[git pull] lguest: paravirt boot code

[git pull] lguest: paravirt boot code

[Rusty Russell]

Hi Linus,

	First attempt at git, so please pull carefully.  I've just put the
three i386 boot changes in the repo for the moment.  If this works I'll pile 
on the 44 lguest patches.

	git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-lguest.git

 Documentation/i386/boot.txt        |   34 +++++++++++++++++++++++++++-
 arch/x86/boot/compressed/head_32.S |   15 ++++++++++-
 arch/x86/boot/compressed/misc_32.c |    3 ++
 arch/x86/boot/header.S             |    7 +++++-
 arch/x86/kernel/asm-offsets_32.c   |    7 +++++
 arch/x86/kernel/head_32.S          |   44 +++++++++++++++++++++++++++++++++--
 include/asm-x86/bootparam.h        |    9 +++++-
 include/linux/linkage.h            |    6 +++++
 8 files changed, 116 insertions(+), 9 deletions(-)

commit f160a3a39ba7a5ac03a0eb64c25d3751c342e0e7
Author: Jeremy Fitzhardinge <jeremy@xensource.com>
Date:   Mon Oct 22 10:56:19 2007 +1000

    i386: paravirt boot sequence
    
    This patch uses the updated boot protocol to do paravirtualized boot.
    If the boot version is >= 2.07, then it will do two things:
    
     1. Check the bootparams loadflags to see if we should reload the
        segment registers and clear interrupts.  This is appropriate
        for normal native boot and some paravirtualized environments, but
        inapproprate for others.
    
     2. Check the hardware architecture, and dispatch to the appropriate
        kernel entrypoint.  If the bootloader doesn't set this, then we
        simply do the normal boot sequence.
    
    Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
    Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
    Acked-by: H. Peter Anvin <hpa@zytor.com>
    Cc: "Eric W. Biederman" <ebiederm@xmission.com>
    Cc: Vivek Goyal <vgoyal@in.ibm.com>
    Cc: James Bottomley <James.Bottomley@HansenPartnership.com>

commit a706681568ed19f147ab78674388c5a0e6ab8c44
Author: Jeremy Fitzhardinge <jeremy@xensource.com>
Date:   Mon Oct 22 10:56:18 2007 +1000

    add WEAK() for creating weak asm labels
    
    Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
    Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
    Acked-by: H. Peter Anvin <hpa@zytor.com>

commit 262ac2158778ac5937ae943d1d43f0adcff1a1d1
Author: Jeremy Fitzhardinge <jeremy@xensource.com>
Date:   Mon Oct 22 10:56:17 2007 +1000

    update boot spec to 2.07
    
    Proposed updates for version 2.07 of the boot protocol.  This includes:
    
    load_flags.KEEP_SEGMENTS- flag to request/inhibit segment reloads
    hardware_subarch	- what subarchitecture we're booting under
    hardware_subarch_data	- per-architecture data
    
    The intention of these changes is to make booting a paravirtualized
    kernel work via the normal Linux boot protocol.  The intention is that
    the bzImage payload can be a properly formed ELF file, so that the
    bootloader can use its ELF notes and Phdrs to get more metadata about
    the kernel and its requirements.
    
    The ELF file could be the uncompressed kernel vmlinux itself; it would
    only take small buildsystem changes to implement this.
    
    Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
    Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
    Acked-by: H. Peter Anvin <hpa@zytor.com>
    Cc: "Eric W. Biederman" <ebiederm@xmission.com>
    Cc: Vivek Goyal <vgoyal@in.ibm.com>

Re: [git pull] lguest: paravirt boot code

[Linus Torvalds]

On Mon, 22 Oct 2007, Rusty Russell wrote:
> 
> 	First attempt at git, so please pull carefully.  I've just put the
> three i386 boot changes in the repo for the moment.  If this works I'll pile 
> on the 44 lguest patches.

Well, it looked ok, but I had already merged the same patches through 
Andrew, so when I pulled, I had no differences. So I ended up unpulling, 
but not because there was anything *wrong* with your tree, just because 
the pull didn't give me anything but a slightly more complex history ;)

		Linus

Re: [git pull] lguest: paravirt boot code

[Rusty Russell]

On Tuesday 23 October 2007 12:28:32 Linus Torvalds wrote:
> On Mon, 22 Oct 2007, Rusty Russell wrote:
> > 	First attempt at git, so please pull carefully.  I've just put the
> > three i386 boot changes in the repo for the moment.  If this works I'll
> > pile on the 44 lguest patches.
>
> Well, it looked ok, but I had already merged the same patches through
> Andrew, so when I pulled, I had no differences.

Well, with that out the way, and some scatterlist fixups, please pull from

	git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-lguest.git

This should be fairly clean: 45 lguest patches including the more generic
drivers (aimed to be used by KVM as well, and hopefully others) and the start
of separation of arch-specific from general lguest code.

