I don't know of any specific documentation, but it's pretty easy: Devices, any kind of device, can export a match, based on specific properties of the subsystem it belongs to. In most cases its the same propery/id that is used inside the kernel, to find a (already loaded) driver which will bind this device. Any unique string, hardware ID's, whatever, are stuffed into a modalias, prefixed by "<subsystem>:" to be unique. PCI and USB are pretty obvious, they just stuff all their hardware ID...

Hi Jesse, Here are patches: [PATCH 1/2] pciehp: Rename duplicate slot name N as N-1, N-2, N-M... [PATCH 2/2] shpchp: Rename duplicate slot name N as N-1, N-2, N-M... Against Linus's 2.6.27-rc4. As noted in the changelogs, these are temporary patches, meant as placeholders until I complete the longer patch series that allows the PCI core to manage the slot names on behalf of the drivers. The longer patch series will keep the same behavior that these temp patches implement, so that's good...

Commit ef0ff95f136f0f2d035667af5d18b824609de320 (shpchp: fix slot name) introduces the shpchp_slot_with_bus module parameter, which was intended to help work around broken firmware that assigns the same name to multiple slots. Commit b3bd307c628af2f0a581c42d5d7e4bcdbbf64b6a (shpchp: add message about shpchp_slot_with_bus option) tells the user to use the above parameter in the event of a name collision. This approach is sub-optimal because it requires too much work from the user. Instead, let...

Hi Greg, While playing around with my slot symlink stuff, I noticed that the following sequence is problematic: 1. clean boot 2. modprobe acpiphp 3. echo 0 > /sys/bus/pci/slots/N/power 4. ??? After step 3, we *should* be seeing pci_release_dev() getting called, but we never do because the refcount on the device is still quite high (5 or 6, on my ia64 system). I'm still trying to track this down, but I did notice, via code inspection, at least one suspicious area: #define for_eac...

In particular, I'd like to know whether this should be fixed by pci_get_dev_by_id() decrementing the refcount of from/dev_start, pci_get_subsys() decrementing 'from', or by bus_find_device() decrementing 'start'. It looks like bus_find_device() is the place where this should logically happen, but the kerneldoc doesn't document the intended behaviour. -- Matthew Wilcox Intel Open Source Technology Centre "Bill, look, we understand that you're interested in selling us this operating system, ...

Wow, thanks Greg, that was fast. I applied this to my for-linus branch; I'll ask Linus to pull it for 2.6.27. Thanks, -- Jesse Barnes, Intel Open Source Technology Center --

otherwise Xen is _completely_ unusable with 5 or more VCPUs. ( when !CONFIG_HAVE_SPARSE_IRQ) based on Alex's patch also add +1 offset after redir_entries Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> Cc: Alex Nixon <alex.nixon@citrix.com> Index: linux-2.6/arch/x86/kernel/io_apic.c =================================================================== --- linux-2.6.orig/arch/x86/kernel/io_apic.c +++ linux-2.6/arch/x86/kernel/io_apic.c @@ -3625,16 +3625,21 @@ int __init probe_nr...

later could set CONFIG_HAVE_SPARSE_IRQ to true by default. YH --

hope we can kill nr_irqs/NR_IRQS YH --

-- Michal Hocko L3 team SUSE LINUX s.r.o. Lihovarska 1060/12 190 00 Praha 9 Czech Republic --

When DEBUG_SHIRQ is selected, a spurious IRQ is issued before the setup_irq() initializes the desc->depth. An IRQ handler may call disable_irq_nosync(), but then setup_irq() will overwrite desc->depth, and upon enable_irq() we'll catch this WARN: ------------[ cut here ]------------ Badness at kernel/irq/manage.c:180 NIP: c0061ab8 LR: c0061f10 CTR: 00000000 REGS: cf83be50 TRAP: 0700 Not tainted (2.6.27-rc3-23450-g74919b0) MSR: 00021032 <ME,IR,DR> CR: 22042022 XER: 20000000 TASK =...

Hi Linus, please pull from 'for-linus' branch of git://git390.osdl.marist.edu/pub/scm/linux-2.6.git for-linus In particular the patch from Eric is noteworthy since it gets ext4 working again on s390. The shortlog: Eric Sandeen (1): [S390] fix ext2_find_next_bit Gerald Schaefer (1): [S390] dcssblk: fix race in dcssblk_add_store() Heiko Carstens (1): [S390] Remove unneeded spinlock initialization. Jan Glauber (3): [S390] qdio: prevent oopsing if qdio_establis...

Hi LKML, I've upgraded my kernel few days earlier and suddenly my cpufreqd stopperd working. My configuration is this: AC adapter connected: set scaling_max_freq and scaling_min_freq to maximal available frequency AC adapter disconnected: set scaling_max_freq to maximal available, scaling_min_freq to minimal available. When I connect AC adapter, everything works fine but when I disconnect my adapter, everything remains on maximal frequency and in logs I can see: cpufreqd: cpufreqd_set_...

