ftp://ftp.kernel.org/pub/linux/kernel/people/alan/2.4/
2.4.1-ac7
o Rebalance the 2.4.1 VM (Rik van Riel)
| This should make things feel a lot faster especially
| on small boxes .. feedback to Rik
o Silence osf syscall error printk (Ivan Kokshaysky)
o Don't trust ARC irq routing on ruffian (Ivan Kokshaysky)
o Report the right module on 3c59x for pcmcia (Arjan van de Ven)
o Update i82365 driver to add locks, delays, and (Arjan van de Ven)
'bouncing' on the card detect
o Get the name right on ide-cs (v ide_cs) and do (Arjan van de Ven)
resource claims
o Merge parport_cs (David Hinds)
o Merge sedlbauer_cs (Marcus Niemann)
o Fix a bug in the Cyrix pirq routing (me)
2.4.1-ac6
o Fix eepro100 reporting on lockup fix (Ion Badulescu)
o Clean up i810 error message (me)
o Fix S390 build bug (me)
o Update version id on cpqarray driver (Charles White)
o Further aic7xxx fixes (Doug Ledford)
| again please report aic7xxx stuff to Doug
o Further maxiradio cleanups (Dimitromanolakis Apostolos)
o Change ide to use mdelay cleanly (Petr Vandrovec)
| Still broken for PROMISE if no IDE_CS
o Fix duplicated ncpfs fix (Petr Vandrovec)
o Improve inode hash function (Dave Miller)
o Correct 62 misspellings of transferred (Andre Dahlqvist)
o Update AC97 codec setup and tables (Jeff Garzik)
2.4.1-ac5
o Fix zero page corruption (Ben La Haise)
o Elevator corruption fixes (Jens Axboe, Linus)
o Fix fdatasync possible corruption problem (Arjan van de Ven)
o Further KSLI ethernet fixes (Eric Sandeen)
o Merge the correct version of the pm fixes (me)
| noted by Mikael Pettersson
o Account for inode/dcache in free memory (Rik van Riel)
o Add info on how to check reiserfsprogs versions (Steven Cole)
o Disable write combining on serverworks LE chips (Mark Rusk)
o Fix via audio crashes (Jeff Garzik)
o Fix ip accounting rules bug (Rusty Russell)
o Handle USB printers that use device not (Johannes Erdfelt)
interface descriptors
o Fix wheel on graphire usb tablet (Peter Hofmann)
o Clean up maxiradio driver (Francois Romieu)
o Fix visor USB size reporting on buffers (Greg Kroah-Hartmann)
o Update USB serial documentation (Greg Kroah-Hartmann)
o Fix locking on etherworks3 ethernet (Jeff Garzik)
o Fix empeg USB driver problems (Gary Brubaker)
o Generic USB serial driver fixes (Greg Kroah-Hartmann)
o Update USB serial configure.help (Greg Kroah-Hartmann)
o Add more device support to mct_u232 USB (Cornel Ciocirlan)
o Fix typo in asm-ppc/semaphore.h (Andre Dahlqvist)
o Report reiserfs tools in ver_linux (Steven Cole)
o Fix resource leaks in NCR_53c406, atari_scsi (Rasmus Andersen)
and qlogicisp
o Move pci_enable_device earlier for hamachi (Dave Jones)
o Type 6 drives are apparently floppy 2.88M (Dave Jones)
o Remove duplicate pci_enable_device in ne2kpci (Dave Jones)
2.4.1-ac4
o Fix sk_in use counting in svcsock.c (Neil Brown)
| Not yet a complete and final agreed solution
o Add support for KLSI USB ethernet (Brad Hards,
Stephane Alnet, 'the Zapman', and co)
o Update aic7xxx driver (Doug Ledford)
| Please test this carefully and cc reports to Doug
o Add help for CONFIG_INPUT (Steven Cole)
o 3c523 driver update (Tom Sightler)
o Fix reiserfs Changes entry further (Steven Cole)
o Limit ide scatter gather to 128 blocks (Jens Axboe)
o Merge hppa config.in changes (Matthew Wilcox)
o Fix tx timeout recovery on via rhine (Manfred Spraul)
o Fix stale comments in fs/block_dev.c (Tigran Aivazian)
o Further defxx driver work (Maciej Rozycki)
o winbond 840 reported wrong setting value (Maciej Rozycki)
o Guillemot Maxi radio support (Dimitromanolakis Apostolos)
o Allow sleeping in pm callbacks but with locking (me)
working
2.4.1-ac3
o Remove ancient dead net/Changes file (Janice Girouard)
o Merge Linus 2.4.2pre1
o Resync xirc2ps with Dave Hinds tree (dilinger)
o Finish sorting out ramfs problems (Mike Galbraith)
o Update AWE32 documentation (Andre Dahlqvist)
o Remove reference to dead PPP documentation (Andre Dahlqvist)
o Make max_map tunable (Werner Almesberger)
o Fix dead references to java support in some (Andre Dahlqvist)
arch/config
o Make shmfs estimate size limits if none set (Christoph Rohland)
o Revert Crusoe hanging pci hanging changes
| Im still chasing something weird in this
| area that some of the pci changes I have fixes...
o Merge HPPA hackers into CREDITS (Mathew Wilcox)
o Merge some of the HPPA updates (Mathew Wilcox)
o Add Reiserfs tools to changes (Steven Cole)
o Fix i2o Configure.help typo (YOSHIMURA Keitaro)
o SuperH HD64465 host bridge support (Greg Banks)
o Fix modversion.h includes (Keith Owens)
o Tlan driver probing updates (Jeff Garzik)
o Change media drivers to use new style module (me)
locking
| Janitorial job - fix the last ones that
| don't use module_*() and dump the init code
2.4.1-ac2
o Fix matrox G450 framebuffer support (Petr Vandrovec)
o Fix description of DMA-mapping.txt (Dave Miller)
o Fix accidental revert of classifier bug (Dave Miller)
o Fix accidental revert of isdn change
o Fix datagram hang on shutdown (Alexey Kuznetsov)
o Fix 64bit build of clntproc (Michal Jaegermann)
| wants a tidier solution yet
o Fix ide toc caching bug introduced in 2.4.0 (Fredrik Vraalsen)
| this should fix the DVD playback problems
o Swapfs renaming and final bits (Christoph Rohland)
o Further APIC/NMI updates (Mikael Pettersson)
o Add further kernel doc contributions (John Levon)
o ACPI battery tweaks (Pavel Machek)
o Further ramfs fixes (Ingo Oeser)
o ROMFS fixes (Mike Galbraith)
o CS4281 fixes (Thomas Woller)
o Shift to authors official fixes for acenic (Jes Sorensen)
o Update the usb host<->host network drivers (David Brownell)
| Experimental but he wanted feedback so if you
| have one beat it up a bit
2.4.1-ac1
o Resync with Linus 2.4.1
o Fix recursive make_request crash (Ingo Molnar)
o Updated VIA IDE driver (Vojtech Pavlik)
| Please exercise due care and caution testing this
| bit...
