Post-halloween doc updates.
I couldn't resist the irony of posting a post-halloween document just before halloween. Quite a few changes since last posting here, and with each posting I tend to get a lot of useful feedback for the next round of changes.
Dave
The post-halloween document. v0.46
(aka, 2.6 - what to expect)
Dave Jones <[email protected]>
(Updated as of 2.6.0test9)This document explains some of the new functionality to be found in the 2.6 Linux kernel, some pitfalls you may encounter, and also points out some new features which could really use testing.
Note, that "contact [email protected]" below also implies that you should also
cc: [email protected].
Latest version of this document can always be found at
http://www.codemonkey.org.uk/docs/post-halloween-2.6.txt
Thanks to many [far too many to list] people for valuable feedback.
Note, that this document is somewhat x86-centric, but most features documented here affect all platforms anyway.
Translations:
Spanish - http://www.terra.es/personal/diegocg/post-halloween-2.6.es.txt
German - http://www.kubieziel.de/computer/halloween-german.html
Polish - http://soltysiak.com/linux/post-halloween-2.6.pl.txtpt_BR - www.maluco.com.br/thiago/docs/post-halloween-2.6-pt_BR.txt
Applying patches.
~~~~~~~~~~~~~~~~~
- In 2.4 and previous kernels, the recommended way to apply patches was to use a command line such as ...
gzip -cd patchXX.gz | patch -p0
In 2.6, Linus started adding an extra path element to the diffs, so using -p1 in the untarred 'to be patched' directory is necessary.
Known gotchas.
~~~~~~~~~~~~~~
Certain known bugs are being reported over and over. Here are the workarounds.
- Blank screen after decompressing kernel?
Make sure your .config has
CONFIG_INPUT=y
CONFIG_VT=y
CONFIG_VGA_CONSOLE=y
CONFIG_VT_CONSOLE=y
A lot of people have discovered that taking their .config from 2.4 and running make oldconfig to pick up new options leads to problems, notably with CONFIG_VT not being set.
- An additional bug biting some people is that NICs fail to receive packets (usually notable by a NIC not getting a DHCP lease for eg, despite being sent one by the server). Booting with "noapic" "acpi=off" or a combination of both fixes this for most people. Additional breakage reports should go to Jeff Garzik <[email protected]>
- (Possibly linked to above bug) VIA APIC routing is currently broken. boot with 'noapic'.
- Can't load any modules? You need updated tools (See modules section below).
Regressions.
~~~~~~~~~~~~
(Things not expected to work just yet)
- The hptraid/promise drivers for proprietary RAID formats are currently non functional, and will probably be converted to use device-mapper.
- Some filesystems still need work (Intermezzo, UFS, HFS, HPFS..)
- A number of drivers don't compile currently due to them needing various work to convert them to the new APIs
- UMSDOS fs is currently missing, pending rewrite.
- The format of /proc/stat changed, which could break some applications that still depend on the old layout.
Currently the only known application to break is the java 'DOTS' app. (http://bugme.osdl.org/show_bug.cgi?id=277)
- Some people seem to have trouble running rpm, most notably Red Hat 9 users. This is a known bug of rpm.
Workaround: run "export LD_ASSUME_KERNEL=2.2.5", before running rpm. This is thought to be a bug related to db4 and O_DIRECT interaction.
Stuff needing forward porting from 2.4.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- HFSPlus
- SuperH 64
Removed features.
~~~~~~~~~~~~~~~~~
- khttpd is gone.
- Older Direct Rendering Manager (DRM) support (For XFree86 4.0) has been removed. Upgrade to XFree86 4.1.0 or higher.
- LVM1 has been removed. See Device-mapper below.
- The system call table is no longer exported. Any module that relied on this previously will no longer work.
- Soundmodem hamradio support has been removed. Its functionality has been superceded by a userspace replacement.
http://www.baycom.org/~tom/ham/soundmodem
- Direct booting from floppy is no longer supported.
You should now use a boot loader program instead.
- Callout tty devices (/dev/cua) have been deprecated since 2.1.90pre2. Support is now removed.
Deprecated/obsolete features.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- devfs has been obsoleted in favour of udev (http://www.kernel.org/pub/linux/utils/kernel/hotplug/)
- boot time root= parsing changed.
ramdisks are now ram<n> instead of rd<n> and cm206 is cm206cd (instead of
cm206).
- usbdevfs will be going away in 2.7. The same filesystem can
be mounted as 'usbfs' in recent 2.4 kernels, and in 2.5.52
and above, which is what the filesystem will furthermore be
known as.
- elvtune is deprecated (as are the ioctl's it used).
Instead, the io scheduler tunables are exported in sysfs (see below)
in the /sys/block/<device>/queue/iosched directory.
Jens wrote a document explaining the tunables of the new scheduler at
http://www.lib.uaa.alaska.edu/linux-kernel/archive/2002-Week-44/att-deadline-iosched.txt
- Calling syscalls by numeric values is deprecated, and will go away
in the next development series.
Modules.
~~~~~~~~
- The in-kernel module loader got reimplemented.
- You need replacement module utilities from
http://www.kernel.org/pub/linux/kernel/people/rusty/modules/
- A backwards compatible set of module utilities is also available from the same URL in RPM format.
- Debian sid or Conectiva snapshot users can just use 'apt-get install module-init-tools'
- Modules now free stuff marked with __init or __initdata.
- For Red Hat users, there's another pitfall in "/etc/rc.sysinit". During startup, the script sets up the binary used to dynamically load modules stored at "/proc/sys/kernel/modprobe". The initscript looks for "/proc/ksyms", but since it doesn't exist in 2.6 kernels, the binary used is "/sbin/true" instead.
This, eventually, will keep modules from working. Red Hat users will have to patch the "/etc/rc.sysinit" script to set "/proc/sys/kernel/modprobe" to "/sbin/modprobe", even when "/proc/ksyms" doesn't exist.
- Modules now have a .ko suffix instead of .o
Kernel build system.
~~~~~~~~~~~~~~~~~~~~
- The build system is much improved compared to 2.4.
You should notice quicker builds, and less spontaneous rebuilds of files on subsequent builds from already built trees.
- There are new graphical config tools.
"make xconfig" now requires the qt libraries.
"make gconfig" uses gtk libraries.
- Make menuconfig/oldconfig has no user-visible changes other than speed, whilst numerous improvements have been made.
- Several new debug targets exist: 'allyesconfig' 'allnoconfig' 'allmodconfig'.
- Note: The new configuration system is not CML2 related.
- Also note: Whilst some ideas were taken from it, Keith Owens' kbuild-2.5 project was not integrated.
- "make" is now the preferred command, without a target; it does <arch-zimage> and modules.
- "make -jN" is now the preferred parallel-make execution.
Do not bother to provide "MAKE=xxx"
- The build is now much less verbose. If you want to see exactly what's going on, try "make V=1" or set KBUILD_VERBOSE=1 in your environment.
- 'make kernel/mm.o' will build the named file, provided a corresponding source exists. This also works for (non-composite) modules. (FIXME: broken for modules right now?)
- 'make kernel/' will compile all files in a subdirectory and below.
- There is no need to run 'make dep' at any stage.
- 'make help' provides a list of typical targets, including debugging targets.
IO subsystem.
~~~~~~~~~~~~~
- You should notice considerable throughput improvements over 2.4 due to much reworking of the block and the memory management layers.
- Report any regressions in this area to Jens Axboe <[email protected]> and Andrew Morton <[email protected]>
- Two different IO elevators are available. The default is the anticipatory IO scheduler. You can select the deadline scheduler by booting with "elevator=deadline" on the kernel command line.
- For some workloads the anticipatory scheduler is around 10% slower than deadline. Most notably, database workloads which seek all over the disk performing reads and synchronous writes. Database folks will likely want to boot with elevator=deadline to get that last bit of performance back.
- Assorted changes throughout the block layer meant various block device drivers had a large scale cleanup whilst being updated to newer APIs.
- The size and alignment of O_DIRECT file IO requests now matches that of the device, not the filesystem. Typically this means that you can perform O_DIRECT IO with 512-byte granularity rather than 4k. But if you rely upon this, your application will not work on 2.4 kernels and will not work on some devices.
block device size support.
~~~~~~~~~~~~~~~~~~~~~~~~~~
- Thanks to work done by Peter Chubb, block devices can now access up to 16TB on 32-bit architectures, and up to 8EB on 64-bit architectures.
- To use the new BLKGETSZ64 ioctls, you'll need updated file-utils. (Currently only jfsutils 1.0.20 has this change, patches for other filesystems are still pending merging)
- The old 'struct statfs' is not able to describe large devices - the statfs() system call will now return -EOVERFLOW for such devices. A new system call called statfs64() with a new structure has been added to support large devices. It it unknown at time of writing how many userspace utilities have been converted to take advantage of this syscall when available.
POSIX ACLs & Extended attributes.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Userspace tools available at http://acl.bestbits.at
VM Changes.
~~~~~~~~~~~
- Version zero swap partitions are no longer supported (everything is using v1 now anyway - rerun mkswap if in doubt).
Linux 2.0.x requires v0 swap space, Linux v2.1.117 and later support v1. mkswap(8) can format swap space in either format.
- The actual 'reverse mappings' part of Rik van Riel's rmap vm was merged. VM behaviour under certain loads should improve.
- VM misbehaviour should be reported to [email protected]
- The VM explicitly checks for sparse swapfiles, and aborts if one is found.
- /proc/sys/vm/swappiness defines the kernel's preference for pagecache over mapped memory. Setting it to 100 (percent) makes it treat both types of memory equally. Setting it to zero makes the kernel very much prefer to reclaim plain pagecache rather than mapped-into-pagetables memory.
- The bdflush() syscall is now officially deprecated. The syscall does nothing, and prints a stern warning to users. The functionality is replaced by the pdflush daemons.
- Due to various changes, swap files should be just as fast as swap partitions.
- In 2.4, up to 64 swap files were possible. In 2.6, this number is reduced to 32. Like 2.4, these files can be up to 64GB in size, though you will need a recent util-linux to have a mkswap utility that supports >2GB
Kernel preemption.
~~~~~~~~~~~~~~~~~~
- The much talked about preemption patches made it into 2.6. With this included you should notice much lower latencies especially in demanding multimedia applications.
- Note, there are still cases where preemption must be temporarily disabled where we do not. These areas occur in places where per-CPU data is used.
