Dear applicant, we have found a suitable job for you in the Gulf area which matches your interests and your qulifications, to apply kindly log on to http://www.middle-east-jobs.blogspot.com/ and submit your C.V as soon as possible Kind regards, Recruitment dept --

I don't think there is any such consensus, so I'm not expecting a technical explanation. As best as I can tell, the primary explanation for why so many A/V vendors are doing it seems to be that it's a damn effective business model, and that may have as much of an effect on its widespread use as any technical merits or demerits. Think about it: you get users to buy your A/V, then you tell them they need to pay a monthly subscription fee to get the latest virus signatures updates. It's like ...

I'd be OK implementing it with an option, rather than a magic char. I was trying to avoid adding too many options, since many kernel developers prefer fewer options where possible. But the magic char makes the code less straightforward. If we ever move towards supporting both prefix and suffice splicing (or even complicated in-the-middle splicing), then the magic char is easier to develop into that. But so far, I can only come up with reasonable cases for append and override, and I don't ...

Different arches handle their command lines differently, but with some elbow grease (and some buy-in from the different arch maintainers), we could try unifying the approach here. One difficulty is that the other arches' command lines are not currently "broken", so there's no real incentive to change them. The only thing novel thing I'm adding here is the addition of the leading '!' to allow for an override. This is needed in some x86 cases I'm familiar with, but I've haven't seen any ...

You're right, I had forgotten about the suffix splicing and my brain is Yes, though I doubt we're in danger of introducing any real backwards compatibility issues with the magic '!' at the beginning. -- Mathematics is the supreme nostalgia of our time. --

Add support for a built-in command line for x86 architectures. The Kconfig help gives the major rationale for this addition. Note that this option is available for many other arches, and its use is widespread in embedded projects. This patch addresses concerns that were raised with an earlier version, regarding precedence between the built-in command line and the command line provided by the boot loader. See the thread at http://lkml.org/lkml/2006/6/11/115 for details. The default ...

I think I agree, and the override option would make it easier to make generic. -hpa --

Nice solution. Where is this relative to early boot option checking? -- Mathematics is the supreme nostalgia of our time. --

This patch contains the scheduled removal of the obsolete SERIAL_COLDFIRE driver. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- Documentation/feature-removal-schedule.txt | 8 drivers/serial/Kconfig | 16 drivers/serial/Makefile | 1 drivers/serial/mcfserial.c | 1965 --------------------- drivers/serial/mcfserial.h | 74 5 files changed, 2 insertions(+), 2062 ...

ptrace_untrace() can now become static. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 22 Jul 2008 include/linux/ptrace.h | 1 - kernel/ptrace.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) 819e9b617ffe996a61a150598050973911acd7e7 diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h index c6f5f9d..fe0d5a4 100644 --- a/include/linux/ptrace.h +++ b/include/linux/ptrace.h @@ -94,7 +94,6 @@ extern void ...

This patch makes the needlessly global probe_serial_gsc() static. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 18 Jun 2008 e0053c1295ce33ec15d2788140070660e95a07f0 diff --git a/drivers/serial/8250_gsc.c b/drivers/serial/8250_gsc.c index 0416ad3..418b4fe 100644 --- a/drivers/serial/8250_gsc.c +++ b/drivers/serial/8250_gsc.c @@ -111,7 +111,7 @@ static struct parisc_driver serial_driver = { .probe = serial_init_chip, }; -int __init ...

This patch makes the following needlessly global code static: - iomap.c: struct iomap_ops[] - memcpy.c: pa_memcpy() Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 15 Jun 2008 arch/parisc/lib/iomap.c | 2 +- arch/parisc/lib/memcpy.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) 45e7cd0b2c678dda4f5ba4ac6d0ed628df9381bd diff --git a/arch/parisc/lib/iomap.c b/arch/parisc/lib/iomap.c index 9abed07..5069e8b 100644 --- ...

This patch contains the following cleanups: - make the following needlessly global code static: - entry.S: resume_userspace - process.c: pm_idle - process.c: default_idle() - smp.c: send_IPI_allbutself() - time.c: timer_interrupt() - time.c: struct irq0 - traps.c: set_eit_vector_entries() - traps.c: kstack_depth_to_print - traps.c: show_trace() - traps.c: die_lock - remove the following unused code: - head.S: startup_32 - process.c: hlt_counter - process.c: ...

This patch fixes the following build error: <-- snip --> ... Building modules, stage 2. MODPOST 1083 modules ERROR: "empty_zero_page" [fs/ext4/ext4dev.ko] undefined! ... make[2]: *** [__modpost] Error 1 <-- snip --> Reported-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 11 Jun 2008 arch/m32r/kernel/m32r_ksyms.c | 2 ++ 1 file changed, 2 insertions(+) ae7a1cecb2071e861cbcd6ee362ed93dd43793d5 ...

a2091_{detect,release}() can become static. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 10 Jun 2008 drivers/scsi/a2091.c | 6 ++++-- drivers/scsi/a2091.h | 3 --- 2 files changed, 4 insertions(+), 5 deletions(-) 22558d3b8c3dcf29bb54d76256a64db89017dedc diff --git a/drivers/scsi/a2091.c b/drivers/scsi/a2091.c index 07d572f..f36029c 100644 --- a/drivers/scsi/a2091.c +++ b/drivers/scsi/a2091.c @@ -23,6 +23,8 @@ #define DMA(ptr) ...

Commit 0f475b2abed6cbccee1da20a0bef2895eb2a0edd (ocfs2/net: Silence build warnings) made sense as far as it fixed compile warnings, but it was not required that it made the functions global. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 5 Jun 2008 fs/ocfs2/cluster/tcp.c | 44 ++++++++++++++++++++++++++------ fs/ocfs2/cluster/tcp_internal.h | 32 ----------------------- 2 files changed, 37 insertions(+), 39 ...

- maintainer has not been active for years - maintainer email bounces - URL no longer exists - covered by the FRAMEBUFFER LAYER entry Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 20 May 2008 fc7a740aef47772e5417c71f52aa9726eb5c716d diff --git a/MAINTAINERS b/MAINTAINERS index 2f74092..af69cab 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1822,13 +1822,6 @@ M: zippel@linux-m68k.org L: linux-kernel@vger.kernel.org S: Maintained -HGA ...

This patch removes a dead URL. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 20 May 2008 deb4137671451a172bc7b9ed62cae9fbfd1decba diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 729cdbd..97d58e8 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -224,8 +224,7 @@ config ALPHA_MIATA bool "Miata" help The Digital PersonalWorkStation (PWS 433a, 433au, 500a, 500au, 600a, - or 600au). There is an Installation HOWTO for this ...

This patch removes a CVS keyword that wasn't updated for a long time from a comment. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 20 May 2008 ecbe2e714c6f467a94d2240223dfa335cdfb501e diff --git a/kernel/dma.c b/kernel/dma.c index d2c60a8..f903189 100644 --- a/kernel/dma.c +++ b/kernel/dma.c @@ -1,4 +1,4 @@ -/* $Id: dma.c,v 1.7 1994/12/28 03:35:33 root Exp root $ +/* * linux/kernel/dma.c: A DMA channel allocator. Inspired by linux/kernel/irq.c. ...

This patch removes CVS keywords that weren't updated for a long time from comments. Signed-off-by: Adrian Bunk <bunk@kernel.org> Acked-by: Jan Kara <jack@suse.cz> --- This patch has been sent on: - 20 May 2008 fs/dquot.c | 2 -- include/linux/quota.h | 2 -- include/linux/quotaops.h | 3 --- 3 files changed, 7 deletions(-) f2096053eb97000dbd556a4a2465c879bdf38670 diff --git a/fs/dquot.c b/fs/dquot.c index 5ac77da..920e430 100644 --- a/fs/dquot.c +++ ...

This patch removes CVS keywords that weren't updated for a long time from comments. Signed-off-by: Adrian Bunk <bunk@kernel.org> --- This patch has been sent on: - 20 May 2008 fs/reiserfs/procfs.c | 3 --- 1 file changed, 3 deletions(-) a41d213debad92b6672c39b577baf58641b5c54c diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index b9dbeec..37173fa 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -8,8 +8,6 @@ /* proc info support a la one created by ...

There's no point in printing some ancient version number forever. Signed-off-by: Adrian Bunk <bunk@kernel.org> Acked-by; Rene Herman <rene.herman@gmail.com> Acked-by; Bjorn Helgaas <bjorn.helgaas@hp.com> Acked-by; Adam M Belay <abelay@MIT.EDU> --- This patch has been sent on: - 20 May 2008 acef517cfbe3cd741f8884311ef6fc2dfbeb1a4d diff --git a/drivers/pnp/core.c b/drivers/pnp/core.c index 20771b7..d924798 100644 --- a/drivers/pnp/core.c +++ b/drivers/pnp/core.c @@ -209,7 +209,6 @@ ...

This patch makes the needlessly global anon_vma_cachep static. Signed-off-by: Adrian Bunk <bunk@kernel.org> Acked-by: Rik van Riel <riel@redhat.com> --- This patch has been sent on: - 5 May 2008 - 22 Apr 2008 - 14 Apr 2008 - 31 Mar 2008 - 25 Feb 2008 include/linux/rmap.h | 12 ------------ mm/rmap.c | 12 +++++++++++- 2 files changed, 11 insertions(+), 13 deletions(-) 8d2d63d71ecbf2db78f31025acb136c9dd4db6a2 diff --git a/include/linux/rmap.h ...

Hello, I want to be able to convert starttime from /proc/#/stat to absolute time and it seems that I need HZ to do that as it is measured in jiffies. How can I get HZ? I have seen some discussion about this but haven't found a definite answer. procps is using some hacks do it and I would prefer to avoid them. I only need a solution for modern kernels, say 2.6.20+ if that matters. -- Best Regards, Piotr Jaroszyński

On Wed, Aug 6, 2008 at 1:48 PM, Piotr Jaroszyński `man sysconf`, in particular you want sysconf(_SC_CLK_TCK).

You can get the information via two ways. At compile time, that is the=20 variable CLK_TCK and this way has been obsoleted. The other way is at=20 runtime via sysconf(_SC_CLK_TCK) and that is still valid. You don't need=20 the real HZ, because the kernel converts it to USER_HZ, which is currently= =20 100 on x86 for historic reasons. Sven

Here's the latest proc(5) man page from June 25, 2008. Look for _SC_CLK_TCK about a third of the way down. http://www.kernel.org/doc/man-pages/online/pages/man5/proc.5.html

On Wed, 6 Aug 2008 16:40:02 -0400 (EDT) What the VFS will do is - lock the page - put the page into a BIO and send it down to the block layer - later, wait for IO completion. It does this by running lock_page[_killable](), which will waiting for the page to come unlocked. The page comes unlocked via the device driver, usually within the IO completion interrupt. A common cause of userspace lockups during IO errors is that the driver layer screwed up and didn't run the ...

This bug is certainly present in 2.6.27-rc2, and it may be present in several earlier kernel versions too, perhaps as far back as 2.6.25. What happens is that in the presence of certain I/O errors, user processes can hang. Here is a procedure for recreating the bug: Make sure that CONFIG_USB_GADGET and CONFIG_USB_GADGET_DUMMY_HCD are enabled as modules, as well as CONFIG_USB_FILE_STORAGE. Apply the patch below and build the drivers. Then create a blank image file with 1 MB of zeros and load ...

It's expected behaviour, but it is indeed misleading. Here's the reason why it happens: In the kernel we're accounting time based on CONFIG_HZ (which I suspect is 1000 in your case) but are exporting values based on USER_HZ (100, historic reasons) to userspace. So we're effectively dividing the values by 10. Well, that division obviously leaves a remainder in most cases, which is dropped. You see in the code that for the summary we first add all in-kernel values up and then do the ...

I meant on top of the list_splice_tail() of course :) Luis --

Hi Linus. Please pull following kbuild related fixes. o Fix "make O=..." builds for arm o update to for newer cscope o patch-kernel fix (but I doubt it has many uses) o small Kconfig clean-up Sam The following changes since commit 0967d61ea0d8e8a7826bd8949cd93dd1e829ac55: Linus Torvalds (1): Linux 2.6.27-rc2 are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild-fixes.git master Denis ChengRq (1): kbuild: a better ...

commit 9e98966c7bb94355689478bc84cc3e0c190f977e (tty: rework break handling) forgot to update one exit point of rs_break() in the Amiga serial driver. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> --- drivers/char/amiserial.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/drivers/char/amiserial.c +++ b/drivers/char/amiserial.c @@ -1254,7 +1254,7 @@ static int rs_break(struct tty_struct *t unsigned long flags; if (serial_paranoia_check(info, tty->name, ...

I just realized that I managed to screw up my From: email address in git-send-email command line. Please reply to maxk@qualcomm.com. Max --

The warn-when-statically-allocated-kobjects-are-used patch crashes my machine during boot when it calls kzalloc(), presumably because the call occurs too early. This patch (as1126) fixes it by using a static string instead of a dynamically-allocated string. Signed-off-by: Alan Stern <stern@rowland.harvard.edu> --- Index: usb-2.6/lib/kobject.c =================================================================== --- usb-2.6.orig/lib/kobject.c +++ usb-2.6/lib/kobject.c @@ -38,10 +38,9 @@ ...