Signed-off-by: Alex Nixon <alex.nixon@citrix.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Ingo Molnar <mingo@elte.hu> --- arch/x86/kernel/paravirt.c | 8 ++++++++ arch/x86/kernel/process_32.c | 10 ++++++++-- arch/x86/kernel/process_64.c | 9 +++++++-- arch/x86/kernel/smpboot.c | 8 ++++---- include/asm-x86/paravirt.h | 27 +++++++++++++++++++++++++++ include/asm-x86/smp.h | 13 +++++++++++-- 6 files changed, 65 insertions(+), 10 deletions(-) ...

The native case of wb_invd_halt uses inline asm to ensure the compiler doesn't reorder the sequence of instructions, which has the side-effect of skirting around the paravirt hooks for those instructions. Thus, create a new hook for this case. Signed-off-by: Alex Nixon <alex.nixon@citrix.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Ingo Molnar <mingo@elte.hu> --- arch/x86/kernel/paravirt.c | 1 + include/asm-x86/paravirt.h | 6 ++++++ include/asm-x86/processor.h | ...

Signed-off-by: Alex Nixon <alex.nixon@citrix.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Ingo Molnar <mingo@elte.hu> --- arch/x86/xen/irq.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c index bb04260..4e3f7f7 100644 --- a/arch/x86/xen/irq.c +++ b/arch/x86/xen/irq.c @@ -122,12 +122,19 @@ static void xen_halt(void) xen_safe_halt(); } +static void xen_wbinvd_halt(void) +{ + native_wbinvd();...

Add the new play_dead into smpboot.c, as it fits more cleanly in there alongside other CONFIG_HOTPLUG functions. Signed-off-by: Alex Nixon <alex.nixon@citrix.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Ingo Molnar <mingo@elte.hu> --- arch/x86/kernel/process_32.c | 38 -------------------------------------- arch/x86/kernel/process_64.c | 28 ---------------------------- arch/x86/kernel/smpboot.c | 25 +++++++++++++++++++++++++ include/asm-x86/smp.h | 1 +...

Note the changes from 2.6.18-xen CPU hotplugging: A vcpu_down request from the remote admin via Xenbus both hotunplugs the CPU, and disables it by removing it from the cpu_present map, and removing its entry in /sys A vcpu_up request from the remote admin only re-enables the CPU, and does not immediately bring the CPU up. A udev event is emitted, which can be caught by the user if he wishes to automatically re-up CPUs when available, or implement a more complex policy. Signed-off-by: Alex Nixon &l...

This patch applies to 2.6.27-rc4 + current -mm patches. From: Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Currently each arch/*/vmlinux.lds file separately mentions the CONSTRUCTORS section. If it is not mentioned, constructors (such as needed for the gcov infrastructure) are never called. If it is mentioned outside of the same section as DATA_DATA, it can lead to linker problems: Moving the CONSTRUCTORS section to DATA_DATA fixes this problem for powerpc and reduces code duplicat...

From: Kay Sievers <kay.sievers@vrfy.org> With 22454cb99fc39f2629ad06a7eccb3df312f8830e we added only the first entry of the device table. We need to loop over the whole device list. Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> --- scripts/mod/file2alias.c | 16 +++++++++++----- 1 files changed, 11 insertions(+), 5 deletions(-) diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index 4fa1f3a..4c9890e 10...

From: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com> Thanks to Magnus Damm for pointing that out. Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com> Cc: Magnus Damm <magnus.damm@gmail.com> Acked-by: Hans J. Koch <hjk@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> --- drivers/uio/uio_pdrv.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/uio/uio_pdrv.c b/drivers/uio/uio_pdrv.c index aa3bff8..0b4ef39 ...

From: Kay Sievers <kay.sievers@vrfy.org> Otherwise we leak references, which is not a good thing to do. Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> --- block/genhd.c | 22 ++++++++++++++++------ 1 files changed, 16 insertions(+), 6 deletions(-) diff --git a/block/genhd.c b/block/genhd.c index f34befc..656c2c7 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -313,8 +313,10 @@ static void *part_start(struct seq_f...

From: Rafael J. Wysocki <rjw@sisk.pl> PM: Remove WARN_ON from device_pm_add Fix message in device_pm_add() saying that the device will not be added to dpm_list, although in fact the device is going to be added to the list regardless of the ordering violation. Remove the WARN_ON(true) triggered in that situation, because it is hit by USB very often and spams the users' logs. This patch fixes bug #11263 Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Greg Kroah-Har...

From: Alan Stern <stern@rowland.harvard.edu> This patch (as1124) fixes a couple of bugs in the PM core. The new dev->power.status field should be initialized regardless of whether CONFIG_PM_SLEEP is enabled, and similarly dpm_sysfs_add() should be called whenever CONFIG_PM is enabled. The patch separates out the call to dpm_sysfs_add() from the call to device_pm_add(). As a result device_pm_add() can no longer return an error, so its return type is changed to void. Signed-off-by: Al...

Or -- for future reference -- using WARN() with non-constant initial parameter: if (WARN(!class-p, "%s called ...\n" ...)) return -EINVAL; Some of those macros now have return values, supportin C code that looks a bit more compact. - Dave --