o Fix case where threaded apps might write to (Ben LaHaise)
freed kernel memory
o Fix ACPI oopses on tecra (apparently bios bugs) (Pavel Machek)
o AHA152x fixes from maintainer (Juergen Fischer)
o Fix case where scsi could hang on boot waiting (Rogier Wolff)
for a disk spinup
o Further maestro3 pm work (Zach Brown)
o Further NTFS fixes (Yuri Per)
o Add GNU make to the list of URLs in Changes (Steven Cole)
o Make dmx3191d enable device before touching it (Rasmus Andersen)
o Make the sbpcd driver actually useful in 2.4 (Paul Gortmaker)
o Make buslogic enable device before touching it (Rasmus Andersen)
o Fix tty module locking mishandling (Maciej Rozycki)
o Workaround code for APIC problems with ne2k (Maciej Rozycki)
| this will break original 82489DX devices for now
| ie _very_ early dual pentium boards
o Fix iptos netfilter bug (Rusty Russell)
o Fix get/set_fpu_mxcsr to check xmm ont fxsr (Doug Ledford)
o Fix name_to_kdev_t symbol (Adam J Richter)
o Update magic sysrq docs (Jeremy Dolan)
o Support for ETinc PCIsync boards (Francois Romieu)
o Mass duplicated word spelling fixes (Dave Jones)
o Update sb driver to use spinlocks (Chris Rankin)
o Fix leak in bmac driver (Hans Grobler)
o Fix kmalloc check in atm/common (Hans Grobler)
o Fix buffer leak in defxx (Hans Grobler)
o Fix kmalloc check in netrom driver (Hans Grobler)
|BTW side exercise - how about using vmalloc here ?
o Ditto for rose (Hans Grobler)
|Ditto for comment ;)
o Fix lockd 64bit handling (H J Lu)
o Tidy pci_match_device ifdefs (Rasmus Andersen)
o Fix qla1280 handling of registration failure (Rasmus Andersen
Rakesh Rakesh)
o Config include fixes (Niels Jensen)
o MatroxFB updates (Petr Vandrovec)
o Tidy fat_read_super to use get_hardsect_size (Tigran Aivazian)
o Fix m68k bitops ffs() (Geert Uytterhoeven)
o Fix ip_nat_standalone ksyms stuff (Rusty Russell)
o Fix copy_from_user mishandling in ip_fw_compat (Rusty Russell)
o Fix romfs for 2.4ac maxbytes (Mike Galbraith)
o filemap/aging updates (Rik van Riel)
o Enable device before reading irq in ne2k-pci (Martin Diehl)
o Remove surplus nr_ioapics definition (Rasmus Andersen)
o S/390 build fixes (Florian Laroche)
o Advansys driver fixes/portability (Arnaldo Carvalho
de Melo)
o Fix out of message handling error in i2o_block (Jason Lai)
o Fix bit granularity of 32 in ACPI driver (Adam J Richter)
o Fix unsafe casting for ARM on NFS root mount (Russell King)
o Fix mxcsr masking on pentium IV (Doug Ledford)
o Update u14/eata drivers to 6.03 (Dario Ballabio)
o Fix signed/unsigned mess in sysctl handlers (me)
2.4.0-ac12
o Merge Linus -pre10
| This replaces our ppc and most net
| protocol diffs
o Fix escaped waitpid prototypes (Dave Miller)
o smctr driver fixes (Jeff Garzik)
o Fix hga probing (Paul Gortmaker)
o Fix 8139too to enable pci before using pci vals (Jeff Garzik)
o maestro3 crash on pm fix
o Further lance cleaning (Arnaldo Carvalho
de Melo)
o depca init cleanup (Jeff Garzik)
o Remove aironet dead code, add probe table (Jeff Garzik)
o hp100 driver cleanup (Arnaldo Carvalho
de Melo)
o Make tms380 driver work (Jeff Garzik)
o Blacklist IBM drivers on HPT366 for the moment (David Woodhouse)
o Fix write_room on empeg serial usb (Gary Brubaker)
o Update natsemi driver (Jeff Garzik)
o Set last_rx on ppp_generic (Jeff Garzik)
o Fix modular tga (Matt Wilson)
o set dev->last_rx at right place in plip (Jeff Garzik)
o drop SIOCATEOR, fix endian bugs in DECnet (Steve Whitehouse)
o Add ISAPnP support to smc-ultra (Alexander Sotirov)
o Report errors on scsi_unregister_module (Oliver Neukum)
o Clean up starfire driver/fix mem ordering (Jeff Garzik, Jes)
o Fix tulip memleak, enable pci before using (Jeff Garzik)
o Yellowfin bss not data segment bits (Jeff Garzik)
o Quieten DMI reporting (me)
o Fix alpha wait4 error (Martin Schinschak)
o Fix md warning (Peter Samuelson)
o Fix via AGP support (Jeff Hartmann)
o Fusion driver updates (Steve Ralston)
o Fix bogus net core warnings on irda (me)
| needs more pondering before a final solutiomn
o Error negative size sysctl (me)
o Fix af_unix crash on big buffers (me)
| partly based on code from Andrew Morton
2.4.0-ac11
o Raid5 corruption fix (Neil Brown)
o Add Etrax 'cris' architecture support (Axis)
o APIC crash fixes (Ingo Molnar)
o Jochen Hein moved (Jochen Hein)
o Fix mm/slab.c doc (Matthew Wilcox)
o Major NTFS updates (Anton Altaparmakov)
o Make ewrk3 driver work (Nathan Hand)
o Fix vfb driver line length reporting (Geert Uytterhoeven)
o Allow xirc2 config on kernel command line (David Luyer)
o via audio mmap support, ioctl fixes (Rui Sousa)
o Fix ncpfs limits (Petr Vandrovec)
o Fix bios reading in i91xx scsi driver (Trevor Hemsley)
o S/390 updates (Holger Smolinski)
o Add pci dma mapping to epic100 (Francois Romieu)
o Resync with Linus 2.4.1pre9
- Fix DRM bugs in pre9 (Linus Torvalds)
- Fix HPFS tests in pre9 (me)
o Remove dead dsp56k/qpmouse inits (Hans Grobler)
o uart401 module locking fixes (Chris Rankin)
o Fix cs46xx build error in non module (Hans Grobler)
o Update hdparm url (Andre Dahlqvist)
o ibmmca updates (Michael Lang)
o Fix smctr build problems (Hans Grobler)
o Typo fixes (Ulrich Kunitz)
o Fix depca to new style module stuff (Hans Grobler)
| more love and attention still needed
o New kmalloc checks in buz.c (Hans Grobler)
o mct_u232 had wrong device id data (Adam J Richter)
o aty128fb error path fixes (Hans Grobler)
o Add radeon config help (Andre Dahlqvist)
o Fix acpi header (Adam J Richter)
o Add missing externs to bttv header (Hans Grobler)
o Add missing externs to bridge (Hans Grobler)
o Fix include/linux/rtc.h typo (John Fremlin)
o Remove unreachable code from atm proc (Hans Grobler)
o wanrouter proc fixes (Arnaldo Carvalho
de Melo)
o Fix naming of GPL all over the code (Andre Dahlqvist)
o sis900 new module locking (Jeff Garzik)
o Remove unneeded private byte count from sb1000 (Jeff Garzik)
o NCPfs didnt set s->maxbytes (Petr Vandrovec)
o DEC lance cleanup (Jeff Garzik)
o Tulip update (Jeff Garzik)
o Multiple drivers last_rx and skb deref fixes (Jeff Garzik)
o Make bonding new style (Jeff Garzik)
o Remove dead definitions from 8390 code (Jeff Garzik)
o 8139too updates (Jeff Garzik)
o 82596 driver updates (Jeff Garzik)
o Set last_rx on acenic (Jeff Garzik)
o Update roadrunner to new module locking (Jeff Garzik)
2.4.0-ac10
o Merge Linus 2.4.1-8
o Add s->s_maxbytes to reiserfs (me)
o Remove EHASHCOLLISION and make reiserfs thus (me)
