This patch adds the Intel Tolapai SMBus Controller text to i2c-i801 documentation. Signed-off-by:  Jason Gaston <jason.d.gaston@intel.com> --- linux-2.6.23-rc4/Documentation/i2c/busses/i2c-i801.orig 2007-08-30 15:56:39.000000000 -0700 +++ linux-2.6.23-rc4/Documentation/i2c/busses/i2c-i801 2007-08-30 15:57:32.000000000 -0700 @@ -13,7 +13,8 @@ * Intel 631xESB/632xESB (ESB2) * Intel 82801H (ICH8) * Intel ICH9 - Datasheets: Publicly available at the Intel website + * Intel ...

This updated patch adds the Intel Tolapai LPC and SMBus Controller DID's. Signed-off-by:  Jason Gaston <jason.d.gaston@intel.com> --- linux-2.6.23-rc4/arch/i386/pci/irq.c.orig 2007-08-27 18:32:35.000000000 -0700 +++ linux-2.6.23-rc4/arch/i386/pci/irq.c 2007-08-28 16:58:31.000000000 -0700 @@ -550,6 +550,7 @@ case PCI_DEVICE_ID_INTEL_ICH9_3: case PCI_DEVICE_ID_INTEL_ICH9_4: case PCI_DEVICE_ID_INTEL_ICH9_5: + case PCI_DEVICE_ID_INTEL_Tolapai_0: r->name = "PIIX/ICH"; r->get ...

This patch adds the Intel Tolapai RAID controller DID's for SATA support. Signed-off-by:  Jason Gaston <jason.d.gaston@intel.com> --- linux-2.6.23-rc4/drivers/ata/ahci.c.orig 2007-08-27 18:32:35.000000000 -0700 +++ linux-2.6.23-rc4/drivers/ata/ahci.c 2007-08-28 16:58:11.000000000 -0700 @@ -411,6 +411,8 @@ { PCI_VDEVICE(INTEL, 0x292f), board_ahci_pi }, /* ICH9M */ { PCI_VDEVICE(INTEL, 0x294d), board_ahci_pi }, /* ICH9 */ { PCI_VDEVICE(INTEL, 0x294e), board_ahci_pi }, /* ICH9M */ + { ...

In the iforce driver we currently get this warning drivers/input/joystick/iforce/iforce-packets.c: In function 'iforce_get_id_packet': drivers/input/joystick/iforce/iforce-packets.c:249: warning: unused variable 'status' if CONFIG_JOYSTICK_IFORCE_USB is not defined. The warning is easy to avoid by simply moving the variable inside the only case in the switch that actually use it. Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com> --- diff --git ...

Hi, Hmm, would this not still give a warning when JOYSTICK_IFORCE_USB=y? [ I didn't know mixing code and declarations (not at top of statement block) was accepted style in the kernel ... ] IMHO either you should at least wrap that case inside a {} of its own (so that the int status; is at top of a statement block), or else, preferably, just add "__maybe_unused" to the first declaration that you removed just now. Satyam -

There are two disjoint directories. I am wondering why there would be cache coherency issues in this case? Is this Linus nfs implementation specific or "working" as in "I can mount the directory and do my work". And there has Fine. These are all very nice theories, but I just want to report this regression and hope it won't cause any big problems for any users out there. -

I am re-sending this after help from Ian and git-bisect. To me it's a show-stopper: I cannot find an acceptable workaround that I can implement. The problem: upgrading to 2.6.23-rc4 from 2.6.22 causes several autofs mounts to fail silently - they just not appear when they should. I believe it's caused by the NFS change that forces multiple mounts from different directories under the same server side filesystem to have the same mount options by default, otherwise it returns EBUSY. For ...

"better" as in: "I now have a chance to notice, when my 'read-only Yes. The default values for acregmin, acregmax, acdirmin, acdirmax are not 600. If /a/tools and /a/share are on the same filesystem on the server, then the NFS client should warn you that you are about to do something that may result in cache coherency problems instead of silently allowing it, and then leaving you to debug the coherency issue. If you know what you are doing, then there is an option which allows you Your ...

Well, it depends on how you define "better". In this particular scenario, the maps read as follows: tools -fstype=nfs,udp,rw,intr,nosuid,nodev,rsize=8192,wsize=8192,mountvers=3,nfsve rs=3,actimeo=600 fs1.domain.com:/a/tools share -fstype=nfs,udp,rw,intr,nosuid,nodev,rsize=8192,wsize=8192,mountvers=3,nfsve rs=3 fs1.domain.com:/a/share The only difference is in the actimeo (I don't even know what it means). Is this enough to fail a mount? More importantly, it is a regression. My ...

-- Thanks, Steve -

Please use the expected (canonical) patch format. See Documentation/SubmittingPatches: 14) The canonical patch format Additional help: http://www.zip.com.au/~akpm/linux/patches/stuff/tpp.txt http://linux.yyz.us/patch-format.html Thanks. --- ~Randy *** Remember to use Documentation/SubmitChecklist when testing your code *** -

Signed-off-by: Rob Landley <rob@landley.net> Index two unindexed documentation files. --- linux-2.6.23-rc4/Documentation/arm/00-INDEX 2007-08-27 20:32:35.000000000 -0500 +++ linux-new/Documentation/arm/00-INDEX 2007-08-30 14:27:15.000000000 -0500 @@ -20,3 +20,7 @@ - description of the virtual memory layout nwfpe - NWFPE floating point emulator documentation +Porting + - Symbol definitions for porting Linux to a new ARM machine. +Setup + - Kernel initialization parameters on ARM ...

Change the interface to use kilobytes instead of pages. Page sizes can vary across platforms and configurations. A new strategy routine has been added to the resource counters infrastructure to format the data as desired. Suggested by David Rientjes, Andrew Morton and Herbert Poetzl Tested on a UML setup with the config for memory control enabled. Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com> --- Documentation/controllers/memory.txt | 22 +++++++++++++- ...

commit 1d8715b388c978b0f1b1bf4812fcee0e73b023d7 was added between 2.6.22.4 and 2.6.22.5 to cure a locking problem, but it seems to have introduced another (worse?) one. With a jffs2 filesystem (on block2mtd) on a 2.6.22.5 kernel, if I do anything that appends to a file with many small writes, I get what looks like a deadlock between the writer and the jffs2 gc thread. For example: # while true; do echo >> /some/file/on/jffs2; done will result in the bash hanging in D state, with these ...