It seems to be impossible to mount packet-written CD-RWs writably in 2.6.26.x, even as root. Things worked in 2.6.25.x. hades:/tmp# ls -l /dev/pktcdvd total 0 brw-rw-rw- 1 root root 254, 0 2008-08-05 00:45 cdrw crw-r--r-- 1 root root 10, 63 2008-08-05 00:44 control brw-rw-rw- 1 root cdrom 254, 0 2008-08-05 00:45 pktcdvd0 hades:/tmp# mount -o rw /dev/pktcdvd/cdrw /mnt/pcdrw mount: block device /dev/pktcdvd/cdrw is write-protected, mounting read-only hades:/tmp# mount -o rw ...

Hello, Fsync batching in ext3 is somewhat flawed when it comes to disks that are very fast. Now we do an unconditional sleep for 1 second, which is great on slow disks like SATA and such, but on fast disks this means just sitting around and waiting for nothing. This patch measures the time it takes to commit a transaction to the disk, and sleeps based on the speed of the underlying disk. Using the following fs_mark command to test the speeds ./fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 ...

Linus, please pull from the for-linus branch at git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6.git for-linus to receive a FireWire subsystem fix-for-a-fix. David Moore (1): firewire: Preserve response data alignment bug when it is harmless drivers/firewire/fw-cdev.c | 29 ++++++++++++++++++++--------- 1 files changed, 20 insertions(+), 9 deletions(-) commit 8401d92ba46a1e859464cbd9c9ee304f6e361da3 Author: David Moore <dcm@acm.org> Date: Tue Jul ...

BTW, in one of your oops, I see: note: cron[1207] exited with preempt_count 268435459 I smell some kind of stack corruption here which is corrupting thread_info (in the above case preempt_count in the thread_info). Similarly, if the status field(in thread_info) gets corrupted(setting TS_USEDFPU) without proper math state allocated(present in thread_struct), we can end up oops in __switch_to. But you seem to say, reverting recent fpu patches make the problem go away. hmm, just wondering ...

We (the -stable team) are announcing the release of the 2.6.26.2 kernel. It contains lots of bugfixes, all over the map. Any users of the 2.6.26 kernel series should upgrade to this version. For details on the fixes, see the changelog entries and the diffstat below. I'll also be replying to this message with a copy of the patch between 2.6.26.1 and 2.6.26.2 The updated 2.6.26.y git tree can be found at: git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-2.6.26.y.git and ...

We (the -stable team) are announcing the release of the 2.6.25.15 kernel. It contains lots of bugfixes, all over the map. Any users of the 2.6.25 kernel series should upgrade to this version. For details on the fixes, see the changelog entries and the diffstat below. I'll also be replying to this message with a copy of the patch between 2.6.25.14 and 2.6.25.15 The updated 2.6.25.y git tree can be found at: ...

Makes sense. I'll add it, since I'm touching that area anyway. Max --

This patch makes sure that , if kmalloc fails in "rebuild_sched_domains", then we've a place to go. If anything else, please let me know. Thanks. Signed-off-by: Md.Rakib H. Mullick (rakib.mullick@gmail.com) --- linux-2.6.27-rc2.orig/kernel/cpuset.c 2008-08-06 16:23:26.000000000 +0600 +++ linux-2.6.27-rc2/kernel/cpuset.c 2008-08-06 23:07:59.504441992 +0600 @@ -615,7 +615,8 @@ void rebuild_sched_domains(void) if (dattr) { *dattr = SD_ATTR_INIT; update_domain_attr_tree(dattr, ...

--Boundary-01=_UzdmI9aKqTUd3wx Content-Type: text/plain; charset="iso-8859-15" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hello, the build (.config attached) failed, make ends with : =2E.. CC arch/x86/lib/usercopy_32.o AR arch/x86/lib/lib.a LD vmlinux.o MODPOST vmlinux.o WARNING: modpost: Found 12 section mismatch(es). To see full details build your kernel with: 'make CONFIG_DEBUG_SECTION_MISMATCH=3Dy' GEN .version CHK ...

Looks good so far, even slimmer than the dynamic alloc ones i'd done. Will test with ide-floppy on the weekend. -- Regards/Gruss, Boris. --

On Wed, Aug 6, 2008 at 7:14 PM, Bartlomiej Zolnierkiewicz <bzolnier@gmail.com> wrote: Please disregard above comments -- I found a better way to deal with rq stack. --

I don't know but I will see about getting it in there. --

In principle it's easy. We append a 3-byte nop to the lock-taking instructions. We invent an AMD_MUTEX_BUG synthetic cpuid feature bit and add boot-time code to detect it. We use the alternatives() infrastructure to replace that nop with lfence at boot-time if AMD_MUTEX_BUG is present. I think the hardest part is locating all lock-taking code sequences. Also I think I'll start by writing a user-space test program that does a stress-test of the plain lock;rmw;unlobk sequence to see ...

Cc: Borislav Petkov <petkovbb@gmail.com> Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com> --- drivers/ide/ide-floppy.c | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) Index: b/drivers/ide/ide-floppy.c =================================================================== --- a/drivers/ide/ide-floppy.c +++ b/drivers/ide/ide-floppy.c @@ -738,7 +738,7 @@ static int idefloppy_get_sfrp_bit(ide_dr return 1; floppy->srfp = pc.buf[8 + 2] & ...

From: Krzysztof Helt <krzysztof.h1@wp.pl> The correct id is the id of the main host (5591) not the id of the PCI-to-PCI bridge AGP (0001). Output from "lspci -nv" shows that only the former has AGP capabilities flag set: 00:00.0 0600: 1039:5591 (rev 02) Flags: bus master, medium devsel, latency 64 Memory at ec000000 (32-bit, non-prefetchable) [size=32M] Capabilities: [c0] AGP version 1.0 00:02.0 0604: 1039:0001 (prog-if 00 [Normal decode]) Flags: bus ...

expr_isvalid() in conditional.c was off-by-one and allowed invalid expression type COND_LAST. However, it is this header file that needs to be fixed. That way the if-statement's disjunction's second component reads more naturally, "if expr type is greater than the last allowed value" ( rather than using ">=" in conditional.c): if (expr->expr_type <= 0 || expr->expr_type > COND_LAST) Signed-off-by: Vesa-Matti Kari <vmkari@cc.helsinki.fi> --- security/selinux/ss/conditional.h | 2 +- 1 ...

Oops, wrong list. Sorry. --

Some devices emit a ACPI_NOTIFY_DEVICE_CHECK while physically unplugging even if the software undock has already been done and dock_present() check fails. However, the internal flags need to be cleared (complete_undock()). Also, even notify userspace if the dock station suddently went away without proper software undocking. Signed-off-by: Holger Macht <hmacht@suse.de> --- diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index bb7c51f..7d2edf1 100644 --- a/drivers/acpi/dock.c +++ ...

Rename is_single_threaded() to is_wq_single_threaded() so that a new is_single_threaded() can be created that refers to tasks rather than waitqueues. Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: James Morris <jmorris@namei.org> --- kernel/workqueue.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 4a26a13..e4e2bd3 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -83,21 +83,21 ...

Document credentials and the new credentials API. Signed-off-by: David Howells <dhowells@redhat.com> --- Documentation/credentials.txt | 563 +++++++++++++++++++++++++++++++++++++++++ 1 files changed, 563 insertions(+), 0 deletions(-) create mode 100644 Documentation/credentials.txt diff --git a/Documentation/credentials.txt b/Documentation/credentials.txt new file mode 100644 index 0000000..3caf1ea --- /dev/null +++ b/Documentation/credentials.txt @@ -0,0 +1,563 @@ + ...

Attach creds to file structs and discard f_uid/f_gid. file_operations::open() methods (such as hppfs_open()) should use file->f_cred rather than current_cred(). At the moment file->f_cred will be current_cred() at this point. Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: James Morris <jmorris@namei.org> --- arch/mips/kernel/vpe.c | 4 ++-- drivers/isdn/hysdn/hysdn_procconf.c | 6 ++++-- fs/coda/file.c | 2 +- fs/file_table.c ...

Make inode_has_perm() and file_has_perm() take a cred pointer rather than a task pointer. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: James Morris <jmorris@namei.org> Acked-by: Serge Hallyn <serue@us.ibm.com> --- security/selinux/hooks.c | 140 ++++++++++++++++++++++++++++++---------------- 1 files changed, 92 insertions(+), 48 deletions(-) diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 27d1779..fb76940 100644 --- ...

Separate the task security context from task_struct. At this point, the security data is temporarily embedded in the task_struct with two pointers pointing to it. Note that the Alpha arch is altered as it refers to (E)UID and (E)GID in entry.S via asm-offsets. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: James Morris <jmorris@namei.org> Acked-by: Serge Hallyn <serue@us.ibm.com> --- arch/alpha/kernel/asm-offsets.c | 11 +- arch/alpha/kernel/entry.S | 10 + ...

Disperse the bits of linux/key_ui.h as the reason they were put here (keyfs) didn't get in. Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: James Morris <jmorris@namei.org> --- include/keys/keyring-type.h | 31 ++++++++++++++++++++ include/linux/key-ui.h | 66 ------------------------------------------- security/keys/internal.h | 31 ++++++++++++++++++++ security/keys/keyring.c | 1 + security/keys/request_key.c | 2 + 5 files changed, 64 ...

Use RCU to access another task's creds and to release a task's own creds. This means that it will be possible for the credentials of a task to be replaced without another task (a) requiring a full lock to read them, and (b) seeing deallocated memory. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: James Morris <jmorris@namei.org> Acked-by: Serge Hallyn <serue@us.ibm.com> --- drivers/connector/cn_proc.c | 16 +++++++---- fs/binfmt_elf.c | 8 ++++- ...

Pass credentials through dentry_open() so that the COW creds patch can have SELinux's flush_unauthorized_files() pass the appropriate creds back to itself when it opens its null chardev. The security_dentry_open() call also now takes a creds pointer, as does the dentry_open hook in struct security_operations. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: James Morris <jmorris@namei.org> --- fs/ecryptfs/ecryptfs_kernel.h | 3 ++- fs/ecryptfs/kthread.c | 9 ...

Change most current->e?[ug]id to current_e?[ug]id(). Change some task->e?[ug]id to task_e?[ug]id(). In some places it makes more sense to use RCU directly rather than a convenient wrapper; these are addressed by a later patch (see the patch entitled "CRED: Use RCU to access another task's creds and to release a task's own creds"). Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: James Morris <jmorris@namei.org> Acked-by: Serge Hallyn <serue@us.ibm.com> --- ...

Make execve() take advantage of copy-on-write credentials, allowing it to set up the credentials in advance, and then commit the whole lot after the point of no return. This patch and the preceding patches have been tested with the LTP SELinux testsuite. This patch makes several logical sets of alteration: (1) execve(). The credential bits from struct linux_binprm are, for the most part, replaced with a single credentials pointer (bprm->cred). This means that all the ...

I've brought my patchset up to date with regards the recent merge melee and built the patches on top of the next branch of James's security testing tree as per his request. A tarball of these patches can be retrieved from: http://people.redhat.com/~dhowells/cow-creds-7.tar.bz2 I've been testing these patches with the LTP syscalls and SELinux test scripts. --- There are three parts to this project: (1) Implement COW credentials. (2) Pass the cred pointer through the vfs_xxx() ...

Change current->fs[ug]id to current_fs[ug]id() so that fsgid and fsuid can be separated from the task_struct. Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: James Morris <jmorris@namei.org> Acked-by: Serge Hallyn <serue@us.ibm.com> --- arch/ia64/kernel/perfmon.c | 4 ++-- arch/powerpc/platforms/cell/spufs/inode.c | 4 ++-- drivers/isdn/capi/capifs.c | 4 ++-- drivers/usb/core/inode.c | 4 ++-- fs/9p/fid.c ...

That'd be ok. Somehow I think it's not worth the effort though. If the machine was able to boot to the point that it can run user-space (request_firmware) with the old microcode it can probably sustain itself a bit longer while waiting for the workqueue thread to be scheduled. You probably have more exciting things to do ;-) The latest patches look ok btw. Max --

Referring to: http://lkml.org/lkml/2008/8/3/114 Dmitry, thanks for your work. Rebased Dmitry's patch set to x86-microcode and incorporated changes into AMD microcode patch loader. No issues with cpu-hotplug anymore, tested on Intel as well as AMD. --Peter --

Signed-off-by: Peter Oruba <peter.oruba@amd.com> --- arch/x86/kernel/microcode_amd.c | 106 ++++++++++++++++++++++++-------------- 1 files changed, 67 insertions(+), 39 deletions(-) diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index 33b2a21..37d26ec 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -14,6 +14,7 @@ * License version 2. See file COPYING for details. */ +/* #define DEBUG */ /* pr_debug */ #include ...

