Hi!
I have made a NetBSD-current kernel which boots on Xen. It still has some
problems but it''s good enough to boot multi-user and allows logins.
While working on this, I noticed a few problems and I''m wondering if
these
are corrected in Xen versions >1.1:
- the count for the initial l2 pagetable seems to be wrong: The page is
pin''ed and it''s used as a l2 pagetable but it''s count
is 0x40000000. If
you switch to another table, you can''t unpin in and if you unpin it
first,
it will stay typed as a l2 pagetable. Additionally, if you unpin it, you
can make it writable then while it''s still in use...
- Xen completely locks up if my idle loop doesn''t call the yield
function,
but consists of:
* clear EVENTS_MASTER_ENABLE_BIT
* check NetBSD runqueues
* set EVENTS_MASTER_ENABLE_BIT
* check for missed events
* loop
Not sure yet, why that is... It only happens occasionally but I never
managed to complete a boot until I added the yield call. I''m using the
hypervisor callback code with critical region fixup from mini-os which I
think is identical to the one used in Linux.
I''ve implemented a network driver and a console driver (output only).
There''s no support for dom0 operations yet and there''s no
driver for
harddisks yet. Also a couple minor things like cpu speed detection and
setting the clock are missing. Right now there''s also still some
problems
with pagetables where either hypervisor calls are used to update inactive
pagetables or vice versa to a lesser extent. I think that will be solved
once I use pinning.
I hope to commit this to the NetBSD tree eventually. Or I''ll make
patches
available after some cleaning up.
Finally, if someone could get me a xen-1.2 and/or xen-unstable tree out of
bitkeeper, that would be much appreciated.
--
Christian Limpach <chris@pin.lu>
[11] text 0xc0100000 data 0xc0276c24 bss 0xc02808c4 end 0xc02c163c esym
0xc02f177c
[11] NetBSD 1.6ZH (XENO) #704: Tue Jan 27 18:52:02 CET 2004
[11] chris@marble:/devel/netbsd/src-current-xen/compile/XENO
[11] start_info: 0xc02bbd80
[11] nr_pages: 4000
[11] shared_inf: 0xc0300000 (was 0xc10a0000)
[11] pt_base: 0xc109f000
[11] mod_start: 0xc0281000
[11] mod_len: 196924
[11] net_rings:
[11] 264000
[11]
[11] blk_ring: 0x258000
[11] dom_id: 11
[11] flags: 0x0
[11] cmd_line:
ip=172.20.4.17:172.20.4.13:172.20.1.1:255.255.128.0::eth0:off
bootdev=xennet0 nfsroot=marble:/netboot/qube
[11] NetBSD Xen console attached.
Loaded initial symtab at 0xc02c1640, strtab at 0xc02db47c, # entries 6576
Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004
The NetBSD Foundation, Inc. All rights reserved.
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
NetBSD 1.6ZH (XENO) #704: Tue Jan 27 18:52:02 CET 2004
chris@marble:/devel/netbsd/src-current-xen/compile/XENO
total memory = 13948 KB
avail memory = 13576 KB
[11] Xen reported: 501.148 MHz processor.
mainbus0 (root)
cpu0 at mainbus0: (uniprocessor)
cpu0: Intel Celeron (Mendocino) (686-class), 20.00 MHz, id 0x665
cpu0: features 183fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR>
cpu0: features 183fbff<PGE,MCA,CMOV,PAT,PSE36,MMX>
cpu0: features 183fbff<FXSR>
cpu0: I-cache 16 KB 32b/line 4-way, D-cache 16 KB 32b/line 4-way
cpu0: L2 cache 128 KB 32b/line 4-way
cpu0: ITLB 32 4 KB entries 4-way, 2 4 MB entries fully associative
cpu0: DTLB 64 4 KB entries 4-way, 8 4 MB entries 4-way
cpu0: 8 page colors
xenc0 at mainbus0: Xen Virtual Console Driver
xennet0 at mainbus0: Xen Virtual Network Driver
xennet0: MAC address aa:00:00:24:1c:9b
npx0 at mainbus0: using exception 16
IPsec: Initialized Security Association Processing.
boot device: xennet0
root on xennet0
mountroot: trying nfs...
nfs_boot: trying static
nfs_boot: client_addr=172.20.4.17
nfs_boot: gateway=172.20.1.1
nfs_boot: netmask=255.255.128.0
nfs_boot: server=172.20.4.13
nfs_boot: root=marble:/netboot/qube
root on marble:/netboot/qube
root time: 0x4016a704
root file system type: nfs
init: copying out path `/sbin/init'' 11
Thu Jan 1 00:01:05 UTC 1970
Starting file system checks:
Setting tty flags.
Setting sysctl variables:
Starting network.
Hostname: qube
IPv6 mode: host
Configuring network interfaces:
..
Building databases...
Starting syslogd.
Mounting all filesystems...
Creating a.out runtime link editor directory cache.
Checking quotas:
done.
/etc/rc: WARNING: No swap space configured!
Starting virecover.
Starting local daemons:
..
Starting sshd.
Starting inetd.
