Kernel Traffic #322 For 3 Sep 2005 By Zack Brown Table Of Contents * Standard Format * Text Format * XML Source * Mailing List Stats For This Week * Threads Covered 1. 26 Jul 2005 - 2 Aug 2005 (46 Scheduling Obsolete OSS Drivers For posts) Removal 2. 27 Jul 2005 - 1 Aug 2005 (10 Support For Sharp Zaurus SL-5500 LCD posts) Power Up/Down 3. 27 Jul 2005 - 29 Jul 2005 (26 Linux 2.6.13-rc3-mm2 Released posts) 4. 28 Jul 2005 - 2 Aug 2005 (18 Linux 2.6.13-rc4 Released posts) 5. 30 Jul 2005 - 31 Jul 2005 (19 Wireless Security Lock Driver To posts) Detect Physical Proximity 6. 2 Aug 2005 - 4 Aug 2005 (4 SquashFS Not NFS-Compatible; Fix On posts) The Way 7. 2 Aug 2005 - 3 Aug 2005 (3 New Documentation/kprobes.txt File posts) 8. 2 Aug 2005 (17 New Stable Review Cycle Started For posts) 2.6.12.4 9. 3 Aug 2005 - 4 Aug 2005 (5 SMB Maintainership posts) Mailing List Stats For This Week We looked at 2060 posts in 12MB. See the Full Statistics. There were 679 different contributors. 266 posted more than once. The average length of each message was 93 lines. +-----------------------------------------------------------------------------+ |The top posters of the | The top subjects of the week were: | | week were: | | |-----------------------+-----------------------------------------------------| |81 posts in 323KB by |78 posts in 319KB for "power consumption hz100, | |pavel machek |hz250, hz1000: new numbers" | |78 posts in 359KB by |63 posts in 311KB for "[patch 2.6.13-rc4] fix | |akpm@osdl.org |get_user_pages bug" | |61 posts in 310KB by |58 posts in 245KB for "revert yenta free_irq on | |ingo molnar |suspend" | |56 posts in 279KB by |45 posts in 217KB for "[patch] real-time preemption, | |steven rostedt |-rt-2.6.13-rc4-v0.7.52-01" | |55 posts in 169KB by |35 posts in 135KB for "[2.6 patch] schedule obsolete | |lee revell |oss drivers for removal" | +-----------------------------------------------------------------------------+ These stats generated by mboxstats version 2.8 1. Scheduling Obsolete OSS Drivers For Removal 26 Jul 2005 - 2 Aug 2005 (46 posts) Archive Link: "[2.6 patch] schedule obsolete OSS drivers for removal" Topics: PCI, Sound: ALSA, Sound: OSS People: Adrian Bunk, i810_audio, Lee Revell, Jeff Garzik, Thomas Sailer Adrian Bunk said, "This patch schedules obsolete OSS drivers (with ALSA drivers that support the same hardware) for removal." Thomas Sailer approved of the patch, while Jeff Garzik cautioned that before this patch was applied, it should be confirmed that the ALSA drivers really did support the same hardware in each case. He confirmed the via82cxxx driver himself, but said the i810_audio driver was not ready, because "ALSA doesn't have all the PCI IDs (which must be verified -- you cannot just add the PCI IDs for some hardware)" . Adrian confirmed that he had tried to confirm each driver, though he admitted that he could have missed something. Elsewhere, Lee Revell asked how many obsolete OSS drivers were included in Adrian's patch, and Adrian replied he hadn't counted them all. 2. Support For Sharp Zaurus SL-5500 LCD Power Up/Down 27 Jul 2005 - 1 Aug 2005 (10 posts) Archive Link: "[patch] Support powering sharp zaurus sl-5500 LCD up and down" People: Pavel Machek, Andrew Morton Pavel Machek said, "This adds support for powering Zaurus's video up and down. PDA without screen is kind of useless, so it is quite important... I'll have to figure out how to really control the frontlight, because LCD without that is quite hard to read." Andrew Morton asked if Pavel would maintain this code, and Pavel agreed. 3. Linux 2.6.13-rc3-mm2 Released 27 Jul 2005 - 29 Jul 2005 (26 posts) Archive Link: "2.6.13-rc3-mm2" Topics: Kernel Release Announcement People: Andrew Morton, Sam Ravnborg Andrew Morton announced Linux 2.6.13-rc3-mm2, saying: ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.13-rc3/ 2.6.13-rc3-mm2/ * Lots of fixes and updates all over the place. There are probably over 100 patches here which need to go into 2.6.13. * A reminder that -mm commit activity may be monitored by subscribing to the mm-commits list. Do echo subscribe mm-commits | mail majordomo@vger.kernel.org Sam Ravnborg asked if the mailing list were archived anywhere, and Andrew replied, "Not to my knowledge." 