freebsd-current Digest, Vol 79, Issue 14

Ward Cinnamon wardcinnamon at charter.net
Sun Nov 21 15:41:30 PST 2004


Unsubscribe me.
On Sun, 2004-11-21 at 12:01 +0000, freebsd-current-request at freebsd.org
wrote:
> Send freebsd-current mailing list submissions to
> 	freebsd-current at freebsd.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://lists.freebsd.org/mailman/listinfo/freebsd-current
> or, via email, send a message with subject or body 'help' to
> 	freebsd-current-request at freebsd.org
> 
> You can reach the person managing the list at
> 	freebsd-current-owner at freebsd.org
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of freebsd-current digest..."
> 
> 
> Today's Topics:
> 
>    1. Giantless VFS. (Jeff Roberson)
>    2. Floppy Boot ?? (Ege M?kan)
>    3. Re: Floppy Boot ?? (Lowell Gilbert)
>    4. Re: Replacing passwd? (Dick Davies)
>    5. Re: Floppy Boot ?? (Michael C. Shultz)
>    6. Re: Problem with HP DL380 (M. Warner Losh)
>    7. Re: Problem with HP DL380 (Wilkinson, Alex)
>    8. 	Re: interrupt moderation for if_* [was: serious networking
>       (em) ...] (Dan Nelson)
>    9. Re: Replacing passwd? (Dan Nelson)
>   10. Re: ATAng support for reading from register on ATAREQUEST
>       command? (S?ren Schmidt)
> 
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Sat, 20 Nov 2004 00:24:10 -0500 (EST)
> From: Jeff Roberson <jroberson at chesapeake.net>
> Subject: Giantless VFS.
> To: current at freebsd.org
> Message-ID: <20041120000743.G18094 at mail.chesapeake.net>
> Content-Type: TEXT/PLAIN; charset=US-ASCII
> 
> I have a patch that I would like people to test and review.  It's
> available here:
> 
> http://www.chesapeake.net/~jroberson/smpffs.diff
> 
> The short description:
> This patch removes Giant from the read(), write(), and fstat() syscalls,
> as well as page faults, and bufdone (io interrupts) when using FFS.  It
> adds a considerable amount of locking to FFS and softupdates.  You may
> also use non ffs filesystems concurrently, but they will be protected by
> Giant.  If you are using quotas you should not yet run this patch.  I have
> done some buildworlds, but any heavy filesystem activity would be
> appreciated.
> 
> Long description:
> There is now a per mount-point mutex in struct ufsmount that covers ffs's
> struct fs related allocation routines.  The rest of the filesystem was
> already covered by the buffer locks on cgs, indirs, etc, as well as a few
> mutexes that were already in place.  I made great attempts to minimize the
> number of lock operations for the common cases which resulted in a couple
> of functions which may be entered with the UFS lock held, but return
> without it held.  Where this is not already done, it will be documented in
> comments.
> 
> The softupdate lk lock has been turned into a mutex, and it now protects
> all global worklists, inode and page dep hash buckets, etc.
> interlocked_sleep() is gone now that BUF_LOCK() and msleep() take
> interlock arguments.  getdirtybuf() has been slightly changed to solve
> some sleep related races that I disucssed with Kirk.  For now there is a
> single softupdates mutex, but eventually it will be made per-mountpoint
> and potentially merged with the ufsmount lock.
> 
> On the infrastructure side, the buffer cache was locked some time ago, and
> now the vm is safe to run without Giant, so there are simply a lot of
> removed GIANT_REQUIRED lines.  I added some locking to vn_start_write()
> and friends, and made some changes for LORs there.  There is a pair of new
> macros called VFS_(UN)LOCK_GIANT(mp) which inspects a new flag on the
> mount-point to determine if we have to acquire giant.  This is done very
> early in the syscall path before we call any VOPs.  The mountpoints should
> be safe to inspect as we always own a reference to the vnode in the cases
> that I have unwound for now.
> 
> Cheers,
> Jeff
> 
> ------------------------------
> 
> Message: 2
> Date: Sat, 20 Nov 2004 23:11:24 +0200
> From: Ege M?kan<egemukan at gmail.com>
> Subject: Floppy Boot ??