Thu Jan 1 00:01:16 UTC 1970
NetBSD/i386 (qube) (console)
login:
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/xen-devel
Cool. Is there any way I could take a look at your changes? You''ll want to bring it up to date to use unstable: bk clone bk://xen.bkbits.net/xeno-unstable.bk -Kip On Tue, 27 Jan 2004, Christian Limpach wrote:> Hi! > > I have made a NetBSD-current kernel which boots on Xen. It still has some > problems but it''s good enough to boot multi-user and allows logins. > > While working on this, I noticed a few problems and I''m wondering if these > are corrected in Xen versions >1.1: > > - the count for the initial l2 pagetable seems to be wrong: The page is > pin''ed and it''s used as a l2 pagetable but it''s count is 0x40000000. If > you switch to another table, you can''t unpin in and if you unpin it first, > it will stay typed as a l2 pagetable. Additionally, if you unpin it, you > can make it writable then while it''s still in use... > > - Xen completely locks up if my idle loop doesn''t call the yield function, > but consists of: > * clear EVENTS_MASTER_ENABLE_BIT > * check NetBSD runqueues > * set EVENTS_MASTER_ENABLE_BIT > * check for missed events > * loop > Not sure yet, why that is... It only happens occasionally but I never > managed to complete a boot until I added the yield call. I''m using the > hypervisor callback code with critical region fixup from mini-os which I > think is identical to the one used in Linux. > > > I''ve implemented a network driver and a console driver (output only). > There''s no support for dom0 operations yet and there''s no driver for > harddisks yet. Also a couple minor things like cpu speed detection and > setting the clock are missing. Right now there''s also still some problems > with pagetables where either hypervisor calls are used to update inactive > pagetables or vice versa to a lesser extent. I think that will be solved > once I use pinning. > > I hope to commit this to the NetBSD tree eventually. Or I''ll make patches > available after some cleaning up. > > Finally, if someone could get me a xen-1.2 and/or xen-unstable tree out of > bitkeeper, that would be much appreciated. > > -- > Christian Limpach <chris@pin.lu> > > [11] text 0xc0100000 data 0xc0276c24 bss 0xc02808c4 end 0xc02c163c esym > 0xc02f177c > [11] NetBSD 1.6ZH (XENO) #704: Tue Jan 27 18:52:02 CET 2004 > [11] chris@marble:/devel/netbsd/src-current-xen/compile/XENO > [11] start_info: 0xc02bbd80 > [11] nr_pages: 4000 > [11] shared_inf: 0xc0300000 (was 0xc10a0000) > [11] pt_base: 0xc109f000 > [11] mod_start: 0xc0281000 > [11] mod_len: 196924 > [11] net_rings: > [11] 264000 > [11] > [11] blk_ring: 0x258000 > [11] dom_id: 11 > [11] flags: 0x0 > [11] cmd_line: > ip=172.20.4.17:172.20.4.13:172.20.1.1:255.255.128.0::eth0:off > bootdev=xennet0 nfsroot=marble:/netboot/qube > [11] NetBSD Xen console attached. > Loaded initial symtab at 0xc02c1640, strtab at 0xc02db47c, # entries 6576 > Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004 > The NetBSD Foundation, Inc. All rights reserved. > Copyright (c) 1982, 1986, 1989, 1991, 1993 > The Regents of the University of California. All rights reserved. > NetBSD 1.6ZH (XENO) #704: Tue Jan 27 18:52:02 CET 2004 > chris@marble:/devel/netbsd/src-current-xen/compile/XENO > total memory = 13948 KB > avail memory = 13576 KB > [11] Xen reported: 501.148 MHz processor. > mainbus0 (root) > cpu0 at mainbus0: (uniprocessor) > cpu0: Intel Celeron (Mendocino) (686-class), 20.00 MHz, id 0x665 > cpu0: features 183fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR> > cpu0: features 183fbff<PGE,MCA,CMOV,PAT,PSE36,MMX> > cpu0: features 183fbff<FXSR> > cpu0: I-cache 16 KB 32b/line 4-way, D-cache 16 KB 32b/line 4-way > cpu0: L2 cache 128 KB 32b/line 4-way > cpu0: ITLB 32 4 KB entries 4-way, 2 4 MB entries fully associative > cpu0: DTLB 64 4 KB entries 4-way, 8 4 MB entries 4-way > cpu0: 8 page colors > xenc0 at mainbus0: Xen Virtual Console Driver > xennet0 at mainbus0: Xen Virtual Network Driver > xennet0: MAC address aa:00:00:24:1c:9b > npx0 at mainbus0: using exception 16 > IPsec: Initialized Security Association Processing. > boot device: xennet0 > root on xennet0 > mountroot: trying nfs... > nfs_boot: trying static > nfs_boot: client_addr=172.20.4.17 > nfs_boot: gateway=172.20.1.1 > nfs_boot: netmask=255.255.128.0 > nfs_boot: server=172.20.4.13 > nfs_boot: root=marble:/netboot/qube > root on marble:/netboot/qube > root time: 0x4016a704 > root file system type: nfs > init: copying out path `/sbin/init'' 11 > Thu Jan 1 00:01:05 UTC 1970 > Starting file system checks: > Setting tty flags. > Setting sysctl variables: > Starting network. > Hostname: qube > IPv6 mode: host > Configuring network interfaces: > .. > Building databases... > Starting syslogd. > Mounting all filesystems... > Creating a.out runtime link editor directory cache. > Checking quotas: > done. > /etc/rc: WARNING: No swap space configured! > Starting virecover. > Starting local daemons: > .. > Starting sshd. > Starting inetd. > Thu Jan 1 00:01:16 UTC 1970 > > NetBSD/i386 (qube) (console) > > login: > > > > > > ------------------------------------------------------- > The SF.Net email is sponsored by EclipseCon 2004 > Premiere Conference on Open Tools Development and Integration > See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. > http://www.eclipsecon.org/osdn > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/xen-devel >------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
> I have made a NetBSD-current kernel which boots on Xen. It still has some > problems but it''s good enough to boot multi-user and allows logins.Good job! It''s really good to hear that another OS is up and running on Xen. Out of interest, how many man hours do you think it took you? I''m very impressed you managed to figure the Xen i/f out on your own without any help from Keir and the rest of us.> While working on this, I noticed a few problems and I''m wondering if these > are corrected in Xen versions >1.1:There''s been a few fixes and improvements for the hypervisor i/f in the 1.2 tree, but I don''t think it will take long to forward port NetBSD. We''ll investigate your specific bug reports.> I hope to commit this to the NetBSD tree eventually. Or I''ll make patches > available after some cleaning up.Great! We''d be very pleased to put the patches in the xen distribution in the meantime, as soon as you''re ready.> Finally, if someone could get me a xen-1.2 and/or xen-unstable tree out of > bitkeeper, that would be much appreciated.There''s a pretty recent src/binary tar ball in: www.cl.cam.ac.uk/netos/xen/downloads/xeno-1.2.tar.gz It''s still in the form of a BK repository, but all of the files are checked out. BTW: I offered to buy dinner for anyone that did an OS port. I hope you''ll take me up on that offer... Best, Ian ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
> I have made a NetBSD-current kernel which boots on Xen. It still has some > problems but it''s good enough to boot multi-user and allows logins.That''s excellent news!> While working on this, I noticed a few problems and I''m wondering if these > are corrected in Xen versions >1.1:<snip> Well not sure off hand if these problems are fixed (I had a quick scan through the 1.0 world but it seems like ancient history :-) I did notice a somewhat odd behaviour in the pinning case (dropping reference counts) but it''s possible that I just misremember how it used to work.. Anyway: strongly recommend you move to 1.2 or 1.3-devel as there are a large number of improvements and bug fixes in a variety of areas. It also makes getting technical feedback on the list a bit easier since most of us are using >= 1.2.> Finally, if someone could get me a xen-1.2 and/or xen-unstable tree out of > bitkeeper, that would be much appreciated.It''s all free to access, e.g. just type one of bk clone bk://xen.bkbits.net/xeno-1.2.bk bk clone bk://xen.bkbits.net/xeno-unstable.bk Currently writes to the repository need to be filtered through one of us; email or similar works just fine. We may move to a more decentralized scheme in the future. cheers, S. ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