4. Linux 2.6.13-rc4 Released 28 Jul 2005 - 2 Aug 2005 (18 posts) Archive Link: "Linux 2.6.13-rc4" Topics: Disks: SCSI, FS: NTFS, Kernel Release Announcement, Sound: ALSA People: Linus Torvalds, Jesper Juhl, Andrew Morton Linus Torvalds announced Linux 2.6.13-rc4, saying: as many of you are aware, we were talking (not enough) about the release process at LKS this year. This ain't it. This is just the regular old release "process", with some LKS backlog put in for good measure. But the good news is, that I'll try the new release process after 2.6.13 is out, which is hopefully not too far away. Which means that we should try to let people know about the fact that if they want to merge stuff, they should do so in the first two weeks after the 2.6.13 release, and no later (also, no earlier either, by now). So if you have a favourite kernel developer, please wake him up with a friendly kick to the head and explain this concept to him in small easy-to-understand words, and tell him that we're in the freeze process for 2.6.13 now, and that he should be gathering up the patches, and make sure they get to me _after_ 2.6.13 is out, but at that point do it in a timely manner. Ok? In the meantime, here's the 2.6.13-rc4 update, with a diffstat so horribly ugly that I won't even show it (the kernel list would eat it as too big anyway), and I'll have to go fix my code that generates it. Oh, and in case you wonder, it's ugly because a cris architecture update with long filenames that really causes git-apply to output som rather nasty-looking diffstats ;) ALSA, IB, NTFS, SCSI (qla2xxx) and the cris architecture update. Jesper Juhl said, "For the bennefit of those of us who were not at LKS, could someone elaborate a bit on "the new release process" ?" Andrew Morton gave a link to a Linux Weekly News summary (http://lwn.net/Articles/144281/) of the process. 5. Wireless Security Lock Driver To Detect Physical Proximity 30 Jul 2005 - 31 Jul 2005 (19 posts) Archive Link: "[PATCH] Wireless Security Lock driver." Topics: USB People: Brian Schau, Pavel Machek, Alistair John Strachan Brian Schau said: I've attached a gzipped version of my Wireless Security Lock patch for v2.6.13-rc4. A Wireless Security Lock (WSL or weasel :-) is made up of two parts. One part is a receiver which you plug into any available USB port. The other part is a transmitter which at fixed intervals sends "ping packets". A "ping packet" usually consists of an ID and a flag telling if the transmitter has just been turned on. Both devices lights up in a nice way when a "ping packet" is send/ received. The whole idea of this is that when the transmitter is brought out of range one could have a process (such as xscreensaver) lock the display. The WSL is a toy gadget. For more information see: http://www.schau.com/l/wsl/index.html The WSL driver touches these files: drivers/usb/Makefile (1 line) drivers/usb/input/Kconfig (10 lines) drivers/usb/input/Makefile (1 line) drivers/usb/input/hid-core.c (2 lines) drivers/usb/input/wsl.c (224 lines) This is my first driver for Linux - feel free to harass me if I am not following procedures or you think the driver is lame. Better still, educate me :-) Pavel Machek replied, "Idea is good... but why don't you simply use bluetooth (built into many notebooks) and bluetooth-enabled phone? Probably could be done in userspace, too :-)." Alistair John Strachan replied: There's a script to this on the gentoo wiki via BlueZ. http://gentoo-wiki.com/TIP_Bluetooth_Proximity_Monitor I personally think the problem with this approach is that most phones have bluetooth enabled explicitly as an option, it doesn't run all the time, or default on. Primarily this is because bluetooth can drain your phone's battery (though, I don't know by how much, if you're not actually transferring data over it). A CR2032 cell, in a specific piece of kit, is going to last for a lot longer than a phone battery. But Brian also replied to Pavel, saying: WSLs are already reality. Sitecom is a producer of these and you can get another brand from ThinkGeek. Sitecom device: http://www.sitecom.com/products_info.php?product_id=293&grp_id= 1 ThinkGeek: http://www.thinkgeek.com/gadgets/security/698d/ Why in kernel? Well, the device is based on the Cypress Ultra Mouse. So with a non WSL aware kernel the events from the WSL will be merged into the standard mouse input queue which will make your mouse pointer move uncontrollable - it'll jump across the screen in a couple of steps every 3 seconds or so. Quite amusing but not very handy! The problem is