More details are available here. I've also suggested to do ucode-updates as early as possible from start_secondary() (or whatever low-level code). The second patch (do via workqueue) was just a proof-of-concept (to Currently, it runs from cpu-hotplug-notifier(CPU_ONLINE, ...) [ and crashes with .26+ due to a reason you may find via the aforementioned link ] - by this moment, at least kernel threads may I'm not sure what's "schedulable context" here terms but the existing code makes ...

forgot a link: http://lkml.org/lkml/2008/8/3/114 -- Best regards, Dmitry Adamushko --

On Wed, 06 Aug 2008 13:31:18 -0700 you're not going to like the answer, but it's "as soon as possible". Unlike normal boot, hotplug is a case where the bios hasn't been able -- If you want to reach me at my work email, use arjan@linux.intel.com For development, discussion and tips for power savings, visit http://www.lesswatts.org --

Signed-off-by: Peter Oruba <peter.oruba@amd.com> --- include/asm-x86/microcode.h | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h index 072bef1..fb08022 100644 --- a/include/asm-x86/microcode.h +++ b/include/asm-x86/microcode.h @@ -1,5 +1,5 @@ -extern int microcode_init(void *opaque, struct module *module); -extern void microcode_exit(void); +int microcode_init(void *opaque, struct module *module); +void ...

I don't use the webcam on my EeePC; I just happened to try it while testing 2.6.27-rc2. I don't know about any other version so this isn't neccessarily a regression. I think I tested it using Kopete, then later I got this warning on resume. I definitely didn't have Kopete open when I suspended though. Thanks Alan

Hi, running the libhugetlbfs test suite I met the following bug: BUG: sleeping function called from invalid context at include/linux/pagemap.h:294 in_atomic():1, irqs_disabled():0 CPU: 0 Not tainted 2.6.27-rc1 #3 Process private (pid: 4531, task: 000000003f68e400, ksp: 000000002a7e3be8) 0700000033a00700 000000002a7e3bf0 0000000000000002 0000000000000000 000000002a7e3c90 000000002a7e3c08 000000002a7e3c08 0000000000016472 0000000000000000 000000002a7e3be8 0000000000000000 ...

Looks good Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com> -- Warm Regards, Balbir Singh Linux Technology Center IBM, ISTL --

hi, mtrr is wrong on my machine. the fixes of 2.6.27-rc1 do not seem to work. bios is the newest version, mainboard vendor says it's not their fault. mainboard is tyan i5000pw. please tell me how i can help and cc me on answers, i am not subscribed. here is some data: seaburg ~ # grep -E "(^\(WW\)|^\(EE\)|^\(NI\)|^\(??\))" /var/log/Xorg.0.log (WW) VESA(0): Failed to set up write-combining range (0xd8000000,0x2000000) seaburg ~ # dmesg|grep mtrr Command line: root=/dev/sda1 ...

Commit e0597d70012c82e16ee152270a55d89d8bf66693 (sd: Identify DIF protection type and application tag ownership) says that if a disk is formatted with Inquiry bit PROTECT=1, it is required to support Read Capacity(16). But my SD cards, accessed by builtin cardreader and generic USB storage, disagree. Therefore fall back to the familiar Read Capacity if Read Capacity(16) fails: without even showing the "failed" message since I expect this will be common. Signed-off-by: Hugh Dickins ...

How about we flip it around? Unconditionally try READ CAPACITY 16 first, then if that fails, try READ CAPACITY? I suppose there's always the possibility that a drive will go tits-up if it receives the RC16 command, so maybe we'll need a blacklist. -- Intel are signing my paycheques ... these opinions are still mine "Bill, look, we understand that you're interested in selling us this operating system, but compare it to ours. We can't possibly take such a retrograde step." --

>>>>> "Hugh" == Hugh Dickins <hugh@veritas.com> writes: Hugh> Commit e0597d70012c82e16ee152270a55d89d8bf66693 (sd: Identify Hugh> DIF protection type and application tag ownership) says that if Hugh> a disk is formatted with Inquiry bit PROTECT=1, it is required Hugh> to support Read Capacity(16). But my SD cards, accessed by Hugh> builtin cardreader and generic USB storage, disagree. Argh. That's really broken. USB storage has no business setting PROTECT. Can you send me the sg_inq ...

OK, so this is the solution until we get USB devices claiming SCSI-3 and mucking with the protection bit ... James --

>>>>> "Hugh" == Hugh Dickins <hugh@veritas.com> writes: Hugh> Almost: I had to #include <scsi/scsi.h> to get SCSI_2 for the Hugh> build somewhere (and a SCSI_2 seems to be worth 1 more than Hugh> anyone else's 2), then it worked just fine for me thanks. The DIF drives in question claim rev. 3 and are configured correctly with James' patch in place. Acked-by: Martin K. Petersen <martin.petersen@oracle.com> -- Martin K. Petersen Oracle Linux Engineering --

sg_inq? I assume that's the same as the inquiry_len bytes at inquiry: 00000000: 00 8d 00 01 1f 01 00 00 47 65 6e 65 72 69 63 2d ........Generic- 00000010: 4d 75 6c 74 69 2d 43 61 72 64 20 20 20 20 20 20 Multi-Card 00000020: 31 2e 30 30 1.00 Ah, sg_inq is a command which spells that out, here you go: standard INQUIRY: PQual=0 Device_type=0 RMB=1 version=0x00 [no conformance claimed] [AERC=0] [TrmTsk=0] NormACA=0 HiSUP=0 ...

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com> --- arch/x86/mm/pat.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) Recently I've run a kernel w/o PAT support and wondered why there was a file "x86/pat_memtype_list" in debugfs. Of course it's empty if PAT is disabled ... so just get rid of this interface if PAT is disabled. Please apply. (patch against v2.6.27-rc2) Andreas diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 2fe3091..647b1c4 ...

use correct dynamic spinlock init function. Signed-off-by: Karsten Keil <kkeil@suse.de> --- drivers/isdn/mISDN/tei.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/isdn/mISDN/tei.c b/drivers/isdn/mISDN/tei.c index 6fbae42..d8af45c 100644 --- a/drivers/isdn/mISDN/tei.c +++ b/drivers/isdn/mISDN/tei.c @@ -1287,7 +1287,7 @@ create_teimanager(struct mISDNdevice *dev) if (!mgr) return -ENOMEM; INIT_LIST_HEAD(&mgr->layer2); - mgr->lock = ...

found by Randy Dunlap. Signed-off-by: Karsten Keil <kkeil@suse.de> --- Documentation/isdn/README.mISDN | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/Documentation/isdn/README.mISDN b/Documentation/isdn/README.mISDN index cd8bf92..5271e66 100644 --- a/Documentation/isdn/README.mISDN +++ b/Documentation/isdn/README.mISDN @@ -1,6 +1,6 @@ -mISDN is a new modular ISDN driver, in the long term it should replace -the old I4L driver architecture for passiv ISDN ...

irq_radix_revmap() currently serves 2 purposes, irq mapping lookup and insertion which happen in interrupt and process context respectively. Separate the function into its 2 components, one for lookup only and one for insertion only. Fix the only user of the revmap tree (XICS) to use the new functions. Also, move the insertion into the radix tree of those irqs that were requested before it was initialized at said tree initialization. Mutual exclusion between the tree ...

Please pull from git://git.kernel.org/pub/scm/linux/kernel/git/kkeil/ISDN-2.6.git master Remove references to external zaptel modules and some cosmetic cleanups. Signed-off-by: Karsten Keil <kkeil@suse.de> --- drivers/isdn/hardware/mISDN/hfc_multi.h | 4 - drivers/isdn/hardware/mISDN/hfcmulti.c | 109 +++++++------------------------ 2 files changed, 24 insertions(+), 89 deletions(-) diff --git a/drivers/isdn/hardware/mISDN/hfc_multi.h b/drivers/isdn/hardware/mISDN/hfc_multi.h index ...

This patch serie, against 2.6.27-rc2, adds support for the PXA-255 based Arcom/Eurotech Viper SBC. It is roughly based on the old 2.6.16.28 port made by Arcom, though all bugs are of course mine. It's been heavily tested over the last two months. Most of these patches are quite straightforward, except for the generic PCMCIA patch that may be questionable. As usual, I'll be glad to receive any comment you may have about this patch. Thanks, M. arch/arm/configs/viper_defconfig | ...

Signed-off-by: Marc Zyngier <marc.zyngier@altran.com> --- drivers/pcmcia/Kconfig | 2 +- drivers/pcmcia/Makefile | 1 + drivers/pcmcia/pxa2xx_viper.c | 183 +++++++++++++++++++++++++++++++++++++++++ 3 files changed, 185 insertions(+), 1 deletions(-) create mode 100644 drivers/pcmcia/pxa2xx_viper.c diff --git a/drivers/pcmcia/Kconfig b/drivers/pcmcia/Kconfig index e0f8840..de57e7e 100644 --- a/drivers/pcmcia/Kconfig +++ b/drivers/pcmcia/Kconfig @@ -220,7 +220,7 @@ ...

Signed-off-by: Marc Zyngier <marc.zyngier@altran.com> --- arch/arm/mach-pxa/Kconfig | 6 + arch/arm/mach-pxa/Makefile | 1 + arch/arm/mach-pxa/viper.c | 952 ++++++++++++++++++++++++++++++++++++++ include/asm-arm/arch-pxa/irqs.h | 23 + include/asm-arm/arch-pxa/viper.h | 91 ++++ 5 files changed, 1073 insertions(+), 0 deletions(-) create mode 100644 arch/arm/mach-pxa/viper.c create mode 100644 include/asm-arm/arch-pxa/viper.h diff --git ...

Hi, I tried to upgrade several of our HP DL385 running a 64 bit kernel to 2.6.26.1 from 2.6.25.11. Every single one of them died within at most a few days, sometimes even within minutes. The systems continue to do TCP/IP/pings/etc, but everything that interacts with the filesystem/disks just hangs. It seems to happen faster with more disk/filesystem load. Nothing is printed to the console. The kernel config of the working 2.6.25.11 and of several 2.6.26.1 kernels, none of wich worked ...

Hello everyone, the following patch removes the use of int cpus_nonempty variable from 'update_flag' function. Thanks. Signed-off-by: Md.Rakib H. Mullick (rakib.mullick@gmail.com) --- linux-2.6.27-rc2.orig/kernel/cpuset.c 2008-08-06 16:23:26.000000000 +0600 +++ linux-2.6.27-rc2/kernel/cpuset.c 2008-08-06 18:03:50.000000000 +0600 @@ -1110,7 +1110,7 @@ static int update_flag(cpuset_flagbits_t { struct cpuset trialcs; int err; - int cpus_nonempty, balance_flag_changed; + int ...

I can easily reproduce this with modprobe rcutorture; sleep 10s; rmmod rcutorture; cat /proc/lockdep_chains > /dev/null Greetings, Eric --

Whoops, sorry ... I see what was going on. My bad. --- g26.orig/drivers/mtd/devices/mtd_dataflash.c 2008-08-06 04:12:45.000000000 -0700 +++ g26/drivers/mtd/devices/mtd_dataflash.c 2008-08-06 04:11:55.000000000 -0700 @@ -616,7 +616,7 @@ static char *otp_setup(struct mtd_info * #else -static char *otp_setup(struct mtd_info *device) +static char *otp_setup(struct mtd_info *device, char revision) { return " (OTP)"; } --

On Wed, Aug 6, 2008 at 5:42 PM, Jeff Chua <jeff.chua.linux@gmail.com> It works. Booted with 16CPUs. 32GB RAM. CPU0 L7345 1.86GHz 0C CPU1 L7345 1.86GHz 0C CPU2 L7345 1.86GHz 0C CPU3 L7345 1.86GHz 0C CPU4 L7345 1.86GHz 0C CPU5 L7345 1.86GHz 0C CPU6 L7345 1.86GHz 0C CPU7 L7345 1.86GHz 0C CPU8 L7345 1.86GHz 0C CPU9 L7345 1.86GHz 0C CPU10 L7345 1.86GHz 0C CPU11 L7345 1.86GHz 0C CPU12 L7345 1.86GHz 0C CPU13 L7345 1.86GHz 0C CPU14 L7345 1.86GHz 0C CPU15 L7345 1.86GHz ...

should be CC Bryan Wu as well ... -mike --

Hi, Hm? I'm sorry, I don't understand what you mean by that. Do you think Oh right. I believed that Bryan Wu is the maintainer who would eventually apply the patch, therefore I put him as a direct recipient of the e-mail (To-field) and that he would add his Signed-off-by line which would make the Cc-line redundant. I guess it makes some sense to also list the maintainer in the patch Ccs since the patch itself may be separate from the raw e-mail (e.g. in mail archives, which generally don't ...