 Documentation/lguest/Makefile         |   26 +-
 Documentation/lguest/lguest.c         | 1629 ++++++++++++++++++---------------
 Documentation/lguest/lguest.txt       |   72 +-
 arch/i386/Kconfig                     |   32 +-
 arch/i386/Makefile                    |    3 +
 arch/x86/kernel/asm-offsets_32.c      |    1 +
 arch/x86/lguest/Kconfig               |   14 +
 arch/x86/lguest/Makefile              |    1 +
 arch/x86/lguest/boot.c                | 1070 ++++++++++++++++++++++
 arch/x86/lguest/i386_head.S           |  115 +++
 arch/x86/xen/Kconfig                  |    5 +-
 drivers/Kconfig                       |    2 +-
 drivers/Makefile                      |    1 +
 drivers/block/Kconfig                 |    6 +
 drivers/block/Makefile                |    2 +-
 drivers/block/lguest_blk.c            |  421 ---------
 drivers/block/virtio_blk.c            |  308 +++++++
 drivers/char/Kconfig                  |    4 +
 drivers/char/Makefile                 |    2 +-
 drivers/char/hvc_lguest.c             |  177 ----
 drivers/char/virtio_console.c         |  225 +++++
 drivers/kvm/Kconfig                   |    4 +
 drivers/lguest/Kconfig                |   13 +-
 drivers/lguest/Makefile               |   10 +-
 drivers/lguest/core.c                 |  568 +-----------
 drivers/lguest/hypercalls.c           |  177 ++---
 drivers/lguest/interrupts_and_traps.c |  125 ++-
 drivers/lguest/io.c                   |  626 -------------
 drivers/lguest/lg.h                   |  189 ++---
 drivers/lguest/lguest.c               | 1108 ----------------------
 drivers/lguest/lguest_asm.S           |   93 --
 drivers/lguest/lguest_bus.c           |  218 -----
 drivers/lguest/lguest_device.c        |  373 ++++++++
 drivers/lguest/lguest_user.c          |  138 +---
 drivers/lguest/page_tables.c          |  250 +++---
 drivers/lguest/segments.c             |   28 +-
 drivers/lguest/switcher.S             |  350 -------
 drivers/lguest/x86/core.c             |  577 ++++++++++++
 drivers/lguest/x86/switcher_32.S      |  351 +++++++
 drivers/net/Kconfig                   |    6 +
 drivers/net/Makefile                  |    2 +-
 drivers/net/lguest_net.c              |  555 -----------
 drivers/net/virtio_net.c              |  435 +++++++++
 drivers/virtio/Kconfig                |    8 +
 drivers/virtio/Makefile               |    2 +
 drivers/virtio/config.c               |   13 +
 drivers/virtio/virtio.c               |  189 ++++
 drivers/virtio/virtio_ring.c          |  313 +++++++
 include/asm-x86/Kbuild                |    3 +
 include/asm-x86/bootparam.h           |  108 ++--
 include/asm-x86/e820.h                |   28 +
 Documentation/lguest/Makefile         |   26 +-
 Documentation/lguest/lguest.c         | 1629 ++++++++++++++++++---------------
 Documentation/lguest/lguest.txt       |   72 +-
 arch/i386/Kconfig                     |   32 +-
 arch/i386/Makefile                    |    3 +
 arch/x86/kernel/asm-offsets_32.c      |    1 +
 arch/x86/lguest/Kconfig               |   14 +
 arch/x86/lguest/Makefile              |    1 +
 arch/x86/lguest/boot.c                | 1070 ++++++++++++++++++++++
 arch/x86/lguest/i386_head.S           |  115 +++
 arch/x86/xen/Kconfig                  |    5 +-
 drivers/Kconfig                       |    2 +-
 drivers/Makefile                      |    1 +
 drivers/block/Kconfig                 |    6 +
 drivers/block/Makefile                |    2 +-
 drivers/block/lguest_blk.c            |  421 ---------
 drivers/block/virtio_blk.c            |  308 +++++++
 drivers/char/Kconfig                  |    4 +
 drivers/char/Makefile                 |    2 +-
 drivers/char/hvc_lguest.c             |  177 ----
 drivers/char/virtio_console.c         |  225 +++++
 drivers/kvm/Kconfig                   |    4 +
 drivers/lguest/Kconfig                |   13 +-

Re: [git pull] lguest: paravirt boot code

[Christian Borntraeger]

Am Dienstag, 23. Oktober 2007 schrieb Rusty Russell:
>        git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-lguest.git
> This should be fairly clean: 45 lguest patches including the more generic
> drivers (aimed to be used by KVM as well, and hopefully others) and the start
> of separation of arch-specific from general lguest code.