compatible with existing glibc/apps
o Clean up oaknet driver (Hans Grobler)
o PCnet32 && / & bug fix (Anton Blanchard)
o c101 driver cleanups (Hans Grobler)
o cs4281 leak fixes (Hans Grobler)
o Update ppc entry code (Cort Dougan)
o smctr cleanup (Hans Grobler)
o unregister hdlc fixes (Francois Romieu)
o PPP async fixes (Paul Mackerras)
o IEEE1394 fixes (Andreas Bombe)
o Fix ac97 mixer crash (Darko Koruga)
o Fix 8xx ethernet driver init fail path (Hans Grobler)
o Fix affinity procfs crash on non SMP (Ingo Molnar)
o Fix raidhotremove bug (Ingo Molnar)
o NMI watchdog for K7 (Petr Vandrovec)
o Fix 386 boot on 2.4 kernels (Robert Kaiser)
o Fix resource leak in ctc on error (Arnaldo Carvalho
de Melo)
o Fix kmalloc fail handling in iucv (Arnaldo Carvalho
de Melo)
o Fix dasd kmalloc fail handling (Arnaldo Carvalho
de Melo)
o Fix tape34xx kmalloc fail handling (Arnaldo Carvalho
de Melo)
o Fix video1394 kmalloc and resource stuff (Hans Grobler)
o Cleanup ioc3 for new style network stuff (Hans Grobler)
o Cleanup acorn ethernet for new style (Hans Grobler)
o Cleanup qpmouse to new style (Hans Grobler)
o Fix last_rx/rx_bytes updates in net drivers (Jeff Garzik)
o Further ibm tape fixes (Carsten Otte)
o Do a reset on OHCI errors (David Brownell)
o Fix wavelan_cs kmalloc bugs (Arnaldo Carvalho
de Melo)
o Partition handling fixes (Andrzej Krzysztofowicz)
o tgafb modular cleanups (Ardrzej Krzysztofowicz)
o acsi driver cleanup (Arnaldo Carvalho
de Melo)
o Fix scsi disk name reporting (Doug Gilbert)
o ipfilter mss clamping (Marc Boucher)
o cciss driver kmalloc/cleanups (Arnaldo Carvalho
de Melo)
o Fix bugs in alternate uhci drivers (Johannes Erdfelt)
o Remove dead dsp56k init (Hans Grobler)
o USB storage updates (Johannes Erdfelt)
o Fix swapfs stuff, mem= > 4gig (Christoph Rohland)
2.4.0-ac9
o Remove duplicated 8139 fixes (Jeff Garzik)
o Drop out PS/2 mouse changes (me)
o Fix raid5 bug (Neil Brown)
o Fix mmio reservation leak in starfire (Ion Badulescu)
o Update gmac driver to new style (Hans Grobler)
o Fix misuse of dev_kfree_skb on cycx_x25 (Arnaldo Carvalho
de Melo)
o IPDDP cleanup/fixes (Hans Grobler)
o Remove = 0 inits from epic100 (Arnaldo Carvalho
de Melo)
o Fix resource failure leaks on depca (Arnaldo Carvalho
de Melo)
o Document ultrix partition option (Steven Cole)
o Fixed unused config option on cadet radio (Russell Kroll)
o Lose static = 0 inits on bmac (Arnaldo Carvalho
de Melo)
o Fix eql driver to use save/restore flags (Arnaldo Carvalho
de Melo)
o Document sysctl interfaces (John Levon)
o Clean up 6pack and reduce default footprint (Hans Grobler)
o Fix the handle alignment issues in NFS (Trond Myklebust)
o Chkconfig fixes (Niels Jensen)
o fusion driver updates (Steve Ralston)
o Clean up com20020-pci driver leaks (Hans Grobler)
o tmpfs/shmfs (Christoph Rohland)
2.4.0-ac8
o Fix PS/2 mouse ack/echo handling behaviour (Julian Bradfield)
| Let me know if you see 'odd' ps/2 stuff (Chris Hanson)
| in 2.4.0ac8 not in ac7
o Merge Linus 1pre3. Drop out some of my vm
diffs in favour of his
o PC110 pad move to new driver style (Hans Grobler)
o Clean up/fix leaks in ncr885e (Hans Grobler)
o Move dsp56k to new style module stuff (Hans Grobler)
o check->request_region, resource leak fixes (Hans Grobler)
for qlogicisp
o Fix iounmap leak in iphase (Hans Grobler)
o Fix iounmap leaks in ymf_pci (Hans Grobler)
o Fix s390mach.c for non SMP (Ulrich Weigand)
o Export queued_sectors (Jens Axboe)
o Fix raid5 build after Linus merge (Andrea Arcangeli)
o Documentation and chkconfig update (Niels Jensen)
o Fix iounmap leaks in oaknet (Arnaldo Carvalho de Melo)
o Clean up mac89x0 (Arnaldo Carvalho de Melo)
o Fix leaks on error in myri_sbus (Arnaldo Carvalho de Melo)
o Convert macsonic.c to new style (Arnaldo Carvalho de Melo)
o RCPCI further fixes (Rasmus Andersen)
2.4.0-ac7
o Export a KMALLOC_MAXSIZE for drivers to check (Hans Grobler)
| this is needed to verify things like firmware
| sizes passed by users
o Fix highmem compile issues (Ingo Molnar)
o Fix kmalloc check missing in hades-pci (Hans Grobler)
o Fix kmalloc fail crash in sdla_ppp (Hans Grobler)
o cfi locking fixes (Hans Grobler)
o Fix missing spin_unlock_irq in hd6457x.c (Hans Grobler)
o Fix lmc_main missing skb_unlock on error case (Hans Grobler)
o Handle out of memory on lanstreamer (Hans Grobler)
o Bring cs46xx.c into working state for non (Hans Grobler)
module. Fix locking
o Fix filesystem locking documentation (Al Viro)
o Fusion driver updates (Steve Ralston)
o Correct netfilter url (Rusty Russell)
o rcpci45 fix the pci_table name (again) (Hans Grobler)
o Fix scsi option ordering bug noted by (Michael Zieger)
o Config.h include updates (Niels Jensen)
o LFS handling cleanup, move some checks to (Al Viro)
vmtruncate
o Fix missing s->maxbytes setup for procfs (me)
o Replace epic100 patches with alternatives (Jeff Garzik)
o eepro fixes for older cards (Aristeu Sergio
Rozanski Filho)
o Buz error handling fix (Hans Grobler)
o DGRS driver cleanups/kmalloc checks (Arnaldo Carvalho
de Melo)
o Fix ioremap leak in zr36120 (Hans Grobler)
o FIx iounmap leaks in Stradis driver (Hans Grobler)
o Further mtd fixes (David Woodhouse)
o Update yellowfin driver (Jeff Garzik, from
Don Beckers drivers)
o Fix iounmap bugs in vga16 (Hans Grobler)
o TCP odd error fix (Dave Miller)
o ll_rw_blk enhancements (Jens Axboe)
o DMFE driver cleanup (Pavel Rabel)
o iucv fix for S/390 build when non SMP (Ulrich Weigand)
o Merge linus -pre2
o Fix ixj kmalloc checks (Ingo Molnar)
o Fix null pointer check in ibm partition code (Ingo Molnar)
o Fix kmalloc check in pc_keyb (Ingo Molnar)
o Fix kmalloc check in atari_pamsnet (Ingo Molnar)
o Fix kmalloc check in 3c515 (Ingo Molnar)
o Tidy up defxx/fix module locks etc (Jeff Garzik)
o Fix kmalloc check in atari_bionet (Ingo Molnar)
o Fix kmalloc check in olympic driver (Ingo Molnar)
o Fix kmalloc checks in avmb1 driver (Ingo Molnar)
o Tokenring needs to be an object file as its (Jeff Garzik)
using initcalls
2.4.0-ac6
o Sunrpc locking fix ()
o Made agpgart smarter about i815 (Charles McLachlan)
o Speed up truncate for shmem and clean up (Christoph Rohland)
o Fix kmalloc test in udf (Ingo Molnar)
o Fix ramfs kmalloc testing (Ingo Molnar)
o Fix irq and sense handling bugs in S/390 (Holger Smolinksi)
o Fix string.h for userspace accidental include (me)
| noted by Ulrich Weigand
o Red Hat office move (David Woodhouse)
o Fix missing highmem includes (Jens Axboe)