> I have made a NetBSD-current kernel which boots on Xen. It still has some > problems but it''s good enough to boot multi-user and allows logins.I presume it''s running a single domain 0? It would be very nice to get the user-space domain builder written to support it, so we can have multiple NetBSD, Linux and XP images all running together on one box... Our view is that domain building is quite OS specific, hence its nice to be able to do most of the intricate work of setting up a domain (installing the image and ramdisks, initial page tables etc) in the nice environment of a user-space application running in another domain [*]. It''s a pain pulling yourself up by your bootstraps in assembler within a domain... Hence, we''ve developed OS-specific domain building tools for Linux and Win XP. It probably makes the sense to do the same for NetBSD, unless the initial memory map is very similar to Linux. User space domain building was pretty grim in 1.0, so its definitely worth forward porting to 1.2 first. Best, Ian [*] One feature of Xen is that you can use the control interface in one domain to arbitrarily manipulate/modify another domain. At the moment, there''s only a notion of a single privileged domain, but in future we intend to be able to delegate privileges to domains to enable them to manages a specified set of other domains e.g. for domain building, debugging, migration, checkpointing, fault injection etc.] ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
Christian Limpach
2004-Jan-28 14:54 UTC
Re: [Xen-devel] NetBSD port and a couple of remarks
> Out of interest, how many man hours do you think it took you? I''m > very impressed you managed to figure the Xen i/f out on your own > without any help from Keir and the rest of us.I think I spent 8-10 hours each day between 1/11 and 1/20 and after that it was more sporadic, another 10 on 1/22 and about 4 each on 1/25 and 1/27. I''d think between 95 and 120 hours total. I also noted some milestones: - 1/11/2004: started with linux/mini-os install/experimentation - 1/12/2004: add mini-os startup to NetBSD startup and make it load/start - 1/12/2004: add arch/xeno tree - 1/14/2004: add minimal console output support - 1/17/2004: add timer event processing - 1/17/2004: started work on xennet network driver - 1/18/2004: pmap pd/pt/pa/ma cleanup - 1/19/2004: hardclock call from timer event works - 1/20/2004: boot init until cnopen over nfs - 1/22/2004: add console device - 1/22/2004: integrate Xen events into NetBSD interrupt system - 1/25/2004: add npx support - 1/27/2004: multi-user boot over nfs I spend quite some time figuring out how some of the i386 cpu features actually work (pagetable/GDT/LDT/TSS), how they are used in NetBSD and how to rewrite using the Xen hypercalls.> There''s been a few fixes and improvements for the hypervisor i/f > in the 1.2 tree, but I don''t think it will take long to forward > port NetBSD. > > We''ll investigate your specific bug reports.Good!> > I hope to commit this to the NetBSD tree eventually. Or I''ll makepatches> > available after some cleaning up. > > Great! We''d be very pleased to put the patches in the xen > distribution in the meantime, as soon as you''re ready.Thanks. I''ll probably have a patch ready in a few days and I''ll also know if/how it gets into the NetBSD tree.> There''s a pretty recent src/binary tar ball in: > www.cl.cam.ac.uk/netos/xen/downloads/xeno-1.2.tar.gz > > It''s still in the form of a BK repository, but all of the files > are checked out.Thanks!> BTW: I offered to buy dinner for anyone that did an OS port. I > hope you''ll take me up on that offer...Oh, excellent! ;-) christian ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
Christian Limpach
2004-Jan-28 16:23 UTC
Re: [Xen-devel] NetBSD port and a couple of remarks
> Well not sure off hand if these problems are fixed (I had a quick scan > through the 1.0 world but it seems like ancient history :-) I did notice > a somewhat odd behaviour in the pinning case (dropping reference counts) > but it''s possible that I just misremember how it used to work..I use the following patch: (against the xen-1.1 tar.gz) diff -pru ../../xeno-1.1.bk/xen/common/domain.c xen/common/domain.c --- ../../xeno-1.1.bk/xen/common/domain.c 2003-10-29 18:27:39.000000000 +0100 +++ xen/common/domain.c 2004-01-22 00:43:16.000000000 +0100 @@ -354,6 +378,7 @@ int final_setup_guestos(struct task_stru net_ring_t *shared_rings; net_vif_t *net_vif; int i; + struct pfn_info *page; if ( (p->flags & PF_CONSTRUCTED) ) return -EINVAL; @@ -362,6 +387,9 @@ int final_setup_guestos(struct task_stru * mem mappings - set them up. */ phys_l2tab = meminfo->l2_pgt_addr; + page = frame_table + (phys_l2tab >> PAGE_SHIFT); + get_page_type(page); + get_page_tot(page); l2tab = map_domain_mem(phys_l2tab); memcpy(&l2tab[DOMAIN_ENTRIES_PER_L2_PAGETABLE], &idle_pg_table[DOMAIN_ENTRIES_PER_L2_PAGETABLE], This will result in a $40000001 count and then after a switch to another table and an unpin of the initial table, the count will be 0.> Anyway: strongly recommend you move to 1.2 or 1.3-devel as there are > a large number of improvements and bug fixes in a variety of areas. It > also makes getting technical feedback on the list a bit easier since > most of us are using >= 1.2.yes, xen-1.1 was the latest version available outside of bitkeeper.> > Finally, if someone could get me a xen-1.2 and/or xen-unstable tree outof> > bitkeeper, that would be much appreciated. > > It''s all free to access, e.g. just type one of > > bk clone bk://xen.bkbits.net/xeno-1.2.bk > bk clone bk://xen.bkbits.net/xeno-unstable.bkyes, except that I don''t want to deal with the bitkeeper licence and figure out whether I''m allowed to use it or not. christian
> > Well not sure off hand if these problems are fixed (I had a quick scan > > through the 1.0 world but it seems like ancient history :-) I did notice > > a somewhat odd behaviour in the pinning case (dropping reference counts) > > but it''s possible that I just misremember how it used to work.. > > I use the following patch: (against the xen-1.1 tar.gz) > diff -pru ../../xeno-1.1.bk/xen/common/domain.c xen/common/domain.c > --- ../../xeno-1.1.bk/xen/common/domain.c 2003-10-29 > 18:27:39.000000000 +0100 > +++ xen/common/domain.c 2004-01-22 00:43:16.000000000 +0100 > @@ -354,6 +378,7 @@ int final_setup_guestos(struct task_stru > net_ring_t *shared_rings; > net_vif_t *net_vif; > int i; > + struct pfn_info *page; > > if ( (p->flags & PF_CONSTRUCTED) ) > return -EINVAL; > @@ -362,6 +387,9 @@ int final_setup_guestos(struct task_stru > * mem mappings - set them up. > */ > phys_l2tab = meminfo->l2_pgt_addr; > + page = frame_table + (phys_l2tab >> PAGE_SHIFT); > + get_page_type(page); > + get_page_tot(page); > l2tab = map_domain_mem(phys_l2tab); > memcpy(&l2tab[DOMAIN_ENTRIES_PER_L2_PAGETABLE], > &idle_pg_table[DOMAIN_ENTRIES_PER_L2_PAGETABLE], > > This will result in a $40000001 count and then after a switch to another > table and an unpin of the initial table, the count will be 0.Ack ; as I noted in the previous email, the dropping of ref counts in the unpin code looked suspicious.> > It''s all free to access, e.g. just type one of > > > > bk clone bk://xen.bkbits.net/xeno-1.2.bk > > bk clone bk://xen.bkbits.net/xeno-unstable.bk > > yes, except that I don''t want to deal with the bitkeeper licence and figure > out whether I''m allowed to use it or not.Ok - I''ve put a bzball at http://www.cl.cam.ac.uk/Research/SRG/netos/xen/downloads/xeno-1.2.tar.bz2 cheers, S.
