Frank, could you comment (and test) following changes to the nozomi driver? I did them before you post the updated version and rediffed them a while ago, I hope I haven't done any mistake while doing it. Thanks. -- nozomi, remove unneded stuff - tty_termios* are set to tty struct, but tty_register_driver overwrites it with its own values (or NULLs), no need to have these pointers stored - [rt]x_data are updated but never used Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- ...

nozomi, remove void * casts Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- commit 92ca82bcbd5dde95096dac24d0f6a9beab68e003 tree 748ac937f772410b364245897492b693491743f6 parent 1ab5d64fc4a2dd7a50b8971c67b9793511b4c47a author Jiri Slaby <jirislaby@gmail.com> Mon, 05 Nov 2007 15:28:46 +0100 committer Jiri Slaby <jirislaby@gmail.com> Sat, 10 Nov 2007 00:15:05 +0100 drivers/char/nozomi.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers...

Especially on this I would like to see feedback. Unlock and lock the spinlock just around the tty_flip_buffer_push would be much more easier, but won't it break anything? -- nozomi, fix tty_flip_buffer_push tty_flip_buffer_push call may deadlock when invoking it while holding spinlock used in e.g. throttle too. Nozomi sets low_latency tty and that means, that ldisc flush will be called immediately and it can call some nozomi functions back. Solve it by invoking tty_flip_buffer_push after rele...

nozomi, tty cleanup - init and deinit tty driver at module load/unload. When the OS (user) loads the driver the hardware usually is ready to driver. - merge (unify) MAX_PORT, NTTY_TTY_MINORS into NOZOMI_MAX_PORTS - remove struct nozomi_devices, it was used only as list entries Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- commit d0b01ce89a7b18ba37ea6192eea6a98cdc01d62e tree a61f0a99633772b863130ec1e8a2be25891b3578 parent 9f9d7197e901ea00771812b7e903b14b95f54e40 author Jiri Slab...

nozomi, reorder and cleanup probe, remove - remap space after requesting region, to not map something we cannot use anyway - init spin lock before request_irq, because due to shared irq debug, isr might be called immediately, where the lock is being held - remove INIT_LIST_HEAD before list_add, it's useless, because list_add re-sets prev and next - reorder cleanup to be the same as fail path of probe (or in other words, the reverse of probe) - no need to call nozomi_setup_private_data in ...

nozomi, tty index cleanup - don't store unneeded copy of tty->index into port structure, tty->index is available everywhere - mod tty->index by MAX_PORT where expected (otherwise array index out of bounds) Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- commit 3afb47133874a0979f57928d7450612fa982a6c5 tree 8d9ea282ddbffed0d75c946b3ae046e3bdcc7456 parent 14e887763c076e45f4f768348361a37c10709902 author Jiri Slaby <jirislaby@gmail.com> Sun, 28 Oct 2007 22:15:22 +0100...

nozomi, expand some functions nozomi_setup_interrupt and tty_do_close are used only in one place and has no pros of being in separate functions. move tty_do_close contents into ntty_close (it contained only tty_do_close call before) and nozomi_setup_interrupt (only request_irq) into nozomi_card_init. Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- commit 29335b619cc5ee27417d50982f029d440570f67c tree 9cef37ef05e53453cd4d245a34c038ca12433253 parent 79aeb51662902086507927cb7fcd682ae319...

riscom8, remove wakeup anf hangup bottomhalves Both of them may be called directly from the code, don't add special code and variables and schedule a work for them. Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> --- commit 9d50b0aba24ca0b601820663939987990db134c6 tree 46524b3fe1a7a7d4b7d414bf3c4984b61875e64e parent 961f0508a584436d4e41ddf5ccf47ab4a7b6a3de author Jiri Slaby <jirislaby@gmail.com> Mon, 05 Nov 2007 16:40:16 +0100 committ...

serial167, remove bottomhalf - Cy_EVENT_OPEN_WAKEUP is simple wake_up - Cy_EVENT_HANGUP is wake_up + tty_hangup, which schedules its own work - Cy_EVENT_WRITE_WAKEUP is tty_wakeup which may be called directly too Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> --- commit 961f0508a584436d4e41ddf5ccf47ab4a7b6a3de tree 056c19b77aae9ca6c77b1d7d314166f4bfb177e9 parent 695ec3fa93346e03d33392504ca99aaaeb4cb5f7 author Jiri Slaby <jirislaby@gmail...