On Thu, 30 Aug 2007, Randy Dunlap wrote: ... i just the other day submitted a patch to remove that backward compatibility, and the m68k portion of it has already been acked by geert uytterhoeven. rday -- ======================================================================== Robert P. J. Day Linux Consulting, Training and Annoying Kernel Pedantry Waterloo, Ontario, CANADA http://crashcourse.ca ======================================================================== -

Robert's patch. -hpa -

Could you mention it in feature-removal-schedule.txt? (People check that for warning of upcoming changes that impact existing code. They may not notice something elsewhere after they've got it working...) (Also, sorting feature-removal-schedule.txt by date would be really nice. The sucker's gotten huge...) And while I'm looking at this, is there some other place that entries in this file go once they _have_ been removed, so people can see "oh yeah, that went away in 2.6.23) when ...

Acked-by: Rob Landley <rob@landley.net> Rob -- "One of my most productive days was throwing away 1000 lines of code." - Ken Thompson. -

Ok, I must admit I didn't check with sparse since it seemed pointless - we usually never cast void pointers to other pointer types, specifically because the C language nicely guarantees that the right thing will happen without the cast. Sometimes we have to cast them to integer types, su sure we need the cast there. But what on earth makes a "zd_addr_t *" so special that we have to explicitly cast a "void *" to that type? I see it's defined as typedef u32 __nocast zd_addr_t; in ...

http://marc.info/?l=linux-netdev&m=117113743902549&w=1 -

--- ~Randy *** Remember to use Documentation/SubmitChecklist when testing your code *** -

I thought I had seen that bug. Module init function execution does not seem serialized enough, so the init function of one module seems to be able to be called in parallel with several other modules in turn being loaded, executing their init functions and even becoming live first class citizens. Function sys_init_module in Linux 2.6.22.x and 2.6.23-rc4 kernel/module.c does not hold module_mutex when executing the init functions of the modules. __________________________________ ...

Hi Clemens, what do you mean by "overwrites it"? It just probably merges your vma with the newly created one, right? -- Jiri Kosina -

I don't seem to get what is the issue here. Your mapping is not removed, only the VMAs are merged together into one larger VMA if they have neighbouring address ranges and compatible protection bits. See vma_merge(). -- Jiri Kosina -

the thing is that they are not. the kernel chooses to REPLACE my mapping. consider the user-space code: mmap(0xaaaa0000, 0x3000, MAP_FIXED, ...); mmap(0xaaaa1000, 0x4000, MAP_FIXED, ...); here, the second call to mmap will shorten the first mapping to 0x1000 bytes and create one big vma with size 0x5000 bytes. is there a way to tell it that the second mmap MUST fail? -

Hi everyone! I have a strange problem where I don't even know if there is a solution to it at the moment: I'm working on a new way of doing memory-management and currently I allocate memory mappings (at non-fixed locations) in user-memory (i.e. < TASK_SIZE) in addition to the regular pages mapped for the apps (e.g. heap memory, etc.) It all works perfectly well (creating & deleting the additional mappings), however, when the kernel feels like it needs to allocate a mapping in user-space ...

On Thu, 30 Aug 2007 09:41:41 -0700 (PDT) have you seen this? module loading is pretty much serialized, so that no 2 modules are being loaded in parallel... -

No, if the switch doesn't suport NP-IV, the 24XX f/w will come up in Public Loop mode (FL). -

-----Original Message----- From: Matthew Jacob [mailto:lydianconcepts@gmail.com] Sent: Thursday, August 30, 2007 9:47 AM To: Seokmann Ju Cc: Abhishek Kane; linux-kernel@vger.kernel.org; Linux Driver; linux-scsi@vger.kernel.org port and it has implemented based on the multiport ID (MID) capability Yes, NPIV is optional, if the switch is NPIV capable then you can create virtual ports, otherwise it works as it does today. NPIV is a standard technology that allows sharing a single physical ...

From: David Howells <dhowells@redhat.com> Initialise s_flags in get_sb_mtd_aux() from the flags parameter. Signed-off-by: David Howells <dhowells@redhat.com> --- drivers/mtd/mtdsuper.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/drivers/mtd/mtdsuper.c b/drivers/mtd/mtdsuper.c index aca3319..9b430f2 100644 --- a/drivers/mtd/mtdsuper.c +++ b/drivers/mtd/mtdsuper.c @@ -70,6 +70,8 @@ static int get_sb_mtd_aux(struct file_system_type *fs_type, int flags, ...

If the ATTR_KILL_S*ID bits are set then any mode change is only for clearing the setuid/setgid bits. For CIFS, skip the mode change and let the server handle it. Signed-off-by: Jeff Layton <jlayton@redhat.com> --- fs/cifs/inode.c | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index dd41677..279f3c5 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1538,6 +1538,11 @@ int cifs_setattr(struct dentry *direntry, struct iattr ...

This one looks fine to, but you'll probably need an additional patch to ecryptfs so it clears ATTR_MODE before calling notify_change for the lower inode. -

Make notify_change not clear the ATTR_KILL_S*ID bits in the ia_vaid that gets passed to the setattr inode operation. This allows the filesystems to reinterpret whether this mode change is simply intended to clear the setuid/setgid bits. This means that notify_change should never be called with both ATTR_MODE and either of the ATTR_KILL_S*ID bits set, since the filesystem would have no way to know what part of the mode change was intentional. If it is called this way, consider it a ...

On Thu, 30 Aug 2007 11:46:50 -0300 any specific reason you're using ndiswrapper rather than the native linux driver for this device? Your oops is all over the ndiswrapper code..... -

[ Adding relevant Cc:'s ] -

I have two systems exhibiting a similar problem, even after swapping out hard drives. Both systems have Intel ICH8 controllers. One is a server class board with a Xeon, the other is a desktop class board with a Core Duo. Both are Intel brand boards. The server box is unable to use a drive in it's 4th, 5th or 6th SATA port. The drive appears fine, and works fine for a few minutes, but eventually starts spitting out errors: [6494068.025887] ata6: SATA link up 1.5 Gbps (SStatus 113 SControl ...