o Fix nfs_flushd deadlock (Andrew Morton)
o Honour owner in mpu401 (Chris Rankin)
o Fix raid5 kmalloc check (Ingo Molnar)
o Export mmu_cr4_features (Adam J Richter)
o Update ide floppy maintainer (Paul Bristow)
o Fix IP_ADD_MEMBERSHIP case (Stefan Jonsson)
o Wavelan resource leak fixes (Hans Grobler)
o Fix spinlock error introduced from 2.4.1pre (Benjamin Redelings)
o Fix u32 classifier possible hang (Dave Miller)
o Further warning fixes (Rich Baum)
o RCPCI driver further cleanups (Rasmus Andersen)
o Remove unneeded test from rlimit code (Hans Grobler)
o Generate header file tags as well as code (Hans Grobler)
o Fix ppp_generic label problem (William Lee Irwin)
o Fix errors failing to restore IRQ's on smctr (Hans Grobler)
o Fix bulkmem kmalloc check error (Ingo Molnar)
o Fix pci kmalloc fail handling error (Ingo Molnar)
o Fix dac960 kmalloc check (Ingo Molnar)
| new driver rev from LNZ due anyway however
o Fix pcmcia cs kmalloc check error (Thiago Rondon)
o Fix pcmcia ds kmalloc check error (Thiago Rondon)
o Bootmem.c uses phys_to_virt but misses include (Bjorn Wesen)
o Fix sknet kmalloc check (Ingo Molnar)
o Fix lmc kmalloc check (Ingo Molnar)
o Atarilance kmalloc check (Ingo Molnar)
o Make some symbols static (Dan Aloni)
o Fix sgiseeq kmalloc fail handling (Ingo Molnar)
o ISDN kmalloc null check fix (Ingo Molnar)
o ATM kmalloc fix (Ingo Molnar)
o Apa1480 kmalloc null check fix (Ingo Molnar)
o Sunlance kmalloc check fix (Ingo Molnar)
o Baget lance kmalloc check fix (Ingo Molnar)
o Update Jes Sorensen's email addr (me)
o Fix athlon crash on boot with local apic/nmi (Ingo Molnar)
o Further ds fix (Dan Aloni)
(Can the pcmcia folk verify that in fact you
could just move it)
o Fix iucv kmalloc bogon (Ingo Molnar)
o Fix sun3 video kmalloc check (Thiago Rondon)
o Further raid5 fixes (Ingo Molnar)
o Netfilter updates (Rusty Russell)
| (so come to http://www.linux.conf.au and say thanks)
o Update audio locking fixes (Chris Rankin)
o Remove ymf_sb driver now ymfpci handles all (Pete Zaitcev)
2.4.0-ac5
o Fix qnx build error (Frank Davis)
o Further generic_file_write fix (me)
| no signal on short write
| write data on write overlapping max fs size
o 3c515 dereferenced freeed skbuffs (Hans Grobler)
o opl3sa2 driver update (Scott Murray)
o Uninline strstr to fix gcc compile problems (me)
| as in 2.2
o Fix dmfe oops if no card found (Andrew Morton)
o Fix df reporting on ntfs (Anton Altaparmakov)
| alternate fix to one by Willem Dekker
o Fix error path on macii_init that left irqs off (Hans Grobler)
o Fix memory/resource leaks in tlan driver (Hans Grobler)
o Fix vmalloc end on highmem (Ingo Molnar)
o Allow dac960 root device specifiers (Leonard Zubkoff)
o Fix missing NULL kmalloc check in fore200e (Hans Grobler)
o AF_UNIX cleanup continued (Hans Grobler)
o Chkconfig fixes (Niels Jensen)
o Fix kmalloc check missing on ppc proc/rtas (Hans Grobler)
o Add recovery for get_block failures (Al Viro)
o Fix partially mapped page handling in gfw (Al Viro)
o Use s_maxbytes in lseek (Al Viro)
o Correct various ext2 items (Al Viro)
o Fix memory leak in 3c527 driver (Hans Grobler)
o Clean up ipc formatting etc except shm (Ingo Molnar)
o Add mising BSDCOMP documentation (Kai Germaschewski)
o Fix unchecked allocation in isdn_ppp.c (Hans Grobler)
o Revised ad1848 patch (Chris Rankin)
o Fix missing kmalloc check on hdlc (Krzysztof Halasa)
o Make mnt name behaviour predictable on oom (Ingo Molnar)
o Configure doc fixes (Jeremy Dolan)
o Fix amifb endif typo (Rich Baum)
o Fix bug where mtd driver left irqs off on error (Hans Grobler)
o Adjusted ext2 max size rule (Andreas Dilger)
o Fix non SMP build of S/390 tree (Bernhard Rosenkraenzer)
o Uniprocessor APIC support/NMI wdog etc (Mikael Pettersson,
Maciej W. Rozycki
Ingo Molnar)
o IXJ driver cleanups/fixes/updates (David Huggins-Daines)
o Fix endian and other minor partition bugs (Andrzej M. Krzysztofowicz)
o Fix nasty irda bug which could leave ints off (Hans Grobler)
o FIx bmac case where ints could be left off (Arnaldo Carvalho
de Melo)
o Remove surplus break from de620 (Hans Grobler)
o Fix iph5526 dereference of free skb (Hans Grobler)
o Remove invalid netfilter url from docs (David Rees)
o Fix madvise crash (Andrew Morton)
o Fix memleak in eepro driver error path (Hans Grobler)
o Ultrastor driver used wrong type for save_flags (Thiago Rondon)
o Fix spin_unlock missing in s390 error path (Hans Grobler)
o MTD update (David Woodhouse)
o Tidy softdog driver (Hans Grobler)
o Fix sunos syscall memory leak (Hans Grobler)
o Fix surplus remove_flags in 53c7xx,8xx driver (Arnaldo Carvalho
de Melo)
o Tidy isicom, fix missing restore_flags (Arnaldo Carvalho
de Melo)
o Fix missing restore_flags in sscape (Arnaldo Carvalho
de Melo)
o Make ixj use dynamic board structures (David Huggins-Daines)
o Fix missing spin_unlock in i2o block (Hans Grobler)
o Fix ad1848 missing restore_flags (Arnaldo Carvalho
de Melo)
o Fix missing spin_unlock in ymfpci (Hans Grobler)
o Adaptec 1542 SCSI command line options (Dmitry Potapov)
o Fix missing ksym (Eyal Lebedinsky)
o Fix megaraid driver (Anwar Payyoorayil)
o Sparc xor fix (Anton Blanchard)
o Fix error returns on truncate/open (Al Viro)
o Fix missing restore_flags in n_r3964 (Arnaldo Carvalho
de Melo)
o mxser driver capable/return fixes (Arnaldo Carvalho
de Melo)
o Fix missing __restore_flags on IDE (Arnaldo Carvalho
de Melo)
o Fix missing spin_unlock_irqrestore in EMU10K (Hans Grobler)
o Resync with Linus tree
2.4.0-ac4
o Fix dereference of freed skbuff in iphase (Hans Grobler)
o Fix dereference of freed skbuff in isdn_ppp (Hans Grobler)
o Fix dereference of freed skbuff in comx (Hans Grobler)
o Fix dereference of freed skbuff in atarilance (Arnaldo Carvalho
de Melo)
o Fix missing NULL check of dev_alloc_skb in (Hans Grobler)
hdlc layer
o Fix vesafb typo (Dag Wieers)
o Z85230 driver cleanup (Hans Grobler)
o Remove spare restore_flags in de600 (Hans Grobler)
o Catch failed kmallocs in ppc ethernet (Hans Grobler)
o AF_PACKET socket cleanup (Hans Grobler)
o ATAPI IDE format facility (Sam Varshavchik)
| Don't rely on the abi or anything yet, Gadi
| the maintainer has yet to comment on it
o Fix smc9194 crash on out of memory (Hans Grobler)
o Documentation fixes (Dag Wieers)
o Fusion driver updates (Steve Ralston,
Eddie Dost,
Arnaldo Carvalho
de Melo)
o Fix ramfs hangs (me)
o Fix assorted LFS problems and missing rlimit (me)
checks. In theory file size rules are now