- If you get "xxx exited with preempt count=n" messages in syslog, don't panic, these are non fatal, but are somewhat unclean. (Something is taking a lock, and exiting without unlocking)
- If you DO notice high latency with kernel preemption enabled in a specific code path, please report that to Andrew Morton <[email protected]> and Robert Love <[email protected]>.
The report should be something like "the latency in my xyz application hits xxx ms when I do foo but is normally yyy" where foo is an action like "unlink a huge directory tree".
Process scheduler improvements.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Another much talked about feature. Ingo Molnar reworked the process scheduler to use an O(1) algorithm. In operation, you should notice no changes with low loads, and increased scalability with large numbers of processes, especially on large SMP systems.
- Scheduler is now Hyperthreading SMP aware and will disperse processes over physically different CPUs, instead of just over logical CPUs.
- Robert Love wrote various utilities for changing behaviour of the scheduler (binding processes to CPUs etc). You can find these tools at
http://tech9.net/rml/schedutils
- The behavior of sched_yield() changed a lot. A task that uses this system call should now expect to sleep for possibly a very long time. Tasks that do not really desire to give up the processor for a while should probably not make heavy use of this function. Unfortunately, some GUI programs (like Open Office) do make excessive use of this call and under load their performance is poor. It seems this new 2.6 behavior is optimal but some user-space applications may need fixing.
- The above applies to use of yield() in the kernel, too.
- 2.6 adds system calls for manipulating a task's processor
affinity: sched_getaffinity() and sched_setaffinity()
- Regressions to [email protected] and [email protected]
- Debian users who encounter effects such as skips in mp3 playback, jerky mouse movement may want to stop the
X server from renicing itself to -10
You can alter this permanently with 'dpkg-reconfigure xserver-common'; if you elect not to have /etc/X11/Xwrapper.config managed by debconf, simply edit it directly.
- Balancing of IRQs between multiple CPUs should be handled using the irqbalance (http://people.redhat.com/arjanv/irqbalance/) program.
- David Mosberger maintains a webpage containing some current 'known gotchas' of the O(1) scheduler at http://www.hpl.hp.com/research/linux/kernel/o1.php
PCI.
~~~~
- PCI domain support has been added. For most people, this just means that all PCI slot names are extended with "0000:" on the front, but for people with bigger servers it means they're able to access all their PCI devices.
- More hotplug drivers have been added, including a fake PCI hotplug driver so people without specialised hardware can test hotplug features.
Random.
~~~~~~~
- /dev/hwrandom got support for some new hardware (now also backported to 2.4) such as the HW RNG on newer VIA Cyrix CPUs.
- rng-tools can be found at http://sourceforge.net/projects/gkernel
Fast userspace mutexes (Futexes).
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Rusty Russell added functionality that allows userspace to have fast mutexes that only use syscalls when there is contention. Used by NPTL.
- Bert Hubert has written some documentation on this functionality at http://ds9a.nl/futex-manpages
epoll
~~~~~
Davide Libenzi wrote an event based poll replacement that got included in 2.6. More info available at
http://www.xmailserver.org/linux-patches/nio-improve.html
http://lwn.net/Articles/13587/
Threading improvements.
~~~~~~~~~~~~~~~~~~~~~~~
- Ingo Molnar put a lot of work into threading improvements for 2.6. Some of the features of this work are:
- Generic pid allocator (arbitrary number of PIDs with no slowdown, unified pidhash).
- Thread Local Storage syscalls
- sys_clone() enhancements (CLONE_SETTLS, CLONE_PARENT_SETTID, CLONE_SETTID, CLONE_CLEARTID, CLONE_DETACHED)
- POSIX thread signals stuff (atomic signals, shared signals, etc.) - Per-CPU GDT - Threaded coredumping support - sys_exit() speedups (O(1) exit) - Generic, improved futexes, vcache - New, threading related ptrace features - exit/fork task cache - /proc updates for threading - API changes for threading.
- Users should notice a significant speedup in basic thread operations. This is true to a lesser extent even for old-threading userspace libraries such as LinuxThreads.
- Regressions should go to Ingo Molnar <[email protected]> and [email protected]. Regressions could happen in the area of signal handling and related threading semantics, plus coredumping.
- Native Posix Threading Library (NPTL).
Ulrich Drepper worked closely with Ingo on the threading enhancements, and developed a 1:1 model threading library. You can find out more about NPTL at
http://people.redhat.com/drepper/nptl-design.pdf
Enhanced coredumping.
~~~~~~~~~~~~~~~~~~~~~
- 2.6 offers you the ability to configure the way core files are named through a /proc/sys/kernel/core_pattern file. You can use various format identifiers in this name to affect how the core dump is named.
%p - insert pid into filename
%u - insert current uid into filename
%g - insert current gid into filename
%s - insert signal that caused the coredump into the filename %t - insert UNIX time that the coredump occurred into filename %h - insert hostname where the coredump happened into filename %e - insert coredumping executable name into filename
You should ensure that the string does not exceed 64 bytes.
- Multithreaded processes can now dump core
Input layer.
~~~~~~~~~~~~
- Possibly the most visible change to the end user. If misconfigured, you'll find that your keyboard/mouse/other input device will no longer work. 2.6 offers a much more flexible interface to devices such as keyboards.
- The downside is more confusing options.
In the "Input device support" menu, be sure to enable at least the following.
--- Input I/O drivers
< > Serial i/o support
< > i8042 PC Keyboard controller
[ ] Keyboards
[ ] Mice
(Also choose the relevant keyboard/mouse from the list)
- If you find your keyboard/mouse still don't work, edit the file drivers/input/serio/i8042.c, and replace the #undef DEBUG with a #define DEBUG
When you boot, you should now see a lot more debugging information. Forward this information to Vojtech Pavlik <[email protected]>
- If you use a KVM switcher, and experience problems, booting with the boot time argument 'psmouse_noext' should fix your problems.
- Users of multimedia keys without X will see changes in how the kernel handles those keys. People who customize keymaps or keycodes in 2.4 may need to make some changes in 2.6
- Users wanting support for the PC speaker need to enable CONFIG_INPUT_PCSPKR, or you won't get a single beep.
- Synaptics touchpad users may be interested to check out
http://w1.894.telia.com/~u89404340/touchpad/
PnP layer.
~~~~~~~~~~
- Support for plug and play devices such as early ISAPnP cards has improved a lot in the 2.6 kernel. The new code behaves more closely to the code handling PCI devices (probe, remove etc callbacks), and also merges PnP BIOS access code.
- Report any regressions in plug & play functionality to
Adam Belay <[email protected]>
ALSA.
~~~~~
- The advanced linux sound architecture got merged into 2.6.
This offers considerably improved functionality over the older OSS drivers, but requires new userspace tools.
- Several distros have shipped ALSA for some time, so you may already have the necessary tools. If not, you can find them at http://www.alsa-project.org/
- ALSA can emulate OSS interface using the snd_pcm_oss/snd_pcm_mixer modules, if your card produces nothing but silence, you may need to run alsamixer to unmute channels wich /dev/mixer doesn't see
- Note that the OSS drivers are also still functional, and still present. Many features/fixes that went into 2.4 are still not applied to these drivers, and it's still unclear if they will remain when 2.6 ships. The long term goal is to get everyone moved over to (the superior) ALSA.
AGP.
~~~~
- The agpgart driver got a long overdue cleanup which involved splitting it into an agpgart core, and per-chipset drivers. You may need to adjust your modules configuration to autoload the chipset drivers on loading the agpgart module.
- Generic AGP 3.0 support is now included.
DRI.
~~~~
- Direct rendering in 2.6 hasn't had much (if any?) testing on older versions of XFree86. Feedback on whether 4.1 works would be useful.
Faster system calls.
~~~~~~~~~~~~~~~~~~~~
- Systems that support the SYSENTER extension (Basically Intel Pentium-II and above, and AMD Athlons) now have a faster method of making the transition from userspace to kernelspace when a syscall is performed.
- Pentium Pro also has SYSENTER, but due to errata, is unusable.
- Without an updated glibc, this will not be noticable.
- VMWare 4 users may get crashes due to this.
Zwane Mwaikambo wrote a patch for a "nosysenter" option which is worth googling for if there isn't a vmware update available.
- Regressions to [email protected] and [email protected]
procps.
~~~~~~~
- The 2.6 /proc filesystems changed some statistics, which confuse older versions of procps. Rik van Riel and Robert Love have been maintaining a version of procps during the development of 2.6 which tracks changes to /proc which you can find at http://tech9.net/rml/procps/
- Alternatively, the procps by Albert Cahalan now supports the altered formats since v3.0.5 -- http://procps.sf.net/
- The /proc/meminfo format changed slightly which also broke gtop in strange ways. Likely this also broke some of the KDE/GNOME panel applets.
Framebuffer layer.
~~~~~~~~~~~~~~~~~~
- James Simmons has reworked the framebuffer/console layer considerably for 2.6. Support for some cards is still lagging a little, but it should be functionally no different than previous incarnations.
- boot time arguments may have changed depending on your driver. an example of the change is..
append = "video=radeon:1024x768-24@100"
needs to become..
append = "video=radeonfb:1024x768-24@100"- Current userspace tools (fbset for eg) are not yet updated,
and won't function as expected.
- The VESA framebuffer now enables MTRRs for the framebuffer memory range during initialisation (Note: PCI cards only).
If you notice screen corruption, please report this, along with an lspci output, so your card can be blacklisted.
- Any problems should go to <[email protected]>
IDE.
~~~~
- The IDE code rewrite was subject to much criticism in early 2.5.x, which put off a lot of people from testing. This work was then subsequently dropped, and reverted back to a 2.4.18 IDE status.
Since then additional work has occurred, but not to the extent of the first cleanup attempts.
- Known problems with the current IDE code.
o Simplex IDE devices (eg Ali15x3) are missing DMA sometimes o Most PCMCIA devices have unload races and may oops on eject o Modular IDE does not yet work, modular IDE PCI modules sometimes
oops on loading
o ide_scsi is completely broken in 2.6 currently. Known problem. If you need it either use 2.4 or fix it 8)
- IDE disk geometry translators like OnTrack, EZ Partition, Disk Manager are no longer autodetected. The only way forward is to remove the translator from the drive, and start over, or use boot parameters depending on the type of remapper used :-
hdx=remap63 - add 63 to each sector (For OnTrack DM)
hdx=remap - remap 0->1 (For EZDrive)
IDE TCQ.
~~~~~~~~
- Tagged command queueing for IDE devices has been included.
- Not all combinations of controllers & devices may like this, so handle with care.
READ AS: ** Don't use IDE TCQ on any data you value.