> To: <freebsd-current at freebsd.org>
> Message-ID: <003801c4cf45$86860680$9590aec3 at redbaron>
> Content-Type: text/plain;	charset="iso-8859-9"
> 
> Hi, 
> 
> Is there way to install FreeBSD 5.3 without install boot loader to HDD but with creating boot floppies to start your system which is installed on your HDD ??
> 
> Thanx 
> 
> Ege.
> 
> ------------------------------
> 
> Message: 3
> Date: 20 Nov 2004 17:13:13 -0500
> From: Lowell Gilbert <freebsd-current-local at be-well.ilk.org>
> Subject: Re: Floppy Boot ??
> To: Ege M?kan<egemukan at gmail.com>
> Cc: freebsd-current at freebsd.org
> Message-ID: <444qjkkw3q.fsf at be-well.ilk.org>
> Content-Type: text/plain; charset=iso-8859-1
> 
> Ege Mkan <egemukan at gmail.com> writes:
> 
> > Is there way to install FreeBSD 5.3 without install boot loader to HDD but with creating boot floppies to start your system which is installed on your HDD ??
> 
> This question would be more appropriate on freebsd-questions, but the
> answer would be the same on a -CURRENT system.  I think you can just
> take the installation boot floppy and set up the boot.config file to
> boot the appropriate disk and the appropriate kernel.
> 
> There are many "boot manager" programs that can be installed to
> floppies, as well.
> 
> ------------------------------
> 
> Message: 4
> Date: Sat, 20 Nov 2004 22:23:25 +0000
> From: Dick Davies <rasputnik at hellooperator.net>
> Subject: Re: Replacing passwd?
> To: Dan Nelson <dnelson at allantgroup.com>
> Cc: FreeBSD Current Users <freebsd-current at freebsd.org>
> Message-ID: <20041120222325.GC17297 at lb.tenfour>
> Content-Type: text/plain; charset=us-ascii
> 
> * Dan Nelson <dnelson at allantgroup.com> [1110 17:10]:
> > In the last episode (Nov 18), Ivan Voras said:
> > > I've setup pam_ldap and nss_ldap and samba3 and smbldap, and it works
> > > fine, but it would be nice to replace /usr/bin/passwd with
> > > smbldap-passwd (which changes both NTLM and Unix password fields in
> > > LDAP, while passwd is unaware of LDAP, at least according to man
> > > page).  Is there a clean way of doing it so I don't have to replace
> > > it by hand after each installworld?
> > 
> > passwd just uses PAM to set passwords
> 
> When did that come in? I can't get passwd(1) to change an ldap password, I get:
> 
> rasputnik at eris:rasputnik$ passwd
> passwd: Sorry, `passwd' can only change passwords for local or NIS users.
> rasputnik at eris:rasputnik$ uname -a
> FreeBSD eris 5.3-STABLE FreeBSD 5.3-STABLE #4: Mon Oct 25 18:03:11 BST 2004     root at eris:/usr/obj/usr/src/sys/ERIS  i386
> rasputnik at eris:rasputnik$ 
> 
>  - this is an nss_ldap nsswitched account, logged in over ssh (pam_ldap auth).
> 
> -- 
> This must be Thursday. I never could get the hang of Thursdays. - Arthur Dent
> Rasputin :: Jack of All Trades - Master of Nuns
> 
> ------------------------------
> 
> Message: 5
> Date: Sat, 20 Nov 2004 14:23:20 -0800
> From: "Michael C. Shultz" <ringworm at inbox.lv>
> Subject: Re: Floppy Boot ??
> To: freebsd-current at freebsd.org
> Message-ID: <200411201423.20426.ringworm at inbox.lv>
> Content-Type: text/plain;  charset="iso-8859-9"
> 
> On Saturday 20 November 2004 01:11 pm, Ege Mkan wrote:
> > Hi,
> >
> > Is there way to install FreeBSD 5.3 without install boot loader to
> > HDD but with creating boot floppies to start your system which is
> > installed on your HDD ??
> >
> > Thanx
> >
> > Ege.
> 
> Take a look at /usr/ports/sysutils/grub. It lets you boot a variety
> of OS's using a floppy as a bootloader.