right and properly enforced except for those
folks not using generic_file_write who need to
do their homework 8)
2.4.0-ac3
o Add support for the newer 3c905 cards (Andrew Morton)
o Drop unused field from scc.h (Hans Grobler)
o Remove dead sysctl stuff from econet (Hans Grobler)
o Fix documentation indexes (Paul Gortmaker)
o Fix post free reference of an skb in lance (Paul Gortmaker)
o Tidy appletalk code (Hans Grobler)
o Fix bootup vesafb hang (David Wragg)
o TCP 'reset_xmit_timer' fix (Dave Miller)
o Tidy up cursor positioning on menuconfig (Kirk Reiser)
o Add missing wait.h includes to some asm/semaphore
(Hans Grobler)
o AF_UNIX socket cleanup (Hans Grobler)
o Update sd locking fixes (Oliver Neukum)
o Add module locking to audio coprocessor calls (Chris Rankin)
o Minor further X.25 tidy (Hans Grobler)
o Fix scsi ioctl/scan crash on out of memory (Douglas Gilbert)
o Soundscape patches (Chris Rankin)
o M68K fixes for mem stats and stram (Geert Uytterhoeven)
o Set MSG_TRUNC correctly on atm sockets (Matti Aarnio)
o Add infrastructure for parport autoloading (Adam J Richter)
o Make lp driver use capable not old suser() (Tim Waugh)
o Fix thread/unload race on i2o block (me)
o Fix drivers that use asm/delay not linux/delay (Geert Uytterhoeven)
o Further warning fixes (Rich Baum)
o Netfilter config/Makefile fixes (Dave Miller)
o Merge updated cs4281 driver and tidy it (Tom Woller)
| some cleanups by me, possibly broken it ;)
o Fix bagetlance reference of freed buffer (Arnaldo Carvalho de Melo)
o ISDN small fixes (Andrea Baldoni,
Daniel Stodden)
o ESS Maestro 3 driver (Zach Brown)
2.4.0-ac2
o Clean up strip driver (Hans Grobler)
o Fix fore atm makefile (Jan Rekorajski)
o Fix m68k lance mismerge (Geert Uytterhoeven)
o Fix tty documentation typos (Hans Grobler)
o Fix ohci1394 build (Arjan van de Ven)
o Remove dead lapbether inits (Hans Grobler)
o Workaround the acpi recursive variable name (Bill Wendling)
Makefile problem
o Further minor S/390 merge (Ulrich Weigand)
o Fix DRM build problem on ATI Rage 120/no AGP (Gareth Hughes)
o Fix mac address setting in 8139too (Ben Greear)
o AGP oops fix/ALi cleanup (Bill Crawford)
o Further DECnet cleanups (Hans Grobler)
o S/390 last fixes (Ulrich Weigand)
o Fix missing arlan symbol (Hans Grobler)
o Do basic IPX/SPX cleanups (Hans Grobler)
2.4.0-ac1
o Resync with Linus
o Fix serial compile bug (Bill Notthingham)
o Clean up lapbether (Hans Grobler)
o Fix endian handling in ne.c (Geert Uytterhoeven)
o Fix root umount handling (Chris Mason &
Al Viro)
o Bring wan drivers up to scratch for 2.4 (Krzysztof Halasa)
o SD module locking fix (Oliver Neukum)
o Merge S/390 32/64bit ports (IBM)
| some rough edges to tidy up yet - guys can
| you change the DMA ifdefs to match 2.2 style..
2.4.0prerelease-ac6
o Cleanup econet (Hans Grobler)
o Further amateur radio cleanups (Hans Grobler)
o Fix irda/SMP deadlocks (Marc Zyngier)
o Further YAM fixes (Hans Grobler)
o Fix rio500 locking bug (Greg Kroah-Hartmann)
o Fix isdn net leak on error (Arnaldo Carvalho de Melo)
o Fix proc_get_inode export (for comx) (Hans Grobler)
o Fix locking error on get_swap_page (Marcelo Tosatti)
o Fix further warnings, and other stuff new gcc (Arjan van de Ven)
shows up
o Add isapnp module device tables to drivers (Bill Nottingham)
[Added to ns558, serial, ide-pnp, cadet,
3c509,3c515, aironet4500,ne,sb1000, aha1542,
NCR5380, ad1816, awe_wave, sb, ixj]
o Resync with Linus prepatch
2.4.0prerelease-ac5
o Resync with Linus prepatch
o One liner microcode driver fix (Tigran Aivazian)
o Fix ACPI ksyms problems (Keith Owens)
o Correctly resync ide-cd fixes (Byron Stanoszek)
o Fix i2o block driver race (Arjan van de Ven)
o Acorn makefile/driver fixes (Russell King)
o Make cyberfb use pci_get_drvdata (Russell King)
o Kill redundant ARM timer irq code (Russell King)
o Remove some ARM hacks from fbmem.c (Russell King)
o Fix config bugs with fusion, indenting (Andrzej M. Krzysztofowicz)
o Handle bootmem order changes in arm (Russell King)
o SA1100 update (Russell King)
o Handle ALI AGP flushes (Ian Hastie)
o Merge some of the PPC changes (Cort Dougan)
2.4.0prerelease-ac4
o DecNET updates (Steve Whitehouse)
o Devices.txt typo fix (Roberto Nibali)
o Fix 15-23bit direct colour in logos (Geert Uytterhoeven)
o Correct framebuffer device.txt (Geert Uytterhoeven)
o Small mkiss fixes (Hans Grobler)
o Fix write off end of disk bug (Jari Ruusu)
2.4.0prerelease-ac3
o Fix cs46xx driver crash (David Huggins-Daines)
2.4.0prerelease-ac2
o Fix further CVS gcc compile warnings (Rich Baum)
2.4.0prerelease-ac1
o Merge with Linux prerelease
2.4.0test13pre7-ac1
o Merge Linus pre7
o Fix eepro100 on machines with unsigned char (Russell King)
o Give the FIQ on the ARM its own handlers (Russell King)
o Update ARM mm code (Russell King)
o Fix ARM optimisations (Russell King)
o Update arm initd patch (Russell King)
o Improve ARM I/O operations (Russell King)
o ARM boot code updates (Russell King)
o ARM scsi driver updates (Russell King)
o Update ARM makefiles to new style (Russell King)
o Add missing sections to arm link script for glue(Russell King)
o Update ARM io includes (Russell King)
o Clean up frame pointer printing on ARM traps (Russell King)
o Update ARM machine definitions (Russell King)
o Move the ARM task unmapped base definition (Russell King)
o Remove ARM specific hacks from char/mem.c (Russell King)
o Fix BIOS32 code for ARM (Russell King)
o Back out bogus SMP halt change (Andi Kleen)
o Update logo palette handling (Geert Uytterhoeven)
o Drop out the compiler selector (2.96/7 seem to work)
2.4.0test13pre6-ac1
o Merge Linus pre6
2.4.0test13pre5-ac1
o Merge Linus pre5
2.4.0test13pre4-ac2
o Further quota build fix (Jarno Paananen)
o Fix various combinations that don't build (Arjan van de Ven)
o Further Fusion driver updates (Steve Ralston)
o Alpha makefile fixes (Dave Gilbert)
2.4.0test13pre4-ac1
o Merge Linus pre4
o Fix network register/hotplug/publish problems (Andrew Morton)
o Hopefully fix quotaless compile (me)
o Help for irda options question (Steven Cole)
2.4.0test13pre3-ac4
o Fix frame size on toshoboe (Christian Gennerat)
o Quota fixes/updates (Jan Kara)
o Fix keyspan usb config (Hugh Blemings)
o Fix module handling in usb serial (Greg Kroah-Hartmann)
o Fix sparc64 build of fusion drivers (Eddie Dost)
o Clean up config.h includes (Niels Jensen)
2.4.0test13pre3-ac3
o Fix the patch file. Some stuff got corrupted.