Hi, I am trying to boot a current 2.6 kernel as a guest under XEN in HVM mode. With linux-2.6.22.(5) as the guest this works fine but current git kernels fail very early in the boot process. The XEN host is a debian kernel with XEN enabled (2.6.18-4-xen-686). The XEN guest is a vanilla kernel from kernel org booting with LILO 22.6.1 in HVM mode. I bisected the problem and the bisection points at this commit: | # bad: [4fd06960f120e02e9abc802a09f9511c400042a5] Use the new x86 setup code ...

All of this point to a bug in the Xen emulation host. You're not saying if you're on an Intel or AMD host, but on Intel, hardware virtualization doesn't actually kick in until you're in protected mode with all segments configured properly. This means all of pmjump.S runs in an interpreter. A bug in that interpreter would cause what you observe. -hpa -

Well, rpcsec_gss isn't inherently password based, and you can authenticate in some way that doesn't actually give away your password And in theory a kernel could provide *some* guarantees against root, right? (Is there some reason a unix-like kernel must provide such things as /dev/kmem?) --b. -

BTW: even when this task is done, a creative root can still find ways to subvert the security (he can read /dev/mem, replace the kernel with a compromised one, ....). The bottom line is that if you can't trust root, don't even log in. Trond -

NFSv3 should work fine with krb5 too, but that won't solve your problem with setuid: kerberos saves the TGT in a file on /tmp, so root can still suid and grab your cred (and the same goes for CIFS). We've got people working on fixing this problem using David Howells' keyrings, but it will probably be a while until we've solved all the upcall issues, and it will probably take even longer to push the kerberos changes back to the official MIT etc distros. Cheers Trond -

Hm? I do not see this problem with CIFS. The user may have local root, but on the server, he only has his non-root account on the server, and as such, can only operate on the server using this non-root fsuid. Did I miss something? (Especially the /dev/mem thing Jan -- -

If you don't know where the list is created, how do you now that it contains a ramdisk tag? There is a chance, that the kernel does not even get a ram disk tag at all. This would also explain, why it is not parsed ;-) The tag list is supplied by your bootloader. How the bootloader creates the list and how to change the order depends on your bootloader. You probably have to read (and patch) it's source code. regards, Uli -- ------- ROAD ...the handyPC Company - - - ) ) ) Uli ...

Hi Uli, thanks for your reply. just have a question, where is the tag list, how can i modify this? regards, yang -

Hello, I have a very simple question about i2c transfers. I'm wondering if I'm allowed to initiate some very short i2c transfers in an interrupt handler. Thanks for your answers. -- Francis -- Francis -

Andrew, do you still strongly oppose to having ARCH_HAS_RANDOMIZE_BRK macro instead please? Thanks, -- Jiri Kosina SUSE Labs -

you might want to use a weak function instead ? void __weak arch_randomize_brk(void) { /* nada by default */ } and overload it only for x86 architectures by now. BTW, how about minimize the arch specific code by slighly changing its prototype ? unsigned long arch_randomize_brk(struct mm_struct *mm) { unsigned long range_end = mm->brk + 0x02000000; return randomize_range(mm->brk, range_end, 0) ? : mm->brk; } Franck -

does it really make sense to stick stubs into no-mmu ports which cannot utilize the ELF binfmt ? -mike -

Hi all, I may completely miss out on this one, but what's the point in knowing? --=20 $ cat ~/.signature Freddy Spierenburg <freddy@dusktilldawn.nl> http://freddy.snarl.nl/ GnuPG: 0x7941D1E1=3DC948 5851 26D2 FA5C 39F1 E588 6F17 FD5D 7941 D1E1 $ # Please read http://www.ietf.org/rfc/rfc2015.txt before complain!

Here's a quick answer, not the best, but quick. I took a user space flash memory driver I'm doing at work and compiled it on my R5000 at home using gcc 4.1 and the MIPS3 abi, stopping with a .o file. I also ran the source through cpp and a couple of grep passes to strip out junk that wasn't really code. This driver may be somewhat typical of what you would run into as it has quite a few inline functions and such. The driver.o was about 23000 bytes. Forgetting about the ...

make allyesconfig make -- Jesper Juhl <jesper.juhl@gmail.com> Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html -

Why don't you simply count the number of non-blank non-comment lines in the source files that you are building, build the kernel and then count the number of instructions in the resulting binary ? -- Jesper Juhl <jesper.juhl@gmail.com> Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html -

According to my supplier, herre is the data sheet for the DOMs: http://www.pqimemory.com/documents/domdata.pdf PIO mode 2 is mentioned. Even DMA seems to be supported. Or am I mistaken? Could there be a bug in my south bridge? Regards. -

Using SMP kernel 2.6.22.6pre-CFS-v20.5 on Pentium D (IA-32). I think this bug (or whatever you want to call it) got triggered when you first allocate several megabytes of memory in a kernel module and then free them, and then run e.g. X and when memory gets tight, you end up with this situation... Top 2 /proc/vmstat Biggest Winners: pgrefill_normal:49900/second pgrefill_high:20810/second $ vmstat 1 procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------ r b ...

hello, with kernel 2.6.22.5 using the new pata_pdc202xx_old driver it doesn detect the cable right on this seagate drive (i tried switching ports / cables ...etc) it detects the maxtor drive fine no matter what port its on, the seagate doesnt detect correctly on either port ata1.00: ATA-6: ST3120026A, 3.06, max UDMA/100 ata1.00: 234441648 sectors, multi 16: LBA48 ata1.00: limited to UDMA/33 due to 40-wire cable ata1.00: configured for UDMA/33 ata2.00: ATA-7: Maxtor 6B200R0, max ...

hmmh, not sure about this ... Russell? regards, Andre -

No idea. The support for external clocking came from someone else (I was never able to work it out myself). I can only assume that the code as it stands does work. So let's remain bug-compatible until there's a proven problem. -- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: -

From: Andre Haupt <andre@finow14.de> Signed-off-by: Andre Haupt <andre@finow14.de> --- Index: linus/drivers/atm/idt77252.c =================================================================== --- linus.orig/drivers/atm/idt77252.c 2007-08-30 11:33:22.000000000 +0200 +++ linus/drivers/atm/idt77252.c 2007-08-30 11:37:00.000000000 +0200 @@ -2092,7 +2092,7 @@ idt77252_rate_logindex(struct idt77252_d { u16 afp; - afp = idt77252_int_to_atmfp(pcr < 0 ? -pcr : pcr); + afp = ...