> 
> -Mike
> 
> ------------------------------
> 
> Message: 6
> Date: Sat, 20 Nov 2004 17:27:38 -0700 (MST)
> From: "M. Warner Losh" <imp at bsdimp.com>
> Subject: Re: Problem with HP DL380
> To: le at FreeBSD.org
> Cc: freebsd-current at FreeBSD.org
> Message-ID: <20041120.172738.38711956.imp at bsdimp.com>
> Content-Type: Text/Plain; charset=us-ascii
> 
> In message: <20041120104224.E566 at korben.prv.univie.ac.at>
>             Lukas Ertl <le at FreeBSD.org> writes:
> : On Sat, 20 Nov 2004, M. Warner Losh wrote:
> : 
> : > : Setting hw.pci.do_powerstate=0 in /boot/loader.conf fixes this problem.
> : >
> : > Dang.  I missed the earlier part of this thread.  You shouldn't need
> : > to do this if I coded things up right.  But, alas, you do.  Can you
> : > send me the details?
> : 
> : No, I think you did everything right.  If I understand it correctly, then 
> : hw.pci.do_powerstate puts a device into C3 state if no driver attaches. 
> : That's just what happens here.  That iLO service processor of the DL380s 
> : shows up as normal PCI device, but exports the console to the network even 
> : without specific driver (as there is none for FreeBSD), so that it works 
> : independent of the OS.
> : 
> : So, since there is no driver, the device gets powered down as soon as the 
> : kernel initializes its ACPI parts, and then the iLO console stops working.
> 
> It may be the case that the base device class needs a 'dummy' driver
> to prevent things like this from happening.  We have a hack in the
> code right now for displays until the situation there is straightened
> out.
> 
> Warner
> 
> ------------------------------
> 
> Message: 7
> Date: Sun, 21 Nov 2004 11:10:02 +1030
> From: "Wilkinson, Alex" <alex.wilkinson at dsto.defence.gov.au>
> Subject: Re: Problem with HP DL380
> To: freebsd-current at freebsd.org
> Message-ID: <20041121004001.GB7395 at squash.dsto.defence.gov.au>
> Content-Type: text/plain; charset="us-ascii"
> 
> what is meant by 'C3 state' ?
> 
>  - aW
> 
> 
> 0n Sat, Nov 20, 2004 at 10:46:25AM +0100, Lukas Ertl wrote: 
> 
> > On Sat, 20 Nov 2004, M. Warner Losh wrote:
> > 
> > >: Setting hw.pci.do_powerstate=0 in /boot/loader.conf fixes this problem.
> > >
> > >Dang.  I missed the earlier part of this thread.  You shouldn't need
> > >to do this if I coded things up right.  But, alas, you do.  Can you
> > >send me the details?
> > 
> > No, I think you did everything right.  If I understand it correctly, then 
> > hw.pci.do_powerstate puts a device into C3 state if no driver attaches. 
> > That's just what happens here.  That iLO service processor of the DL380s 
> > shows up as normal PCI device, but exports the console to the network even 
> > without specific driver (as there is none for FreeBSD), so that it works 
> > independent of the OS.
> > 
> > So, since there is no driver, the device gets powered down as soon as the 
> > kernel initializes its ACPI parts, and then the iLO console stops working.
> > 
> > cheers,
> > le
> > 
> > -- 
> > Lukas Ertl                         http://homepage.univie.ac.at/l.ertl/
> > le at FreeBSD.org                     http://people.freebsd.org/~le/
> > _______________________________________________
> > freebsd-current at freebsd.org mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-current
> > To unsubscribe, send any mail to "freebsd-current-unsubscribe at freebsd.org"
> > 
> 
> ------------------------------
> 
> Message: 8
> Date: Sat, 20 Nov 2004 20:00:50 -0600
> From: Dan Nelson <dnelson at allantgroup.com>
> Subject: 	Re: interrupt moderation for if_* [was: serious networking
> 	(em) ...]
> To: "Bjoern A. Zeeb" <bzeeb-lists at lists.zabbadoz.net>
> Cc: Shunsuke SHINOMIYA <shino at fornext.org>
> Message-ID: <20041121020050.GB94473 at dan.emsphone.com>
> Content-Type: text/plain; charset=us-ascii
> 
> In the last episode (Nov 20), Bjoern A. Zeeb said:
> > On Fri, 19 Nov 2004, Jeremie Le Hen wrote:
> > > >  Hi, Jeremie, how is this? To disable Interrupt Moderation,
> > > >  sysctl hw.em?.int_throttle_valve=0.