2.4.0test13pre3-ac2 adds
o Resync with the powerpc folks (Cort Dougan)
o Parport experimental label fix (Tim Waugh)
o Make uhci return the same error code as the (David Brownell)
other USB hub controllers
o Merge Fusion drivers (Steve Ralston)
2.4.0test13pre3-ac1 adds
o Fix leak in link() syscall (Christopher Yeoh)
o Fix ramfs deadlock (Al Viro)
o Fix udf deadlock (Al Viro)
o Improve parport docs (Tim Waugh)
o Document some of the macros (Tim Waugh)
o Fix ppa timing issues (Tim Waugh)
o Mark the parport fifo code as experimental (Tim Waugh)
o Resynch ppa changelog (Tim Waugh)
| Tim please double check as I got offsets
o Add documentation to the PCI api (Jani Monoses)
o Fix inode.c documentation (Jani Monoses)
o Fix ext2 modular build (Jeff Raubitschek)
o Fix bug in scripts/Configure.in matching (Matthew Wilcox)
o Update SiS video drivers (Can-Ru Yeou)
o Yamaha audio doc fix (Pavel Roskin)
o Fix timeout problms with rocktports at 249 days
2.4.0test12-ac1 adds
o ARM bootup/initd fixes (Russell King)
o Fix ymf_sb setup bug (Pavel Roskin)
o Correctly print names of md10+ (me)
[Based on code from Roberto Ragusa]
o Fix sound crashes in various drivers (Tjeerd Mulder)
o Update epic100 to new pci api (Francois Romieu)
o Fix IOC/SIOC ioctl problems in ac97 code (Dick Streefland)
To merge
o Fix Ruffian Alpha boot (Ivan Kokshaysky)
o Bridge handling patches needed for Alpha (Ivan Kokshaysky /
Richard Henderson)
o Epic100 update
o Support mixed pnp and legacy sb cards
o Hopefully fix the bugs in the FAT and HPFS file systems that
caused fs corruption
o Fix cramfs vanishing data bug
o Power management locking fixes
o filemap posix compliance fix
o Fix pte handling race
o Remove unneeded inits to 0 in ide code (Bartlomiej Zolnierkiewicz)
o IDE documentation fixes (Bartlomiej Zolnierkiewicz)
Submitted to Linus
o Add the powermac extras to the input and (Franz Sirl)
keyboard drivers
scripts in 2.4test
o Fix kd_mksound declaration (Geert Uytterhoeven)
o SMC token ring driver update (Jay Schulist)
o Update USB documentation (Greg Kroah-Hartmann)
o Cleanup ramdisk namespace (Jeff Garzik)
o Ramdisk missing blkdev_put
Merged by Linus from -ac or direct
o Add clocking option to maestro (broken laptop (me)
stuff again)
o Put back the module locking in soundcore (David Schleef)
that someone disabled
o Abyss driver cleanup (Jeff Garzik)
o Fix most of the tq changes (Mohammad A. Haque)
o DOC1000 driver fixes (David Woodhouse)
o Switch tvaudio and msp3400 to use up_and_exit (David Woodhouse)
o usb-uhci was using constants not flags for (Jeff Garzik)
pci interface
o Small fix for kdoc (Tim Waugh)
o Fix nubus build (Geert Uytterhoeven)
o atari/sun3lance update (Geert Uytterhoeven)
o Amiga gayle pcmcia fixups (Geert Uytterhoeven)
o Fixes for amiga scsi drivers (Geert Uytterhoeven)
o Simplify amiga irq handling code (Geert Uytterhoeven)
o Amiga sound/fb driver update (Geert Uytterhoeven)
o Amiga/Mac/Atari keyboard driver changes (Geert Uytterhoeven)
o Integrate atari stram with bootmem (Geert Uytterhoeven)
o Restore atafb_fix that someone deleted (Geert Uytterhoeven)
o m68k include updates for 64bit structs (Geert Uytterhoeven)
o Add driver for MVME147 onboard scsi (Geert Uytterhoeven)
o Enable Q40 ide interface (Geert Uytterhoeven)
o Replace init with initdata in places on m68k (Geert Uytterhoeven)
o MMU code changes for m68k (Geert Uytterhoeven)
o dma_addr_t and other minor updates for m68k (Geert Uytterhoeven)
o m68k ptrace update (Geert Uytterhoeven)
o Fix pmc551 when used without bugfix enabled (David Woodhouse)
o Fix endianness on ftl layer (David Woodhouse)
o Fix atm build (Markus Kossmann)
o Update 8139too driver (Jeff Garzik)
o Fix readdir returns on procfs (Matt Kraai)
o Make SET_MODULE_OWNER macro safer (Jeff Garzik)
o Hisax needed __init (Jeff Garzik)
o APM updates, fix the Dell 5000e check for APM=m (Stephen Rothwell)
o Fix module initialization oops (Keith Owens)
o Clean up Abyss driver (Jeff Garzik)
o Fix raid linking order (Neil Brown)
o Cleanup console_verbose() duplication
o Radio driver cleanups
o BTTV radio config option
o Fix qcam VIDIOCGWIN bugs
o 8390 seperate tx timeout path
o Tulip crash fix on weird eeproms
o ISAPnP hang on boot port fix
o Maestro pci_enable fix
o Fix function prototype in wacom drivr
o Fix SCSI / PCI dependancies (Jeff Garzik)
o m68k config fixes (Geert Uytterhoeven)
o Fix dquot overflow/recovery (Jan Kara)
o Make uid16 macros safer (Andreas Schwab)
o Fix missing Config doc and sound doc error (Thierry Vignaud)
o APM update (Stephen Rothwell)
o Fix SMP build on x86 (Steven Cole)
o Maestro ioctl locking fix (Zach Brown)
o Make console_* static inline not extern (Jeff Garzik)
o Work arounds for broken Dell laptop APM (me)
o Fix aha1542 memory scribbles (Phil Stracchino)
o Fix ide scsi printk (Geert Uytterhoeven)
o Update EATA driver and Ultrastor driver (Dario Ballabio)
o Clean up printk formatting in a few drivers (me)
o Documentation for CONFIG_TOSHIBA
o Updated version of Rusty's kernel-hacking doc
o Updated SubmittingDrivers
o Added SubmittingPatches
o Updated procfs docs
o Updated initrd docs
o Tidy network drivers module locking (Jeff Garzik)
(Some in, a few to go)
o Alpha PCI fixes (update resource not __init, (Ivan Kokshaysky)
off by one on check)
o Fix warning in rclan driver (Rasmus Andersen)
o Clean up rcpci driver (new style pci etc) (Jeff Garzik)
o Fix generic bitops bugs
o Fix pcnet32 printk problems (Vojtech Pavlik)
o Network driver check/request region fixes