Likely. That is the one that will break the user-space automounter as well (and keeps me from .23). I don't care very much about what the default is, but it would be great if the new behaviour could be globally changed at run- (or boot-) time. It will be some time until the new mount option makes it into the distros. Cheers Martin PS: Sorry, but I likely killed the CC list ------------------------------------------------------ Martin Knoblauch email: k n o b i AT knobisoft DOT ...

On Thu, 30 Aug 2007 11:26:04 +0200 It would probably look a lot cleaner if you either provided dummy dma_alloc_coherent/free coherent inlines for SPARC32, fixed it (if sparc32 can do coherent DMA) or if you must ifdef it provide dummy functions in the memalloc code so its a single ifdef -

At Thu, 30 Aug 2007 11:09:39 +0100, It's been indeed inline on 2.6.22 or ealier in include/asm-generic/dma-mapping.h, but it's changed to extern without defining functions. I don't think it's a good idea to have function declarations even though we surely know that there are no real function definitions... Takashi -

I don't see the point of sending patches for old Linux versions such as 2.6.21, unless it's something applicable to the -stable tree. There's no kabi. Michal -

Ah, good, sorry for noise then. In that case I don't think it makes sense to move sched_exit() to do_exit(), of course. This doesn't look suitable for the -stable tree. Oleg. -

might as well make that pid_t, or maybe even a struct pid* and keep a Other than that it looks good, pretty much what I suggested :-) -

Short-living process returns its timeslice to the parent, this affects process that creates a lot of such short-living threads, because its not a parent for new threads. Patch fixes this issue and doesn't break kabi as does the patch from reporter: http://lkml.org/lkml/2007/4/7/21 An example and script for systemtap were modified a bit. Patch was tested on 2.6.21 with results: Pass 1: parsed user script and 54 library script(s) in 330usr/10sys/340real ms. Pass 2: analyzed script: 3 probe(s), 10 ...

I personally am unlikely to make the time to test a specific HBA driver. But perhaps James or one of the other scsi developers (or any developer at all) might like to volunteer to help out with testing, in which case yes, please do send out a card or two. First up, please become familar with the way in which we handle code patches. I'd suggest that you subscribe to the linux-kernel or linux-scsi mailing lists for a while, watch how other people prepare and present their ...

Hi Linus, Please consider pulling from: git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git for-linus or master.kernel.org:/pub/scm/linux/kernel/git/dtor/input.git for-linus to receive updates to the input subsystem. Changelog: --------- Adrian Bunk (1): Input: psmouse - make dummy touchkit_ps2_detect() static Andrew Morton (1): Input: iforce - fix 'unused variable' warning Fernando Luis Vázquez Cao (1): Input: i8042 - fix retrun value of ...

There are designers, and there are builders. Can you tell me who is more important? Thanks! -- Al -

I might agree with you on this wrt hacking around the kernel, but when it comes to introducing new subsystems, then we have a two fold situation: 1. Designing the internals of the new subsystem 2. Interfacing it with the rest of the kernel Part 1 is completely independent of the implementation, it's part 2 that needs intricate implementation knowledge. We recently had an example of this: kexec based hibernation So, what's wrong with tapping into people's design suggestions, and ...

I really didn't expect this kind of a response, but I guess when somebody runs out of arguments, then that's probably one of the responses to expect. Really sad. Thanks anyway! -- Al -

For most users the goal should be to automate behaviour. It should just work. -

Or lazyness on behalf the vendor. You know, some even don't fix their DSDT... Jan -- -

Depends on the materials and connector. The required electrical characteristics for any cable are in the spec, and if you want to know more you'll need to know a lot more than I do about transmission line theory! -

Hard to tell, since some (most?) laptops have sort of a backplane and there might be no real cable you could see because it's all mainboard wire paths already. Jan -- -

We really only care about the first two bus masters (playback and capture). There's no need to have unused BM code lying around, so let's get rid of it. If for some reason we trigger an IRQ for some BM that we're not using.. well, that warrants spitting out an error message (imo). Signed-off-by: Andres Salomon <dilinger@debian.org> --- sound/pci/cs5535audio/cs5535audio.c | 24 +++--------------------- sound/pci/cs5535audio/cs5535audio.h | 31 +------------------------------ 2 files ...

ACcording to 6.3.2.7 of the cs5535/cs5536 data sheets, the ACC_BM[x]_CMD registers are only 8 bits wide. This driver treats them as 32 bits wide, and also has bits in the wrong place. Simple fix to the definitions. Signed-off-by: Andres Salomon <dilinger@debian.org> --- sound/pci/cs5535audio/cs5535audio.h | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/sound/pci/cs5535audio/cs5535audio.h b/sound/pci/cs5535audio/cs5535audio.h index c7a2044..516219a ...

-- Jordan Crouse Systems Software Development Engineer Advanced Micro Devices, Inc. -

-- Jordan Crouse Systems Software Development Engineer Advanced Micro Devices, Inc. -

-- Jordan Crouse Systems Software Development Engineer Advanced Micro Devices, Inc. -

These look good - and they should work fine on the 5535 - none of the patches really change how we touch the hardware, just when we touch the hardware. 5536 is by far the more popular chip now, so I recommend we put these in, and if by some unlucky coincidence a 5535 bug pops up, we'll deal with it when it happens. Jordan -- Jordan Crouse Systems Software Development Engineer Advanced Micro Devices, Inc. -

The patches look good. I'm traveling till Sep 2nd so can't test on 5535 yet. Thanks very much for your efforts Andres. Thanks, jaya Acked-by: Jaya Kumar <jayakumar.alsa@gmail.com> -

From: Divy Le Ray <divy@chelsio.com> cxgb3 used netdev_priv() and dev->priv for different purposes. In 2.6.23, netdev_priv() == dev->priv, cxgb3 needs a fix. This patch is a partial backport of Dave Miller's changes in the net-2.6.24 git branch. Without this fix, cxgb3 crashes on 2.6.23. Signed-off-by: Divy Le Ray <divy@chelsio.com> --- drivers/infiniband/hw/cxgb3/cxio_hal.c | 2 - drivers/net/cxgb3/adapter.h | 2 + drivers/net/cxgb3/cxgb3_main.c | 126 ...