specialix, remove bottomhalves - tqueue is used only for tty_wakeup, call it directly from the code - tqueue_hangup for tty_hangup, it schedules its own work, use it directly too Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> --- commit fb6091801dc98a44bd22f04acd4e66524eb0be9c tree 7c495c5b0bc2982c18f321016939eb7977790709 parent d55ce2f15b7233e7a4e5e1880a656c4160b69866 author Jiri Slaby <jirislaby@gmail.com> Mon, 05 Nov 2007 16:26:...

rocket, switch long delay to sleep Don't busy wait for whole 1s when registering some rocket modems. Sleep instead since we are not in atomic. Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- commit a9b06d9c7a580246dc328e96f31039d932bca47c tree eb2e25299ebce29c89efa19c3014d41f742454c0 parent 5b07d82896b4d783e76f82dab6c4ba162545e414 author Jiri Slaby <jirislaby@gmail.com> Sat, 20 Oct 2007 11:51:04 +0200 committer Jiri Slaby <jirislaby@gmail.com> Fri, 09 Nov 2007 22:21:49 +...

stallion, fix compiler warnings Don't emit warnings on 64 bit platforms from min(). sizeof() on those is not uint, neither 2 pointers difference, cast it to uint by min_t in both cases. Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- commit 7a411a4d1729e34bd7dc6697a8eedec7dd989c32 tree 2422610e5f3cdde71c9a5cec4d134baf9b72daa5 parent f412e3e0d4d012b7b57e381ae85ddfb1b5fc99bf author Jiri Slaby <jirislaby@gmail.com> Tue, 23 Oct 2007 12:47:28 +0200 committer Jiri Slaby <jirisla...

char/serial, remove SERIAL_TYPE_NORMAL redefines Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> --- commit d927dc37e63aa8347e596052615b85ff4936b3e2 tree 0c62b277b0a1b414cd1ac911de0f65b0ebcfd216 parent 585adae2019a87909ec738cbd414a330afbe04a8 author Jiri Slaby <jirislaby@gmail.com> Sat, 20 Oct 2007 22:35:24 +0200 committer Jiri Slaby <jirislaby@gmail.com> Fri, 09 Nov 2007 22:33:04 +0100 drivers/char/epca.h | 1 - dr...

On Fri, 9 Nov 2007 18:36:08 -0500 Acked-by: Alan Cox <alan@redhat.com> I'd just done some of that in my tree too. I have some other changes but they don't overlap (termios and a very very minor auth fix) -

The missing \n was intentional looking at the old code, and meant you got a lot of trace per line for debugging. The KERN_ additions are the breakage here. We might as well just lose them, we have proper debuggers now and they are useless with \n, and pretty useless without. -

sunrpc, remove SPIN_LOCK_UNLOCKED SPIN_LOCK_UNLOCKED is deprecated, use DEFINE_SPINLOCK instead Signed-off-by: Jiri Slaby <jirislaby@gmail.com> --- commit d5e782e62a4fe2663a012571c3444445d9887b02 tree b993038b020d8c619f6ffdad412fbb992c073513 parent 828042d12cc0aa515e049889aa76d4066df100c0 author Jiri Slaby <jirislaby@gmail.com> Thu, 25 Oct 2007 12:00:13 +0200 committer Jiri Slaby <jirislaby@gmail.com> Thu, 25 Oct 2007 12:00:13 +0200 net/sunrpc/xprt.c | 2 +- 2 fil...

cpwatchdog, remove SPIN_LOCK_UNLOCKED SPIN_LOCK_UNLOCKED is deprecated, use __SPIN_LOCK_UNLOCKED with an unique name instead Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Cc: Wim Van Sebroeck <wim@iguana.be> --- commit 4e91ba06c53856bfaab065aaf9fd8f5b0dd59f6f tree ff123ed71958d9331623876dfc20efa86c639498 parent 5290a0e1fffeb61f0062ebca98e45a6a5ba75711 author Jiri Slaby <jirislaby@gmail.com> Thu, 25 Oct 2007 11:56:54 +0200 committer Jiri Slaby <jirislaby@gmail.com> T...