> > This will result in a $40000001 count and then after a switch to another > > table and an unpin of the initial table, the count will be 0. > > > Ack ; as I noted in the previous email, the dropping of ref counts > in the unpin code looked suspicious.The unpin code is fine, but the initial page directory does not have its refcnt incremented to reflect that CR3 is pointing at it. This is fixed in 1.2 and 1.3 but I guess we thought we''d obsoleted 1.1 by then. I''ll check in the fix. -- Keir
> > yes, except that I don''t want to deal with the bitkeeper licence and figure > out whether I''m allowed to use it or not. >Are you doing development work on subversion, arch, or another revision control system? If not, what is the concern? -Kip ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
Christian Limpach
2004-Jan-28 17:07 UTC
Re: [Xen-devel] NetBSD port and a couple of remarks
> > I have made a NetBSD-current kernel which boots on Xen. It still hassome> > problems but it''s good enough to boot multi-user and allows logins. > > I presume it''s running a single domain 0?I''m running Linux in domain 0 and I''m using: ID=`xi_create $MEM test` echo $ID xi_build $ID /boot/nbsd-os 1 initrd=/boot/nbsd-os.sym ip=$IP:172.20.4.13:172.20.1.1:255.255.128.0::eth0:off bootdev=xennet0 nfsroot=marble:/netboot/qube xi_vifinit $ID 0 $IP xi_start $ID I''m abusing the initrd to pass in the kernel''s symbol table. I couldn''t figure out how to make objcopy append the symbol table to the raw image. I''m using a small tool to extract the symbol table from the kernel and use this file as initrd. The kernel will then copy the symbol table to the end of the bss before clearing the bss. It would probably be easy enough to remap the pages used by the symbol table. I didn''t want to modify xi_build, but it would probably be neat if it could read elf images (and thus load the symbol table to the right address). That''s what the regular NetBSD bootloader does.> It would be very nice to get the user-space domain builder > written to support it, so we can have multiple NetBSD, Linux and > XP images all running together on one box... > > Our view is that domain building is quite OS specific, hence its > nice to be able to do most of the intricate work of setting up a > domain (installing the image and ramdisks, initial page tables > etc) in the nice environment of a user-space application running > in another domain [*]. It''s a pain pulling yourself up by your > bootstraps in assembler within a domain...I kind of do because I don''t rely on much setup done by the domain build. I only rely on the pagetables to be setup to map VA0xc0100000 to PA0x00000000, all memory to be mapped and a pointer to the initial pagetables. Then I do the following: - build a phys to machine mapping table - build a new pagetable with only the kernel (+some fixed pages) mapped - switch pagetables and unpin the initial pagetable - hand the unmapped PA range to our memory management system (uvm). I was considering making uvm manage the MA pages directly but the initial load only works with a finite number of memory segments and the collection of MA pages is too fragmented.> Hence, we''ve developed OS-specific domain building tools for > Linux and Win XP. It probably makes the sense to do the same for > NetBSD, unless the initial memory map is very similar to > Linux.I think it''s similar enough. christian ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
Christian Limpach
2004-Jan-28 18:35 UTC
Re: [Xen-devel] NetBSD port and a couple of remarks
> > yes, except that I don''t want to deal with the bitkeeper licence andfigure> > out whether I''m allowed to use it or not. > > > Are you doing development work on subversion, arch, or another revision > control system? If not, what is the concern?I''ve worked in the past on cvs for a client. I usually don''t care much about licences but there''s been so much noise about this one and I take that as an indication that the bitkeeper people have some agenda with that licence and I''d rather not be subject to it. christian ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
> > Are you doing development work on subversion, arch, or another revision > > control system? If not, what is the concern? > > I''ve worked in the past on cvs for a client. I usually don''t care much > about licences but there''s been so much noise about this one and I take that > as an indication that the bitkeeper people have some agenda with that > licence and I''d rather not be subject to it.I''ve just gone looking for the current version of the license -- it''s not actually easy to find from the web pages: http://www.bitkeeper.com/bkl.txt We certainly had concerns when we embarked on using bk, but there''s no denying that it''s a damn good tool, considerably better than cvs/arch/subversion. Different people in the group have tried using both arch and subversion for other large projects, but the experience hasn''t been good. I''m no lawyer, but from my reading of the license there should be no *legal* problem with *anyone* using it for Xen development, providing that a) you don''t mind having your ChangeLog entries logged and viewable to the whole world and b) you intend to make the source "publicly available". The latter point made us slightly nervous as some of the ChangeLog entries for a previous large project were somewhat ''colourful'', but in practice this hasn''t been a problem, so far. Anyhow, I''ve been meaning to write a script to make the nightly src and builds available for download as a tar ball. Best, Ian ======================================================================== BitKeeper License version 1.40, 08/23/02 1. DEFINITIONS BKL: This license in its entirety, also known as the BitKeeper License. You: The licensee of the BitKeeper Software. BitMover: The licensor of the BitKeeper Software. BitKeeper Software: The complete set of executable programs and any accom- panying files, such as documentation, known as the BitKeeper Soft- ware. The set of programs and files must include all files and pro- grams distributed by BitMover as part of the BitKeeper Software. BitKeeper Package: A set of files managed by the same BitKeeper ChangeSet file. There may be multiple instances of the package; each instance is called a repository. Single user BitKeeper Package: A BitKeeper Package wherein all changes to all files are made by the same person and the total number of files does not exceed 1000. Metadata: Information about the data managed by the BitKeeper Software in a BitKeeper Package, such as + The ChangeSet file; + The messages which annotate modifications of the data (also known as check in comments, ChangeLog entries, and/or log messages); + All files contained below the top level BitKeeper directory in a BitKeeper Package, in particular the BitKeeper/html directory and the BitKeeper/etc/config file. Open Logging: The transmission of Metadata about the data managed by the BitKeeper Software, to a functioning Open Logging server in the open- logging.org domain (or an alternative domain as posted on www.bit- keeper.com/logging). Examples of such