There are patches in -mm for revokeat()/frevoke(), which can be used to implement exactly that. If a device "vanishes" (CD is removed in the middle of loading, USB pend rive yanked out the middle of I/O, NFS server thats gone MIA), A user-space program (maybe HAL) could iterate over the open files and revoke() them, at which point the system can be cleanly unmounted. -

Yeah, sure. But these days disk space is cheap and I tend to put too big swap partitions, and I always regret it later ... Xav -

Normally mlockall is what is used in this sort of situation, that way it doesn't force all swapped data in for every app. It's possible that calling this with lots of swapped pages in the app at the time may have the same problem though. -- Robert Hancock Saskatoon, SK, Canada To email, remove "nospam" from hancockr@nospamshaw.ca Home Page: http://www.roberthancock.com/ -

If the swap device is full, then there is no need for random seeks as the swap pages can be read in disk order. A not so full swap will skip over the unused areas, the time needed should still be limited to the time needed for reading the whole swap device. If this optimization is worth it is another problem though. Helge Hafting -

Good point. This thus-far-undescribed module could make its internals externally visible via one of the kernel's many register_foo() interfaces, but it would be a buggy module if it was doing register_foo(my_foo) before my_foo() was ready to be called. -

How about this one: diff -uprN linux/sound/pci/hda/hda_codec.c linux.new/sound/pci/hda/hda_codec.c --- linux/sound/pci/hda/hda_codec.c 2007-08-30 09:27:42.000000000 +0800 +++ linux.new/sound/pci/hda/hda_codec.c 2007-08-30 09:27:50.000000000 +0800 @@ -514,7 +514,11 @@ static int read_widget_caps(struct hda_c static void init_hda_cache(struct hda_cache_rec *cache, unsigned int record_size); -static inline void free_hda_cache(struct hda_cache_rec *cache); + +static inline void ...

I think the reported meant interrupt driven involuntary preemption. So ret_from_intr() is not the right place to look. But afaict you're still right, see how preempt_schedule*() adds PREEMPT_ACTIVE to the preempt_count, and how that makes the scheduler ignore the task state. -

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=75180df2ed... This (and it's related patches) may be the problem. I can probably tell if you post your map or if you strace the automount process managing the a problem mount point and look for mount returning -

Maybe but there is an NFS change that appears to be in the current 2.6.23-rc kernel and doesn't seem to be in 2.6.22 that is known to break some autofs maps and also breaks amd. I can't seem to locate the commit just now. In the meantime what version of user space autofs and nfs-utils are you running? And can you post your autofs maps please? Ian -

Hey, I obviously approve. And the patch looks simple. Feel free to re-submit after 2.6.23 is out the door, I don't think anybody will really complain. Any NFS user will know why something like this can be really nice. Linus -

Won't this change just cause functions like do_sync_read() and -

Hi, Well, I don't have ppc cross-compiler... so I can't try this for you. What I can tell you is that it doesn't seem like there was any change regarding those functions. But even though, you're selecting the needed Kconfig Macros. -- Best Regards, Felipe Balbi felipebalbi@users.sourceforge.net -

The tragic bit is that we were sold similar DOMs in 2007... Double word is 32 bits, right? Isn't "Cannot perform double-word IO" in contradiction with the following statements? IO_support = 1 (32-bit) Buffer size: 1.0kB bytes avail on r/w long: 4 When you say "the current libata IDE" do you mean PATA_VIA (in my case)? I've avoided this driver because it is marked EXPERIMENTAL. Would there What do the warnings mean? :-) Regards. -

OK, didn't realize it had to be beneath the root node, and that the truncation truncated the front and not the back. I would have done it differently, but this should work. Acked-by: Joel Schopp <jschopp@austin.ibm.com> -

Yes, well, but no. The location code is useful as a geographical location: slots and devices are physically labelled with stickers so you can tell which is which. Handy when you have to unplug stuff. By contrast, the device-tree full_name is mostly just gobldy-gook, with some crazy phb numbering in there that, after four years of staring at them, I still can't reliably do anything useful with. Location codes are nice. --linas -

Nope. Userspace programs should not depend on ibmebus' way of naming the devices; especially since some overly long loc_codes tended to be truncated and thus rendered useless. I have tested IBM's DLPAR tools While there's no such guarantee (as in "officially signed document"), yes, I expect future GX devices to also appear beneath the OFDT root node. For the existing devices, the unit addresses are already part of the device name, so I save the need to use sprintf() again. Plus, I ...

At Wed, 29 Aug 2007 17:34:19 +0200, The problem is that nid = 0 is used. So, your patch just hides another bug. Please check whether the bug still exists with git.kernel.org perex/alsa.git mm branch. If still there, build with CONFIG_SND_DEBUG and show the autoconfig output. thanks, -

I just compiled 2.6.23-rc4 on this setup to check if we made any progress. After reboot i see: [SNIP] NET: Registered protocol family 17 Using IPI Shortcut mode Freeing unused kernel memory: 148 freed init[1]: segfault at 85890000 eip 85890000 esp bfcfcb8c error 4 init[1]: segfault at 85890000 eip 85890000 esp bfcfcb8c error 4 init[1]: segfault at 85890000 eip 85890000 esp bfcfcb8c error 4 init[1]: segfault at 85890000 eip 85890000 esp bfcfcb8c error 4 init[1]: segfault at 85890000 eip ...

binutils 2.15.95, gcc 3.3.6 and I could update to 4.0.4 or something more recent I guess. And yes, it's only alpha. Of which file do you want the objdump? Jan -

Could you give me an asm dump via objdump of one of these functions? I wonder what is going on there? Seeing the code generated may give us a hint what is going on. Likely an old compiler that has troubles performing constant folding <sigh>. One solution would be to use a newer compiler? And yes, the page allocator pass through patch in mm would fix this. Or define CONFIG_BROKEN_CONSTANT_FOLDING for alpha and then use this patch: --- include/linux/slub_def.h | 4 ++++ 1 file ...

