF?
>
> Thanks for any help in advance,
> Serban Giuroiu
>
> Get your own FREE E-mail address at http://www.linuxfreemail.com
> Linux FREE Mail is 100% FREE, 100% Linux, 100% better, and 100% yours!
>
> -
> To unsubscribe from this list: send the line "unsubscribe
linux-c-programming" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>
>
> --------------------------------------------------------------------------
-----
>    Achtung: diese Newsgruppe ist eine unidirektional gegatete
Mailingliste.
>      Antworten nur per Mail an die im Reply-To-Header angegebene Adresse.
>                    Fragen zum Gateway -> newsmaster@inka.de.
> --------------------------------------------------------------------------
-----

Re: Newbie Questions

[Glynn Clements]

Gyzmobro@aol.com wrote:

> Linux Free Mail has been down so I haven't had a chance to respond to this 
> thread and am now forced to use a proprietary email system. I hope I haven't 
> missed parts of this thread and I don't have the last email, but here I go:
> 
> >From what you've said, I would have to write a while loop that goes through
> the file one line at a time and checks to see if it's
> at the EOF through an if statement. How would I make
> the program go through the file line by line? Is this 
> if statement right:?
> 
> char c_check_eof[10];
> 
> if(fgets(c_check_eof, 10, fp) == "EOF")
> {
>  break;
> }

No. It should be:

	char line[80];	/* large enough to hold one line of input */

	...

	for (;;) {
		if (!fgets(line, sizeof(line), fp))
			break;

		/* if we get here, "line" holds one line of input */
		...
	}

-- 
Glynn Clements <glynn.clements@virgin.net>

Re: Newbie Questions

[Gyzmobro]

Linux Free Mail has been down so I haven't had a chance to respond to this 
thread and am now forced to use a proprietary email system. I hope I haven't 
missed parts of this thread and I don't have the last email, but here I go:

From what you've said, I would have to write a while loop that goes through
the file one line at a time and checks to see if it's
at the EOF through an if statement. How would I make
the program go through the file line by line? Is this 
if statement right:?

char c_check_eof[10];

if(fgets(c_check_eof, 10, fp) == "EOF")
{
 break;
}

Serban Giuroiu

Newbie questions

[Charles Steinkuehler]

OK, I have to ask for help.

I'm trying to compile MTD on a 2.2.19 kernel and am getting all turned
around.  I've found several patch files (in the CVS patches directory, and
on the FTP site), but these all see to be ancient (like a year old or so).

Running the patchin.sh script in the patches directory seems to hook the CVS
code into my kernel tree, but the kernel build system is left unmodified.

Finally, a simple "make" in the top level of the CVS directory doesn't work,
I think because I can't get my kernel patched properly.

I'm no stranger to writing code, or patching/compiling kernels, I'm just
seeing too many half-finished trails to figure out which one I'm supposed to
follow...

So, how does one compile MTD for a 2.2.x kernel?

Charles Steinkuehler
charles@steinkuehler.net