It's likely bad combinations will be blacklisted as and when discovered.
- If you didn't choose the "TCQ on by default" option, you can enable it by using the command
echo "using_tcq:32" > /proc/ide/hdX/settings
(replacing 32 with 0 disables TCQ again).
- Report success/failure stories to Jens Axboe <[email protected]> with inclusion of hdparm -i /dev/hdX, and lspci output.
SCSI.
~~~~~
- Various SCSI drivers still need work, and don't even compile.
- Various drivers currently lack error handling.
These drivers will cause warnings during compilation due to missing abort: & reset: functions.
- Note, that some drivers have had these members removed, but still lack error handling. Those noticed so far are ncr53c8xxx, sym53c8xx
- large dev_t support allowing thousands of disks to be supported (was 128 or 256 in the 2.4 series)
- major code cleanup, initially to support the block layer (bio) improvements have led to:
- better throughput (?) [less double handling of data]
- per HBA locks (there was a single io_request_lock in
the 2.4 series)
- more flexible interface to HBA drivers
- better hotplug support, especially for USB mass storage and ieee1394 sbp2 devices [well it's work_in_progress]
- improved error processing and scanning code (support for large, sparse lun spaces)
- lots of scsi driver internals available via sysfs
v4l2.
~~~~~
- The video4linux API finally got its long awaited cleanup.
- xawtv, bttv and most other existing v4l tools are also compatible with the new v4l2 layer. You should notice no loss in functionality.
- See http://bytesex.org/v4l/ for more information.
Quota reworking.
~~~~~~~~~~~~~~~~
The new quota system needs new tools. Supports 32 bit uids.
http://www.sf.net/projects/linuxquota/
CD Recording.
~~~~~~~~~~~~~
- Jens Axboe added the ability to use DMA for writing CDs on ATAPI devices. Writing CDs should be much faster than it
was in 2.4, and also less prone to buffer underruns and the like.
- With a recent cdrecord, you also no longer need ide-scsi in order to use an IDE CD writer.
- Ripping audio tracks off of CDs now also uses DMA and should be notably faster. You can also find an updated cdda2wav at:
*.kernel.org/pub/linux/kernel/people/axboe/tools/
- Send good/bad reports of audio extraction with cdda2wav and burning with the cdrecord to Jens Axboe <[email protected]>
- Currently only 'open by device name' works in cdrecord. cdrecord -dev=/dev/hdX -inq
- More info at http://lwn.net/Articles/13538/ & http://lwn.net/Articles/13160/
USB:
~~~~
- USB host controller drivers got renamed in 2.6. They are now uhci-hcd for UHCI controllers.
ohci-hcd for OHCI controllers.
ehci-hcd for EHCI (USB 2.0) controllers.
- Very little user visible changes, the only noticable 'major' change is that there is now only one UHCI driver. As noted elsewhere, usbdevfs got renamed to usbfs.
- USB-storage has changed behaviour. A device which is disconnected and then reconnected is not reassociated with the old /dev node.
- USB storage also got several performance enhancements.
Nanosecond stat:
~~~~~~~~~~~~~~~~
The stat64() syscall got changed to return jiffies granularity. This allows make(1) to make better decisions on whether or not it needs to recompile a file. Not all filesystems may support such precision.
Filesystems:
~~~~~~~~~~~~
A number of additional filesystems have made their way into 2.6. Currently it supports: ext2, ext3, reiserfs, jfs, xfs, minix, romfs, iso9660, udf, msdos, vfat, ntfs (ro), adfs, amiga ffs, apple macintosh hfs, BeOS befs (ro), bfs, efs (ro), cramfs, free vxfs, os/2 hpfs, qnx4fs, sysvfs, ufs.
Whilst these have had testing out of tree, the level of testing after merging is unparalleled. Be wary of trusting data to immature filesystems. A number of new features and improvements have also been made to the existing filesystems from 2.4.
Reports of stress testing with the various tools available would be beneficial.
Generic VFS changes.
~~~~~~~~~~~~~~~~~~~~
- Since Linux 2.5.1 it is possible to atomically move a subtree to another place. The call is...
mount --move olddir newdir
- Since 2.5.43, dmask=value sets the umask applied to directories only. The default is the umask of the current process.
The fmask=value sets the umask applied to regular files only. Again, the default is the umask of the current process.
- Directories can now be marked as synchronous using chattr +S, so that all changes will be immediately written to disk. Note, this does not guarantee atomicity, at least not for all filesystems and for all operations. You *can* be guaranteed that system calls will not return until the changes are on disk; note though that this does have has some significant performance impacts.
devfs.
~~~~~~
- devfs got somewhat stripped down and a lot of duplicate functionality got removed. You now need to enable CONFIG_DEVPTS_FS=y and mount the devpts filesystem in the same manner you would if you were not using devfs.
EXT2.
~~~~~
- 2.5.49 included an extension to ext2 which will cause it to not attach buffer_head structures to file or directory pagecache at all, ever. This is for the big highmem machines. It is enabled via the `-o nobh' mount option.
- The ext2 filesystem is now using finer-grained locking which yields reduced context switch rates and higher throughput on large SMP machines.
EXT3.
~~~~~
- The ext3 filesystem has gained indexed directory support, which offers considerable performance gains when used on filesystems with directories containing large numbers of files.
- In order to use the htree feature, you need at least version 1.32 of e2fsprogs.
- Existing filesystems can be converted using the command
tune2fs -O dir_index /dev/hdXXX
- The latest e2fsprogs can be found at
http://prdownloads.sourceforge.net/e2fsprogs
- The ext2 and ext3 filesystems have new file allocations policies (the "Orlov allocator") which will place subdirectories closer together on-disk. This tends to mean that operations which touch many files in a directory tree are much faster if that tree was created under a 2.6 kernel.
Reiserfs.
~~~~~~~~~
- Reiserfs now supports inode attributes such as immutable. (Also included in 2.4.17, so not really 'new').
- Relocated/non-standard size journal support (also backported to 2.4.22pre3)
- Support for writes larger than 4KB in size, which means speedups on large file writes, esp in append mode, should also be more SMP friendly.
- Variable blocksize support. (Ie, you can choose any blocksize in the range of 1024 .. PAGE_CACHE_SIZE, must be power of 2).
NFS.
~~~~
- Basic support has been added for NFSv4 (server and client)
- Additionally, kNFSD now supports transport over TCP. This experimental feature is also backported to 2.4.20
- Interoperability reports with other OS's would be useful.
- v1.0.3 of nfs-utils supports the newer 2.6 kernels change of kdev_t type. You can grab it at http://nfs.sourceforge.net
- Problems to [email protected]
NTFS.
~~~~~
- A new, rewritten NTFS driver got merged for 2.6. It has the following main benefits over the old driver:
- SMP and reentrant safe
- support bigger than 4 kB cluster sizes
- full support for sparse files on W2K/XP/W2K3
- mmap() support
- More stable, and much faster than the previous NTFS driver.
- Still read-only, but with safe file overwrite support without changes to the file size
- More information is available at http://linux-ntfs.sf.net
sysfs.
~~~~~~
In simple terms, the sysfs filesystem is a saner way for drivers to export their innards than /proc.
This filesystem is always compiled in, and can be mounted just like another virtual filesystem. No userspace tools beyond cat and echo are needed.
mount -t sysfs none /sys
See Documentation/filesystems/sysfs.txt for more info.
JFS.
~~~~
IBM's JFS got merged for 2.6. (And backported to 2.4.20, but it was still a new feature here first. You can read more about JFS at
http://www-124.ibm.com/developerworks/oss/jfs/index.html
XFS.
~~~~
The SGI XFS filesystem has been merged, and has a number of userspace features. Users are encouraged to read http://oss.sgi.com/projects/xfs
for more information.
The various utilities for creating and manipulating XFS volumes can
be found on SGI's ftp server:
ftp://oss.sgi.com/projects/xfs/download/download/cmd_tars/xfsprogs-2.5.4.src.tar.gz
CIFS.
~~~~~
Support utilities and documentation for the common internet file system (CIFS) can be found at http://us1.samba.org/samba/Linux_CIFS_client.html
FAT.
~~~~
CVF (Compressed VFAT) support has been removed. This means you will no longer be able to access DriveSpace partitions.
HugeTLBfs.
~~~~~~~~~~
Files in this filesystem are backed by large pages if the CPU supports them. See Documentation/vm/hugetlbpage.txt for more details.
Internal filesystems.
~~~~~~~~~~~~~~~~~~~~~/proc/filesystems will contain several filesystems that are not mountable in userspace, but are used internally by the kernel to keep track of things. Amongst these filesystems are futexfs and eventpollfs.
Kernel Asynchronous I/O (AIO) Support
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Support for kernel AIO has been included in 2.6.
AIO enables even a single application thread to overlap I/O operations with other processing, by providing an interface for submitting one or more i/o requests in one system call (io_submit) without waiting for completion, and a separate interface (io_getevents) to reap completed i/o operations associated with a given completion group.
The following is a quick summary of what works today as expected:
- AIO read and write on raw (and O_DIRECT on blockdev)
- AIO read and write on files opened with O_DIRECT on ext2, ext3, jfs, xfs
And what doesn't work as expected or is not currently supported:
- AIO read and write on files opened without O_DIRECT (i.e. normal buffered filesystem AIO). On ext2, ext3, jfs, xfs and nfs, these do not return an explicit error, but quietly default to synchronous or rather non-AIO behaviour (i.e io_submit waits for i/o to complete in these cases). For most other filesystems, -EINVAL is reported.
- AIO fsync (not supported for any filesystem)
- AIO read and write on sockets (doesn't return an explicit error, but quietly defaults to synchronous or rather non-AIO behaviour)
You need to install libaio-0.3.92 (available at
http://www.kernel.org/pub/linux/kernel/people/bcrl/aio/)
if you are writing AIO applications which use the native AIO interfaces.
More info is available at http://lse.sf.net/io/aio.html
Profiling.
~~~~~~~~~~
- A system wide performance profiler (Oprofile) has been included in 2.6. With this option compiled in, you'll get an oprofilefs filesystem which you can mount, that the userspace utilities talk to. You can find out more at http://oprofile.sf.net/
- You need a fixed readprofile utility for 2.6.
Present in util-linux as of 2.11z
Improved BIOS table support.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Linux now supports various new BIOS extensions.
Simple boot flag support.
~~~~~~~~~~~~~~~~~~~~~~~~~
The SBF specification is an x86 BIOS extension that allows improved system boot speeds. It does this by marking a CMOS field to say
"I booted okay, skip extensive POST next reboot".