Robert, just checked it out. At least with the "6i", you do not want to disable the WBC :-) Performance really goes down the toilet for all cases. Do you still have a pointer to that bulletin? Cheers Martin ------------------------------------------------------ Martin Knoblauch email: k n o b i AT knobisoft DOT de www: http://www.knobisoft.de -

There are some issues with us using mlockall. Admittedly, most/all of them are not the kernels problem (but a fast swapoff would be a good workaround): We're using python 2.4, so mlock() itself isn't really an option (we don't realistically have access to the address regions hidden behind the language). mlockall() is a possibility, but the fact that all allocations above a particular limit will fail would potentially cause us problems given that it's hard to control python's memory usage for ...

Andrew rewrote swapfile support in 2.5, making it use FIBMAP at swapon time: so that in 2.6 swapfiles are as deadlock-free and as efficient (unless the swapfile happens to be badly fragmented) as raw disk partitions. There's certainly scope for a study of I/O patterns in swapping, it's hard to imagine that improvements couldn't be made (but also easy to imagine endless disputes over different kinds of workload). But most people would appreciate an improvement in active swapping, and not care ...

On Thu, 30 Aug 2007 04:07:11 +0530 I have no strong opinion. But how about Mega bytes ? (too big ?) There will be no rounding up/down problem. -Kame. -

updated patch based on Jeff Garzik's comments. - check adapter firmware version and use appropriate interface accordingly - add new PCI device IDs and use PCI_VDEVICE macro - update driver version string - remove unused data structures - remove unnecessary typecasts Signed-off-by: HighPoint Linux Team <linux@highpoint-tech.com> --- drivers/scsi/hptiop.c | 63 +++++++++++++---- drivers/scsi/hptiop.h | 229 ++------------------------------------------------------------- 2 files changed, ...

Tested here, works fine. Luca -

i've just tested the minimal fix here: http://bugzilla.kernel.org/show_bug.cgi?id=8958#c7 rgds -daniel -

Not at all. all sysctls are under /proc/sys/ -

Ah, right. Good point. (I was thinking of the problem of splitting /proc/$PID directories from the rest of the stuff in /proc.) Rob -- "One of my most productive days was throwing away 1000 lines of code." - Ken Thompson. -

It's not hard to do read-side, right? Take the list of sysctl's, and create a program which reads it via the binary interface and the /proc interface, and verify they are the same. Testing write-side, where we have to worry about permission tests, making sure the correctr value is set, locking issues, etc., is admittedly more difficult. My guess though many programs/libraries are reading from the sysctl interface than writing to it. - Ted -

Yes. So all we really need to do is split them apart and do the nfs multiple mount thing to maintain backwards compatibility when mounting proc. The code is close enough to split apart now that it probably would not be to difficult to finish the job. Eric -

Ahh. Great. Keep me posted. -

[2/4] ext2: fix rec_len overflow - prevent rec_len from overflow with 64KB blocksize Signed-off-by: Takashi Sato <sho@tnes.nec.co.jp> Signed-off-by: Mingming Cao <cmm@us.ibm.com> --- fs/ext2/dir.c | 46 ++++++++++++++++++++++++++++++++++++---------- include/linux/ext2_fs.h | 13 +++++++++++++ 2 files changed, 49 insertions(+), 10 deletions(-) Index: linux-2.6.23-rc3/fs/ext2/dir.c =================================================================== --- ...

The next 4 patches support large block size (up to PAGESIZE, max 64KB) for ext2/3/4, originally from Takashi Sato. http://marc.info/?l=linux-ext4&m=115768873518400&w=2 It's quite simple to support large block size in ext2/3/4, mostly just enlarge the block size limit. But it is NOT possible to have 64kB blocksize on ext2/3/4 without some changes to the directory handling code. The reason is that an empty 64kB directory block would have a rec_len == (__u16)2^16 == 0, and this would cause an ...

Hmmm... So we need to check the page size and generate multiple requests in submit_bh? -

My current view of the IO stack is the following: -------------- -------------- ------------ ---------- ---------- --------- |NET_PCI_BACK| |BLK_PCI_BACK| |9P_PCI_BACK| |NET_FRONT| |BLK_FRONT| |9P_FRONT| -------------- -------------- ------------ ---------- ---------- --------- ------------- --------------- ------------------- |KVM_PCI_BUS| |hypercall_ops| |shared_mem_virtio| ------------- ...

O.K. as a first step i'll limit 5210 code only then, just an option like "support older 5210 chipsets" which is going to be off by default instead of 3 options. It's not just saving space, it's also saving some runtime checks. It's not really a gain in performance though, most checks are done during initialization and dfs setup, i just thought it would be usefull to save as much cpu as possible. -- GPG ID: 0xD21DB2DB As you read this post global entropy rises. Have Fun ;-) Nick -

If you're doing these checks in a hotpath something is badly wrong with your architecture. -

On Thu, Aug 30, 2007 at 10:26:52AM +0200, Jarek Poplawski wrote: ...or maybe it's OK... Sorry. Jarek P. -

In theory, that could work, but in practice relying on functions that the Linux kernel offers in GPLv2-only headers etc. will make the result GPLv2 anyway, and disentangling it would be a nightmare. johannes

Since I didn't apply 1/5, this one didn't apply either. It seems fine overall, so if you rediff I'll be happy to apply. John -- John W. Linville linville@tuxdriver.com -

Sorry, I've lost the original patch from two separate mailboxes... The proper behavior depends on how we end up tying filesystems to user namespaces, which isn't actually decided yet. The way I was recommending doing that was: A filesystem is tied to a user namespace. If a uid in another naemspace is to be allowed to access the filesystem, it will actually - through a key in it's keyring (which acts like a capability) - be mapped to a uid in the filesystem's uid namespace. So in Eric's ...

Maybe before proceeding further with the discussion I'd like to understand following: What are these user namespaces supposed to be good for? I imagine it so that you have a machine and on it several virtual machines which are sharing a filesystem (or it could be a cluster). Now you want UIDs to be independent between these virtual machines. That's it, right? Now to continue the example: Alice has UID 100 on machineA, Bob has UID 100 on machineB. These translate to UIDs 1000 and 1001 on ...