collected information may be seen at http://www.openlogging.org. Conforming Software: BitKeeper Software that: (i) passes all of the current, unmodified, regression tests for the BitKeeper Software; (ii) performs all licensing functions, such as Open Logging, identically to the current version of the BitKeeper Software as distributed by BitMover, Inc. 2. LICENSE GRANTS Licensees may freely install, use, copy, and distribute Conforming Soft- ware. 3. LICENSEE OBLIGATIONS (a) Maintaining Open Logging Feature: You hereby warrant that you will not take any action to disable or otherwise interfere with the Open Log- ging feature of the BitKeeper Software. You hereby warrant that you will take any necessary actions to ensure that the BitKeeper Software successfully transmits the Metadata to an Open Logging server within 21 days of the creation of said Metadata. By transmitting the Meta- data to an Open Logging server, You hereby grant BitMover, or any other operator of an Open Logging server, permission to republish the Metadata sent by the BitKeeper Software to the Open Logging server. (b) Accessing Others'' BitKeeper Package: You may only use the BitKeeper Software to access a BitKeeper Package created by BitMover or third parties if you comply with the license of the BitKeeper Package, which can be found at the BitKeeper/etc/REPO LICENSE file within the Bit- Keeper Package and/or by running bk repo license. (c) Maintaining Open Source: It is the intent of BitMover that Your use of BitKeeper under this license is for the purpose of maintaining Open Source. By accepting this license, You agree that You are prepared to demonstrate Your conformance, at the request of BitMover, by making your BitKeeper repositories publicly available via the BitKeeper pro- tocol within 15 days from the time of such request. In the event that You do not wish to make BitKeeper repositories publicly available, You have 15 days in which to negotiate a waiver, convert said repositories to closed use, or cease use of said repositories. (c) Notwithstanding any other terms in this License, this License is not available to You if You and/or your employer develop, produce, sell, and/or resell a product which contains substantially similar capabili- ties of the BitKeeper Software, or, in the reasonable opinion of Bit- Mover, competes with the BitKeeper Software. (d) Inclusion with another product having source and/or configuration man- agement features: Inclusion of the BitKeeper Software for use with a system having substantially similar capabilities of the BitKeeper Software requires prior written permission from BitMover. 4. NON-CONFORMING USE 4.1. Single user packages For single user BitKeeper Packages, Open Logging is optional. 4.2. Closed Use Closed use is the use of the BitKeeper Software without participating in BKL licensing restrictions such as Open Logging. Closed use of the Bit- Keeper Software requires that you (or your organization) purchase closed use licenses for all users of the BitKeeper Software within your organiza- tion. This license, the BKL, does not convey authority to make closed use of the BitKeeper Software. 4.3. Logging Waivers Certain sites which do not wish to participate in Open Logging, such as educational or research institutes, may apply for, and may be granted, a written waiver from BitMover, Inc. After applying for a written waiver, such an institution may use the BitKeeper Software without Open Logging, for up to 90 days, or until a response is received from BitMover, Inc., whichever comes first. Should BitMover not grant your waiver request, you have the option of converting to Open Logging, immediately terminating your use of the BitKeeper Software or continuing your use after purchasing closed use license[s]. 4.4. Damages Use, copying, or distribution of non-conforming software is a violation of copyrights held by BitMover on the BitKeeper Software. Damages for copy- right infringement are the greater of actual damages or statutory damages, which are currently up to $150,000 per infringement. This license is not available to You if You and/or your company have any unresolved copyright disputes with BitMover. 5. DISCLAIMER OF WARRANTY COVERED CODE IS PROVIDED UNDER THIS LICENSE ON AN ``AS IS'''' BASIS, WITHOUT WARRANTY OR INDEMNIFICATION OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OR INDEMNITIES CONCERNING INTEL- LECTUAL PROPERTIES (E.G. PATENTS OR COPYRIGHTS), WARRANTIES THAT THE COV- ERED CODE IS FREE OF DEFECTS, MERCHANTABLE, FIT FOR A PARTICULAR PURPOSE OR NON-INFRINGING. SHOULD ANY PORTION OF BITKEEPER SOFTWARE PROVE DEFECTIVE IN ANY RESPECT, YOU ASSUME THE COST OF ANY RESULTING DAMAGES, NECESSARY SERVICING, REPAIR OR CORRECTION. THIS DISCLAIMER OF WARRANTY CONSTITUTES AN ESSENTIAL PART OF THIS LICENSE. NO USE OF BITKEEPER SOFTWARE IS AUTHORIZED HEREUNDER EXCEPT SUBJECT TO THIS DISCLAIMER. 6. TERMINATION + This License and the rights granted hereunder will terminate automati- cally if you fail to comply with terms herein. Provisions which, by their nature, should remain in effect beyond the termination of this License shall survive. + If any of the licensing requirements, such as Open Logging, are found to be unenforceable, then this license automatically terminates unless You continue to comply with all of the licensing requirements. + Should You or your organization choose to institute patent, copyright, and/or intellectual property litigation against BitMover, Inc. with respect to the BitKeeper Software, then this License and the rights granted hereunder will terminate automatically as of the date such liti- gation is filed. + If this License is terminated for any reason, You must delete all copies of the BitKeeper Software and cease using the BitKeeper Software. 7. LIMITATION OF LIABILITY TO THE FULL EXTENT ALLOWED BY APPLICABLE LAW, BITMOVER''S LIABILITY TO YOU FOR CLAIMS RELATING TO THIS LICENSE, WHETHER FOR BREACH OR IN TORT, SHALL BE LIMITED TO ONE HUNDRED PERCENT (100%) OF THE AMOUNT HAVING THEN ACTUALLY BEEN PAID BY YOU TO BITMOVER FOR ALL COPIES LICENSED HEREUNDER OF THE PAR- TICULAR ITEMS GIVING RISE TO SUCH CLAIM, IF ANY. IN NO EVENT WILL BITMOVER BE LIABLE FOR ANY INDIRECT, PUNITIVE, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH OR ARISING OUT OF THIS LICENSE (INCLUDING, WITHOUT LIMITATION, LOSS OF PROFITS, USE, DATA, OR OTHER ECONOMIC ADVANTAGE), HOWEVER IT ARISES AND ON ANY THEORY OF LIABIL- ITY, WHETHER IN AN ACTION FOR CONTRACT, STRICT LIABILITY OR TORT (INCLUDING NEGLIGENCE) OR OTHERWISE, WHETHER OR NOT SUCH PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE AND NOTWITHSTANDING THE FAILURE OF ESSENTIAL PURPOSE OF ANY REMEDY. 8. MISCELLANEOUS 8.1. Merger This License represents the complete agreement between You and BitMover regarding the BitKeeper Software covered by this License. 8.2. Assignment BitMover may assign this License, and its rights and obligations hereunder, at its sole discretion. 8.3. Severability If any provision of this License is held to be unenforceable, such provi- sion shall be reformed only to the extent necessary to make it enforceable. 8.4. Governing Law/Jurisdiction This License shall be governed by the laws of the US and the State of Cali- fornia, as applied to contracts entered into and to be performed in Cali- fornia between California residents. By using this product, you submit to the jurisdiction of the courts in the Northern District of California. BKL Copyright (C) 1999-2002 BitMover, Inc. Page 1 ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