Userspace tool is at http://www.codemonkey.org.uk/projects/sbf/sbf.c
More info on SBF is at http://www.microsoft.com/hwdev/resources/specs/simp_bios.asp
EDD Support.
~~~~~~~~~~~~
- Support for BIOS Enhanced Disk Drive Services (EDD) was added, which exports information on what the BIOS thinks is the boot drive and other useful info to /sys/firmware/edd
- Matt Domsch is interested in hearing success/fails on this code with some simple tests decribed at http://domsch.com/linux/edd30/results.html
Improved system monitoring.
~~~~~~~~~~~~~~~~~~~~~~~~~~~
- lm_sensors.
- Shipped in vendors kernels for years, lm_sensors is now part of mainline. It does however have a different interface. (/sysfs instead of /proc
- http://www.xs4all.nl/~thospel/ASIS/bin/psensors is a handy script for parsing the new sysfs fields.
- IPMI. (Intelligent Platform Management Interface)
- IPMI is a standard for monitoring the hardware in a system.
- Project home page: http://openipmi.sourceforge.net
- Specification: http://www.intel.com/design/servers/ipmi/spec.htm
x86 CPU detection.
~~~~~~~~~~~~~~~~~~
- The CPU detection code got a pretty hefty shake up. To be certain your CPU has all relevant workarounds applied, be sure to check that it was detected correctly. cat /proc/cpuinfo will tell what the kernel thinks it is.
- Likewise, the x86 MTRR driver got a considerable makeover.
Check that XFree86 sets up MTRRs in the same way it did in 2.4 (Failures will get logged in /var/log/XFree86.log)
- Early PII Xeon processors and possibly other early PII processors require microcode updates either from the BIOS or the microcode driver to work around CPU bugs the O(1) scheduler exposes.
You can find the relevant microcode tools at
http://www.urbanmyth.org/microcode/
- Any regressions in both should go to [email protected] Cc: [email protected]
Extra tainting.
~~~~~~~~~~~~~~~
Running certain AMD processors in SMP boxes is out of spec, and will taint the kernel with the 'S' flag. Running 2 Athlon XPs for example may seem to work fine, but may also introduce difficult to pin down bugs. In time it's likely this tainting will be extended to cover other out of spec cases.
Additionally, the new modules interface will taint the kernel if you try to 'force' a module to load with insmod -f.
Power management.
~~~~~~~~~~~~~~~~~
- 2.6 contains a more up to date snapshot of the ACPI driver. Should you experience any problems booting, try booting with the argument "acpi=off" to rule out any ACPI interaction. ACPI has a much more involved role in bringing the system up in 2.6 than it did in 2.4
- The old "acpismp=force" boot option is now obsolete, and will be ignored due to the old "mini ACPI" parser being removed.
- software suspend is still in development, and in need of more work. Use with SMP and/or PREEMPT not advised.
- The ACPI code will do basic sanity checks on the DMI structure in the BIOS to determine the date it was written. BIOSes older than year 2000 are assumed to be broken. In some circumstances, this assumption is wrong. If you see a message saying ACPI is disabled for this reason, try booting with acpi=force. If things work fine, send the output of dmidecode
(http://www.nongnu.org/dmidecode/) to [email protected] with an explanation of why your BIOS shouldn't be blacklisted.
CPU frequency scaling.
~~~~~~~~~~~~~~~~~~~~~~
Certain processors have the facility to scale their voltage/clockspeed. 2.6 introduces an interface to this feature, see Documentation/cpufreq for more information. This functionality also covers features like Intel's speedstep, and the Powernow! feature present in mobile AMD Athlons. In addition to x86 variants, this framework also supports various ARM CPUs. You can find a userspace daemon that monitors battery life and adjusts accordingly at: http://sourceforge.net/projects/cpufreqd
Background polling of MCE.
~~~~~~~~~~~~~~~~~~~~~~~~~~
The machine check handler has been extended so that it regularly polls for any problems on AMD Athlon, and Intel Pentium 4 systems. This may result in machine check exceptions occuring more frequently than they did in 2.4 on out of spec systems (Overclocking/inadequate cooling/underated PSU etc..).
LVM2 - DeviceMapper.
~~~~~~~~~~~~~~~~~~~~
The LVM1 code got removed wholesale, and replaced with a much better designed 'device mapper'.
- This is backwards compatible with the LVM1 disk format.
- Device mapper does require new tools to manage volumes however. You can get these from ftp://ftp.sistina.com/pub/LVM2/tools/
Debugging options.
~~~~~~~~~~~~~~~~~~
During the stabilising period, it's likely that the debugging options in the kernel hacking menu will trigger quite a few problems. Please report any of these problems to [email protected] rather than just disabling the relevant CONFIG_ options.
Merging of kksymoops means that the kernel will now spit out automatically decoded oopses (no more feeding them to ksymoops). For this reason, you should always enable the option in the kernel hacking menu labelled "Load all symbols for debugging/kksymoops".
Testing with CONFIG_PREEMPT will also increase the amount of debug code that gets enabled in the kernel. Kernel preemption gives us the ability to do a whole slew of debugging checks like sleeping with locks held, scheduling while atomic, exiting with locks held, etc.
Compiler issues.
~~~~~~~~~~~~~~~~
- The recommended compiler (for x86) is still 2.95.3.
- When compiled with a modern gcc (Ie gcc 3.x), 2.6 will use additional optimisations that 2.4 didn't. This may shake out compiler bugs that 2.4 didn't expose.
- Do not use gcc 3.0.x on x86 due to a stack pointer handling bug.
- gcc 2.96 is not supported with CONFIG_FRAME_POINTER=y due to a stack pointer handling bug.
Security concerns.
~~~~~~~~~~~~~~~~~~
Several security issues solved in 2.4 may not yet be forward ported to 2.6. For this reason 2.6.x kernels should not be tested on untrusted systems. Testing known 2.4 exploits and reporting results is useful.
SELinux.
~~~~~~~~
NSA Security-Enhanced Linux (SELinux) got merged in 2.6.
It is disabled by default and can be enabled with a boot time parameter selinux=1.
You can obtain SELinux tools and an example policy configuration from
http://www.nsa.gov/selinux
Networking.
~~~~~~~~~~~
- ebtables
The bridging firewall code got merged. To manage these you'll
need the ebtables tool available from
http://users.pandora.be/bart.de.schuymer/ebtables/
More on bridge-nf can be found at http://bridge.sourceforge.net
- Bridged packets can now be 'seen' by iptables.
- IPSec
Linux finally has IPSec support in mainline. Use the KAME tools port on
http://sourceforge.net/projects/ipsec-tools
For more info see http://www.lib.uaa.alaska.edu/linux-kernel/archive/2002-Week-44/1127.html
Also Bert Hubert has a howto at http://lartc.org/howto/lartc.ipsec.html
Additionally, ipsec-utils is at http://sourceforge.net/projects/ipsec-tools
Herbert Xu also has patches against FreeSWAN 2.00 to allow its userspace
to use the 2.6 IPSec functionality. They can be downloaded from
http://gondor.apana.org.au/~herbert/freeswan/
An additional HOWTO is at http://www.ipsec-howto.org
- Some applications may trigger the kernel to spit out warnings about
'process xxx using obsolete setsockopt SO_BSDCOMPAT' .
- Bind 9.2.2 checks for #ifdef SO_BSDCOMPAT in <asm/socket.h> correctly,
so a recompile is all that is needed.
- bind9-host from debian testing triggers, though the 'host' package doesn't.
- process `snmpd' is using obsolete setsockopt SO_BSDCOMPAT
- process `snmptrapd' is using obsolete setsockopt SO_BSDCOMPAT
- ntop uses obsolete (PF_INET,SOCK_PACKET)
- Users of boxes with >1 NIC may find that for eg, eth0 and eth1 refer to
the opposites of what they did in 2.4. This is a bug that will be fixed
before 2.6.0. One option (or management workaround) for this is to use 'nameif' to name Ethernet interfaces. There is a HOWTO for doing this at
<http://xenotime.net/linux/doc/network-interface-names.txt>
- Support for various new RFCs.
- RFC3173 (IP Payload Compression).
- RFC3041 (IPv6 Privacy Extensions).
- RFC2473 (IPv6 in IPv6 tunnels).
- RFC2960 (SCTP - see below).
- Linux reaches congestion collapse when subjected to heavy network load.
NAPI fixes this amongst other things and therefore improving network performance.
More info at http://www.cyberus.ca/~hadi/usenix-paper.tgz and ftp://robur.slu.se/pub/Linux/net-development/NAPI/
- IPVS (IP Virtual Server)
http://www.linuxvirtualserver.org/
- SCTP (Stream Control Transport Protocol)
SCTP is an IP based, message oriented reliable transport protocol with congestion control, support for transparent multi-homing and multiple ordered streams of messages. RFC2960 defines the core protocol.
More information about the protocol can be found at
http://www.ietf.org/rfc/rfc2960.txt
and about the Linux kernel implementation at
http://lksctp.sourceforge.net
- ANSI/IEEE 802.2 LLC type 2 Support
Full implementation of LLC 1 and 2 stack, used by Appletalk, IPX and Token Ring, also needed for the out of the tree, not yet functional NetBEUI stack and for the for Linux SNA.
This is based on the stack released under the GPL by Procom Inc. for the 2.0.30 Linux kernel.
Crypto
~~~~~~
- A generic crypto API has been merged, offering support for various algorithms (HMAC,MD4,MD5,SHA-1,SHA256,SHA384,SHA512,DES,Triple DES EDE, Blowfish, Twofish, Serpent, AES, CAST5, CAST6)
- This functionality is used by IPSec and the crypto-loop. It's possible that it will later also be available for use in userspace through a crypto device, possibly compatible with the OpenBSD crypto userspace.
- The in-kernel loopback device can now do crypto using the CryptoAPI. May need new userspace tools.
Ports.
~~~~~~
- 2.6 features support for several new architectures.
- x86-64 (AMD Hammer)
- ppc64
- UML (User mode Linux)
See http://user-mode-linux.sf.net for more information.
- uCLinux: m68k(w/o MMU), h8300 and v850. sh also added a uCLinux option.
- The 64 bit s390x port got collapsed into a single port, appearing as a config option in the base s390 arch.
- In the opposite direction, arm26 was split out from arm.
