[PATCH] udev - allow all files in a directory as the config
I was on the train for 5 hours today and the TODO is almost empty :)
So, at least four people wanted this feature, then here is a actual
working patch.
We may specify now in udev.conf:
udev_rules="/etc/udev/"
and udev will scan the whole directory for files ending with *.rules,
sort it in lexical order and create our rule list from all of the files.
A plain given file will still work and the same applies to the *.permissions.
I sort the files in our usual linked list, cause klibc has no scandir().
[PATCH] udev - create all partitions of blockdevice
Here is the first try to create all partitons of a blockdevice, since
removable media devices may need to acces the expected partition to
revalidate the media.
It uses the attribute syntax introduced with the last %s{file} patch.
I'm using this with my multi-slot-flash-card-reader:
On Sun, Feb 15, 2004 at 03:36:00AM +0100, Kay Sievers wrote:
>
> Since we have %s{file} it may be nice to allow SYSFS{file}.
> This patch allows:
>
> BUS="usb", SYSFS{idProduct}="a511", NAME="video%n"
>
> compared to the current:
>
> BUS="usb", SYSFS_idProduct="a511", NAME="video%n"
>
> The curent syntax is still supported.
> Looks a bit nicer and less hackish, I think.
Better patch with infrastructure to easily implement KEY{attribute}
for every other key. The first user is the SYSFS{file} key.
Both versions, brackets or underscore is supported for the attribute.
[PATCH] Adding '%s' format specifier to NAME and SYMLINK
On Thu, Feb 12, 2004 at 05:34:57PM -0800, Greg KH wrote:
> On Tue, Feb 10, 2004 at 09:14:20AM +0100, Hannes Reinecke wrote:
> > Hi all,
> >
> > this patch makes the format for NAME and SYMLINK a bit more flexible:
> > I've added a new format specifier '%s{<SYSFS_var>}', which allows for
> > the value of any sysfs entry found for this device to be inserted.
> > Example (for our S/390 fcp adapter):
> >
> > BUS="ccw", SYSFS_devtype="1732/03", NAME="%k" \
> > SYMLINK="zfcp-%s{hba_id}-%s{wwpn}:%s{fcp_lun}"
> >
> > I know this could also be done with an external program, but having this
> > incorporated into udev makes life easier, especially if run from
> > initramfs. Plus it makes the rules easier to follow, as the result is
> > directly visible and need not to be looked up in some external program.
> >
> > Comments etc. welcome.
>
> Oops, sorry I missed this for the 017 release. I'll look at it tomorrow
> and get back to you. At first glance it looks like a good thing.
>
> Oh, you forgot to update the documentation, that's important to do if
> you want this change to make it in :)
I took a part of the code and made a version that uses already implemented
attribute finding logic.
The parsing of the format length '%3x' and the '%x{attribute}' is a fuction now,
maybe there are more possible users in the future.
incremental to udev-016/extras/multipath-0.0.16.3,
* add a GROUP_BY_SERIAL flag. This should be useful for
controlers that activate their spare paths on simple IO
submition with a penalty. The StorageWorks HW defaults to
this mode, even if the MULTIBUS mode is OK.
* remove unused sg_err.c
* big restructuring : split devinfo.c from main.c. Export :
* void basename (char *, char *);
* int get_serial (int, char *);
* int get_lun_strings (char *, char *, char *, char *);
* int get_evpd_wwid(char *, char *);
* long get_disk_size (char *);
Now we see clearly what is expected from an external package
like scsi_id.
* stop passing struct env as param
> Hello,
>
> incremental to udev-016/extras/multipath,
>
> * don't rely on the linux symlink in the udev/klibc dir since
> udev build doesn't use it anymore. This corrects build breakage
> * remove make_dm_node fn & call. Rely on udev for this.
>
> The first patch is to be applied.
> The second is conditioned by udev dealing correctly with devmap names.
>
> For this I can suggest a CALLOUT rule like this :
> KERNEL="dm-[0-9]*", PROGRAM="/tmp/name_devmap %M %m", NAME="%k",
> SYMLINK="%c"
>
> With name_devmap like :
> #!/bin/sh
> /usr/sbin/dmsetup ls|/bin/grep "$1, $2"|/usr/bin/awk '{print $1}'
>
ok I coded the suggested tool.
it works with the following rule :
KERNEL="dm-[0-9]*", PROGRAM="/sbin/devmap_name %M %m", NAME="%k", SYMLINK="%c"
I don't know if it's right to keep this tools packaged with multipath because
it's widely more general.