I understand people''s concern. However, Bitmover is no Rambus. As I already stated in a private mail to Christian, Larry''s only agenda is feeding his family and paying his employees. -Kip On Wed, 28 Jan 2004, Ian Pratt wrote:> > > Are you doing development work on subversion, arch, or another revision > > > control system? If not, what is the concern? > > > > I''ve worked in the past on cvs for a client. I usually don''t care much > > about licences but there''s been so much noise about this one and I take that > > as an indication that the bitkeeper people have some agenda with that > > licence and I''d rather not be subject to it. > > I''ve just gone looking for the current version of the license -- > it''s not actually easy to find from the web pages: > http://www.bitkeeper.com/bkl.txt > > We certainly had concerns when we embarked on using bk, but > there''s no denying that it''s a damn good tool, considerably > better than cvs/arch/subversion. Different people in the group > have tried using both arch and subversion for other large > projects, but the experience hasn''t been good. > > I''m no lawyer, but from my reading of the license there should be > no *legal* problem with *anyone* using it for Xen development, > providing that a) you don''t mind having your ChangeLog entries > logged and viewable to the whole world and b) you intend to make > the source "publicly available". > > The latter point made us slightly nervous as some of the > ChangeLog entries for a previous large project were somewhat > ''colourful'', but in practice this hasn''t been a problem, so far. > > Anyhow, I''ve been meaning to write a script to make the nightly > src and builds available for download as a tar ball. > > Best, > Ian > > ========================================================================> > > BitKeeper License version 1.40, 08/23/02 > > 1. DEFINITIONS > > BKL: This license in its entirety, also known as the BitKeeper License. > > You: The licensee of the BitKeeper Software. > > BitMover: The licensor of the BitKeeper Software. > > BitKeeper Software: The complete set of executable programs and any accom- > panying files, such as documentation, known as the BitKeeper Soft- > ware. The set of programs and files must include all files and pro- > grams distributed by BitMover as part of the BitKeeper Software. > > BitKeeper Package: A set of files managed by the same BitKeeper ChangeSet > file. There may be multiple instances of the package; each instance > is called a repository. > > Single user BitKeeper Package: A BitKeeper Package wherein all changes to > all files are made by the same person and the total number of files > does not exceed 1000. > > Metadata: Information about the data managed by the BitKeeper Software in a > BitKeeper Package, such as > > + The ChangeSet file; > > + The messages which annotate modifications of the data (also known > as check in comments, ChangeLog entries, and/or log messages); > > + All files contained below the top level BitKeeper directory in a > BitKeeper Package, in particular the BitKeeper/html directory and > the BitKeeper/etc/config file. > > Open Logging: The transmission of Metadata about the data managed by the > BitKeeper Software, to a functioning Open Logging server in the open- > logging.org domain (or an alternative domain as posted on www.bit- > keeper.com/logging). Examples of such collected information may be > seen at http://www.openlogging.org. > > Conforming Software: BitKeeper Software that: > > (i) passes all of the current, unmodified, regression tests for the > BitKeeper Software; > > (ii) performs all licensing functions, such as Open Logging, identically > to the current version of the BitKeeper Software as distributed by > BitMover, Inc. > > 2. LICENSE GRANTS > > Licensees may freely install, use, copy, and distribute Conforming Soft- > ware. > > 3. LICENSEE OBLIGATIONS > > (a) Maintaining Open Logging Feature: You hereby warrant that you will not > take any action to disable or otherwise interfere with the Open Log- > ging feature of the BitKeeper Software. You hereby warrant that you > will take any necessary actions to ensure that the BitKeeper Software > successfully transmits the Metadata to an Open Logging server within > 21 days of the creation of said Metadata. By transmitting the Meta- > data to an Open Logging server, You hereby grant BitMover, or any > other operator of an Open Logging server, permission to republish the > Metadata sent by the BitKeeper Software to the Open Logging server. > > (b) Accessing Others'' BitKeeper Package: You may only use the BitKeeper > Software to access a BitKeeper Package created by BitMover or third > parties if you comply with the license of the BitKeeper Package, which > can be found at the BitKeeper/etc/REPO LICENSE file within the Bit- > Keeper Package and/or by running bk repo license. > > (c) Maintaining Open Source: It is the intent of BitMover that Your use of > BitKeeper under this license is for the purpose of maintaining Open > Source. By accepting this license, You agree that You are prepared to > demonstrate Your conformance, at the request of BitMover, by making > your BitKeeper repositories publicly available via the BitKeeper pro- > tocol within 15 days from the time of such request. In the event that > You do not wish to make BitKeeper repositories publicly available, You > have 15 days in which to negotiate a waiver, convert said repositories > to closed use, or cease use of said repositories. > > (c) Notwithstanding any other terms in this License, this License is not > available to You if You and/or your employer develop, produce, sell, > and/or resell a product which contains substantially similar capabili- > ties of the BitKeeper Software, or, in the reasonable opinion of Bit- > Mover, competes with the BitKeeper Software. > > (d) Inclusion with another product having source and/or configuration man- > agement features: Inclusion of the BitKeeper Software for use with a > system having substantially similar capabilities of the BitKeeper > Software requires prior written permission from BitMover. > > 4. NON-CONFORMING USE > > 4.1. Single user packages > > For single user BitKeeper Packages, Open Logging is optional. > > 4.2. Closed Use > > Closed use is the use of the BitKeeper Software without participating in > BKL licensing restrictions such as Open Logging. Closed use of the Bit- > Keeper Software requires that you (or your organization) purchase closed > use licenses for all users of the BitKeeper Software within your organiza- > tion. This license, the BKL, does not convey authority to make closed use > of the BitKeeper Software. > > 4.3. Logging Waivers > > Certain sites which do not