- x86 architecture also got 'subarch' support to support 'strange' x86 boxes (usually big boy toys). Currently supported subarchs include
- ES7000
- PC9800 (incomplete merge)
- VISWS (Was in 2.4, but now maintained again)
- Voyager. (http://www.hansenpartnership.com/voyager/)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Revision history:
0.46 - Update akpm's email address.
Mention LLC in the networking section.
Added link to 2.6 migration guide.
0.45 - DEVFS is obsolete.
Add pt_BR translation.
0.44 - lm_sensors.
x86 subarch support.
Add link to Joseph Pranevich's 2.6 doc.
Mention chattr +S
Update dmidecode URL
Fix sbf.c URL
Fix typos (s/proprietory/proprietary/)
add URL to patched cdda2wav.
Mention which filesystems does 2.6 actually supports.
Add a small note about SELinux
Add a few more RFCs which we now support in the Networking section Change IPVS to IPVS (IP Virtual Server)
Mention additional encryption algorithms that crypto api supports Add a few things to TODO section
Add URL to AIO.
Polish translation URL added.
Fix up a stray [email protected] that I missed.
Update spanish translation URL.
0.43 - oprofile URL changed.
Update Linus' email
Drop report of buggy Red Hat 3.2.2-5 gcc.
Remove reference to Jens' hacked cdrecord, standard version is now ok. PC Speaker note added.
Mention elevator=deadline for database folks.
Merge two deprecated sections.
Update URL to xfsprogs.
Mention Conectiva module-init-tools
Mention acpi=force, and DMI blacklist.
DRI CFT on X4.1
Pentium Pro SYSENTER is broken.
Update Alans list of IDE bugs
Add geometry translator workarounds.
Mention swapfile limitations.
Update Reiserfs new features.
Move some features marked as 'deprecated' to 'removed'. Add 'needs forward porting' section.
Pointer to synaptics touchpad driver webpage.
Update ipsec utils URL.
Mention usb-storage changes.
Modules are now .ko instead of .o
Mention how to change elevator on command line.
Document statfs64()
ext3 data=journal mode should now be fixed.
Fix iosched pathname.
Document USB changes.
Add SCTP to Networking.
s/2.5/2.6/ in most of the doc.
Mention AIO.
Add additional IPSEC howto URL.
Deprecate numeric sysctls
0.42 - CONFIG_LOG_BUF_SHIFT is now sanity checked by Kconfig
Hyperthreading scheduler improvements.
ALSA OSS emulation.
32bit uids in new quota.
Not all filesystems support nanosecond stat.
Note that NTFS still isn't R/W
Mention CryptoLoop.
0.41 - V=0 is now default. Document V=1
s/Redhat/Red Hat/ everywhere.
Added vmware sysenter note.
MTRR for vesafb
Various grammar fixes.
Selectable elevators.
PCI domains
0.40 - Callout tty devices are removed.
Added note about modules in Red Hat 9
0.39 - irqbalance note added.
Added ntop, snmp tools obsolete messages.
Added link to David Mosberger's O(1) page.
Mention Herbert Xu's FreeSWAN patches.
Add CONFIG_VGA_CONSOLE to the list of gotchas.
Add note about 2.4 .config's to gotchas.
Reword devpts note.
0.38 - Fixed URL to nameif
0.37 - devfs users now need to mount devpts.
mention h8300 port.
mentioned NTFS rewrite.
0.36 - Include Doug Gilberts 'positive SCSI spin'.
Mention NAPI.
Reword the CPU bug workaround that the O(1) scheduler exposes. Added 'Known Gotchas' section
0.35 - Note about KDE panel applets.
mount --move, dmask, fmask
Removed note about oprofile utilities being underdeveloped. Mention ext2 locking, and ext2/ext3 orlov allocator.
0.34 - Remove people.redhat.com NPTL URLs on Ulrich Dreppers request.
Added note about s390x going away.
Various kbuild updates.
Note about swap files.
Added note about -p1 vs -p0
Lots of typo fixes from Randy Dunlap.
RPM from RH9 seems to have problems.
0.33 - Networking RFCs section added.
0.32 - Added Soundmodem userspace replacement URL.
0.31 - ext3 data=journal breakage noted.
0.30 - Athlon powernow is now supported.
0.29 - Mention NIC renumbering and ACPI/APIC NIC bugs.
0.28 - SO_BSDCOMPAT obsolete messages, nfsutils.
0.27 - radeon -> radeonfb
0.26 - Added info about readprofile.
0.25 - Added cdrecord example. Added URL to Spanish translation. 0.1->0.24 - Unrecorded history
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Other Links.
http://www.kernelnewbies.org/status/
http://bugzilla.kernel.org/
http://www.kniggit.net/wwol26.html
http://thomer.com/linux/migrate-to-2.6.html
TODO: (Please mail me a few lines about these if you are
the owner of these).
PCI IDs (new_id, agpgart try_unsupported)
gadgetfs
libsysfs
kdev_t changes?
ISDN rewrite?
AFS
DVB
Hangcheck timer
/proc/sysrq-trigger
libata
Quoted 5 lines · Dave Jones
- An additional bug biting some people is that NICs fail to receive packets (usually notable by a NIC not getting a DHCP lease for eg, despite being sent one by the server). Booting with "noapic" "acpi=off" or a combination of both fixes this for most people. Additional breakage reports should go to Jeff Garzik <[email protected]>
As this is a symptom of irq routing failure, I dunno if bug reports should necessarily go to me :)
Quoted 3 lines
Stuff needing forward porting from 2.4.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- HFSPlus
I think Roman Zippel updated this recently?
Quoted 2 lines
- Direct booting from floppy is no longer supported. You should now use a boot loader program instead.
hmmm, what does this mean?
"make bzdisk" continues to work... it requires syslinux, however, so that the makefile may install the syslinux bootloader on the floppy.
Quoted 8 lines
Modules.
~~~~~~~~
- For Red Hat users, there's another pitfall in "/etc/rc.sysinit". During startup, the script sets up the binary used to dynamically load modules stored at "/proc/sys/kernel/modprobe". The initscript looks for "/proc/ksyms", but since it doesn't exist in 2.6 kernels, the binary used is "/sbin/true" instead.
Quoted 4 lines
This, eventually, will keep modules from working. Red Hat users will have to patch the "
/etc/rc.sysinit" script to set "/proc/sys/kernel/modprobe" to "/sbin/modprobe", even when "/proc/ksyms" doesn't exist.
Is this all still true?
Bill Nottingham graciously updated modutils to support 2.6.x, and locally all my Red Hat systems (with the updated modutils) work out of the box.
Quoted 9 lines
IDE.
~~~~
- The IDE code rewrite was subject to much criticism in early 2.5.x, which put off a lot of people from testing. This work was then subsequently dropped, and reverted back to a 2.4.18 IDE status.
Since then additional work has occurred, but not to the extent of the first cleanup attempts.
- Known problems with the current IDE code.
o Simplex IDE devices (eg Ali15x3) are missing DMA sometimes
I think this is fixed now... Bart?
Quoted 5 lines
o Most PCMCIA devices have unload races and may oops on eject
o Modular IDE does not yet work, modular IDE PCI modules sometimes oops on loading
oide_scsiis completely broken in 2.6 currently. Known problem. If you need it either use 2.4 or fix it 8)
also perhaps add a mention of new and spiffy Serial ATA drivers :)
Quoted 7 lines
CD Recording.
~~~~~~~~~~~~~
- Jens Axboe added the ability to use DMA for writing CDs on ATAPI devices. Writing CDs should be much faster than it
was in 2.4, and also less prone to buffer underruns and the like.
- With a recent cdrecord, you also no longer need ide-scsi in order to use an IDE CD writer.
Maybe add a pointer in the IDE section, pointing to this section? Since they both mention ide-scsi...
Quoted 8 lines
Generic VFS changes.
~~~~~~~~~~~~~~~~~~~~
- Directories can now be marked as synchronous using chattr +S, so that all changes will be immediately written to disk. Note, this does not guarantee atomicity, at least not for all filesystems and for all operations. You *can* be guaranteed that system calls will not return until the changes are on disk; note though that this does have has some significant performance impacts.
Also, too, this is filesystem-specific, right?
Quoted 6 lines
devfs.
~~~~~~
- devfs got somewhat stripped down and a lot of duplicate functionality got removed. You now need to enableCONFIG_DEVPTS_FS=yand mount the devpts filesystem in the same manner you would if you were not using devfs.
Wasn't this mentioned elsewhere in the document?
Quoted 3 lines
Improved BIOS table support.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Linux now supports various new BIOS extensions.
This is a bit vague. Even I have no idea what this refers to :)
Quoted 9 lines
Power management.
~~~~~~~~~~~~~~~~~
- 2.6 contains a more up to date snapshot of the ACPI driver. Should you experience any problems booting, try booting with the argument "acpi=off" to rule out any ACPI interaction. ACPI has a much more involved role in bringing the system up in 2.6 than it did in 2.4
- The old "acpismp=force" boot option is now obsolete, and will be ignored due to the old "mini ACPI" parser being removed.
- software suspend is still in development, and in need of more work.
Quoted 8 lines
Use with SMP and/or PREEMPT not advised.
- The ACPI code will do basic sanity checks on the DMI structure in the BIOS to determine the date it was written. BIOSes older than year 2000 are assumed to be broken. In some circumstances, this assumption is wrong. If you see a message saying ACPI is disabled for this reason, try booting with acpi=force. If things work fine, send the output of dmidecode
(http://www.nongnu.org/dmidecode/) to [email protected] with an explanation of why your BIOS shouldn't be blacklisted.
I dunno how much there is to add, but I just have a general feeling that software suspend and ACPI sleep state support has progressed since this was written.
Quoted 3 lines
Compiler issues.
~~~~~~~~~~~~~~~~
- The recommended compiler (for x86) is still 2.95.3.
I'm not sure this is still the case, in practice. Recent times have seen people breaking 2.95.x, which did not support the C99/C++ style of mixing variable declarations and code. People would forget this, and we only find out a few days later that the 2.95.x build was broken.
Quoted 10 lines
Networking.
~~~~~~~~~~~
- Some applications may trigger the kernel to spit out warnings about 'process xxx using obsolete setsockoptSO_BSDCOMPAT' .
- Bind 9.2.2 checks for #ifdefSO_BSDCOMPATin <asm/socket.h> correctly, so a recompile is all that is needed.
- bind9-host from debian testing triggers, though the 'host' package doesn't.
- process `snmpd' is using obsolete setsockoptSO_BSDCOMPAT
- process `snmptrapd' is using obsolete setsockoptSO_BSDCOMPAT
- ntop uses obsolete (PF_INET,SOCK_PACKET)
Wasn't there a recent lkml thread relating to this?