o MDAcon cleanup (Pavel Rabel)
o Tidy up mad16 driver (Pavel Rabel)
o ACPI updates (Andrew Grover)
o Fix FPU emulation compile (Adam Richter)
o M68K/PPC makefile fixes (Geert Uytterhoeven)
o Work around a funny in the Solaris NFS client (Neil Brown)
o Fix building of network modules (Peter Samuelson)
o Fix media makefiles (me)
o FPU emulator source set for m68k (Geert Uytterhoeven)
o Fix m68k build with rmw disabled (Geert Uytterhoeven)
o Fix sun3 scsi, mmu and includes (Geert Uytterhoeven)
o M68k setup update (Geert Uytterhoeven)
o Revert accidental amifb change (Geert Uytterhoeven)
o Remove obsolete bits for q40 (Geert Uytterhoeven)
o Tidy m68k includes (Geert Uytterhoeven)
o I2C bus driver updates (Frodo Looijaard)
o Fix reference counting in ATM (Patrick van de Lageweg)
o Update Changes to give correct modutils rev (Steven Cole)
o Fix NLS config.in bug for SMB
o Fix xconfig/menuconfig problems with config (Andrzej Krzysztofowicz)
o Add firestream ATM driver (Patrick van de Lageweg)
o Rename block_til_ready in generic_serial (Patrick van de Lageweg)
o Fix i810 tco locking (me)
o Tidy riscom8 and sx namespace (Jeff Garzik)
o Fix pcmcia ordering on socket remove (David Woodhouse)
o Merge aha152x delay fixes
o Fix warning in sim710 driver (Pavel Rabel)
o Improve the ALSxxx sound driver documentation (Jonathan Woithe)
o Tidy the tachyon 5526 driver (Rasmus Andersen)
o Clean old old compile time config stuff from (Pavel Rabel)
mad16 driver
o Push Davicom support into the main tulip driver (Tobias Ringstrom)
o Merge bttv 0.7.50 (Gerd Knorr)
o Clean it up to use pci_pci_quirks properly (me)
o OSST scsi driver for Onstream drives (Willem Riede)
o Merge typo/doc fixes from 2.2.18
o Further NetROM tidies (Hans Grobler)
o Further rose fixes (Hans Grobler)
o Documentation/script fixes (Tim Waugh)
o BPQ ethernet tidy (Hans Grobler)
o Updated AX.25 tidy (Hans Grobler)
o Update credits to add Hans Grobler (Hans Grobler)
o Handle TLB flush reruns caused by APIC rexmit (me)
o Fix Yam driver for Linux 2.4test (Hans Grobler)
o Fix AF_ROSE sockets for 2.4 (Hans Grobler)
o Fix AF_NETROM sockets for 2.4 (Hans Grobler)
o Tidy AF_AX25 sockets for 2.4 (Hans Grobler)
o Teach kernel-doc about const (Jani Monoses)
o First block of mkiss driver fixes (Hans Grobler)
o Update acenic patches (Jes Sorensen)
o Acenic update
o Watchdog header to use __u32 etc (Eric Brower)
o Correct md name printing (Luca Berra)
o Add Steven Cole to the credits file (Steven Cole)
o DC390 update (Kurt Garloff)
o X.25 ifdef cleanups (Henner Eisen)
o Update OSST driver to 0.9.4.3 (Kurt Garloff)
o Fix make xconfig failure on irda (Steven Cole)
o Fix crashes on unload of msr and cpuid drivers
o Remove crud from epca driver (me)
o Merge support for CPU's >2Ghz from 2.2.18
o Merge core loops_per_jiffy support
o Merge first batch of driver fixes from 2.2.18
o Make smp cpu halt synchronous (Andi Kleen)
o Fix eepro module warnings (Aristeu Filho)
o Fix most of the netfilter oops cases (David Miller)
o Fix appletalk config entry (William McGonigle)
o CCISS root= table (Charles White)
o Rusty's fixes/review of unsafe set_bit usage
(A few left to go)
o RCPCI45 PCI cleanup fixes (mark 2) (Rasmus Andersen)
o Fix 8139too signal handling and task scribble (Andrew Morton)
o Fix signal handling for usermode helper (Shuu Yamaguchi)
o Fix tty DoS bug (Andrew Morton)
o Wireless include update (Jean Tourrilhes)
o Resync mac ethernet drivers (Cort Dougan)
o Remove bogus asserts in 8139too driver (Jeff Garzik)
o Fix radio drivers (Russell Kroll)
o Fix rcpci build error (Hans Grobler)
o Fix incorrect preprocessor use in umsdos (Andreas Franck)
o DRM makefile fix (Keith Owens)
o IDE 2.4.0-prerelease*1231.patch (Andre Hedrick)
o Fixes for CVS gcc and semaphores (Andreas Franck)
o Better atm linking fix (Jan Rekorajski)
o Macintosh IDE updates (Geert Uytterhoeven)
o Update 68k ksyms (Geert Uytterhoeven)
o Fix m68k keyboard ioctls (Geert Uytterhoeven)
o Fix fbdev config.in allow PM2 modular (Geert Uytterhoeven)
o Update m68k ethernet drivers (Geert Uytterhoeven)
o 2.4 Y2K fixes for Amiga clock (Geert Uytterhoeven)
o Fix sun/mac scsi drivers (Geert Uytterhoeven)
o Fix fb init order (Geert Uytterhoeven)
o Fix m68k miscellaneous stuff (Geert Uytterhoeven)
o Update m68k lance driver (Geert Uytterhoeven)
o Fix m68k asm constraints (Geert Uytterhoeven)
o Fix m68k config (Geert Uytterhoeven)
o Amiga serial update/serial console support (Geert Uytterhoeven)
o Update m68k to use loops_per_jiffy (Geert Uytterhoeven)
o Add support functions needed by gcc (Geert Uytterhoeven)
o Fix amiga resource management (Geert Uytterhoeven)
o Fix raid buffer leak (Neil Brown)
o Additional knfsd locking (Neil Brown)
o Fix loops per jiffy oddments (Geert Uytterhoeven)
o Fixed lost video patch in -ac (Geert Uytterhoeven)
o Tidy up LAPB code (Hans Grobler)
o Tidy up X.25 code (Hans Grobler)
o General warning/minor bug fixes (Arjan van de Ven)
o Remove extra codec reset from i810 audio (Anwar Payyoorayil)
| should fix failed VRA on some boards
o Fix page allocator recursion (Rik van Riel)
o Fix CMOS locking for 2.4.x (Paul Gortmaker)
o FAT cache locking for SMP ('manmower')
o Skip older dm9100's from tulip driver (me)
o Further iee1394 build fixes (Andreas Bombe)
o Fix i810 divide by zero bug (Anwar)
o Remove dead pi and pt drivers (Hans Grobler)
o SCC driver update (Hans Grobler)
o Adjust csr0 on tulip for known iffy chipsets (me)
| Thanks to Don Becker and others for the chipset list
| and knowing what the problem was.