Just fyi Eric, Note that given the amount of churn going on due to pid and network namespaces, I was seeing completion of user namespaces as something to be done sometime next year. In the meantime I was only going to do something with capabilities to restrict root in user namespaces (which I think will take the form of per-process non-expandable cap_bsets, which I plan to start basically right now). But I'll gladly do the userns enhancements earlier if it's actually wanted. They promise ...

There can be arbitrary number of listeners (potentially from different namespaces if I understand it correctly) listening to broadcasts. So I think we should pass some universal identifier rather than try to find out who is listening etc. I think such identifiers would be useful for other things too, won't they? BTW: Do you have some idea, when would be the infrastructure clearer? Whether it makes sence to currently proceed with UIDs and later change it to something generic or whether I ...

Hi Darrick, OK, that's a bit different then, but I'm still not sure that there is much value in exporting these values in sysfs, in particular if there is no way to reset them. I am also not happy with the names you proposed: power1_max_input and power1_min_input are somewhat confusing IMHO, I'd suggest power1_input_highest and power1_input_lowest to make them clearly different from the min and max limits we have for other sensor types. If we have them at all, of course. -- Jean ...

Christoph that code was written over 6-7 years ago, feel free to provide a patch for it to use the slab allocator, we only turn this code on though when developing new drivers, I don't think it ever gets used outside of that.. we don't need the overhead of the memory tracking in normal use.. Please stop implying that all the code in the drm is magically new and that we should be reusing kernel infrastructure that wasn't even written when the code first appeared.. or provide ...

So that you can actually pass in a const struct task_struct * without having to cast it back to [non-const]. Why one would have a const struct task_struct * in the first place is a different matter. But see http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=77adbfbf4c... Jan -- -

"those callers". There was _exactly one_ caller, and that was an out-of-tree module. There were not any in-kernel callers before, and it did not generate any warning. That is perhaps why no one had constified it before me. This does I think it is equally 'necessary'. Jan -- -

Or we can use a vprintk-like function to parse the format string and serialize the information into trace buffers. I prefer this latter option because, overall, it will localize the probes in a few bytes of functions instead of duplicating the memory and instruction cache required by multiple serializing functions. I have the code ready, but I do not want to flood LKML with patches neither.... Mathieu -- Mathieu Desnoyers Computer Engineering Ph.D. Student, Ecole Polytechnique de ...

Andrew, any chance to get this into -mm ASAP so we can have it in 2.6.24? Just in case anyone wonders what this is usefulfor I've ported my hacking spu tracing code to it, and if markers get in I'd push a cleaned up version of this in the tree of the benefit of everyone trying to follow what's going on in the spufs code. Similarly I'd like to port the xfs tracing code over to it which is very helpful in trying to debug filesystem issues. Note that in this patch the actual logging code is ...

In fact no: although CONFIG_SUSPEND and CONFIG_HIBERNATION were certainly involved in the logic which arrived at missetting CONFIG_HOTPLUG_CPU, they were correctly prompted for themselves. Hugh -

Thanks for working that out: your patch certainly works for me, and would be good to see in 2.6.23, to prevent lots of users getting an unnecessary CONFIG_HOTPLUG_CPU forever after. I wonder what config But please add a sentence to your patch description, something like: make oldconfig from a 2.6.22 .config with CONFIG_HOTPLUG_CPU not set was in some configurations setting CONFIG_HOTPLUG_CPU=y without asking, even when there was no actual requirement for ...

Yes, but a common error is "could not mount root", which is way after the keyboard has been initialized. *In fact*, you can start initramfs, have your PgUp/PgDown keys available, and when you exit out of initramfs without a valid root filesystem, you lose the keys again. Jan -- -

Except that __buf goes out of scope before printk() is called... Jan -- -

Or even pr_debug. Hey, can we make this const, if it does not change at runtime? Jan -- -

The problem is that we don't really have a maintainer for the aic7xyz drivers any more. Volunteers welcome. NOT IT! -- 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." -

Take it before someone else does! Jan -- -

One must consult the documentation of that device. This wear leveling is low level and most devices do not export any information about it. Recent SSDs start to export some values through SMART that let you monitor the state. Some companies think that hiding is better than exposing... -

Well it's just 'breaking' old behavior in certain error cases. I.e. if audit fails, or no handler is found for the binary, we never reach compute_creds (which is called from within the binary loader), so in the past keep_capabilities would have remained 1 until something was actually executed. Now in all likelyhood the process would try to exec something else, but if it should happen to decide to setuid() instead, with your patch keep_capabilities will have been unexpectedly set to 0 during ...

Thank you very much for your response Lennart, but unfortunately this did not solve my problem - the system (got four machines in my test setup) still generate the error and make the kernel wait forever. regards Espen M. Rutger -

We don't care if ebp is on the stack, we care about ebp + 4. Without this, lguest (with CONFIG_DEBUG_LOCKDEP) can touch a page unmapped by CONFIG_DEBUG_PAGEALLOC. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> diff -r b0b1ab8ecf48 arch/i386/kernel/traps.c --- a/arch/i386/kernel/traps.c Fri Aug 31 03:25:06 2007 +1000 +++ b/arch/i386/kernel/traps.c Fri Aug 31 07:57:35 2007 +1000 @@ -100,7 +100,7 @@ print_context_stack(struct thread_info *tinfo, unsigned long addr; ...

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 glibc does nothing but call the code in the vdso. We have a function pointer variable which either has the old vsyscall value or the address of the function in the vdso. Everything else is identical. Unless the interface of the vdso function is different (which it shouldn't) I don't think you can blame glibc. - -- ➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 ...

Just found this duplicated code in 2.6.23-rc4, maybe it was supposed to be something else? The second one was added in the x86_64 vdso patch. arch/x86_64/kernel/vsyscall.c: void update_vsyscall(struct timespec *wall_time, struct clocksource *clock) { unsigned long flags; write_seqlock_irqsave(&vsyscall_gtod_data.lock, flags); /* copy vsyscall data */ vsyscall_gtod_data.clock.vread = clock->vread; vsyscall_gtod_data.clock.cycle_last = ...