Quoted 7 lines · Jeff Garzik
Quoted 5 lines
(usually notable by a NIC not getting a DHCP lease for eg, despite being
sent one by the server). Booting with "noapic" "acpi=off" or a combination
of both fixes this for most people. Additional breakage reports should go to Jeff Garzik <[email protected]>As this is a symptom of irq routing failure, I dunno if bug reports should necessarily go to me :)
Sorry, you just looked guilty 8-)
Removed that last bit.
Quoted 9 lines
Quoted 3 lines
Stuff needing forward porting from 2.4.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- HFSPlusI think Roman Zippel updated this recently?
Quoted 2 lines
- Direct booting from floppy is no longer supported. You should now use a boot loader program instead.
hmmm, what does this mean?
it means you now need syslinux.
Quoted 2 lines
"make bzdisk" continues to work... it requires syslinux, however, so that the makefile may install the syslinux bootloader on the floppy.
Added a note mentioning that.
Quoted 3 lines
Quoted 2 lines
- For Red Hat users, there's another pitfall in "
/etc/rc.sysinit". .... deletiaIs this all still true?
Possibly not, I'll look into this.
Quoted 6 lines
Quoted 5 lines
o Most PCMCIA devices have unload races and may oops on eject
o Modular IDE does not yet work, modular IDE PCI modules sometimes oops on loading
oide_scsiis completely broken in 2.6 currently. Known problem. If you need it either use 2.4 or fix it 8)also perhaps add a mention of new and spiffy Serial ATA drivers :)
See the very last line of of the doc 8-)
Quoted 7 lines
Quoted 5 lines
- Jens Axboe added the ability to use DMA for writing CDs on ATAPI devices. Writing CDs should be much faster than it
was in 2.4, and also less prone to buffer underruns and the like.
- With a recent cdrecord, you also no longer need ide-scsi in order to use an IDE CD writer.Maybe add a pointer in the IDE section, pointing to this section? Since they both mention ide-scsi...
Done.
Quoted 7 lines
Quoted 6 lines
devfs.
~~~~~~
- devfs got somewhat stripped down and a lot of duplicate functionality got removed. You now need to enableCONFIG_DEVPTS_FS=yand mount the devpts filesystem in the same manner you would if you were not using devfs.Wasn't this mentioned elsewhere in the document?
briefly in the 'obsolete' section. the filesystem changes section typically has more info.
Quoted 5 lines
Quoted 3 lines
Improved BIOS table support.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Linux now supports various new BIOS extensions.This is a bit vague. Even I have no idea what this refers to :)
Then read on...
SBF, EDD & IPMI
Quoted 3 lines
I dunno how much there is to add, but I just have a general feeling that software suspend and ACPI sleep state support has progressed since this was written.
last I tried both were quite dismal, that was a few months back though.
Quoted 8 lines
Quoted 3 lines
Compiler issues.
~~~~~~~~~~~~~~~~
- The recommended compiler (for x86) is still 2.95.3.I'm not sure this is still the case, in practice. Recent times have seen people breaking 2.95.x, which did not support the C99/C++ style of mixing variable declarations and code. People would forget this, and we only find out a few days later that the 2.95.x build was broken.
*nod*, more and more distros are now shipping gcc3 as their stock compiler, so it's likely at some point things are going to change.
Quoted 10 lines
Quoted 8 lines
'process xxx using obsolete setsockopt
SO_BSDCOMPAT' .
- Bind 9.2.2 checks for #ifdefSO_BSDCOMPATin <asm/socket.h> correctly, so a recompile is all that is needed.
- bind9-host from debian testing triggers, though the 'host' package doesn't.
- process `snmpd' is using obsolete setsockoptSO_BSDCOMPAT
- process `snmptrapd' is using obsolete setsockoptSO_BSDCOMPAT
- ntop uses obsolete (PF_INET,SOCK_PACKET)Wasn't there a recent lkml thread relating to this?
not that I recall.
Thanks for the comments.
Dave
Quoted 10 lines
Quoted 8 lines
Quoted 3 lines
Compiler issues.
~~~~~~~~~~~~~~~~
- The recommended compiler (for x86) is still 2.95.3.I'm not sure this is still the case, in practice. Recent times have seen people breaking 2.95.x, which did not support the C99/C++ style of mixing variable declarations and code. People would forget this, and we only find out a few days later that the 2.95.x build was broken.
*nod*, more and more distros are now shipping gcc3 as their stock compiler,
That may be partially motivated by the fact that latest GLIBC no longer compiles with GCC 2.95.x
Quoted 1 line
so it's likely at some point things are going to change.
Do you think that GCC > 2.95.x actually produces noticably better code though? I'm not convinced yet, and yet compile times can be a lot higher.
By the way, it might be worth adding something to the input devices section suggesting that people try forcing set 3 if they have a keyboard which has keys that are not independently usable, and it's not a regression since 2.4. I have such a keyboard, and I saw a post recently describing similar behavior with a VMS keyboard.
John.
Quoted 9 lines
- Blank screen after decompressing kernel?
Make sure your .config has
CONFIG_INPUT=y
CONFIG_VT=y
CONFIG_VGA_CONSOLE=y
CONFIG_VT_CONSOLE=y
A lot of people have discovered that taking their .config from 2.4 and running make oldconfig to pick up new options leads to problems, notably withCONFIG_VTnot being set.
Would a couple of "default y" entries in Kconfig be a cure against this?
Quoted 11 lines · Holger Schurig
Quoted 9 lines
- Blank screen after decompressing kernel?
Make sure your .config has
CONFIG_INPUT=y
CONFIG_VT=y
CONFIG_VGA_CONSOLE=y
CONFIG_VT_CONSOLE=y
A lot of people have discovered that taking their .config from 2.4 and running make oldconfig to pick up new options leads to problems, notably withCONFIG_VTnot being set.Would a couple of "default y" entries in
Kconfigbe a cure against this?
Reports of this problem seem to have tailed off over the last few months so its entirely possible that this is now fixed.
Dave
On Thu, 30 Oct 2003 14:15:19 +0000 Dave Jones <[email protected]> wrote:
| The post-halloween document. v0.46 | (aka, 2.6 - what to expect) | Dave Jones <[email protected]> | (Updated as of 2.6.0test9)
| Known gotchas. | ~~~~~~~~~~~~~~ | Certain known bugs are being reported over and over. Here are the | workarounds. | - Can't load any modules? You need updated tools (See modules section below).
- depmod reports Unresolved symbols? depmod from modutils instead of depmod from module-init-tools is first in $PATH (might be different $PATHs as $USER and $ROOT)
| Regressions. | ~~~~~~~~~~~~ | - Some filesystems still need work (Intermezzo, UFS, HFS, HPFS..)
+ EFS (has a blocksize problem, depending on the device that the filesystem is being mounted on)
| Modules. | ~~~~~~~~ | - Modules now have a .ko suffix instead of .o
Some (older) versions of 'mkinitrd' don't search for modules that end with .ko, so update your mkinitrd if this is a problem.
[
or is this invalid since initrd isn't used any more? if that's the case, mention it in here somewhere... need to add anything about initramfs?
]
| Nanosecond stat: | ~~~~~~~~~~~~~~~~ | The stat64() syscall got changed to return jiffies granularity. | This allows make(1) to make better decisions on whether or not it | needs to recompile a file. Not all filesystems may support such precision.
General comment/change: use 'was changed' instead of 'got changed' (above). This is the second or third instance that I've noticed where 'got' should/would better be 'was', so if you would do a global search and conditional replace for those, that would be Good.
| Generic VFS changes. | ~~~~~~~~~~~~~~~~~~~~ | - Since Linux 2.5.1 it is possible to atomically move a subtree to | another place. The call is...
s/call/usage/
| mount --move olddir newdir
| sysfs. | ~~~~~~ | In simple terms, the sysfs filesystem is a saner way for | drivers to export their innards than /proc. | This filesystem is always compiled in, and can be mounted | just like another virtual filesystem. No userspace tools | beyond cat and echo are needed.
'tree' is also nice for viewing it.
| mount -t sysfs none /sys | | See Documentation/filesystems/sysfs.txt for more info. | | | | SELinux. | ~~~~~~~~ | NSA Security-Enhanced Linux (SELinux) got merged in 2.6. | It is disabled by default and can be enabled with a boot time parameter | selinux=1. | You can obtain SELinux tools and an example policy configuration from | http://www.nsa.gov/selinux
Somebody correct me here if needed...
selinux can't just be enabled by using 'selinux=1', if the config options are set for checking that.
The way that I read security/selinux/Kconfig and hooks.c, if SECURITY_SELINUX_BOOTPARAM is enabled, then the 'selinux' boot option is also enabled. However, it can be disabled by using 'selinux=0' as a kernel boot option.
| Networking. | ~~~~~~~~~~~ | - Users of boxes with >1 NIC may find that for eg, eth0 and eth1 refer to | the opposites of what they did in 2.4. This is a bug that will be fixed | before 2.6.0. One option (or management workaround) for this is to use
It will be? It's time to do it then, if it's not already done...
| 'nameif' to name Ethernet interfaces. There is a HOWTO for doing this at | <http://xenotime.net/linux/doc/network-interface-names.txt>
| - SCTP (Stream Control Transport Protocol)
RFC 2960 - Stream Control Transmission Protocol
Overall a very good job, Dave. Thanks for keeping this updated.
--
~Randy
Quoted 8 lines · * Randy.Dunlap ([email protected]) wrote:
| SELinux. | ~~~~~~~~ | NSA Security-Enhanced Linux (SELinux) got merged in 2.6. | It is disabled by default and can be enabled with a boot time parameter | selinux=1. | You can obtain SELinux tools and an example policy configuration from | http://www.nsa.gov/selinux
Quoted 7 lines
Somebody correct me here if needed...
selinux can't just be enabled by using 'selinux=1', if the config options are set for checking that.
The way that I readsecurity/selinux/Kconfigandhooks.c, ifSECURITY_SELINUX_BOOTPARAMis enabled, then the 'selinux' boot option is also enabled. However, it can be disabled by using 'selinux=0' as a kernel boot option.
This is correct. SELinux defaults to not being config'd in. If you config it in it defaults to enabled. If you also config the bootparam you can use that param to disable it, otherwise selinux=1 is redundant as that's the default.
thanks,
-chris
Signature · 1 line
Linux Security Modules http://lsm.immunix.org http://lsm.bkbits.net
Quoted 1 line · Randy.Dunlap
Overall a very good job, Dave. Thanks for keeping this updated.