(Use a wider diff context such as -U10 after the patch has been applied to see the actual uses of the variables which we change.) From 3ef36948a88a968eec1b09859aa251dc6727df4e Mon Sep 17 00:00:00 2001 From: Vegard Nossum <vegard.nossum@gmail.com> Date: Wed, 6 Aug 2008 12:00:23 +0200 Subject: [PATCH] blackfin/sram: use 'unsigned long' for irqflags Using just 'unsigned' will make flags an unsigned int. While this is arguably not an error on blackfin where sizeof(int) == sizeof(long), the patch ...

Hm. I agree that git/eclipse/emacs/etc. information is not very useful. However... For errors found with lockdep, we usually put either the lockdep output in the commit message or say that it was found with lockdep. Having this information in the log is useful because it also establishes a track record for the tool which was used to discover/fix the error. Arguably, the semantic patch itself should be present in the log as well. There are different practices here, but in this case, the ...

This is a tiny series of patches for bringing binfmt_elf_fdpic back in line with binfmt_elf in terms of auxiliary vector layout, particularly for the new entries that have popped up recently. [PATCH 1/3] binfmt_elf_fdpic: Support auxvec base platform string. [PATCH 2/3] binfmt_elf_fdpic: Convert initial stack alignment to arch_align_stack(). [PATCH 3/3] binfmt_elf_fdpic: Wire up AT_EXECFD, AT_EXECFN, AT_SECURE. Changes since the last version include the updated 2/3 patch and a cosmetic fixup ...

Commit 483fad1c3fa1060d7e6710e84a065ad514571739 introduces AT_BASE_PLATFORM, but only implements it for binfmt_elf. Given that AT_VECTOR_SIZE_BASE is unconditionally enlarged for us, and it's only optionally added in for the platforms that set ELF_BASE_PLATFORM, wire it up for binfmt_elf_fdpic, too. Signed-off-by: Paul Mundt <lethal@linux-sh.org> --- fs/binfmt_elf_fdpic.c | 50 ++++++++++++++++++++++++++++++++++++++++++------ 1 files changed, 43 insertions(+), 7 deletions(-) diff ...

Hi Steve, the following is one for you. The kernel is 2.6.26.1 plus the 2.6.26-rt1 patch (yes it applies against 2.6.26.1) and some of the patches that were discussed in the lists. I don't think it is relevant which patches, but if you need an exact list I will provide it. WARNING: at kernel/trace/trace_hist.c:396 tracing_hist_preempt_stop+0x23d/0x301() Modules linked in: Pid: 1, comm: init Not tainted 2.6.26.1-rt1.jk #4 Call Trace: [<ffffffff802397f4>] warn_on_slowpath+0x5d/0x99 ...

(CCing LKML) Hi Jens, Rusty, Trying to implement i/o tracking all the way up to the page cache (so that cfq and the future cgroup-based I/O controllers can schedule buffered I/O properly) I noticed that struct request's ioprio is initialized but never used for I/O scheduling purposes. Indeed there seems to be one single user of this member: virtio_blk. Virtio uses struct request's ioprio in the request() function of the virtio block driver, which just copies the ioprio value to the output ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- drivers/char/random.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/drivers/char/random.c b/drivers/char/random.c index fdbdcfc..872669e 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -558,7 +558,13 @@ struct timer_rand_state { }; static struct timer_rand_state input_timer_state; + +#ifdef CONFIG_HAVE_DYN_ARRAY +static struct timer_rand_state ...

to prepare to use dyn_array support etc. Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/x86/mm/init_32.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 4974e97..b5c1751 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -65,6 +65,7 @@ static unsigned long __meminitdata table_end; static unsigned long __meminitdata table_top; static int __initdata after_init_bootmem; +int ...

use small array with index to handle irq locking for serial port hope 32 slot is enough v2: according to Eric, move irq_no into irq_info, and not clean irq_no Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> --- drivers/serial/8250.c | 45 +++++++++++++++++++++++++++++++++++++++++---- 1 files changed, 41 insertions(+), 4 deletions(-) diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index 3920324..595b956 100644 --- ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- drivers/char/hpet.c | 2 +- drivers/char/random.c | 4 ++-- drivers/char/vr41xx_giu.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c index 2908a0e..8d562ae 100644 --- a/drivers/char/hpet.c +++ b/drivers/char/hpet.c @@ -219,7 +219,7 @@ static void hpet_timer_set_irq(struct hpet_dev *devp) for (irq = find_first_bit(&v, HPET_MAX_IRQ); irq < ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/x86/kernel/io_apic_32.c | 54 +++++++++++++++++++++++++++++++++-------- arch/x86/kernel/io_apic_64.c | 28 ++++++++++++++++----- arch/x86/kernel/setup.c | 6 ++++ 3 files changed, 70 insertions(+), 18 deletions(-) diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c index d5953ff..a0ce64c 100644 --- a/arch/x86/kernel/io_apic_32.c +++ b/arch/x86/kernel/io_apic_32.c @@ -70,7 +70,7 @@ int ...

could have crazy big arrays and allocate them in bootmem at init stage. also also to allocate array according to size we need to use to avoid wasting memory use CONFIG_HAVE_DYN_ARRAY to enable it or not usage: |static struct irq_desc irq_desc_init __initdata = { | .status = IRQ_DISABLED, | .chip = &no_irq_chip, | .handle_irq = handle_bad_irq, | .depth = 1, | .lock = __SPIN_LOCK_UNLOCKED(irq_desc->lock), |#ifdef CONFIG_SMP | .affinity = ...

and at this point it is equal NR_IRQS Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- include/linux/irq.h | 2 ++ kernel/irq/autoprobe.c | 10 +++++----- kernel/irq/chip.c | 20 ++++++++++---------- kernel/irq/handle.c | 3 ++- kernel/irq/manage.c | 16 ++++++++-------- kernel/irq/proc.c | 2 +- kernel/irq/resend.c | 4 ++-- kernel/irq/spurious.c | 4 ++-- 8 files changed, 32 insertions(+), 29 deletions(-) diff --git ...

so don't all get_irq_desc at begining to allocate all. and only call that when needed /proc/interrupts related still use old ways... for compatablilty reason. Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/x86/kernel/io_apic_64.c | 30 +++++++++++++++--------------- arch/x86/kernel/irq_64.c | 5 ++--- arch/x86/kernel/irqinit_64.c | 17 +++++------------ fs/proc/proc_misc.c | 7 +++++-- include/linux/irq.h | 18 ++++++++++++++++++ ...

preallocate size is 32, and if is not enough, get_irq_cfg will more with alloc_bootmem or kzalloc Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/x86/kernel/io_apic_64.c | 182 ++++++++++++++++++++++++++++++++--------- 1 files changed, 142 insertions(+), 40 deletions(-) diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 3734216..8908555 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -57,7 +57,11 @@ #define ...

preallocate 32 irq_2_pin, and use get_one_free_irq_2_pin to get one and link to irq_cfg if needed. so don't waste one for no irq is enabled. Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/x86/kernel/io_apic_64.c | 162 +++++++++++++++++++++++++++++++----------- 1 files changed, 121 insertions(+), 41 deletions(-) diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index 8908555..c8e4613 100644 --- a/arch/x86/kernel/io_apic_64.c +++ ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- include/linux/kernel_stat.h | 4 ++++ kernel/sched.c | 5 ++++- 2 files changed, 8 insertions(+), 1 deletions(-) diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h index cf9f40a..fe1f7fe 100644 --- a/include/linux/kernel_stat.h +++ b/include/linux/kernel_stat.h @@ -28,7 +28,11 @@ struct cpu_usage_stat { struct kernel_stat { struct cpu_usage_stat cpustat; +#ifdef ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/Kconfig | 2 ++ arch/x86/Kconfig | 1 + arch/x86/kernel/vmlinux_32.lds.S | 1 + arch/x86/kernel/vmlinux_64.lds.S | 3 +++ 4 files changed, 7 insertions(+), 0 deletions(-) diff --git a/arch/Kconfig b/arch/Kconfig index 0267bab..c1f9feb 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -103,3 +103,5 @@ config HAVE_CLK The <linux/clk.h> calls support software clock gating and ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- drivers/serial/8250.c | 4 ++-- drivers/serial/amba-pl010.c | 2 +- drivers/serial/amba-pl011.c | 2 +- drivers/serial/cpm_uart/cpm_uart_core.c | 2 +- drivers/serial/m32r_sio.c | 4 ++-- drivers/serial/serial_core.c | 2 +- drivers/serial/serial_lh7a40x.c | 2 +- drivers/serial/sh-sci.c | 2 +- drivers/serial/ucc_uart.c ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- fs/proc/proc_misc.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index ded9698..1968251 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -503,7 +503,7 @@ static int show_stat(struct seq_file *p, void *v) struct timespec boottime; unsigned int *per_irq_sum; - per_irq_sum = kzalloc(sizeof(unsigned int)*NR_IRQS, GFP_KERNEL); + per_irq_sum = ...

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- drivers/pcmcia/at91_cf.c | 2 +- drivers/pcmcia/vrc4171_card.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pcmcia/at91_cf.c b/drivers/pcmcia/at91_cf.c index 6849685..194c8ba 100644 --- a/drivers/pcmcia/at91_cf.c +++ b/drivers/pcmcia/at91_cf.c @@ -273,7 +273,7 @@ static int __init at91_cf_probe(struct platform_device *pdev) goto fail0d; cf->socket.pci_irq = board->irq_pin; } ...

so could spare some memory with small alignment in bootmem also tighten the alignment checking..., and make print out less debug info Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/x86/kernel/setup_percpu.c | 16 +++++++--- include/linux/init.h | 2 +- init/main.c | 65 ++++++++++++++++++++++++++++++--------- 3 files changed, 62 insertions(+), 21 deletions(-) diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index ...

So could remove some duplicated calling to get_irq_desc Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com> --- arch/x86/kernel/io_apic_32.c | 6 +++- arch/x86/kernel/io_apic_64.c | 45 +++++++++++++++++------------------------ arch/x86/kernel/irq_32.c | 3 +- arch/x86/kernel/irq_64.c | 8 ++++-- drivers/mfd/tc6393xb.c | 7 ++++- drivers/parisc/dino.c | 3 +- drivers/parisc/eisa.c | 4 ++- drivers/parisc/gsc.c | 3 +- ...

On Wed, 06 Aug 2008 13:35:55 -0700 A lot of these are range checks so could be replaced by a single valid_irq(irq) test. Alan --

It looks like you might possibly be iterating to something usable. Breaking architectures is not acceptable however. My suggestion for a non-breaking path: - Kill NR_IRQS in then generic code. With Alan's serial patch and my kstat patch it looks like we have the worst of that. - Introduce get_irq_desc and the functions or function modifications that pass a struct irq_desc * into the genirq code. Although I absolutely hate the name get_irq_desc as it implies we are reference ...

[Empty message]

Possibly irq_desc_with_new() or even create_irq_desc in the one or two code paths that care. Everywhere else can reasonably assume that the irq_desc entries have either been setup, or that the irq number is invalid. Eric --

for some platform like 32bit, may not like to mess up with CONFIG_HAVE_SPARSE_IRQ, to have dyn_array to take small nr_irqs with already add for_each_irq_desc(irq, desc)... | #ifndef CONFIG_HAVE_DYN_ARRAY | /* could be removed if we get rid of all irq_desc reference */ | extern struct irq_desc irq_desc[NR_IRQS]; | #else | extern struct irq_desc *irq_desc; | #endif | | #define for_each_irq_desc(irq, desc) \ | for (irq = 0, desc = irq_desc; irq < nr_irqs; irq++, desc ...

Part of what I am thinking is that there should never be a patch that does s/NR_IRQS/nr_irqs/ that is just pointless noise. Either we need a new abstraction like for_each_irq or to update the code to the current best practices. Which means the patches should be uncontroversial and mergeble on their own with no weird dependencies. Eric --

Following patches fix two memory leaks with CPU hotplug. Some per CPU data is allocated each time a CPU is set online. But this space is never freed. Usually this memory leak is not a big deal (for normal CPU hotplug usage). But during stress tests with lots of CPU offline/online cycles this really matters. The entire leak is 40K (10 pages) for each offline/online cycle per CPU. I've verified both fixes performing more than 90000 CPU offline/online cycles. This is not a regression but I ...

Exception stacks are allocated each time a CPU is set online. But the allocated space is never freed. Thus with one CPU hotplug offline/online cycle there is a memory leak of 24K (6 pages) for a CPU. Fix is to allocate exception stacks only once -- when the CPU is set online for the first time. Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com> --- arch/x86/kernel/cpu/common_64.c | 23 +++++++++++++---------- 1 files changed, 13 insertions(+), 10 deletions(-) diff --git ...

[ Andrew/Ben, this fixes the bootup hang we were talking about the other day. ] radeonfb: Fix copyarea for R300 and later. For current chips we need to make sure the DSTCACHE_CTLSTAT and WAIT_UNTIL registers are properly setup. If not, a sequence of copyarea operations can result in the 2D raster processor hanging. Here is a trace on an RV370 (PCI device ID 0x5b64), it records the RBBM_STATUS register, then the source x/y, destination x/y, and width/height used for the ...