Maybe Joe should merge it in the device-mapper package or provide the
functionnality through dmsetup ?
Here is a small improvement. We check for the type of message we receive
and udevsend seems not to need all the credential setup stuff, the
kernel will fill it for us.
udevd now refuses to start as non root, cause it doesn't make any sense.
Here is the badly needed client authorization for udevd.
Since we switched to abstract namespace sockets, we are unable to
control the access of the socket by file permissions.
So here we send a ancillary credential message with every datagram,
to be able to verify the uid of the sender. The sender can't fake the
credentials, cause the kernel doesn't allow it for non root users.
On Mon, Feb 09, 2004 at 05:41:15AM +0100, Kay Sievers wrote:
> It seems that today was just another udev-sunday for me :)
>
> Here is a working patch to compile udevd with klibc.
>
> It's sweet the static binary takes 6 kbytes and it runs
> with only 80 kbytes virtual memory.
>
> I changed a few peaces and added a siginterrupt.c file to klibc.
> We may check with hpa to get the changes upstream?
So here is the next try :)
hpa, for good reason, didn't like my changes to klibc.
He will dump signal() completely from klibc instead, so here we switch to
sigaction() and keep udevd working with klibc.
On Sun, Feb 08, 2004 at 04:36:01PM +0100, Kay Sievers wrote:
> We don't handle NAME="" the right way. Thanks to Emil None <emil71se@yahoo.com>
> for pointing this out. Here is a fix for it and a trivial style cleanup.
Once again, patch to make logging a config option.
Reason for this (since you asked for it):
- In our setup it is easy (although still annoying) .. just edit the
ebuild, add logging support (or remove it) and rebuild. For say a
binary distro, having the logging is useful for debugging some
times, but its more a once of, or rare thing, as you do not add or
change config files every day. Sure, we can have logging by
default, but many do not want ~300 lines of extra debugging in their
logs is not pleasant, and they will complain. Rebuilding the
package for that binary package (given the users it is targeted to)
is usually not within most users grasp.
[PATCH] convert udevsend/udevd to DGRAM and single-threaded
On Fri, Feb 06, 2004 at 01:08:24AM -0500, Chris Friesen wrote:
>
> Kay, you said "unless we can get rid of _all_ the threads or at least
> getting faster, I don't want to change it."
>
> Well how about we get rid of all the threads, *and* we get faster?
Yes, we are twice as fast now on my box :)
> This patch applies to current bk trees, and does the following:
>
> 1) Switch to DGRAM sockets rather than STREAM. This simplifies things
> as mentioned in the previous message.
>
> 2) Invalid sequence numbers are mapped to -1 rather than zero, since
> zero is a valid sequence number (I think). Also, this allows for real
> speed tests using scripts starting at a zero sequence number, since that
> is what the initial expected sequence number is.
>
> 3) Get rid of all threading. This is the biggie. Some highlights:
> a) timeout using setitimer() and SIGALRM
> b) async child death notification via SIGCHLD
> c) these two signal handlers do nothing but raise volatile flags,
> all the
> work is done in the main loop
> d) locking no longer required
I cleaned up the rest of the comments, the whitespace and a few names to match
the whole thing. Please recheck it. Test script is switched to work on subsystem
'test' to let udev ignore it.
* don't rely on the linux symlink in the udev/klibc dir since
udev build doesn't use it anymore. This corrects build breakage
* remove make_dm_node fn & call. Rely on udev for this.
The first patch is to be applied.
The second is conditioned by udev dealing correctly with devmap names.
For this I can suggest a CALLOUT rule like this :
KERNEL=3D"dm-[0-9]*", PROGRAM=3D"/tmp/name_devmap %M %m", NAME=3D"%k", SY=
MLINK=3D"%c"
With name_devmap like :
#!/bin/sh
/usr/sbin/dmsetup ls|/bin/grep "$1, $2"|/usr/bin/awk '{print $1}'
[PATCH] udevd - switch socket path to abstract namespace
As Chris Friesen <chris_friesen@sympatico.ca> suggested, here we switch
the unix domains socket path to abstract namespace and get rid of the
socket file in the filesystem.
Hey, this was new to me today. So here a few words:
Linux supports a abstract namespace for sockets. We don't need a
physical file on the filesystem but only a unique string magically
starting with the '\0' character.
strace with real file:
connect(3, {sa_family=AF_UNIX, path="/udev/.udevd.sock"}, 110)
strace with abstract namespace:
connect(3, {sa_family=AF_UNIX, path=@udevd}, 110)
This patch allows udevsend to be called by the user and not only by the
kernel with its SEQNUM. If no SEQNUM is given, we move the event straight
to the exec queue and don't look if something is missing.