described here: http://www.qbik.ch/usb/devices/showdev.php?id=3095 The WSL kernel driver will translate the device packets to a separate event queue. And you're right. The WSL driver is not a standalone thingy - you'll some userland tools as well. These can be gotten from: http://www.schau.com/l/wsl/index.html The tools contains a patch for xscreensaver (patch submitted to maintainer) and a small WSL monitor program which will monitor in-range/out-of-range signals and disable/enable xscreensaver as needed. Given the existence of the hardware, Brian's work mad sense to Pavel, and the thread petered out. 6. SquashFS Not NFS-Compatible; Fix On The Way 2 Aug 2005 - 4 Aug 2005 (4 posts) Archive Link: "squashfs seems nfs-incompatible" Topics: FS: NFS, FS: SquashFS People: Jan Engelhardt, Phillip Lougher Jan Engelhardt said, "I found out that you cannot mount an exported squash fs. The exports(5) fsid= parameter does not help it [like it did with unionfs]." Phillip Lougher replied, "The exports(5) man page says fsid=num is necessary for filesystems on non-block devices - I don't know whether this includes loopback filesystems. Have you tried exporting a Squashfs filesystem mounted on a real block device?" Jan said, "Loopback is a real block device, and no, fsid= does not help it. I have talked with the unionfs people, because it works for them. After a short flash of idea and comparison, it turns out that squashfs is missing sb->s_export->get_parent (the only requirement as it seems). Includes that you have sb->s_export non-null, of course. sb->s_export can be set within fill_super()." Phillip replied, "Ok, thanks. I'll try and get a fix for it in the next release." 7. New Documentation/kprobes.txt File 2 Aug 2005 - 3 Aug 2005 (3 posts) Archive Link: "[PATCH] Add Documentation/ kprobes.txt" People: Jim Keniston Jim Keniston said, "The enclosed patch creates Documentation/kprobes.txt, a guide to using the existing Kprobes facility for dynamic kernel instrumentation." 8. New Stable Review Cycle Started For 2.6.12.4 2 Aug 2005 (17 posts) Archive Link: "[00/13] -stable review" Topics: Big Memory Support, Bug Tracking, Kernel Build System People: Chris Wright Chris Wright said: This is the start of the stable review cycle for the 2.6.12.4 release. There are 13 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let us know. If anyone is a maintainer of the proper subsystem, and wants to add a signed-off-by: line to the patch, please respond with it. These patches are sent out with a number of different people on the Cc: line. If you wish to be a reviewer, please email stable@kernel.org to add your name to the list. If you want to be off the reviewer list, also email us. Responses should be made by Friday, Aug 5 07:00:00, UTC 2005. Anything received after that time, might be too late. The changelog entries were included at the top of each patch: * [PATCH] kbuild: build TAGS problem with O= make O=/dir TAGS fails with: MAKE TAGS find: security/selinux/include: No such file or directory find: include: No such file or directory find: include/asm-i386: No such file or directory find: include/asm-generic: No such file or directory The problem is in this line: ifeq ($(KBUILD_OUTPUT),) KBUILD_OUTPUT is not defined (ever) after make reruns itself. This line is used in the TAGS, tags, and cscope makes. * Correct handling of fc_remote_port_add() failure case. Immediately return if fc_remote_port_add() fails to allocate resources for the rport. Original code would result in NULL pointer dereference upon failure. * If bailing out because there is nothing to receive in rp_do_receive(), tty_ldisc_deref is not called. Failure to do so increases the ref count=20 and causes release_dev() to hang since it can't get the ref count to 0. * malicious 32bit app can have an elf section at 0xffffe000. During exec of this app, we will have a memory leak as insert_vm_struct() is not checking for return value in syscall32_setup_pages() and thus not freeing the vma allocated for the vsyscall page. Check the return value and free the vma incase of failure. * This is the code to load packet data into a register: k = fentry->k; if (k < 0) { ... } else { u32 _tmp, *p; p = skb_header_pointer(skb, k, 4, &_tmp); if (p != NULL) { A = ntohl(*p); continue; } } skb_header_pointer checks if the requested data is within the linear area: int hlen = skb_headlen(skb); if (offset + len <= hlen) return skb->data + offset; When offset is within [INT_MAX-len+1..INT_MAX] the addition will result in a negative number which is <= hlen. I couldn't trigger a crash on my AMD64 with 2GB of memory, but a coworker tried on his x86 machine and it crashed immediately. This patch fixes the check in skb_header_pointer to handle large positive offsets similar to skb_copy_bits. Invalid data can still be accessed using negative offsets (also similar to skb_copy_bits), anyone using negative offsets needs to verify them himself. Thanks to Thomas V?gtle for verifying the problem by crashing his machine and providing me with an Oops. * [NETFILTER]: Fix deadlock in ip6_queue Already fixed in ip_queue, ip6_queue was missed. * [NETFILTER]: Fix potential memory corruption in NAT code (aka memory NAT) The portptr pointing to the port in the conntrack tuple is declared static, which could result in memory corruption when two packets of the same protocol are NATed at the same time and one conntrack goes away. * [NETFILTER]: Wait until all references to ip_conntrack_untracked are dropped on unload Fixes a crash when unloading ip_conntrack. * [XFRM]: Fix possible overflow of sock->sk_policy Spotted by, and original patch by, Balazs Scheidler. * [PATCH] bio_clone fix Fix bug introduced in 2.6.11-rc2: when we clone a BIO we need to copy over the current index into it as well. It corrupts data with some MD setups. See http://bugzilla.kernel.org/show_bug.cgi?id=4946 Huuuuuuuuge thanks to Matthew Stapleton for doggedly chasing this one down. * sys_get_thread_area does not memset to 0 its struct user_desc info before copying it to user space... since sizeof(struct user_desc) is 16 while the actual datas which are filled are only 12 bytes + 9 bits (across the bitfields), there is a (small) information leak.i This was already committed to Linus' repository. * [VLAN]: Fix early vlan adding leads to not functional device OK, I can see what's happening here. eth0 doesn't detect link-up until after a few seconds, so when the vlan interface is opened immediately after eth0 has been opened, it inherits the link-down state. Subsequently the vlan interface is never properly activated and are thus unable to transmit any packets. dev->state bits are not supposed to be manipulated directly. Something similar is probably needed for the netif_device_present() bit, although I don't know how this is meant to work for a virtual device. * powernow-k8 requires that a data structure for each core be created in the _cpu_init function call. The cpufreq infrastructure doesn't call _cpu_init for the second core in each processor. Some systems crashed when _get was called with an odd-numbered core because it tried to dereference a NULL pointer since the data structure had not been created. The attached patch solves the problem by initializing data structures for all shared cores in the _cpu_init function. It should apply to 2.6.12-rc6 and has been tested by AMD and Sun. Several people posted their approval of various patches; no one had any objections. 9. SMB Maintainership 3 Aug 2005 - 4 Aug 2005 (5 posts) Archive Link: "Is anyone maintaining the smb filesystem?" Topics: FS: smbfs, MAINTAINERS File, Samba People: Adrian Bunk, Tvrtko A. Ursulin, Urban Widmark Adrian Bunk asked: Is anyone maintaining the smb filesystem in the Linux kernel? Emails to the Email address urban@teststation.com of Urban Widmark that is listed in MAINTAINERS are bouncing. Can anyone tell how to reach Urban and/or whether anyone is currently maintaining smbfs? Tvrtko A. Ursulin replied, "It probably won't help you much, but I had the same problem few months ago. There was a bug in smbfs which I tried to discuss with someone, and after failing to contact the maintainer, I sent the fix to Linus. I don't think even he managed to get a response from Urban or someone else. The fix went in so I stopped chasing it. So it looks like smbfs is not maintained." Sharon And Joy Kernel Traffic is grateful to be developed on a computer donated by Professor Greg Benson and Professor Allan Cruse in the Department of Computer Science at the University of San Francisco. This is the same department that invented FlashMob Computing. Kernel Traffic is hosted by the generous folks at kernel.org. All pages on this site are copyright their original authors, and distributed under the terms of the GNU General Public License version 2.0.