wish to participate in Open Logging, such as > educational or research institutes, may apply for, and may be granted, a > written waiver from BitMover, Inc. After applying for a written waiver, > such an institution may use the BitKeeper Software without Open Logging, > for up to 90 days, or until a response is received from BitMover, Inc., > whichever comes first. Should BitMover not grant your waiver request, you > have the option of converting to Open Logging, immediately terminating your > use of the BitKeeper Software or continuing your use after purchasing > closed use license[s]. > > 4.4. Damages > > Use, copying, or distribution of non-conforming software is a violation of > copyrights held by BitMover on the BitKeeper Software. Damages for copy- > right infringement are the greater of actual damages or statutory damages, > which are currently up to $150,000 per infringement. > > This license is not available to You if You and/or your company have any > unresolved copyright disputes with BitMover. > > 5. DISCLAIMER OF WARRANTY > > COVERED CODE IS PROVIDED UNDER THIS LICENSE ON AN ``AS IS'''' BASIS, WITHOUT > WARRANTY OR INDEMNIFICATION OF ANY KIND, EITHER EXPRESSED OR IMPLIED, > INCLUDING, WITHOUT LIMITATION, WARRANTIES OR INDEMNITIES CONCERNING INTEL- > LECTUAL PROPERTIES (E.G. PATENTS OR COPYRIGHTS), WARRANTIES THAT THE COV- > ERED CODE IS FREE OF DEFECTS, MERCHANTABLE, FIT FOR A PARTICULAR PURPOSE OR > NON-INFRINGING. SHOULD ANY PORTION OF BITKEEPER SOFTWARE PROVE DEFECTIVE > IN ANY RESPECT, YOU ASSUME THE COST OF ANY RESULTING DAMAGES, NECESSARY > SERVICING, REPAIR OR CORRECTION. THIS DISCLAIMER OF WARRANTY CONSTITUTES AN > ESSENTIAL PART OF THIS LICENSE. NO USE OF BITKEEPER SOFTWARE IS AUTHORIZED > HEREUNDER EXCEPT SUBJECT TO THIS DISCLAIMER. > > 6. TERMINATION > > + This License and the rights granted hereunder will terminate automati- > cally if you fail to comply with terms herein. Provisions which, by > their nature, should remain in effect beyond the termination of this > License shall survive. > > + If any of the licensing requirements, such as Open Logging, are found to > be unenforceable, then this license automatically terminates unless You > continue to comply with all of the licensing requirements. > > + Should You or your organization choose to institute patent, copyright, > and/or intellectual property litigation against BitMover, Inc. with > respect to the BitKeeper Software, then this License and the rights > granted hereunder will terminate automatically as of the date such liti- > gation is filed. > > + If this License is terminated for any reason, You must delete all copies > of the BitKeeper Software and cease using the BitKeeper Software. > > 7. LIMITATION OF LIABILITY > > TO THE FULL EXTENT ALLOWED BY APPLICABLE LAW, BITMOVER''S LIABILITY TO YOU > FOR CLAIMS RELATING TO THIS LICENSE, WHETHER FOR BREACH OR IN TORT, SHALL > BE LIMITED TO ONE HUNDRED PERCENT (100%) OF THE AMOUNT HAVING THEN ACTUALLY > BEEN PAID BY YOU TO BITMOVER FOR ALL COPIES LICENSED HEREUNDER OF THE PAR- > TICULAR ITEMS GIVING RISE TO SUCH CLAIM, IF ANY. > > IN NO EVENT WILL BITMOVER BE LIABLE FOR ANY INDIRECT, PUNITIVE, SPECIAL, > INCIDENTAL OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH OR ARISING OUT OF > THIS LICENSE (INCLUDING, WITHOUT LIMITATION, LOSS OF PROFITS, USE, DATA, OR > OTHER ECONOMIC ADVANTAGE), HOWEVER IT ARISES AND ON ANY THEORY OF LIABIL- > ITY, WHETHER IN AN ACTION FOR CONTRACT, STRICT LIABILITY OR TORT (INCLUDING > NEGLIGENCE) OR OTHERWISE, WHETHER OR NOT SUCH PARTY HAS BEEN ADVISED OF THE > POSSIBILITY OF SUCH DAMAGE AND NOTWITHSTANDING THE FAILURE OF ESSENTIAL > PURPOSE OF ANY REMEDY. > > 8. MISCELLANEOUS > > 8.1. Merger > > This License represents the complete agreement between You and BitMover > regarding the BitKeeper Software covered by this License. > > 8.2. Assignment > > BitMover may assign this License, and its rights and obligations hereunder, > at its sole discretion. > > 8.3. Severability > > If any provision of this License is held to be unenforceable, such provi- > sion shall be reformed only to the extent necessary to make it enforceable. > > 8.4. Governing Law/Jurisdiction > > This License shall be governed by the laws of the US and the State of Cali- > fornia, as applied to contracts entered into and to be performed in Cali- > fornia between California residents. By using this product, you submit to > the jurisdiction of the courts in the Northern District of California. > > BKL Copyright (C) 1999-2002 BitMover, Inc. Page 1 > > > ------------------------------------------------------- > The SF.Net email is sponsored by EclipseCon 2004 > Premiere Conference on Open Tools Development and Integration > See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. > http://www.eclipsecon.org/osdn > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/xen-devel >------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
On Wed, 28 Jan 2004 13:28:32 -0800 (PST) Kip Macy <kmacy@fsmware.com> wrote:> > (c) Notwithstanding any other terms in this License, this License is not > > available to You if You and/or your employer develop, produce, sell, > > and/or resell a product which contains substantially similar capabili- > > ties of the BitKeeper Software, or, in the reasonable opinion of Bit- > > Mover, competes with the BitKeeper Software.This one is the killer. If you ever touched the source of any VCS or work for a company who might ship one (and many big software companies have similar programs) then you cannot use it. Also it is vague enough that you could basically get sued anytime. They just have to define some product you''re involved in as "competing with bitkeeper" Who knows, maybe Xen will compete with some obscure feature of BitKeeper at some point? -Andi ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
On Wed, 28 Jan 2004, Andi Kleen wrote:> This one is the killer.For the individual in question, certainly.> > If you ever touched the source of any VCS or work for a company who might ship > one (and many big software companies have similar programs) then you cannot use it. > Also it is vague enough that you could basically get sued anytime. They just > have to define some product you''re involved in as "competing with bitkeeper"Well, then Larry needs to clarify the license. However, he has received so much pointless criticism over the years for what in end effect was meant to be a favor to Linus that I wouldn''t be surprised if he didn''t consider it worth his while.> Who knows, maybe Xen will compete with some obscure feature of BitKeeper at some point?I wholly doubt it. ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
On Wed, 28 Jan 2004 15:10:08 -0800 (PST) Kip Macy <kmacy@fsmware.com> wrote:> Well, then Larry needs to clarify the license. However, he has received > so much pointless criticism over the years for what in end effect was > meant to be a favor to Linus that I wouldn''t be surprised if he didn''t > consider it worth his while.He won''t do it. This nasty paragraph was actually added quite late and subject to much discussion. At least for me it tipped over the BitKeeper license from merely being annoying (OpenLogging for my test trees etc.) to totally unacceptable. Of course you can ignore it and hope he or whoever owns BitKeeper at some point won''t sue you, but I wouldn''t bet on that. -Andi ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