That and the flush code for when we're about to manually hit the fb. I'll send a patch later today. Ben. --

Problem found in the following kernels 2.6.26,2.6.25 Distribution slackware 12.1 and 12.0 Compile the module rtl8185 (realtek wireless PCI card) module name r8180 OS hang during boot during UDEV configuration need to turn off computer. kernel 2.6.25 Compile the driver inside the kernel, boot is OK as long as I do not try to configure the network. I was able to ifconfig wlan0 up but the system hang during DHCP, no oops just hang. Need to turn off the computer. I believe to be a ...

Applied to cryptodev-2.6. Thanks a lot Austin! -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt --

crc32c-intel.ko? -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html --

Revised by comments: Add 'static' for limitation namespace; Resend for fixing lines-folded by adjusting evolution config; CRC32c algorithm with the new CRC32 instruction in SSE 4.2 instruction set. The patch detects the availability of the feature, and chooses the most proper way to calculate CRC32c checksum. Byte code instructions are used for compiler compatibility. No MMX / XMM registers is involved in the implementation. Signed-off-by: Austin Zhang ...

Well, it should normally go to comment at the beggining of file. Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html --

Yes I think this should be ENODEV to be consistent with the existing drivers such as padlock-aes.c. I'll make that change in cryptodev. Thanks, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt --

Callers of pci_hp_register() need to be converted from a char name[] to a char *name. This change allows pci_create_slot() to dynamically rename the sysfs name of the slot in the event of a name collision. This change also removes the unused struct task_event from the slot structure. We also remove the shpchp_slot_with_bus module parameter, which essentially reverts commits: shpchp: fix slot name ef0ff95f136f0f2d035667af5d18b824609de320 shpchp: add message about ...

Hello Kenji-san, You are obviously correct about my prior bad patch that simply removed the check in pci_hp_register() for duplicate names. I also talked with Willy, and he thought that the proper way to fix all these issues was not in the individual drivers, but in the PCI core, which I agree with. Here is a patch series that attempts to implement Willy's suggestion. Patches 1--6 convert the callers of pci_hp_register/pci_create_slot from a static char name[] to a dynamically ...

Commit a86161b3134465f072d965ca7508ec9c1e2e52c7 added a check to detect platforms with broken firmware that attempt to assign identical slot names to multiple slots. This approach is suboptimal because it prints a scary message and forces the user to reload a hotplug driver with a module parameter. We can do better here by attempting to rename these duplicate slots on behalf of the user. If firmware assigns the name N to multiple slots, then: The first registered slot is assigned N The ...

Callers of pci_hp_register() need to be converted from a char name[] to a char *name. This change allows pci_create_slot() to dynamically rename the sysfs name of the slot in the event of a name collision. Signed-off-by: Alex Chiang <achiang@hp.com> --- drivers/pci/hotplug/acpiphp.h | 2 +- drivers/pci/hotplug/acpiphp_core.c | 17 ++++++++++------- 2 files changed, 11 insertions(+), 8 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp.h ...

hello, I am trying to set up USB networking between the linux host and developement board.I am using g_ether gadget along with the controller driver.Currently the code is tested with the windows host (xp service pack 2). RNDIS option in g_ether is enabled and it works fine with windows host.ftp,telnet and ping are working. With the linux host i am disabling RNDIS and using the CDC option. Currently i am facing following problems. 1)With RedHat host (2.4.20) on connecting the cable i ...

32bit. High Memory Support (64GB) ---> -*- PAE (Physical Address Extension) Support Memory model (Sparse Memory) ---> -*- 64 bit Memory and IO resources (EXPERIMENTAL) The following is "cut&paste" ... (gmail is not the a good way to do this.) for the 8 cpus that I managed to boot. routine.. 3723.90 BogoMIPS (lpj=7447810) CPU: L1 I cache: 32K, L1 D cache: 32K CPU: L2 cache: 4096K CPU: Physical Processor ID: 6 CPU: Processor Core ID: 0 x86 PAT enabled: cpu 3, old ...

Yes. But, does it matter? Thanks, Jeff. --

You could try booting CONFIG_NR_CPUS=16 with maxcpus=8 (kernel command line option). If it boots you can then try bringing the rest of the cpus online manually echo 1 > /sys/devices/system/cpu/cpu8/online ... echo 1 > /sys/devices/system/cpu/cpu15/online Might get a better OOPS/BUG_ON/etc report. Max --

From: "Jeff Chua" <jeff.chua.linux@gmail.com> Do you have lockdep enabled? If sure, try turning that off. --

I've a Dell R900 with 4 quad-core Xeon processors (total 16CPUs), but can only managed to boot up with CONFIG_NR_CPUS=8. Setting CONFIG_NR_CPUS=16 causes the kernel to hang while booting. Here's the dmesg with CONFIG_NR_CPUS=8 ... CPU6: Intel(R) Xeon(R) CPU L7345 @ 1.86GHz stepping 0b checking TSC synchronization [CPU#0 -> CPU#6]: passed. CPU 7 irqstacks, hard=c0526000 soft=c051e000 Booting processor 7/26 ip 6000 Initializing CPU#7 Calibrating delay using timer specific ...

Ok, booted with CONFIG_NR_CPUS=16 with maxcpus=8, but can't find the rest of the CPUs. I can only find cpu0 to cpu7 ... drwxr-xr-x 4 root root 0 Aug 6 19:38 cpu0 drwxr-xr-x 4 root root 0 Aug 6 19:38 cpu1 drwxr-xr-x 4 root root 0 Aug 6 19:38 cpu2 drwxr-xr-x 4 root root 0 Aug 6 19:38 cpu3 drwxr-xr-x 4 root root 0 Aug 6 19:38 cpu4 drwxr-xr-x 4 root root 0 Aug 6 19:38 cpu5 drwxr-xr-x 4 root root 0 Aug 6 19:38 cpu6 drwxr-xr-x 4 root root 0 Aug 6 ...

Sorry, found it. These are not obvious. I had selected "Subarchitecture Type (PC-compatible)" and could find a place to set CONFIG_X86_GENERICARCH. Just found it under " Subarchitecture Type (Generic architecture)", and then it shows the CONFIG_X86_BIGSMP option. Ok, compiling and testing now. Jeff. --

One trivial thing to try would be to just bisect it. I assume 2.6.26 is fine, so while it will take a few boots to try it out (there's 8111 commits in between, so 13 reboots should do it), the advantage of bisection is that it's fairly straightforward to do even if you don't have any clue where the problem might lurk. And with your machine, recompiling the kernel 13 times shouldn't take that long ;) Linus --

Thanks to all the great helpful suggestions from everyone, and this turns out that I just need to enable the following switches, so I didn't bisect further, and since it's first machine that I've tried with more than 8 CPUs so I wasn't sure whether 2.6.16 has the same Thank you all for the great linux kernel! Jeff. --

It used to. 64-bit kernel used to handle maxcpus option as documented in the Documentation/cpu-hotplug.txt and 32-bit one was broken. I just looked at the latest code and realized that both are now broken. They ignore cpu id > maxcpus instead of not-booting them. I'll send a patch that fixes that tomorrow. Max --

Ok, may be not as bad as I thought. These are not in .config meaning, Thanks, Jeff. --

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi-- I'm having a problem resuming (from RAM) if the i915 and drm modules are loaded (these are needed for running X on my machine). Vanilla kernel versions 2.6.25.x and 2.6.26.x. I get the same behavior with Debian linux-image-2.6.25-2-686 kernel. I can't believe I'm the first to encounter this. suspend-to-ram and resuming works fine with earlier kernels (e.g. 2.6.17), but I haven't tested exhaustively. Possibly related is a failure in ...

Talking about wireless driver updates - has anybody looked at the RaLink wireless driver? It's in the newer EeePC's (901 and 1000), and it actually has a driver from the company which the comments make clear is GPLv2 too, no apparent oddness or anything: http://www.ralinktech.com.tw/data/drivers/2008_0708_RT2860_Linux_STA_v1.7.0.0.tar.bz2 which admittedly apparently has some trivial problems (with a trivial patch as seen ...

I looked at the driver and got it working. However my 901 seems to have an issue with the antennas. I can do scans, but associating with an access point or even thinking about transmit keeps failing. Even with external antennas this still doesn't work. So either my EeePC is broken or it needs some ACPI magic to enable it. Personally I think mine is broken since the PCIe hotplug stuff works for the wireless card and that seems to be their way of doing the rfkill. The driver itself ...

I've send a mail to Ralink with a request if they could step away from the "port the windows driver to linux" idea and get the Linux driver "kernel-ready" early on. That won't solve the problem now, but hopefully it does mean the drivers for Well it basically is the Windows driver ported to Linux. ;) Ivo --

Here are the pending networking bits. There was going to be the addition of the ath9k wireless driver but there was some fallout that is being worked on right now (linux/list.h changes needed lkml review, ath9k driver triggered some gcc aborts, all kinds of fun stuff :-) so hopefully it will make it in the next pass. I tried to wait an extra day for it to be resolved, but that was optimistic. Highlights: 1) Intel driver guys triggered an OOPS when setting MTU during a busy transmit ...

Hi, I am having some major problems with the newest kernels on my Thinkpad R61i and I'm doing my best trying to debug the problems. One problem is waking up from suspend to RAM. Earlier kernels worked fine (e.g. 2.6.25), but with the latest the system is unusable after waking up. When trying to wake the computer up it first is completely unresponsive with a black screen. After about 40 seconds it comes back alive and shows the locked screensaver. But whatever I do I just get error ...

Same here, sata is dead after the resume. Thinkpad X60. -- Jens Axboe --

Known bug && patch available. http://bugzilla.kernel.org/show_bug.cgi?id=11178 -- tejun --

Actually it is better. When I travelled from brussels they were so Well, I'm afraid that AV industry with its "lets enumerate badness" model... just does not protect against anything at all... Maybe we should be thinking more like "only ever run exectuables from trusted sources" -- like from your distro -- ? Fortunately that's the way most linux users work today. Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) ...

Last I heard, all sync levels were on their way out (Arjan). Rene. --

On Tue, 05 Aug 2008 20:25:29 -0400 That's not my worry. My real worry is that the anti-virus companies have been working with an enforcement policy that has been evolving slowly from the DOS days, while today's threat model has changed considerably. I do not see how the proposed hooks would close off a system sufficiently to claim anything approaching security. The way forward is to: 1) define a threat model 2) figure out what infrastructure is needed for protection 3) come up with ...

... and which also doesn't into account some of the facilities which Linux has, that DOS/Windows does not have. Part of the problem I suspect is that the AV folks have managed to get CIO's believe that all computer systems need to have anti-virus software, of the same design that is needed for DOS/Windows systems. This state of delusion is so bad that apparently some AV engineers aren't even willing to reason from first principles what is necessary or not to maintain a secure system. And ...

From: Alexander Beregalov <a.beregalov@gmail.com> Applied, thanks Alexander. --

This seems to be the agreed work-around for this. --

Hi Matthew! Ah, THIS patch. Yes, a more intelligent version of it (that doesn't scream blood murder on thinkpads without PWMS) needs to go into thinkpad-acpi along with your patch. -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh --

Well, for what is it worth, thinkpad-acpi has a knob (brightness_mode) which can be used. Set it to CMOS mode (see docs). From what I recall, it should do what your patch does. -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh --

Woah, it works perfectly fine here. Thanks :) Cheers, --=20 Yves-Alexis

The drivers need to know if you are going to be using odd sizes regularly so they can pick between - I do this fine who cares (most chips) - Er uh wtf its not 512 byts (some chip state machines) - FIFO off (performance hit) for this disk - FIFO managed for the odd command thats a funny size - Various other levels of software managed controller thumping It's not a passive thing and we'd want to do it post identify on the drive pair as it'll often need per channel decisions (eg on ...

Put a 512 byte and a 2K sector sized disk on the same channel and its Because I'd like to do better than that - and FIFO is only one of the cases. I think your proposal basically works - providing you identify both drives before worrying about sector sizes. Most of our other logic is done that way - the drives are all identified then the controller makes decisions. The setup time method then either vetos the disk or accepts it The existing mode_filter can be used to implement stuff like ...

ATA-1 says that word 106 is reserved (and further that reserved means the drive shall return 0). I don't have a spec earlier than that. I suspect the ATA spec writers are assuming that no drive puts random bits there. all-0 and all-1 make sense, but alternating 0 and 1 are unlikely. At least, it's good enough for us for checking words 48, 76, 78, Obviously it is going to change per command -- because different commands have different sizes. I was thinking that we could call the driver ...

Ah, .flags = ASYNC_BOOT_AUTOCONF I guess? + I'm sure you will need to set up more entries of your uart_port structure. --

Not quite, but it got me on the right track. I needed to set port->type before calling uart_add_one_port(). Now it gets into my code. Previously, it had never even called any of my code. None of it. Unfortunately, I ran it and it segfaulted. Turns out that Greg's tinyserial driver didn't implement ops->set_termios() and that was the cause. Implementing that got everything working. Now I get in my dmesg: [ 73.813893] ttytiny: Tiny serial driver [ 83.714486] Enter: tiny_startup [ ...