> > >
> > > Great, I would have called it "int_throttle_ceil", but that's a
> > > detail and my opinion is totally subjective.
> > >
> > > >  However, because this patch is just made now, it is not fully tested.
> > >
> > > I'll give it a try this weekend although I won't be able to make
> > > performance mesurements.
> > 
> > when thinking about this could there be some "global" way for
> > interrupt moderation; For if_sk there is this PR:
> > http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/41220
> > 
> > something 'unique' to all NIC drivers would be good I guess ?
> 
> You need NIC support for it, just like checksum offloading.  And other
> drivers offer more sysctls; fxp lets you choose the maximum delay and
> maximum packets to hold before firing an interrupt.
> 
> -- 
> 	Dan Nelson
> 	dnelson at allantgroup.com
> 
> ------------------------------
> 
> Message: 9
> Date: Sat, 20 Nov 2004 20:07:45 -0600
> From: Dan Nelson <dnelson at allantgroup.com>
> Subject: Re: Replacing passwd?
> To: Dick Davies <rasputnik at hellooperator.net>
> Cc: FreeBSD Current Users <freebsd-current at freebsd.org>
> Message-ID: <20041121020745.GC94473 at dan.emsphone.com>
> Content-Type: text/plain; charset=us-ascii
> 
> In the last episode (Nov 20), Dick Davies said:
> > * Dan Nelson <dnelson at allantgroup.com> [1110 17:10]:
> > > In the last episode (Nov 18), Ivan Voras said:
> > > > I've setup pam_ldap and nss_ldap and samba3 and smbldap, and it works
> > > > fine, but it would be nice to replace /usr/bin/passwd with
> > > > smbldap-passwd (which changes both NTLM and Unix password fields in
> > > > LDAP, while passwd is unaware of LDAP, at least according to man
> > > > page).  Is there a clean way of doing it so I don't have to replace
> > > > it by hand after each installworld?
> > > 
> > > passwd just uses PAM to set passwords
> > 
> > When did that come in? I can't get passwd(1) to change an ldap password, I get:
> > 
> > rasputnik at eris:rasputnik$ passwd
> > passwd: Sorry, `passwd' can only change passwords for local or NIS users.
> > rasputnik at eris:rasputnik$ uname -a
> > FreeBSD eris 5.3-STABLE FreeBSD 5.3-STABLE #4: Mon Oct 25 18:03:11 BST 2004     root at eris:/usr/obj/usr/src/sys/ERIS  i386
> > rasputnik at eris:rasputnik$ 
> > 
> >  - this is an nss_ldap nsswitched account, logged in over ssh (pam_ldap auth).
> 
> Weird.  There's definitely a lot of PAM code in passwd.c.  What happens
> if you comment out the errx() function that prints that error (line
> 124)?
> 
> -- 
> 	Dan Nelson
> 	dnelson at allantgroup.com
> 
> ------------------------------
> 
> Message: 10
> Date: Sun, 21 Nov 2004 13:00:27 +0100
> From: S?ren Schmidt<sos at DeepCore.dk>
> Subject: Re: ATAng support for reading from register on ATAREQUEST
> 	command?
> To: m.kucenski at computer.org
> Cc: freebsd-current at freebsd.org
> Message-ID: <41A0835B.9040404 at DeepCore.dk>
> Content-Type: text/plain; charset="iso-8859-1"
> 
> Matt Kucenski wrote:
> > I am trying to develop a program that will allow modifications to the Host Protected Area settings and one of the commands (READ NATIVE MAX ADDRESS) returns the native max address back in the register (LBA high, low, mid).
> > 
> > I have been looking at the smartmontools project for pointers on how to write this code and according to that source, this is not possible yet with ATAng.  There is a comment in their code that another command (ATA_CMD_READ_REG) patch has been submitted to ATAng, but it does not appear to have made it into any of the latest sources.
> > 
> > Can anyone offer any information on this?
> 
> The following patch returns the register values in the request you sent 
> through ioctl call. That should do the trick without any new calls..
> 
> 



More information about the freebsd-current mailing list