I don't know if this is really needed, but some people seem trying to
send events trough udevd instead of calling udev directly with their
scripts and confuse the reorder logic with that.
So at least, we may remove this source of confusion and udevsend is much
much faster back than udev itself and it will also block concurrent events
for the same devpath.
this patch corrects ide devices with number greater than 9 being linked
into wrong discs/discX directories (my hda10 device was in discs/disc1
directory)
it adds % into pattern for $DRIVE, so for example, hda10 is not eaten
to hda1, but to hda, and break in for cycle works for it
> I don't mind udevd using glibc, I just want the programs that get run a
> lot of different times (udev and udevsend) to be as small as possible to
> get the best cache results. As udevd sticks around all the time, it's
> not as important. Sound sane to you?
> Here is a small cleanup and better Makefile integration.
> udevd and udevsender are now installed. Just switch HOTPLUG_EXEC from ROOT
> to SENDER before install and udevsend will be called.
>
> We may add the location of the socket and lock file to the config,
> if this is needed.
[PATCH] udevd - cleanup and better timeout handling
On Thu, Jan 29, 2004 at 04:55:11PM +0100, Kay Sievers wrote:
> On Thu, Jan 29, 2004 at 02:56:25AM +0100, Kay Sievers wrote:
> > On Wed, Jan 28, 2004 at 10:47:36PM +0100, Kay Sievers wrote:
> > > Oh, couldn't resist to try threads.
> > > It's a multithreaded udevd that communicates through a localhost socket.
> > > The message includes a magic with the udev version, so we don't accept
> > > older udevsend's.
> > >
> > > No need for locking, cause we can't bind two sockets on the same address.
> > > The daemon tries to connect and if it fails it starts the daemon.
> > >
> > > We create a thread for every incoming connection, handle over the socket,
> > > sort the messages in the global message queue and exit the thread.
> > > Huh, that was easy with threads :)
> > >
> > > With the addition of a message we wakeup the queue manager thread and
> > > handle timeouts or move the message to the global exec list. This wakes
> > > up the exec list manager who looks if a process is already running for this
> > > device path.
> > > If yes, the exec is delayed otherwise we create a thread that execs udev.
> > > n the background. With the return of udev we free the message and wakeup
> > > the exec list manager to look if something is pending.
> > >
> > > It is just a quick shot, cause I couldn't solve the problems with fork an
> > > scheduling and I wanted to see if I'm to stupid :)
> > > But if anybody with a better idea or more experience with I/O scheduling
> > > we may go another way. The remaining problem is that klibc doesn't support
> > > threads.
> > >
> > > By now, we don't exec anything, it's just a sleep 3 for every exec,
> > > but you can see the queue management by watching syslog and do:
> > >
> > > DEVPATH=/abc ACTION=add SEQNUM=0 ./udevsend /abc
>
> Next version, switched to unix domain sockets.
Next cleaned up version. Hey, nobody wants to try it :)
Works for me, It's funny if I connect/disconnect my 4in1-usb-flash-reader
every two seconds. The 2.6 usb rocks! I can connect/diconnect a hub with 3
devices plugged in every second and don't run into any problem but a _very_
big udevd queue.
On Tue, Jan 27, 2004 at 11:02:25AM -0800, Greg KH wrote:
> On Mon, Jan 26, 2004 at 07:28:03PM -0500, Adrian Drzewiecki wrote:
> > Looking over the code, I noticed something odd in
> > namedev.c:strcmp_pattern() --
> >
> > while (*p && (*p != ']'))
> > p ++;
> > return strcmp_pattern(p+1, s+1);
> >
> > If the pattern string is invalid, and is not terminated by a ']', then 'p'
> > will point at \0 and p+1 will be beyond the string.
>
> Yes, I think you are correct.
>
> Hm, Kay, any idea of the proper way to fix this? I've attached a patch
> below, but I don't think it is correct.
>
> while (*p && (*p != ']'))
> p++;
> - return strcmp_pattern(p+1, s+1);
> + if (*p)
> + return strcmp_pattern(p+1, s+1);
> + else
> + return 1;
> }
> }
Sure, it's perfectly correct. I'm wondering how Adrian found this.
We can use the return 1 at the end of the whole function, and asking
for the closing ']' is more descriptive, but it does the same.
projects / systemd / log