Thanks for the feedback. I've updated the working version[1] of the doc with your (and everyone elses so far) suggestions.
Dave
[1] http://www.codemonkey.org.uk/docs/post-halloween-2.6.txt-devel
On Thu, 30 Oct 2003 18:38:56 +0000 Dave Jones <[email protected]> wrote:
| On Thu, Oct 30, 2003 at 08:52:22AM -0800, Randy.Dunlap wrote: | > Overall a very good job, Dave. Thanks for keeping this updated. | | Thanks for the feedback. I've updated the working version[1] of | the doc with your (and everyone elses so far) suggestions. | | Dave | | [1] http://www.codemonkey.org.uk/docs/post-halloween-2.6.txt-devel
Yes, looks good. But please change this one also:
| - SCTP (Stream Control Transport Protocol)
Stream Control Transmission Protocol
--
~Randy
Quoted 4 lines · Randy.Dunlap
| [1] http://www.codemonkey.org.uk/docs/post-halloween-2.6.txt-devel
Yes, looks good. But please change this one also:
| - SCTP (Stream Control Transport Protocol)
Stream Control Transmission Protocol
Done. Good work eagle-eyes 8-)
Dave
Quoted 4 lines · Dave Jones
Modules.
~~~~~~~~
- Debian sid or Conectiva snapshot users can just use 'apt-get install module-init-tools'
It's in Debian sarge, too.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Dave, LKML!
Quoted 1 line · Dave Jones
The post-halloween document. v0.46
...
Quoted 1 line
Input layer.
...
Quoted 3 lines
- If you find your keyboard/mouse still don't work, edit the file
drivers/input/serio/i8042.c, and replace the #undef DEBUG with a #define DEBUG
, recompile and reinstall.
Quoted 3 lines
When you boot, you should now see a lot more debugging information. Forward this information to Vojtech Pavlik <[email protected]>
BYtE
Philipp
Signature · 3 lines
/ / (_)__ __ ____ __ Philipp Hahn / /__/ / _ \/ // /\ \/ / /____/_/_//_/\_,_/ /_/\_\ [email protected]
Quoted 4 lines · Dave Jones
- The format of
/proc/statchanged, which could break some applications that still depend on the old layout. Currently the only known application to break is the java 'DOTS' app. (http://bugme.osdl.org/show_bug.cgi?id=277)
'xosview' is also broken by this change.
Quoted 8 lines · bd
Quoted 4 lines · Dave Jones
- The format of
/proc/statchanged, which could break some applications that still depend on the old layout. Currently the only known application to break is the java 'DOTS' app. (http://bugme.osdl.org/show_bug.cgi?id=277)'xosview' is also broken by this change.
I can confirm this, also xawtv and gnomeradio, neither can apparently see the devices registered in at boot time according to grepping /var/log/messages.
Quoted 5 lines
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Signature · 7 lines
Cheers, Gene AMD K6-III@500mhz 320M Athlon1600XP@1400mhz 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Quoted 6 lines · bd
Quoted 4 lines
- The format of
/proc/statchanged, which could break some applications that still depend on the old layout. Currently the only known application to break is the java 'DOTS' app. (http://bugme.osdl.org/show_bug.cgi?id=277)'xosview' is also broken by this change.
there's a patch around the net that fix it (alternatively, you got it from the xosview source rpm from mandrake contribs)
Quoted 6 lines · Dave Jones
Process scheduler improvements.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Another much talked about feature. Ingo Molnar reworked the process scheduler to use an O(1) algorithm. In operation, you should notice no changes with low loads, and increased scalability with large numbers of processes, especially on large SMP systems.
I think we should mention excellent Con Kolivas contributions to the 2.6 kernel scheduler. He did a great job in tunning the scheduler for maximum interactive feeling.
Quoted 10 lines · Felipe Alfaro Solana
Quoted 6 lines
Process scheduler improvements.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Another much talked about feature. Ingo Molnar reworked the process scheduler to use an O(1) algorithm. In operation, you should notice no changes with low loads, and increased scalability with large numbers of processes, especially on large SMP systems.I think we should mention excellent Con Kolivas contributions to the 2.6 kernel scheduler. He did a great job in tunning the scheduler for maximum interactive feeling.
With no disrespect to Con, I'm not going to do this, or I'll end up spending far too much time adding credits to everyone whoever improved something in 2.5. There's a place for such stuff, and it's aptly named 'CREDITS' in toplevel of src dir.
Dave
Quoted 5 lines · Dave Jones
Quoted 5 lines
Quoted 4 lines
- Another much talked about feature. Ingo Molnar reworked the process scheduler to use an O(1) algorithm. In operation, you should notice no changes with low loads, and increased scalability with large numbers of processes, especially on large SMP systems.
Quoted 8 lines
Quoted 3 lines
I think we should mention excellent Con Kolivas contributions to the 2.6 kernel scheduler. He did a great job in tunning the scheduler for maximum interactive feeling.
With no disrespect to Con, I'm not going to do this, or I'll end up spending far too much time adding credits to everyone whoever improved something in 2.5. There's a place for such stuff, and it's aptly named 'CREDITS' in toplevel of src dir.
Given that Ingo's reworkings led to horrendous interactive performance for many desktop users for a long time during 2.5, it seems a little odd to prefer crediting Ingo over Con given what you've said. How about just mentioning that we have an O(1) scheduler, plus there has been significant work on interactive performance.
john
Signature · 2 lines
Khendon's Law: If the same point is made twice by the same person, the thread is over.
Quoted 5 lines · John Levon
Given that Ingo's reworkings led to horrendous interactive performance for many desktop users for a long time during 2.5, it seems a little odd to prefer crediting Ingo over Con given what you've said. mentioning that we have an O(1) scheduler, plus there has been significant work on interactive performance.
Preference has nothing to do with it. It comes down to one definitive source of information.
Quoted 1 line
From
MAINTAINERS..
SCHEDULER
P: Ingo Molnar M: [email protected] P: Robert Love [the preemptible kernel bits] M: [email protected] L: [email protected] S: Maintained
Shall I add rml there too? How about the other dozen or so people that hacked on O(1) over the last year or so? Get real, I'm not about to waste my time going over attributions, nor further entertain any responses in this thread.
Dave
Re: ide-scsi is working [was: Post-halloween doc updates.]
In reply to Dave JonesHi!!
Dave Jones wrote:
[...]
Quoted 3 lines
IDE.
~~~~
[...]
Quoted 2 lines
o
ide_scsiis completely broken in 2.6 currently. Known problem. If you need it either use 2.4 or fix it 8)
I have 2.6.0-test9 and I just recorded knoppix to CD with ide-scsi. In fact, it never stopped working, in this box, with several 2.6.* kernels.
Hmmmm maybe it's because my cdrecorder is USB?
The box has debian sid and cdrecord is:
puma:/# cdrecord -scanbus
Cdrecord-Clone 2.01a19 (i686-pc-linux-gnu) Copyright (C) 1995-2003 Jörg Schilling
Linux sg driver version: 3.5.29
Using libscg version 'schily-0.7'
scsibus2:
2,0,0 200) 'HP ' 'CD-Writer cd4f ' '1.0A' Removable CD-ROM
2,1,0 201) *
2,2,0 202) *
2,3,0 203) *
2,4,0 204) *
2,5,0 205) *
2,6,0 206) *
2,7,0 207) *
scsibus3:
3,0,0 300) 'HP ' 'psc 2210 ' '1.00' Removable Disk
3,1,0 301) *
3,2,0 302) *
3,3,0 303) *
3,4,0 304) *
3,5,0 305) *
3,6,0 306) *
3,7,0 307) *
puma:/#Just to be clear: I *can* record cd's! :-)
Can I provide any more info?
Regards,
Nuno Silva
Re: ide-scsi is working [was: Post-halloween doc updates.]
Quoted 12 lines · Nuno Silva
Hi!!
Dave Jones wrote:
[...]
Quoted 2 lines
IDE.
~~~~[...]
Quoted 7 lines
Quoted 2 lines
o
ide_scsiis completely broken in 2.6 currently. Known problem. If you need it either use 2.4 or fix it 8)I have 2.6.0-test9 and I just recorded knoppix to CD with ide-scsi. In fact, it never stopped working, in this box, with several 2.6.* kernels.
Hmmmm maybe it's because my cdrecorder is USB?
So you don't even use ide-scsi... and it is broken.
--bartlomiej
Quoted 3 lines · Dave Jones
Deprecated/obsolete features.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- devfs has been obsoleted in favour of udev (http://www.kernel.org/pub/linux/utils/kernel/hotplug/)
s/has been obsoleted/will be olsoleted some day/
many devices need to be added to sysfs,
before udev+sysfs can replace devfs.
Quoted 9 lines · Andreas Jellinghaus
Quoted 3 lines · Dave Jones
Deprecated/obsolete features.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- devfs has been obsoleted in favour of udev (http://www.kernel.org/pub/linux/utils/kernel/hotplug/)s/has been obsoleted/will be olsoleted some day/
many devices need to be added to sysfs,
before udev+sysfs can replace devfs.
So the correct term should be deprecated rather than obsoleted. Which can be taken to mean "please which switch when possible".
-Paul
And APM suspend seems to have broken in -test8. Does it work for anyone?
Ian
Quoted 17 lines · Dave Jones
Power management.
~~~~~~~~~~~~~~~~~
- 2.6 contains a more up to date snapshot of the ACPI driver. Should you experience any problems booting, try booting with the argument "acpi=off" to rule out any ACPI interaction. ACPI has a much more involved role in bringing the system up in 2.6 than it did in 2.4
- The old "acpismp=force" boot option is now obsolete, and will be ignored due to the old "mini ACPI" parser being removed.
- software suspend is still in development, and in need of more work. Use with SMP and/or PREEMPT not advised.
- The ACPI code will do basic sanity checks on the DMI structure in the BIOS to determine the date it was written. BIOSes older than year 2000 are assumed to be broken. In some circumstances, this assumption is wrong. If you see a message saying ACPI is disabled for this reason, try booting with acpi=force. If things work fine, send the output of dmidecode
(http://www.nongnu.org/dmidecode/) to [email protected] with an explanation of why your BIOS shouldn't be blacklisted.
Quoted 2 lines · Ian Soboroff
And APM suspend seems to have broken in -test8. Does it work for anyone?
Doesn't work for me.