Hi Paul, I thought about hierarchical RCU, but I never found the time to implement it. Do you have a concept in mind? Right now, I try to understand the current code first - and some of it doesn't make much sense. There are three per-cpu lists: ->nxt ->cur ->done. Obviously, there must be a quiescent state between cur and done. But why does the code require a quiescent state between nxt and cur? I think that's superflous. The only thing that is required is that all cpus have ...

This works fine for my toolchains at least, so I'll apply it. --

Ok, thanks. I didn't see it as a bugfix because it doesn't make any difference on s390, and nobody else is using arch_prepare_hugepage() so far. But of course this may change, so I should have made two Right, this was missing from the beginning. It affects s390 only, so I'll check if we need a backport. Thanks, Gerald --

This is a trivial cleanup. Anyone doesn't use it any more. Signed-off-by: MinChan Kim <minchan.kim@gmail.com> --- mm/mempolicy.c | 1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 97020c0..36f4257 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -808,7 +808,6 @@ static int migrate_to_node(struct mm_struct *mm, int source, int dest, int do_migrate_pages(struct mm_struct *mm, const nodemask_t *from_nodes, const ...

Yes, I got that :) But if we're not about to sprinkle "if check_ptr(arg)" all And if you hadn't used kzalloc you'll still blow up. I dislike zeroing allocs myself because I have dreams of valgrinding the kernel. gcc would warn about You mean -EINTR? Yes, it should probably be left undefined: the caller Good point. I assumed passing through the value would be useful, but as it's not been after a couple of years, we should make the callback return void. No, all we know is ...

I'm sure the compiler warns about that. -- Intel are signing my paycheques ... these opinions are still mine "Bill, look, we understand that you're interested in selling us this operating system, but compare it to ours. We can't possibly take such a retrograde step." --

Amusingly, yes. -- Horms -> looks for somewhere to hide --

linux-next-20080805/drivers/mtd/devices/mtd_dataflash.c: In function 'add_dataflash_otp': linux-next-20080805/drivers/mtd/devices/mtd_dataflash.c:670: error: too many arguments to function 'otp_setup' make[4]: *** [drivers/mtd/devices/mtd_dataflash.o] Error 1 make[3]: *** [drivers/mtd/devices] Error 2 --- ~Randy Linux Plumbers Conference, 17-19 September 2008, Portland, Oregon USA http://linuxplumbersconf.org/

Ack, looks good, fixes those warnings. --- ~Randy Linux Plumbers Conference, 17-19 September 2008, Portland, Oregon USA http://linuxplumbersconf.org/ --

These warnings have been around for several days now. Is there a patch available but not yet merged for linux-next? linux-next-20080805/drivers/xen/balloon.c:599: warning: initialization from incompatible pointer type linux-next-20080805/drivers/xen/balloon.c:600: warning: initialization from incompatible pointer type linux-next-20080805/drivers/xen/balloon.c:601: warning: initialization from incompatible pointer type linux-next-20080805/drivers/xen/balloon.c:602: warning: initialization from ...

Your (2) sounds reasonable. -- Bill Davidsen <davidsen@tmr.com> "We have more to fear from the bungling of the incompetent than from the machinations of the wicked." - from Slashdot --

Would you have any document about MDB, or project-site url? I think it is hard for anyone to understand what you like to Would you have talked with kdb/kgdb people? Since linux kernel already have kgdb, I think you might better talk about merging your functionality (and stability) to kgdb, or at least Out of curiously, is there anyone except for you using MDB? -- Masami Hiramatsu Software Engineer Hitachi Computer Products (America) Inc. Software Solutions Division e-mail: ...

On Tue, Aug 05, 2008 at 12:26:35AM -0600, jmerkey@wolfmountaingroup.com wrote: Kprobes needs to be the first consumer of these exceptions (via notify_die); if the exception was due to a kprobe, it does its thing without requiring any user intervention. Here, for example, you can get into mdb for a kprobe breakpoint hit. Please move the mdb hooks to after kprobes has been notified. Better still, integrate mdb to use the notify_die infrastructure and use a lower priority than what kprobes does ...

I guess you got complaints from scripts/checkpatch.pl for at least 3/4 of all lines by now. Some or many of the complaints will probably look silly. Have a look at Documentation/CodingStyle then because it gives context for many of the complaints. It also gives some rules of thumb for things which checkpatch.pl cannot check, e.g. legibility of global names. From what I saw, there are _many_ whitespace "deviations" in your code (which perhaps scripts/Lindent can help to adjust), ...

We probably should and i should've probably hit that during testing and maybe hit an OOPS along that path but we don't do REQ_TYPE_BLOCK_PC in ide-floppy yet, as it appears, and since idefloppy_blockpc_cmd() is only called when blk_pc_request(rq) that's why this one doesn't matter here. We should consider it though when we convert to using BLOCK_PC type rq's. -- Regards/Gruss, Boris. --

For kernel-doc notation, the parameters need to follow the function name+short description. After that, a longer description can be added. I.e., please move the parameters up to --

Sorry, I forgot to add my kdoc for list_splice_tail_init(), here it is with it added: -- list.h: Add list_splice_tail() and list_splice_tail_init() If you are using linked lists for queues list_splice() will not do what you would expect even if you use the elements passed reversed. We need to handle these differently. We add list_splice_tail() and list_splice_tail_init(). Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Luis R. Rodriguez ...

On Wed, 6 Aug 2008 15:16:02 +0100 one argument against the namespaces one is that of security; on the one hand I can see the argument of running the virus scanner in its own container, on the other hand I can see the argument of not liking that is a fair list of complaints.. the question is can inotify be fixed.... -- If you want to reach me at my work email, use arjan@linux.intel.com For development, discussion and tips for power savings, visit http://www.lesswatts.org --

It prevents any user from overwriting an existing executable, and it prevents any user from adding an executable file to a common directory (/usr/bin). It also prevents any user from overwriting a different user's data file. What specific threat model are you feeling is still present on Linux today that this proposal is supposed to address? thanks, greg k-h --

I see what you mean, it should have been worded better. Nevertheless that is what was intended by it - to enable caching only on filesystems where I don't have a very good understanding of the VM subsystem I must admit. So in other words with the current kernel file modification time is not necessarily correct - it represents when the file was last opened for modification, not when it was actually modified? (While those two points in time can be arbitrarily separated) How would I use ...

On delivery you'd need to decide what security scheme is actually available on the packet and in what order to interpret any inbound security data. If you had an MLS scheme that uses CIPSO, an integrity mechanism using IPSEC and a DAC scheme that assigns user ids by host address getting the ordering right and every domain registered properly in the networking stack would be a trick. Plus, making sure that any state the security scheme requires is tricky. Maybe it's not actually worse if the ...

On Tue, 5 Aug 2008 14:04:26 -0400 We're getting close. What exactly are the threat models you want to protect against? Once we have the answer to that question, we can figure out whether the current patch set addresses the threat model or whether alternate approaches are needed. -- All rights reversed. --

That is exactly what you cannot do. Otherwise you have to rewrite filesystems not to use the pagecache, and probably have to do disallow or do some horrible hacks to support mmap. The requirement you actually intend is more like what you said before, It is a best effort thing. When you start mmapping, that allows get_user_pages, which in turn may be able to take pages and modify Even just with mmap (we do a much better job of it now, but) you can probably write to a file after mtime if ...

On Wed, 6 Aug 2008 17:12:56 +0100 A universally useful large change is often easier to get into the Linux kernel than a single-purpose hack. -- All Rights Reversed --

Actually local disk file systems can be changed invisibly to the VFS too by directly writing to the block device. This does not change the page cache, but the on disk copy and when a page is pruned from RAM and reloaded VFS will see the new contents without knowing about any change. How would you stop that in your proposal? I assume you could always require a special LKM that forbids block writes for anything mounted, but that has other problems too and one wuld need to be extremly careful of ...

I can't, but everyone is doing it so that is at least an implied Maybe it is not how you imagine it? It is not a database of every possible individual attack but there are more intelligent methods. But I am not an expert in this field to explain it better.. Tvrtko Sophos Plc, The Pentagon, Abingdon Science Park, Abingdon, OX14 3YP, United Kingdom. Company Reg No 2096520. VAT Reg No GB 348 3873 20. --

My guess is that wine is good enough these days to run many Windows worms actually. -Andi --

-----Original Message----- From: Peter Dolding [mailto:oiaohm@gmail.com] Sent: Wednesday, August 06, 2008 8:39 AM To: Press, Jonathan Cc: Rik van Riel; Greg KH; Arjan van de Ven; Eric Paris; linux-kernel@vger.kernel.org; linux-security-module@vger.kernel.org Subject: Re: [malware-list] [RFC 0/5] [TALPA] Intro to a linuxinterfaceforon access scanning On Wed, Aug 6, 2008 at 10:10 PM, Press, Jonathan <Jonathan.Press@ca.com> This buffer overflow risk and other equals are why LSM's exist. It ...

Should be OK from my point of view assuming we keep cache and basic filesystem exclusions in kernel. Otherwise it would be too much work (I am talking about CPU time) to do with each and every interception. Tvrtko Sophos Plc, The Pentagon, Abingdon Science Park, Abingdon, OX14 3YP, United Kingdom. Company Reg No 2096520. VAT Reg No GB 348 3873 20. --

Over engineering in a void for some supposed someday user isn't a good idea either. If there are other people interested in working together to come up with a solution to multiple needs I'm here to help! -Eric --

You did? I didn't see any patches from you here, did I just happen to miss them? If Eric's code was based on yours, that's great, but it didn't imply maintenance, especially as it doesn't look like your original patches much :) thanks, greg k-h --

In the patches I posted, "checks" are done at open and close if the result is not already in the cache. Every write invalidates the cache and thus the next open/close will do a "check." So the longer a process keeps a file open the longer it is susceptible to "unclean" data existing in that file. --

We are not just kernel. http://hal.freedesktop.org/docs/PolicyKit/ Is a good read. Long term Linux design is basically Nuke the root user and protect the core os providing restricted access. So anti system core damage integrates outside kernel space. We are You got to remember syscall trapping can be done for good and evil. You use it to monitor actions another use it to do harm like steal data. We need syscall trapping for debugging. ptrace and the like. Something important you ...

It is much more advanced than that, really. I don't know if ever a whole file is read and in 99% it is just a tiny part of it. I don't know what I am allowed to disclose and also it is not my area of expertise, but if you are interested in how detection actually works maybe we can talk off list and put you in touch with some other people here. It is also wrong to think that you can scan only what has changes because that bit may be harmless itself but present a final part of a malware ...

Open for read is the "traditional" time for scanning, but the sequence (open for write) -> change -> (time passes or close happens) is specifically a good time to do content checking, so as to have the answer to the check available for the open for read. I'd suggest "read" and "write" are the two cases that are interesting, and that we've been using 'open" an "close" for a not very good approximation to them (;-)) --dave -- David Collier-Brown | Always do right. This will ...

While I try to get someone to talk to me, lets consider even the most simplistic threat model. I'm not stating that this is the only threat that vendors wish to protect against, but I think we can all agree it is a meaningful threat. Trying to get the AV vendors to talk about their products actual uses is like pulling teeth. This simple thread shows what I believe to be clear and compelling evidence of the need for an in kernel solution. Lets just consider that we are a high input, high ...

On Wed, 6 Aug 2008 11:33:23 -0400 Disks are slow and files are getting larger by the day. We can do a lot better than scanning a whole file. A mechanism that can notify programs about what file changed and what byte range in the file changed can reduce scanning overhead by only needing to scan the part of the file that changed. More importantly, getting info on which bytes in a file changed will also help backup programs and disk indexing programs. Blocking on open can be useful for ...

On Wed, 6 Aug 2008 08:10:53 -0400 this indeed gets to the heart of the problem. several of us here are trying to argue that scan-on-close isn't very good, BUT that if you do scan-on-change (say with inotify) you do reach the same goal but with much better results. notice the "but" in there. What I hope will happen is that one or more people from the AV side (eg you and tvrtko) will actually read the "but" part rather than just dismissing it outright because of not liking your solution in the ...

Simplely they are following the windows way of doing things. Rootkit the OS no one will stop us. Sorry that RootKitting is not going to work here long term because we will fix Root Kit weaknesses. TPM from IBM will make it even harder. Nice bits of that are in 2.6.27. Allowing LSM stacking solves nothing. IBM's developers credentials is a far better place for file system monitoring to hook in. Its pre caching allows all the alterations they could want. Even better credentials is a ...

How are namespaces and pathnames relevant to exclusions based on filesystem *type*? It is not about checking for /proc but checking if the As small as possible - just one extra field in the inode struct. And really you are talking about a different thing again and moving the argument around. I answered you why your argument about putting a cache in userspace since we do so much there already is wrong, because we don't do a lot (anything really) in userspace for vast majority of opens ...