=> whoever owns BitKeeper at some point won''t sue you, In light of recent events that is a genuinely compelling point. -Kip ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
On Wed, Jan 28, 2004 at 03:48:24PM -0800, Kip Macy wrote:> => whoever owns BitKeeper at some point won''t sue you, > > In light of recent events that is a genuinely compelling point.Given that we have shown our committment to the open source community by turning down overtures from Red Hat, VA Linux, Borland, and multiple VC''s, it''s not a compelling point at all. If I was motivated strictly by money I would have either kept my 3% of Cobalt or kept my position as emp #4 at Google, either of which would have made me far more money than BK will. Money isn''t the reason we''d sell out, we would have done that by now, we''ve had multiple offers, any one of which would have made me enough money that I would never have had to ever deal with another mail message like this one. On the other hand, I''m less than thrilled with being, as Kip pointed out, pointlessly flamed for doing something that has been of tremendous help no matter how you look at it. I''d appreciate it if the people who were happy using BK kept on using it and if the rest of the people were polite enough to air their opinions anywhere but my mailbox. I''ve done my part and gotten off of all public lists so I don''t bother anyone with my opinions, it would be oh-so-pleasant if my good friends in Germany were to return the favor. -- --- Larry McVoy lm at bitmover.com http://www.bitmover.com/lm ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
My fault. I''m sorry. End of discussion. -Kip On Wed, 28 Jan 2004, Larry McVoy wrote:> On Wed, Jan 28, 2004 at 03:48:24PM -0800, Kip Macy wrote: > > => whoever owns BitKeeper at some point won''t sue you, > > > > In light of recent events that is a genuinely compelling point. > > Given that we have shown our committment to the open source community by > turning down overtures from Red Hat, VA Linux, Borland, and multiple VC''s, > it''s not a compelling point at all. If I was motivated strictly by money > I would have either kept my 3% of Cobalt or kept my position as emp #4 > at Google, either of which would have made me far more money than BK will. > > Money isn''t the reason we''d sell out, we would have done that by now, > we''ve had multiple offers, any one of which would have made me enough > money that I would never have had to ever deal with another mail message > like this one. > > On the other hand, I''m less than thrilled with being, as Kip pointed > out, pointlessly flamed for doing something that has been of tremendous > help no matter how you look at it. I''d appreciate it if the people who > were happy using BK kept on using it and if the rest of the people were > polite enough to air their opinions anywhere but my mailbox. I''ve done > my part and gotten off of all public lists so I don''t bother anyone with > my opinions, it would be oh-so-pleasant if my good friends in Germany > were to return the favor. > -- > --- > Larry McVoy lm at bitmover.com http://www.bitmover.com/lm >------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
John Szakmeister
2004-Jan-29 11:14 UTC
Re: [Xen-devel] NetBSD port and a couple of remarks
> We certainly had concerns when we embarked on using bk, but > there''s no denying that it''s a damn good tool, considerably > better than cvs/arch/subversion. Different people in the group > have tried using both arch and subversion for other large > projects, but the experience hasn''t been good.I don''t know who on the list has tried Subversion, but I''d love to invite them to try the latest incarnation of it, as well as provide feedback about the experiences they''ve had. Over the last 6 months, their have been serious speed improvements, and it''s currently pushing for a 1.0 release (which looks to be happening towards the end of February). I''m certainly not advocating that you guys give up the version control that you have right now, it appears that a choice has been made. :-) But the Subversion community would love to hear how we could make the tool better, especially in the way of large projects. -John ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel
Christian Limpach dixit:>I have made a NetBSD-current kernel which boots on Xen. It still has some >problems but it''s good enough to boot multi-user and allows logins.Just announcing... There''s an operating system called MirOS, which is a derivate of OpenBSD, where OpenBSD is a derivate of NetBSD(tm). The OpenBSD people are reluctant even thinking of integrating Xen support. MirOS is run by a five-man developer team, but we''d really like to get Xen supported (in addition to the already existing and working i386, sparc and nearly completed macppc support). I''m the founder and head of the MirOS Project, and myself pretty good at x86 asm (intel syntax) and boot loaders, but don''t have got the time for porting it to Xen, nor much experience with kernel programming (in C and for Unix). If someone would consider porting MirOS BSD/i386 to Xen in the same manner as you did for NetBSD(tm)/i386, I''d be grateful. If code duplication would be avoided whenever possible, it''d be even better. You''ll get complete freedom what to do with the port, our only requirements are: * it''s BSD licenced or similar (as with nbsd) * it''s readable code * it works (at least for Domain N) If you provide more (e.g. Domain 0 support, documentation, continue to hacking on it and not do it as a one-shot project), the better. Your reward is being mentioned as a MirOS contributor, being hated even more by the OpenBSD people ;-) and working with a team of two (active) European and one (more or less active) US American developer, joined by an inactive developer each from Europe and Canada. Sorry I don''t have more to offer, but we''re mostly a fun for ourselfes project. We''re using gcc 3.4.4 (from CVS), binutils mainline (from CVS) and other stuff, and are pretty flexible in other matters as well, so I hope working with us should be easy. Having installed and used OpenBSD at least once before helps, though. Please contact either the public mailing list at miros-discuss@mirbsd.org or the developers-only list at miros-dev@ if you''re interested. bye, //mirabile --> [...] Echtzeit hat weniger mit "Speed"[...] zu tun, sondern damit, daß der > richtige Prozeß voraussagbar rechtzeitig sein Zeitscheibchen bekommt.Wir haben uns[...] geeinigt, dass das verwendete Echtzeit-Betriebssystem[...] weil selbst einfachste Operationen *echt* *Zeit* brauchen. (aus d.a.s.r) ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://ads.osdn.com/?ad_ide95&alloc_id396&op=click _______________________________________________ Xen-devel mailing list Xen-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xen-devel