... what explodes? If others have BKL-removal patches which look like 2.6 ... drivers/rtc/rtc-dev.c | 12 ++++++-- drivers/s390/char/fs3270.c ... to several drivers spidev: BKL pushdown vcs: BKL pushdown xilinx icap: ...

linux-kernel - Jonathan Corbet - May 18 2008 - 18:15

... /video1394.c | 18 +++++++++--- ... > ieee1394: cdev lock_kernel() pushdown I have yet to look at ... drivers in detail whether they need BKL or not. They likely don't. > firewire: cdev lock_kernel() pushdown drivers ...

linux-kernel - Stefan Richter - May 19 2008 - 13:46

... /ieee1394/video1394.c | 18 +++++++++--- >> ... >>> ieee1394: cdev lock_kernel() pushdown >> >> I have yet to look at these ... @s5r6.in-berlin.de> Subject: ieee1394: video1394: reorder module init, prepare BKL removal This ...

linux-kernel - Stefan Richter - May 19 2008 - 16:07

... video1394.c | 18 +++++++++--- > ... >> ieee1394: cdev lock_kernel() pushdown > > I have yet to look at ... drivers in detail whether they need BKL or > not. They likely don't ... a patch. dv1394 and raw1394 do not need the BKL in ...

linux-kernel - Stefan Richter - May 19 2008 - 15:27

... towards getting rid of the BKL at last. > drivers/infiniband/core/ ... then I can merge BKL removal patches for these places that ... is to never add the BKL to these places as part ... just lands on some giant patch. Thanks, Roland -- To unsubscribe from ...

linux-kernel - Roland Dreier - May 19 2008 - 00:00

... that we have a pure "push the BKL down" changeset merged, > and then I can merge BKL removal patches for these places that never > needed the BKL? (I guess I can send you such ... a stack of Knuth output that the BKL is not needed for specific open ...

linux-kernel - Jonathan Corbet - May 19 2008 - 09:37

... in fact, > need to acquire the BKL, so now they all do. ... lwn.net/linux-2.6.git bkl-removal > > Stephen, might it be about ... explodes? > > If others have BKL-removal patches which look like 2.6 ... interact with the kill-the-BKL bits that attack it all ...

linux-kernel - Ingo Molnar - May 19 2008 - 09:17

... .net (Jonathan Corbet) wrote: > > There's a new tree with this stuff: >=20 > git://git.lwn.net/linux-2.6.git bkl-removal >=20 > Stephen, might it be about time to pull this into linux-next and see > what explodes? I have added that to today's ...

linux-kernel - Stephen Rothwell - May 19 2008 - 00:03

... you're sure that this code doesn't need the BKL (and it kind of > looked that way to me), ... has been looked > at. So sending me a patch which does this would be great. OK, will send such a patch after auditing more carefully. Just to be very clear, ...

linux-kernel - Roland Dreier - May 19 2008 - 16:38

... wrote: > OK, will send such a patch after auditing more carefully. Just to be > very clear, the ... it, but, as Alan pointed out, there's more. The BKL currently protects open() calls against concurrency with other opens, with ioctl ...

linux-kernel - Jonathan Corbet - May 19 2008 - 16:42

> That's part of it, but, as Alan pointed out, there's more. The BKL > currently protects open() calls against concurrency with other opens, > with ioctl(), and with driver initialization as well. So it's a matter > of having one's locking and ...

linux-kernel - Roland Dreier - May 19 2008 - 18:18

... Right. Or two ioctls against each other unless one sleepers, or against random other parts of the kernel which still use the BKL - eg fasync, bits of procfs ... > And the point about driver initialization is that open() can be called > as soon as the ...

linux-kernel - Alan Cox - May 20 2008 - 04:26

Stephen Rothwell wrote: > I haven't been following this too closely, so how much of this stuff has > been posted/review/tested? This is the second posting (some of the changesets are new this time around). It all builds and

linux-kernel - Jonathan Corbet - May 19 2008 - 09:46

Roland Dreier wrote: > Thanks. Just to be super-explicit, ioctl() cannot be called on a given > file until the open() for that particular file has returned, right? ioctl() will not be called on a given file descriptor before

linux-kernel - Jonathan Corbet - May 19 2008 - 18:56

On Mon, May 19, 2008 at 04:56:24PM -0600, Jonathan Corbet wrote: > ioctl() will not be called on a given file descriptor before open() is > done, no. If there are other file descriptors open, though, somebody > can be calling ioctl() on them while the

linux-kernel - Jeff Dike - May 19 2008 - 22:10