>  drivers/net/virtio_net.c              |  435 +++++++++
>  drivers/virtio/Kconfig                |    8 +
>  drivers/virtio/Makefile               |    2 +
>  drivers/virtio/config.c               |   13 +
>  drivers/virtio/virtio.c               |  189 ++++
>  drivers/virtio/virtio_ring.c          |  313 +++++++
>  drivers/block/virtio_blk.c            |  308 +++++++
>  drivers/char/virtio_console.c         |  225 +++++

Having virtio in Linus tree would make my life easier wiring it up for 
virtualization on s390.

For all virtio patches:

Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>


That reminds me that I promised Rusty to look at one small issue in virtio
net driver. Nothing that should hold of the merge. I will sent the patch via
Rusty when its ready. 

Re: [git pull] lguest: paravirt boot code

[Rusty Russell]

On Tuesday 23 October 2007 16:55:06 Christian Borntraeger wrote:
> Having virtio in Linus tree would make my life easier wiring it up for
> virtualization on s390.

Indeed.  And given lguest's lack of interface guarantees, it make sense to use 
it as the first virtio guinea pig.  I expect changes as we adapt it for 
others anyway.

> That reminds me that I promised Rusty to look at one small issue in virtio
> net driver. Nothing that should hold of the merge. I will sent the patch
> via Rusty when its ready.

Thanks,
Rusty.

Re: [git pull] lguest: paravirt boot code

[Linus Torvalds]

On Tue, 23 Oct 2007, Rusty Russell wrote:
>
> Well, with that out the way, and some scatterlist fixups, please pull from
> 
> 	git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-lguest.git

Ok. However, please include a proper diffstat. And by proper, I mean:

	git diff -M --stat --summary

because right now your diffstat doesn't contain the final "total number of 
lines added/removed" (which is the easiest way for me to see that there 
isn't anything strange in there), and your diffstat doesn't do renames, so 
renames look like delete/create pairs. For example, you had:

>  arch/x86/lguest/boot.c                | 1070 ++++++++++++++++++++++
>  drivers/lguest/lguest.c               | 1108 ----------------------

but in reality it was not at all thousands of lines changed, but instead a 
rename with much smaller changes:

 drivers/lguest/lguest.c => arch/x86/lguest/boot.c  |  102 +-

which is a lot more readable and tells me much more about what is going 
on.

And at the end, I really do want to see the details like this:

	 70 files changed, 4822 insertions(+), 4401 deletions(-)
	 create mode 100644 arch/x86/lguest/Kconfig
	 create mode 100644 arch/x86/lguest/Makefile
	 rename drivers/lguest/lguest.c => arch/x86/lguest/boot.c (93%)
	 rename drivers/lguest/lguest_asm.S => arch/x86/lguest/i386_head.S (73%)
	 delete mode 100644 drivers/block/lguest_blk.c
	 create mode 100644 drivers/block/virtio_blk.c
	 delete mode 100644 drivers/char/hvc_lguest.c
	 create mode 100644 drivers/char/virtio_console.c
	 delete mode 100644 drivers/lguest/io.c
	 delete mode 100644 drivers/lguest/lguest_bus.c
	 create mode 100644 drivers/lguest/lguest_device.c
	 create mode 100644 drivers/lguest/x86/core.c
	 rename drivers/lguest/{switcher.S => x86/switcher_32.S} (99%)
	 delete mode 100644 drivers/net/lguest_net.c
	 create mode 100644 drivers/net/virtio_net.c
	 create mode 100644 drivers/virtio/Kconfig
	 create mode 100644 drivers/virtio/Makefile
	 create mode 100644 drivers/virtio/config.c
	 create mode 100644 drivers/virtio/virtio.c
	 create mode 100644 drivers/virtio/virtio_ring.c
	 create mode 100644 include/asm-x86/lguest.h
	 create mode 100644 include/asm-x86/lguest_hcall.h
	 delete mode 100644 include/linux/lguest_bus.h
	 create mode 100644 include/linux/virtio.h
	 create mode 100644 include/linux/virtio_blk.h
	 create mode 100644 include/linux/virtio_config.h
	 create mode 100644 include/linux/virtio_console.h
	 create mode 100644 include/linux/virtio_net.h
	 create mode 100644 include/linux/virtio_ring.h

which again gives a nice overview of what the heck is actually going on 
here. All of which you get for free if you use the above git command line.

Anyway, pulled. But just for future reference,

		Linus

Re: [git pull] lguest: paravirt boot code

[Jeff Garzik]

Linus Torvalds wrote:
> 
> On Tue, 23 Oct 2007, Rusty Russell wrote:
>> Well, with that out the way, and some scatterlist fixups, please pull from
>>
>> 	git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-lguest.git

Also, please add a "lguest: " prefix to all future commits.