I have only ever used: git clone git://git.kernel.org/pub/scm/linux/kernel/git/jwessel/linux-2.6-kgdb.git Jason. -

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Trying: git clone http://master.kernel.org/pub/scm/linux/kernel/git/jwessel/linux-2.6-kgdb.git - - -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFG1gnFJICwm/rv3hoRApOoAJ9BHXLsIuxDiOCaAFRfAZGwrDXATQCeLL3O bxtr3qz0soPRghPmtSZgOqc= =kQd1 -----END PGP SIGNATURE----- -

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Why am I getting this when I do: git clone http://master.kernel.org/pub/scm/linux/kernel/git/jwessel/linux-2.6-kgdb.git - ---------------------------------------------------------------------------- error: Couldn't get http://master.kernel.org/pub/scm/linux/kernel/git/jwessel/linux-2.6-kgdb.git/refs/tags... for tags/v2.6.11 The requested URL returned error: 404 error: Could not interpret tags/v2.6.11 as something to pull rm: cannot ...

Welcome to the wonderful world of SCSIfying ATA. (Don't talk about ATAPI, USB/Firewire, it's a different matter.) Jan -- -

USB / Firewire / FC / iSCSI are all SCSI transports and fit within the SCSI subsystem by design. ie. Just like ethernet, DSL, T-1, etc can all carry IP traffic with no conceptual conflict, many media by design carry SCSI traffic. The PATA and SATA physical layer typically carry ATA commands and having them tied into the SCSI stack is an aberration that I hope will be eliminated some day. ATAPI is an exception. Not sure where that would end up in a perfect world. Greg -- Greg ...

I guess eventually all disks will appear the same, just like on BSD and many other systems (probably most other systems). :) -- Len Sorensen -

Is suspend the right thing? The fact that time keeping seems to be reinventing the standard infrastructure for shutting down instead of just reusing it has me a bit confused. Regardless hooking into the shutdown is the right thing However that works. I just know that in the device tree we have shutdown methods that are called on reboot and kexec and the hpet needs one, and that is where the code belongs. I was assuming that CLOCK_EVT_MODE_SHUTDOWN just mapped to the shutdown method of ...

Ok. It looks like you understand this issue. Can you please try calling hpet_disable_int from hpet_set_mode under CLOCK_EVT_MODE_SHUTDOWN. I haven't traced the clock event methods all of the way through but as a first approximation I think that will get things called at the appropriate time with out needing to patch machine_kexec. Which is very much the wrong place to add call any hpet code from. We may also need to make the hpet initialization more robust so we can do something sane ...

There isn't that much that is duplicated - and there are also bits of the /proc/PID/mem code that are not needed in this case, so I'm not really sure if it is worth doing. I did submit a patch a few months ago - see: <http://marc.info/?l=linux-kernel&m=117862109623007&w=2> James Pearson -

got 45min on mini before I had to go elsewhere the amd64 shutdown fine and has been up for more than 3 hours -

Sysfs files support poll, if they are implemented that way. A driver can wake up listeners that wait in poll() for changes. http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=4508a7... Kay -

We have the events that can be sent over all devices, actually (we can use button events for this). But blocking on read might indeed require a separate input device... or a file, or something like that. And we *definately* should implement something like proper select() and poll() support for the position sysfs attribute. HDAPS has also some other crap that needs to go over the input device (mouse activity, keyboard activity flags). ams and hpmdp have the "possible shock imminent" events, ...

What's wrong with the stuff I did in mdps? a misc character device that I also looked at what you did in the patches as well as the modified hdapsd. I'm doing the raw input device right now in the mdps, but I have a suggestion. It seems to me that right now there are at least 3 drivers that provide the same functionality - hdaps, ams and mdps. Why not create the input device that exports raw accelerometer data with a name that is generic - something like accel/input or something along ...

and note that the schedule() call in there is not part of the crash it just happened to be on the kernel stack. Nor is the do_IRQ() entry real. Both are frequent functions (and were executed recently) that's why they were still in the stackframe. Ingo -

On 08/29/2007 09:56 PM, Rene Herman wrote: Rene. -

From: Andreas Gruenbacher <agruen@suse.de> dput must be called before mntput here. Please consider for 2.6.24 and stable. Signed-off-by: Andreas Gruenbacher <agruen@suse.de> Acked-By: David Howells <dhowells@redhat.com> --- fs/afs/mntpt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -235,8 +235,8 @@ static void *afs_mntpt_follow_link(struc err = do_add_mount(newmnt, nd, MNT_SHRINKABLE, &afs_vfsmounts); switch (err) { ...

These changes need some fixing .. The "struct stack_trace" infrastructure is based on CONFIG_STACKTRACE , so at least you would want to ifdef your changes on that config options .. I modified you patch to include the ifdefs, ftp://source.mvista.com/pub/dwalker/misc/debug-preempt-5-levels-of-stack-frames.patch The other problem I noticed is that the save_stack_trace(trace); seems kind of heavy weight to be calling from inside add_preempt_count. If your only going to a depth of 5 you could ...

Natalie Protasevich wrote: I agree. ...to track very concrete feature-related TODO items. One Issues & Resolutions Database which I occasionally work with indeed distinguishes between "bug", "new feature", "improvement", "task", whatever that means. The type of issue is the first question one is asked when one enters a new issue. 'My' Linux kernel subsystem has a long virtual TODO list, with bugs, missing features, and other kinds of items in it. Since there is almost no personnel, ...

Fine with me, my point was simply that they don't belong into a Is there any reasonable query that would be possible through SQL but You need some kind of first level support that does some initial debugging and assigns the bug into the right category, and that will cu Adrian [1] there are special cases where the kernel might deliberately not some specification -- "Is there not promise of rain?" Ling Tan asked suddenly out of the darkness. There had been ...

What still bothers me about your response is that, while you know the race exists and do not disagree with my example, you don't seem to see that that race can eventually lock up the block device by repeatedly It is a weakness all right. Well, - if (q && q->metric) { + if (q && q->metric && !bio->queue) { which fixes that problem. Maybe there is a better fix possible. Thanks for the catch! The original conception was that this block throttling would apply only to the highest ...

You are right. There are no drivers in the tree that can hit this, but it needs to be fixed up. I'll submit the following change: diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c index 0350071..bc18cbb 100644 --- a/crypto/async_tx/async_tx.c +++ b/crypto/async_tx/async_tx.c @@ -80,6 +80,7 @@ dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx) { enum dma_status status; struct dma_async_tx_descriptor *iter; + struct dma_async_tx_descriptor ...