I didn't consider it. Most likely at the end of the day the finding will be, "if you can write directly to the block device you already won since there as so many other things you can do to subvert the system." Admittedly its the first technical point brought up on list that I didn't consider at all (lots of other things brought up on list that need more thought that weren't exactly technical details, don't let me seem like I'm downplaying those) -Eric --

Seriously, it's David Howells at Redhat who's doing that. In your last email i figured it was a typo, but credit where it's due :) -serge --

Failed daemon support has to exist in Credentials for filesystems that cannot store all the security data and needing third party support. This is the issue same features basically required to do the same things. Different reasons. One to virus scan. One to fix up security lacking file systems. From what I see exactly the same set of features are needed. At most only 1 feature is missing a form of copy on write. Adding this at credentials level over the complete system would not be a ...

Peter Dolding wrote: Then in a real sense you've solved much of their problem for them (;-)) After this comes engineering, so that they can re-use the scanning mechanisms they already have, but from a different caller. The requirements are probably that they know - is this an open for read or write (somewhat less time-sensitive)? - is the data present, or do we have to wait? - if so, for what? as of the time they start looking at the file. Having a race-free mechanism using ...

In general this metadata provides more context to the event that happened. For example reporting - log message/UI popup/centralised something can be displayed saying which user running which application was involved with bad stuff. Also we can find out where the user is logged in and send him a message there. It is more descriptive than just failing the access with -EACCESS which When you don't run an userspace client cache should not come into play because nothing will be cached ...

On Tue, 5 Aug 2008 14:38:23 -0400 How can you, as a security professional, argue for a security measure that you know is flawed, when there is a mailing list full of people willing to help figure out what a working security measure would look like? Yes, abandoning some of the old DOS anti-virus security model may cause work, but surely that will be worth it if it leads to a more secure system? -- All rights reversed. --

There's no fixed process name in Linux that cannot be easily faked: Use process name -- every process can change that by writing to its own environment. Use comm name -- there's a prctl to change that and there can be collisions Use path name of binary -- breaks with chroot and name spaces. Also existing binaries can be subverted. Use inode of binary -- can be faked with fuse and breaks when the binaries is copied ... Use dev, inode -- breaks when copying binary and when running on ...

Yeah, those are both leftovers from the 16-bit code. I'll fix them up in --

1byte? Thanks. -- tejun --

Well, as most of machines are 64-bit these days, discussing whether a word means 16 or 32 bits is kind of moot. I usually just think byte, word, dword, qword and and libata and pci assume that too - ata_id_has_dword_io() and PCI configuration accessors. It's ultimately a peripheral issue either way tho. Thanks. -- tejun --

Applied. Sam --

eeepc-laptop uses the hwmon struct after unregistering the device, causing an oops on module unload. Flip the ordering to fix. Signed-off-by: Matthew Garrett <mjg@redhat.com> --- commit 95f4bc41ef256b8e3dfa94cabe1faf0776ac988c Author: Matthew Garrett <mjg59@srcf.ucam.org> Date: Mon Aug 4 ...

To get some more debug data, can you please retest with latest kernel (2.6.27-rc2) using "debugpat" kernel option and provide dmesg output plus contents of <debugfs>/x86/pat_memtype_list? Thanks, Andreas --

A build fix, probably run-away search-'n-replace damage.. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> --- Index: linux-2.6/kernel/lockdep.c =================================================================== --- linux-2.6.orig/kernel/lockdep.c +++ linux-2.6/kernel/lockdep.c @@ -2821,7 +2821,7 @@ static int lock_release_nested(struct ta #ifdef CONFIG_DEBUG_LOCKDEP hlock->prev_chain_key = 0; - hlock_class(hlock) = NULL; + hlock->class_idx = 0; hlock->acquire_ip = 0; ...

I'm confused. Are these two of the competing controllers? Or are the complementary in some way? -- Dave --

I don't think it will cause significant performance loss, because the link between a page and a page_cgroup has already existed, which the memory resource controller prepared. Bio_cgroup uses this as it is, and does nothing about this. And the link between page_cgroup and bio_cgroup isn't protected by any additional spin-locks, since the associated bio_cgroup is guaranteed to exist as long as the bio_cgroup owns pages. I've just noticed that most of overhead comes from the spin-locks when ...

Hi Dave, I have been tracking the memory controller patches for a while which spurred my interest in cgroups and prompted me to start working on I/O bandwidth controlling mechanisms. This year I have had several opportunities to discuss the design challenges of i/o controllers with the NEC and VALinux Japan teams (CCed), most recently last month during the Linux Foundation Japan Linux Symposium, where we took advantage of Andrew Morton's visit to Japan to do some brainstorming on this topic. ...

Would you like to split up IO into read and write IO. We know that read can be very latency sensitive when compared to writes. Should we consider them Won't that get too complex. What if the user has thousands of disks with several Are you suggesting that the IO and memory controller should always be bound Yes, I agree with this step as being the first step. May be extending the Very nice summary -- Warm Regards, Balbir Singh Linux Technology Center IBM, ISTL --

I'd like to add the following item to the goals. 7. Selectable from multiple bandwidth control policy (proportion, I agree with your plan. We keep bio-cgroup improving and porting to the latest kernel. Thanks, Ryo Tsuruta --

Yes, please solve the I/O control problem first. -- Warm Regards, Balbir Singh Linux Technology Center IBM, ISTL --

Ok, so I can reproduce on a few 32 bits configs with ftrace enabled. Looks like some non volatile GPRs get corrupted. I don't know yet if ftrace is the culprit though, I couldn't find anything obviously wrong with the mcount implementation we have. It looks like the corrupted GPR has been saved/restored on the stack and that the corruption is due to the stack itself being written to. It's not clear by whome though and in what circumstances. We'll have to dig more. Cheers, Ben. --

please attach the output of acpidump using the latest pmtools at http://www.lesswatts.org/patches/linux_acpi/ Of course you can open a new bug in bugzilla and attach the output of dmesg, acpidump, lspci -vxxx. http://bugzilla.kernel.org/enter_bug.cgi?product=ACPI --

Hmm, this construct of empty conditional blocks is new to me. What is its purpose? -- Stefan Richter -=====-==--- =--- --==- http://arcgraph.de/sr/ --

I'll post some patches with these changes and submit it as a module. I will of course need to patch the kernel to introduce the die notifiers where they are not and there's the issue of cross call NMI to stop other processors on SMP. That's the real issues, Andi, is the xcall stuff for IPC to control all the processors. --

NMIs have die notifiers too. That said a cleaner way would be to generalize the synchronization code already in kdump and reuse that. -Andi --

Eh, that info is already trivially available from my original mail: uhci_hcd 0000:00:1d.2: new USB bus registered, assigned bus number 3 [...] input: USB Compliant Keyboard as /class/input/input0 input: USB HID v1.10 Keyboard [USB Compliant Keyboard] on usb-0000:00:1d.2-1 input: USB Compliant Keyboard as /class/input/input1 Yes, obviously the keyboard is attached during boot (and has been connected No idea. I could try this I guess if it will really provide useful info. Will include ...

Hi Alan, Meaningless, but annoying. Especially as such "errors" get through to the console when you boot with the 'quiet' boot parameter and thus draw more Thanks for the explanation. Attached boot messages (for -rc2) with usb debug enabled as suggested by Felipe Balbi. Messages have been extracted from kern.log (dmesg had already overflowed). If I filter out messages related to bus 3 (PCI 1d.2) what I see is: [...] ! usb 3-1: new low speed USB device using uhci_hcd and address 2 [ ...

It depends how you look at the problem. I would agree that the use of gdb + kgdb vs an assembly debugger are completely different cases. The kgdb core in the mainline kernel, can actually allow to write such a front end however. The kgdb core has an API for I/O and it is possible to write an I/O module that implements an in kernel assembly debugger. The kgdb test suite is not a great example, but it is a complete example of using the kgdb core directly without a second machine. If there ...

No. First try to integrate them together so you have the best of both from one code base is what I was saying. I specifically said if they are significantly different and can't be reconciled then it could be merged. --

Nick, please note there is/was some mis-communication between the two of you with respect to kgdb, the currently merged GNU debugger interface, and kdb, the SGI kernel debugger. Merkey responded to you as if you asked about differences with KDB while what you did was ask about differences with KGDB. Both KDB and MDB are significantly different from KGDB at least in sofar that the latter is a remote debugger; it requires two machines. KDB and MDB are local. This makes KDB and MDB more ...

It's a little too early for that. Right now it's at the phase "how to make it better integrate with the kernel", with the use of existing hooks, adding the needed hooks to be more complete, working as a module, etc. When that is done then the philosophical aspects can come into play, but it's not there yet. OG. --

That idea sounds familiar, the "suspend2" response, when something new and significantly different is offered, instead of putting it in and letting people choose in configuration, take the position that what is there is good enough, and if the author of the new solution will just drop all their ideas and slap some band-aids on the existing code it will be "gooder enough" without actually offering people a choice of something different. I totally agree with this, the whole idea of a remote ...

Mikael Pettersson writes: > On Mon, 4 Aug 2008 15:56:05 +0200, Arkadiusz Miskiewicz wrote: > >On Monday 04 August 2008, Mikael Pettersson wrote: > >> Arkadiusz Miskiewicz writes: > >> > Hello, > >> > > >> > http://google-perftools.googlecode.com/svn-history/r48/trunk/src/base/= > >at > >> >omicops-internals-x86.cc says > >> > > >> > " // Opteron Rev E has a bug in which on very rare occasions a locked > >> > // instruction doesn't act as a read-acquire barrier if followed ...

This looks good. Could you move the unlock down to the end of serial_link_irq_chain? Eric --

You're welcome. :-) Thanks, Rafael --

It's all done behind the scenes I think (it's an ipw2200 device). There's Ok, that makes things much easier for me :-) But it means that for now the Why is this necessary? Doesn't the radio power down as part of the suspend process? How would I tell what the hardware is doing? I'll update the diff when I can. --phil --

For cards based on the HFC mulit chips we need runtime detection, since not all card support both access methods and this is the only way how distributions can support all cards. For the xhfc this may be a differnt choice, to select the access method on compile time, because the chip is mainly used for embedded systems, so it do not need to have a generic driver, the kerne is usually configured exactly for the hardware. On the other side, if you look into the xhfc chip and documentation, it ...

On Tue, 5 Aug 2008 09:30:07 -0400 We can go ahead step by step. I think PAM support is the first step. The daemon is the second. 1. PAM 2. A daemon for task placement (via netlink ?) I think developping "a daemon for task placement" is important. but cannot be perfect solution for any situations. The third step is 3. Modify applications (in newer version of them.) "should work without any apache modifications" is (maybe) necessary. But for perfect control, it's not enough. We ...

Do you have any measurements comparing performance of the old and the new call? TLB flush is quite time critical so we should be very careful here. -Andi --

No. What benchmark would you suggest? J --

"overkill" -- by what metric? If something is overkill, it means something is excessive. Excess (and deficiency) occur along some scale, some metric. If not by CPU cycles, then by what metric is it overkill? -- I won't rest till it's the best ... Programmer, Linux Scalability Paul Jackson <pj@sgi.com> 1.940.382.4214 --

I was referring to an earlier discussion, which resulted in a patch which includes the line: But you seemed to be saying that the reason it was not best to do so was because it was best not to change more than necessary. Perhaps I still don't understand your metric. Apparently Well, your metric seems clear enough there - minimize effort of code conversion and testing. How about this ... two routines quite identical and parallel, even in their names, except that one is async and ...

Thank you. I'll go ahead and respin the patch. Max --

Sure. What I'm saying is that I do not think it's the best I still do not see a good reason why. IMO exceptions are acceptable. Only domain rebuilds triggered by cpuset fs writes need to be async. I do not see a good technical reason why the rest needs to be converted I'm not sure what you're referring to. There was no back an forth. You suggested reverting the rename and I pointed out that it was not a rename, I simply factored out the part that generates the To be fair the fact that you had ...

Good point about "inline" confusions. I am willing to agree to "sync_rebuild_sched_domains()". -- I won't rest till it's the best ... Programmer, Linux Scalability Paul Jackson <pj@sgi.com> 1.940.382.4214 --

Paul E. McKenney wrote: Sorry about this. percpu_ptr is used for dynamic allocation percpu pointer. It seems that we cannot get a pointer from a static declare percpu data It's not necessary. Actually I like one tail per list which is good for readability. But in my patch, the most work is combining lists, not moving a list to next list, so i use multi-tailed simplify this works and others(etc: "if (rdp->nxtlist)" will be changed to be a more complex and less readability statement if i ...

Right, but the current code will basically force the discrete card drivers to implement backing store for all allocations. Do we want this ? Also, for cards that can handle page-based memory allocations, there is no way to make use of this feature, do we want this too ? Stephane --

On Wed, 6 Aug 2008 18:20:40 +0200 the ATI driver guys already have the interface working. In addition.. since when is "oh you must also make it work for THAT" a requirement? Traditionally, it's up to the second person to make generic code work for them (within reason of course, but I'll argue that the bar of reasonableness has been met here) -- If you want to reach me at my work email, use arjan@linux.intel.com For development, discussion and tips for power savings, visit ...