Otherwise, in the one-line summary, it is impossible to tell what 
portion of the kernel is being updated:

	Use "struct boot_params" in example launcher
	Loading bzImage directly.
	Rename "cr3" to "gpgdir" to avoid x86-specific naming.
	Pagetables to use normal kernel types
	Move register setup into i386_core.c
	Accept elf files that are valid but have sections that can not
		be mmap'ed for some reason.

git tools are all keyed to use that first line as a summary of the 
changeset, without giving any other context whatsoever.  See Linus's 
2.6.X-rcY announcements, for example.

This is also described in akpm's 
http://www.zip.com.au/~akpm/linux/patches/stuff/tpp.txt or my 
http://linux.yyz.us/patch-format.html

	Jeff

Re: [git pull] lguest: paravirt boot code

[Rusty Russell]

On Wednesday 24 October 2007 02:10:21 Linus Torvalds wrote:
> On Tue, 23 Oct 2007, Rusty Russell wrote:
> > Well, with that out the way, and some scatterlist fixups, please pull
> > from
> >
> > 	git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-lguest.git
>
> Ok. However, please include a proper diffstat. And by proper, I mean:
>
> 	git diff -M --stat --summary

Thanks, that is much better.  I was just happy to locate git diff in the 133 
git commands on my system; finding options 4, 8 and 22 was beyond my skill.

So, this time for sure!  I haven't figured out how everyone does those neat 
committer: commits* lists, but I'm sure there's an option here somewhere...

Please pull from:

	git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-lguest.git

 Rusty Russell - lguest: documentation update
 Rusty Russell - lguest: Add to maintainers file.
 Jeff Garzik - lguest: build fix
 Rusty Russell - lguest: clean up lguest_launcher.h
 Rusty Russell - lguest: remove unused "wake" element from struct lguest
 Rusty Russell - lguest: use defines from x86 headers instead of magic numbers
 Rusty Russell - lguest: example launcher header cleanup.

Without the documentation update (which only changes comments and whitespace), 
the diff looks like this:

 Documentation/lguest/lguest.c   |   23 +++++++++++------------
 MAINTAINERS                     |    7 +++++++
 arch/x86/lguest/boot.c          |    6 +++---
 drivers/lguest/lg.h             |    3 ---
 drivers/lguest/x86/core.c       |    2 +-
 include/linux/lguest_launcher.h |   18 ++++--------------
 6 files changed, 26 insertions(+), 33 deletions(-)

Including the documentation update, it looks scarier:

 Documentation/lguest/lguest.c         |  178 +++++++++++++++++++-------------
 MAINTAINERS                           |    7 ++
 arch/x86/lguest/boot.c                |   54 +++++-----
 arch/x86/lguest/i386_head.S           |    8 +-
 drivers/lguest/core.c                 |    5 +-
 drivers/lguest/hypercalls.c           |   11 +-
 drivers/lguest/interrupts_and_traps.c |   37 ++++++--
 drivers/lguest/lg.h                   |    7 +-
 drivers/lguest/lguest_device.c        |   11 ++-
 drivers/lguest/lguest_user.c          |   23 ++--
 drivers/lguest/page_tables.c          |  113 +++++++++++++--------
 drivers/lguest/segments.c             |   48 ++++++----
 drivers/lguest/x86/core.c             |  122 ++++++++++++-----------
 drivers/lguest/x86/switcher_32.S      |   71 ++++++++++----
 include/asm-x86/lguest_hcall.h        |   16 ++--
 include/linux/lguest.h                |    4 +-
 include/linux/lguest_launcher.h       |   24 ++---
 17 files changed, 440 insertions(+), 299 deletions(-)

Cheers,
Rusty.

Re: [git pull] lguest: paravirt boot code

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 382 bytes --]

On Thu, 25 Oct 2007 15:19:05 +1000 Rusty Russell <rusty@rustcorp.com.au> wrote:
>
> So, this time for sure!  I haven't figured out how everyone does those neat 
> committer: commits* lists, but I'm sure there's an option here somewhere...

try "git request-pull ..."

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [git pull] lguest: paravirt boot code

[Linus Torvalds]

On Thu, 25 Oct 2007, Rusty Russell wrote:
> >
> > 	git diff -M --stat --summary
> 
> Thanks, that is much better.  I was just happy to locate git diff in the 133 
> git commands on my system; finding options 4, 8 and 22 was beyond my skill.

Hey, we give you the power. Actually realizing it all and getting used to 
it may take some time. In the meantime, you can use the pre-canned helper 
functions and just cut-and-paste from emails ;)

One of the 133 git commands is actually "git request-pull" that does this 
all for you. I always forget it, because I don't use it myself, and quite 
frankly, it's just a shell-script helper to do this all for you. But if 
you're into that kind of thing, you can actually read that shell-script 
(and others) and learn more that way ;)

		Linus