With this patchset the former ARCH=i386 / ARCH=x86_64 are replaced by ARCH=x86. The rationale behind the patchset are that with a unified x86 architecture this should be reflected in the build commands. With this patch set the 32/64 bit selection is done at configuration time like we know it from parisc and powerpc. Please pull to your cleanup branch: ssh://master.kernel.org/pub/scm/linux/kernel/git/sam/x86.git I leave it to you (x86 maintainers) to decide when to push this to Linus. But...

After unification of the Kconfig files it required only trivial changes to enable "make ARCH=x86". This patch unteach kconfig about SRCARCH and let kbuild compute x86 for `uname -m` equals x86_64 and the i.86 variants. The selection of 32 versus 64 bit is now done as a configuration step like we select SMP, PREMPT etc. Two drawbacks of this patch: 1) People are used to select 32 versus 64 using ARCH={i386,x86_64} 2) we lost the possibility to generate a true allmodconfig for a 64 bit x86 ke...

To ease unification of Kconfig.i386 and Kconfig.x86_64 add X86_32 dependencies to all i386 specific symbols. This patch introduce no functional changes but is one step towards unification. This smaller step is used to ease review of the patch set. Signed-off-by: Sam Ravnborg <sam@ravnborg.org> --- arch/x86/Kconfig.i386 | 90 +++++++++++++++++++++++++++++-------------------- 1 files changed, 53 insertions(+), 37 deletions(-) diff --git a/arch/x86/Kconfig.i386 b/arch/x86/Kconfig.i386 ...

This step introduces the file arch/x86/Kconfig which contains all the menu's from "Power Management" and below. The main part of the new Kconfig file is shared and the remaining i386/x86_64 specific symbols are covered by dependencies. A x86_64 allmodconfig build did not show any differences. Signed-off-by: Sam Ravnborg <sam@ravnborg.org> --- arch/x86/Kconfig | 401 +++++++++++++++++++++++++++++++++++++++++++++++ arch/x86/Kconfig.i386 | 326 +--------------------------------...

Most of the arch settings were equal so combine them in the first part of Kconfig. Signed-off-by: Sam Ravnborg <sam@ravnborg.org> --- arch/x86/Kconfig | 136 +++++++++++++++++++++++++++++++++++++++++++++++ arch/x86/Kconfig.i386 | 119 ----------------------------------------- arch/x86/Kconfig.x86_64 | 115 --------------------------------------- 3 files changed, 136 insertions(+), 234 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 9fbb049..d47b5a2 100644 ...

Merge the two Kconfig files to a single file. Checked using make allmodconfig for x86_64. No changes in build. Signed-off-by: Sam Ravnborg <sam@ravnborg.org> Cc: Adrian Bunk <bunk@kernel.org> Cc: Dave Jones <davej@codemonkey.org.uk> --- arch/x86/Kconfig.i386 | 2 +- arch/x86/Kconfig.x86_64 | 2 +- arch/x86/kernel/cpu/cpufreq/Kconfig | 275 ++++++++++++++++++++++++++++++++ arch/x86/kernel/cpu/cpufreq/Kconfig_32 | 250 ----------------...

I've been investigating a problem recently, in which N runnable CPU-bound tasks on an N-way machine run on only N-1 CPUs. The remaining CPU is almost 100% idle. I have seen it occur with both the CFS and O(1) schedulers. I've traced this down to what seems to be a quirk in the SMP balancer, whereby a high-priority thread which spends most of its time sleeping can artificially inflate the CPU load average calculated for one processor. Most of the time this CPU is idle (nr_running==0) yet its CPU lo...

Linus, please pull the latest scheduler git tree from: git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched.git There are two cross-subsystem groups of fixes: three commits that resolve a KVM build fix on !SMP - acked by Avi to go in via the scheduler git tree because it changes a central include file. The other one is a powerpc CPU time accounting regression fix from Paul Mackerras. The remaining 14 commits: one crash fix (only triggerable via the new control-groups fil...