Now, taking off my "open source co-operative hat" and placing my "reality" hat on, I'd suggest that anyone who finds that APM doesn't work to consider it a dead loss - It's an obsolete technology, and therefore no one is interested in it anymore. I've reported the problem multiple times here and there's been very little, if any, reaction, so this seems to back that up.
Now, imagine the PCMCIA maintainer being unable to test the suspend/ resume functionality during the stability freeze. Oh, sorry, you don't have to imagine that hard. 8)
Quoted 19 lines
Quoted 17 lines · Dave Jones
Power management.
~~~~~~~~~~~~~~~~~
- 2.6 contains a more up to date snapshot of the ACPI driver. Should you experience any problems booting, try booting with the argument "acpi=off" to rule out any ACPI interaction. ACPI has a much more involved role in bringing the system up in 2.6 than it did in 2.4
- The old "acpismp=force" boot option is now obsolete, and will be ignored due to the old "mini ACPI" parser being removed.
- software suspend is still in development, and in need of more work. Use with SMP and/or PREEMPT not advised.
- The ACPI code will do basic sanity checks on the DMI structure in the BIOS to determine the date it was written. BIOSes older than year 2000 are assumed to be broken. In some circumstances, this assumption is wrong. If you see a message saying ACPI is disabled for this reason, try booting with acpi=force. If things work fine, send the output of dmidecode
(http://www.nongnu.org/dmidecode/) to [email protected] with an explanation of why your BIOS shouldn't be blacklisted.
Signature · 4 lines
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 PCMCIA - http://pcmcia.arm.linux.org.uk/
2.6 Serial coreQuoted 4 lines · Russell King
Quoted 2 lines · Ian Soboroff
And APM suspend seems to have broken in -test8. Does it work for anyone?
Quoted 8 lines
Doesn't work for me.
Now, taking off my "open source co-operative hat" and placing my "reality" hat on, I'd suggest that anyone who finds that APM doesn't work to consider it a dead loss - It's an obsolete technology, and therefore no one is interested in it anymore. I've reported the problem multiple times here and there's been very little, if any, reaction, so this seems to back that up.
Well, ok, but the alternatives are ACPI, which has always been spotty, and two competing power management schemes from Patrick and Pavel, neither of which seem to actually work yet. Wouldn't it be nice to have at least one working method of putting a laptop to sleep?
Once I get some free time (maybe next week, who knows) I'll try backing out bits of the -test8 patch to see what broke. In the meantime, -test7 works great.
Ian
Quoted 1 line · Ian Soboroff
Well, ok, but the alternatives are ACPI, which has always been spotty,
Wrong - there are systems out there which do not support ACPI (or maybe the kernel doesn't support the ACPI which is implemented in them.)
Quoted 3 lines
and two competing power management schemes from Patrick and Pavel, neither of which seem to actually work yet. Wouldn't it be nice to have at least one working method of putting a laptop to sleep?
Indeed. With 2.6 there are zero ways I can make my x86 laptop suspend or hibernate.
Quoted 3 lines
Once I get some free time (maybe next week, who knows) I'll try backing out bits of the -test8 patch to see what broke. In the meantime, -test7 works great.
Oh, my problem goes back many kernel versions. It's nothing new. Somewhere during 2.5 development (current thinking is somewhere around 2.5.3x) something changed which makes the APM BIOS believe that the machine isn't in the correct state to suspend.
However, exactly what that is seems to be impossible to track down. Building a kernel minimal driver support results in a system which can't suspend... and by "minimal" that includes dropping stuff like the input layer and console completely from the kernel.
It basically comes down to some really subtle change in what 2.6 is does differently from 2.4.
Signature · 4 lines
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 PCMCIA - http://pcmcia.arm.linux.org.uk/
2.6 Serial coreQuoted 9 lines · Ian Soboroff
Quoted 6 lines · Russell King
Quoted 2 lines · Ian Soboroff
And APM suspend seems to have broken in -test8. Does it work for anyone?
Doesn't work for me.
APM working here nicely with -test9 on Thinkpad A31. Stilling getting 1394 badness when suspending/resuming with my cardbus 1394 controller plugged in. Other than that, works pretty seamlessy with no cardbus cards plugged.
Even radeon with VESA suspend/resume, DRI and X 4.3 (using some precarious vtswitch calls in /etc/apm/event.d/). wlan-ng works okay although it requires rmmod/modprobe to wake up properly after resume.
Quoted 6 lines
Quoted 6 lines
Now, taking off my "open source co-operative hat" and placing my "reality" hat on, I'd suggest that anyone who finds that APM doesn't work to consider it a dead loss - It's an obsolete technology, and therefore no one is interested in it anymore. I've reported the problem multiple times here and there's been very little, if any, reaction, so this seems to back that up.
Hmmm, well for me; a working APM (here) is much prefered to non- working ACPI suspend/resume. I'd prefer not to obsolete APM until there is a working alternative (with mature userspace tools).
~mc
In article <[email protected]>,
Ian Soboroff <[email protected]> wrote:
| Well, ok, but the alternatives are ACPI, which has always been spotty, | and two competing power management schemes from Patrick and Pavel, | neither of which seem to actually work yet. Wouldn't it be nice to | have at least one working method of putting a laptop to sleep?
I have no problem putting the laptop to sleep. On the other hand, waking it up...
Signature · 3 lines
bill davidsen <[email protected]> CTO, TMR Associates, Inc Doing interesting things with little computers since 1979.
Quoted 10 lines · bill davidsen
In article <[email protected]>,
Ian Soboroff <[email protected]> wrote:
| Well, ok, but the alternatives are ACPI, which has always been spotty, | and two competing power management schemes from Patrick and Pavel, | neither of which seem to actually work yet. Wouldn't it be nice to | have at least one working method of putting a laptop to sleep?I have no problem putting the laptop to sleep. On the other hand, waking it up...
Which suspend version are you using? echo -n "4">/proc/acpi/something or echo -n "disk" > sys/power/state?
The second works for me (for a slightly malleable value of "works": resume has never failed on me, suspend works maybe 9 times out of 10, and either immediately resumes without power down (apparently with no harm done, it just didn't work) or panics the rest of the time).
The first has never come close to working (for me).
Rob
Quoted 4 lines
Quoted 2 lines
And APM suspend seems to have broken in -test8. Does it work for anyone?
Doesn't work for me.
works for me. it isn't rock solid as in 2.4 but most times it resumes without problems even using dma on ide controller (in 2.4 i should disable dma before suspend and couldn't enable it later).
Quoted 6 lines
Now, taking off my "open source co-operative hat" and placing my "reality" hat on, I'd suggest that anyone who finds that APM doesn't work to consider it a dead loss - It's an obsolete technology, and therefore no one is interested in it anymore. I've reported the problem multiple times here and there's been very little, if any, reaction, so this seems to back that up.
i guess it isn't exactly the problem. maybe there's no _developers_ with time and skills to work on it. i would do it if i have enough skills. when stock 2.6 comes out we'll have a lot of people complaining about APM because there a lot of notebooks out there that still needs it. including mine.
--
aris
Hi!
Quoted 6 lines
Doesn't work for me.
Now, taking off my "open source co-operative hat" and placing my "reality" hat on, I'd suggest that anyone who finds that APM doesn't work to consider it a dead loss - It's an obsolete technology, and therefore no one is interested in it anymore.
Actually APM is so simple technology (very similar to ACPI S1) that I'm reasonably sure we'll get it working in 2.6.5.
Signature · 2 lines
Pavel Written on sharp zaurus, because my Velo1 broke. If you have Velo you don't need...
Hi Dave,
hi lkml,
Quoted 9 lines · Dave Jones
Threading improvements.
~~~~~~~~~~~~~~~~~~~~~~~
- Ingo Molnar put a lot of work into threading improvements for 2.6. Some of the features of this work are:
- Generic pid allocator (arbitrary number of PIDs with no slowdown, unified pidhash).
- Thread Local Storage syscalls
-sys_clone()enhancements (CLONE_SETTLS,CLONE_PARENT_SETTID,CLONE_SETTID,CLONE_CLEARTID,CLONE_DETACHED)
Quoted 5 lines
- POSIX thread signals stuff (atomic signals, shared signals, etc.) - Per-CPU GDT - Threaded coredumping support - sys_exit() speedups (O(1) exit) - Generic, improved futexes, vcache
vcache is gone. futex pages are swappable now and not pinned anymore.
Good work. I love this document ;-)
Regards
Ingo Oeser
Could you add the following to the Regressions section?
- In 2.4 users of Japanese keyboards were able to type '|' or '\' characters without loading any custom keymap on the console. With the keymap in 2.6, this is not possible anymore. People with these keyboards have to load a keymap with loadkeys rebuilt from the source, since loadkeys in some vendor distributions cannot load keycodes larger than 127. There is a patch to fix this, but it has not been integrated
(http://tinyurl.com/t75a).
Quoted 3 lines · Dave Jones
Modules.
~~~~~~~~
- Modules now have a .ko suffix instead of .o
- Building external modules requires the use of kbuild, so a full kernel src package is needed. See also Documentation/kbuild/modules.txt
Quoted 2 lines
Kernel build system.
~~~~~~~~~~~~~~~~~~~~
I suggest the following ordering and content:
- The build system is much improved compared to 2.4.
You should notice quicker builds, and no spontaneous rebuilds of files on subsequent builds from already built trees.
- There are two new graphical config tools.
"make xconfig" requires the qt libraries.
"make gconfig" requires gtk libraries.
- Make menuconfig/oldconfig has no user-visible changes other than speed, whilst numerous improvements have been made.
- 'make help' provides a list of typical targets, including several new debugging targets ('allyesconfig' 'allnoconfig' 'allmodconfig' + more).
- "make" is now the preferred command, without a target; it builds <arch-zimage> and modules. See 'make help' for what is built by default.
- "make -jN" is now the preferred parallel-make execution.
Do not bother to provide "MAKE=xxx"
- Output files can be directed to an alternative directory. Use 'make O=dir' to locate all output files (including .config) in 'dir'.
- The build is now much less verbose. If you want to see exactly what's going on, try "make V=1" or set KBUILD_VERBOSE=1 in your environment.
- 'make kernel/mm.o' will build the named file, provided a corresponding source exists. This also works for (non-composite) modules.
- 'make kernel/' will recursively build all files in the specified subdirectory and below.
- 'make dep' has been obsoleted and does no longer have any effect.
- Note: The new configuration system is not CML2 related.
- Also note: Whilst some ideas were taken from it, Keith Owens' kbuild-2.5 project was not integrated.
Sam