Write a driver and we'll talk. Until then, I don't think your argument is well supported. If the result is that your driver requires a completely separate interface than the Intel driver, that's OK too -- we can talk about merging them together in some unified mechanism at that point. I'm trying to get from zero to one kernel driver for Intel graphics at this point; that seems a valuable goal in itself. I hope it leads to other people building kernel drivers for other graphics hardware. ...

Right, but it sounds adventurous to merge this work upstream before it is generic enough to work on discrete cards. Right now it only works on intel integrated cards ; integrated cards are one business, discrete are a completely different world and the issues afoot completely different. Should this go upstream, the kernel guys have to keep in mind and accept the possibility that another memory manager might be needed at some point. Stephane --

Sorry, minor correction. diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index 0ba8857..c886113 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -428,12 +428,12 @@ _request_firmware(const struct firmware **firmware_p, const char *name, if (!firmware) return -ENOMEM; - mutex_lock(&fw_lock); tmp = kzalloc(sizeof(struct firmware_list), GFP_KERNEL); if (!tmp) { retval = -ENOMEM; goto error_kfree_fw; ...

Will do. Note that I'm really using quilt, not git, to push patches to linux-next. I only use git to push patches to Linus once in a while. -- Jean Delvare --

Hi Andrew, No, that's not OK, sorry. This goes beyond what I can offer at this point in time. I offered to push selected patches to Linus, with no other guarantee than the fact that said patches have been carefully reviewed and are believed to be correct. I do not have the time to do more than that. If this isn't OK with you, then I will have to withdraw my proposal. It's up to you. -- Jean Delvare --

On Wed, 6 Aug 2008 20:03:01 +0200 <rereads> Ah, I see, this tree is only things-which-Jean-wrote. No probs here. --

The patches in question are here if anyone wants to take a look: http://khali.linux-fr.org/devel/linux-2.6/jdelvare-hwmon/ I'll make a git branch out of them by the end of the day, for Linus to pull from. -- Jean Delvare --

How about a middle ground: ignore, but not silently? Report an error, or send it to the syslog, or both, but ultimately ignore unchangeable options, change what can be changed, and give the user/admin as much as possible. This can be particularly pertinent for XFS root. If it's mounted RO at first, it may (will?) need to become RW at some later point. Failing the remount could result in a system that requires a rescue CD (or lots of headaches for remote administration). --

FWIW, your root filesystem does not need to be rw. I keep mine ro nearly all the time on my laptop, only mounting rw if I need to install software that puts files outside /usr or if I need to modify a config file in /etc. -jasper --

Le Mon, 4 Aug 2008 14:16:41 -0400, You're right, that was stupid. Either should I add the Kconfig dependencies *OR* add the function stubs, not both. The following patch implements the second solution. I've checked that NFS, CIFS, GFS2, OCFS2, AFS and FUSE compile correctly. I've tested NFS only, though. Here is the new patch. Comments and suggestions welcome. Sincerly, Thomas --- Configure out file locking features This patch adds the CONFIG_FILE_LOCKING option which allows to ...

The usage of elfcorehdr_addr has changed recently such that being set to ELFCORE_ADDR_MAX is used by is_kdump_kernel() to indicate if the code is executing in a kernel executed as a crash kernel. However, arch/ia64/kernel/setup.c:reserve_elfcorehdr will rest elfcorehdr_addr to ELFCORE_ADDR_MAX on error, which means any subsequent calls to is_kdump_kernel() will return 0, even though they should return 1. Ok, at this point in time there are no subsequent calls, but I think its fair to say ...

Thanks, but just keep pulling linux-next and I'll try and get them uptodate more frequently :-) -- Jens Axboe --

Hi Jens, OK, will do. --=20 Cheers, Stephen Rothwell sfr@canb.auug.org.au http://www.canb.auug.org.au/~sfr/

The kmsg check is invoked with "make D=1" and reads the source files for all objects that are built by the current configuration and searches for matching kmsg descriptions for the kmsg messages in the source which have a messages id > 0. If a message description can not be found the The kmsg catalog script if called with "make D=1" will check all the messages in the current build. When a message is introduced you have to pick a free number. The checking is done automatically and errors ...

No, I mean use dev_printk() as the base for your logging macro. Add your message tag as the first field after the dev_printk() information. thanks, greg k-h --

Yes,. I think so. It's the original bug for me, while testing I found the mutex deadlock problem. But this week I will have no time to trace it. so if I have time I -- Regards dave --

This is a kbuild bug. The top-level Makefile creates the include2/asm symlink only if the file$(srctree)/include/asm-$ARCH/errno.h is present. And you moved errno.h when you committed your changes so symlink is not created anymore thus it fails. But this is oly with "make O=..." builds which I think kautobuild does not use (since it did not trigger the bug). I cannot recall the exact reason why I did this check before creating the symlink so I will simply create it always. Sam --

And the expression "oh bugger" comes to mind. The breakage in linux-next is because of 2e57d051160dd61776461637f767df19036b1186 and ea35455e0dc17d732436a5b98bd511cab64eb10e. During the last fortnight, I'd moved the "easy to move" ARM headers from include/asm-arm to arch/arm/include/asm. This means that, unknown to me, the symlink was no longer being created. That didn't get picked up in my testing, because I always build using O= builds into already existing object trees, which of course ...

I'm not keen on the idea of introducing another specialised path just for stacks. Testing coverage is tricky enough as it is and problems still slip through occasionally. Maybe going through hugetlbfs is less than ideal, So far, it is working out as a decent model. It is able to track reservations and deal with the differences between SHARED and PRIVATE without massive difficulties. While we could add another specialised path to directly insert the pages into pagetables for private mappings, ...

The catch with 2MB pages on x86 is that x86 CPUs generally have much less 2MB TLB entries than 4K entries. So if you're unlucky and access a lot of mappings you might actually thrash more with them. That is why they are not necessarily an universal win. -Andi --

On Wed, 06 Aug 2008 08:29:49 -0400 You can't kmalloc pci_dev or setup some trivial values. You need to use a proper value. The pci code does for us. Yes, see iommu_area_alloc(). --

Oops -- I meant struct device, not struct pci_dev. Anwyay, Jesse -- is this true? I can no longer do something like: static struct device junk_dev = { .bus_id = "junk device", .coherent_dma_mask = 0xffffffff, .dma_mask = &junk_dev.coherent_dma_mask, }; And then use that as the device target for dma_alloc_coherent? AFAIK, that has always worked for me. Anyhoo -- it is possible that dma_get_seg_boundary returns 0xffffffff. Then they don't handle the ...

On Wed, 6 Aug 2008 15:49:27 -0400 (EDT) because I'm done booting to full UI in that time. it's forever. -- If you want to reach me at my work email, use arjan@linux.intel.com For development, discussion and tips for power savings, visit http://www.lesswatts.org --

Right - so then all that's needed is to move usb/ before most other things that take a while to init, and it has some time to initialise usb devices in the I'm not an expert on what can be done with the Makefile process, perhaps there's an easier way to move things around based on a config option. If host init is moved before device init for everyone, wouldn't there be too many side effects? I've not tried to use usb-storage as root but presumably that works. If that already uses a ...

Whatever that involves... /dev never truly settles; it's always All the major distributions do, as far as I know. (I haven't actually Probably nothing. What do you do on your keyboard while waiting for What if usbhid is the one device driver that is compiled in? :-) Alan Stern --

Why is 5 seconds too long? Too long for what? What you're doing is already a race between hub_thread() and driver initcalls. My suggestion is no worse. "it'll still have to wait..." If by "it" you mean the initcall thread, you're wrong. If by "it" you mean the user, you still aren't necessarily correct; the user can do plenty of other things while waiting for USB devices to initialize. I suppose you could make the hub_thread delay time a module parameter for usbcore, defaulting to ...

Not exactly -- you should move most of usb/ forward so that it comes before usb/host/. Or alternatively move usb/host/ later so that it Doesn't the host init _already_ come before device init? If host init were moved _after_ device init, I don't think there would be a lot of I noticed in your logs earlier that some of your USB drivers were compiled in and others weren't. Maybe if _all_ of them were compiled in you wouldn't experience as much contention and the init delays would ...

On Wed, 6 Aug 2008 16:09:21 -0400 (EDT) because it means that while the rest of boot is done and the system idle, the mouse and keyboard aren't going to work. -- If you want to reach me at my work email, use arjan@linux.intel.com For development, discussion and tips for power savings, visit http://www.lesswatts.org --

Ah, okay. Then if the delay time were set up as a module parameter, you'd be one of the people who leaves the parameter set at its default value of 0. Alan Stern --

Cool, so this should hook into sensors subsystem, right? int is probably better here. -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html --

--- From: Borislav Petkov <petkovbb@gmail.com> Date: Sun, 3 Aug 2008 18:46:35 +0200 Subject: [PATCH] ide-generic: handle probing of legacy io-ports v4 Avoid probing the io-ports in case an IDE PCI controller is present and it uses the legacy iobases. If we still want to enforce the probing, we do ide_generic.probe_mask=0x3f on the kernel command line. The iobase checking code is adapted from drivers/ata/pata_legacy.c after converting hex pci ids into their corresponding macros in ...

So the comment in there saying that in some cases BAR0-3 could contain junk is a bogus? In other words, can we assume that one will always read 0x1f0 from BAR0 and 0x170 from BAR2 in compatibility mode. If so, the check is even simpler: if (pci_resource_start(p, 0) == 0x1f0) *primary = 1; if (pci_resource_start(p, 2) == 0x170) *secondary = 1; -- Regards/Gruss, Boris. --

And to the fact that the value 0x1f0 should only be ever seen in BAR0 and 0x170 in BAR2 even if they would have been read off the chips (some chips have these values reading back even in legacy mode, and even could malfunction if other values are written there), not fixed up there, and certainly not in BAR1 or BAR3, so it's quite pointless to look in these BARs too. WBR, Sergei --

Hello. Would have been probably enough to test only BAR0/2, don't you think? MBR, Sergei --

Thanks for the trial. My original patch has already been merged into another patch "[patch 2/3] cpuidle: Menu governor fix wrong usage of measured_us" sent out by Venkatesh. The new patch should be more intact. Jimmy --

This is a note to let you know that I've just added the patch titled Subject: USB: cxacru: Fix printk format flag in error message to my gregkh-2.6 tree. Its filename is usb-cxacru-fix-printk-format-flag-in-error-message.patch This tree can be found at http://www.kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/patches/ From simon@fire.lp0.eu Wed Aug 6 15:50:52 2008 From: Simon Arlott <simon@fire.lp0.eu> Date: Sat, 12 Jul 2008 22:19:48 +0100 Subject: USB: ...

Hi Milton, hi Greg, I _am_ arguing the correctness of the removed code. That code silently defaults driver_data to 0 when the driver doesn't set dynids.use_driver_data. This assumes that 0 is always a safe value, I have to agree with Milton. Allowing all drivers to use dynamic IDs first, and then limiting the use of the driver_data field to drivers setting a specific flag, doesn't make sense. For one thing, just because the driver sets the flag, doesn't mean that it does the proper checks ...

the following is a patch to add the overhead. :) any comments? YAMAMOTO Takashi Signed-off-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp> --- diff --git a/fs/buffer.c b/fs/buffer.c index 9749a90..a2dc642 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -41,6 +41,7 @@ #include <linux/bitops.h> #include <linux/mpage.h> #include <linux/bit_spinlock.h> +#include <linux/memcontrol.h> static int fsync_buffers_list(spinlock_t *lock, struct list_head *list); @@ -708,12 +709,19 @@ ...

On Wed, 6 Aug 2008 17:20:46 +0900 (JST) Do you have some numbers ? ;) How about changing these to be == void mem_cgroup_test_set_page_dirty() { if (try_lock_page_cgroup(pg)) { pc = page_get_page_cgroup(pg); if (pc ......) { } unlock_page_cgroup(pg) } } == Off-topic: I wonder we can delete this "lock" in future. Because page->page_cgroup is 1. attached at first use.(Obiously no race with set_dirty) 2. deleted at removal. (force_empty is problematic ...

Disabling C1E is not an option. It saves the most power when you have a multicore AMD CPU. Neither C2 nor C3 are declared here. AMD C1E is hardware level power management if all cores are in the C1 state. If you disable C1E and use a NOHZ kernel your power consumption is The sane way to disable it is in the BIOS -- if your BIOS provides such an option. Thomas pointed already out that C1E might be enabled later during boot. So you would have to reset the respective bits whenever you enter ...

ok. but i can't do that in this week. so, I'll do it in next week. honestly, I don't know how to use slabinfo command :-) --

On Sat, 01 Dec 2007 00:27:06 -0800 The separation has now finally been performed in case you want to revisit this patch. Rgds --=20 -- Pierre Ossman WARNING: This correspondence is being monitored by the Swedish government. Make sure your server uses encryption for SMTP traffic and consider using PGP for end-to-end encryption.