o 6pack cleanups (Hans Grobler)
o Documentation cleanups (Hans Grobler)
o Fix unchecked scsi_allocate_request in sg.c (me)
| Spotted by Doug Gilbert
o Fix compile bug in pcxx driver (me)
o Fix spelling of pedant (Tim Waugh)
o Fix mkiss build error (Hans Grobler)
o Patches to fix warnings from gcc 2.97 cvs (Marcel Schmidt)
Superceded by other fixes
o Features is back to flags for compatibility (me)
o MTRR updates (36bit etc)
o Dont crash on boot with a dual cpu board holding a non intel cpu
o CS46xx update
o NFS atomic fixes (Trond Myklebust)
o Fix O_SYNC for ext2fs (Stephen Tweedie)
[ I believe so anyway ]
o Disable PMC511 driver - its obviously broken (me)
o kbuild documentation improvements (Neil Brown)
o Fix ppa and imm hangs on io_request_lock (Tim Waugh)
o Fix pport reverse/forward logic error (Tim Waugh)
o ACPI updates (Andrew Grover)
o E820 handling fixup (Andrea Arcangeli)
o Fix missing memory barrier in bootp/dhcp code (Cort Dougan)
o Fix ACPI driver wakeup races (David Woodhouse)
o Fix drm makefiles (Peter Samuelson)
o Link correctly with ACPI on ACPI_INTERPRETER off
o Shared memory fixes (Christoph Rohland)
o Fix bug in VFAT short name handling (Nicolas Goutte)
o Clean up the i810 driver (Tjeerd Mulder)
o Clean up misleading indenting in partition code (JAmes Antill)
o Support kgcc autodetect
Other
---
Alan Cox <[email protected]>
Red Hat Kernel Hacker
& Linux 2.2 Maintainer Brainbench MVP for TCP/IP
http://www.linux.org.uk/diary http://www.brainbench.com
Doug,
I confirm that ac7 fixed all the aic7xxx problems on my machine.
Thanks,
Tigran
two mistakes:
a) [email protected], not veritas.com! (it was pine, not me -- default
domain etc :)
b) it was ac6 which fixed it, not ac7 (but I am running ac7)
On Thu, 8 Feb 2001, Tigran Aivazian wrote:
> Doug,
>
> I confirm that ac7 fixed all the aic7xxx problems on my machine.
>
> Thanks,
> Tigran
>
>
At 19:33 08/02/01, Tigran Aivazian wrote:
>On Thu, 8 Feb 2001, Tigran Aivazian wrote:
> > I confirm that ac7 fixed all the aic7xxx problems on my machine.
<AOL>me,too</AOL> (-:
AHA2940UW dual channel adapter (on board a SMP Tyan Thunder Pro 100 GX440
mobo).
-ac5 crashed on boot at SCSI init.
-ac6 untested
-ac7 all working fine again
Best regards,
Anton
--
Anton Altaparmakov <aia21 at cam.ac.uk> (replace at with @)
Linux NTFS Maintainer / WWW: http://sourceforge.net/projects/linux-ntfs/
ICQ: 8561279 / WWW: http://www-stu.christs.cam.ac.uk/~aia21/
Tigran Aivazian wrote:
>
> two mistakes:
>
> a) [email protected], not veritas.com! (it was pine, not me -- default
> domain etc :)
>
> b) it was ac6 which fixed it, not ac7 (but I am running ac7)
>
> On Thu, 8 Feb 2001, Tigran Aivazian wrote:
>
> > Doug,
> >
> > I confirm that ac7 fixed all the aic7xxx problems on my machine.
Thanks, I hoped it would ;-) It's amazing what happens when you have a bcopy
in assembly that's missing the source address initialization :-(
--
Doug Ledford <[email protected]> http://people.redhat.com/dledford
Please check my web site for aic7xxx updates/answers before
e-mailing me about problems
On Thu, 8 Feb 2001, Doug Ledford wrote:
> Thanks, I hoped it would ;-) It's amazing what happens when you have a bcopy
> in assembly that's missing the source address initialization :-(
Yes! The output from the description of my SCSI hds when the driver
initialised was highly amusing (containing extremely random garbage)...
(-;
Best regards,
Anton
--
Anton Altaparmakov <aia21 at cam.ac.uk> (replace at with @)
Linux NTFS maintainer / WWW: http://sourceforge.net/projects/linux-ntfs/
ICQ: 8561279 / WWW: http://www-stu.christs.cam.ac.uk/~aia21/
On Thu, 8 Feb 2001, Alan Cox wrote:
> ftp://ftp.kernel.org/pub/linux/kernel/people/alan/2.4/
>
> 2.4.1-ac7
> o Rebalance the 2.4.1 VM (Rik van Riel)
> | This should make things feel a lot faster especially
> | on small boxes .. feedback to Rik
I'd really like feedback from people when it comes to this
change. The change /should/ fix most paging performance bugs
because it makes kswapd do the right amount of work in order
to solve the free memory shortage every time it is run.
This, in turn, should make it far less likely that user processes
will *ever* need to call try_to_free_pages() themselves, unless
the system really goes into overload mode.
It would be good to know if this change really fixes the bug or
if it only helps for certain workloads and not for others. I'd
really like to close the following bug but need confirmation
that it works first ;)
http://distro.conectiva.com/bugzilla/show_bug.cgi?id=1178
regards,
Rik
--
Linux MM bugzilla: http://linux-mm.org/bugzilla.shtml
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com/
On Thu, Feb 08 2001, Rik van Riel wrote:
> On Thu, 8 Feb 2001, Alan Cox wrote:
>
> > ftp://ftp.kernel.org/pub/linux/kernel/people/alan/2.4/
> >
> > 2.4.1-ac7
> > o Rebalance the 2.4.1 VM (Rik van Riel)
> > | This should make things feel a lot faster especially
> > | on small boxes .. feedback to Rik
>
> I'd really like feedback from people when it comes to this
> change. The change /should/ fix most paging performance bugs
> because it makes kswapd do the right amount of work in order
> to solve the free memory shortage every time it is run.
Rik,
Just installed ac7 and after some 30 minutes of unpacking
kernel-sources and diffing patches, I left my computer unattended
for about 1 hour. When I came back the system was unusable (like it
was frozen), and /var/log/messages just displayed messages of the
type:
Feb 8 22:54:40 fry kernel: Out of Memory: Killed process 455 (xmms).
...
The OOM killer killed most of my apps, and finally X. I had to reboot
in order to get the system back. I've been running ac1-ac6 since they
came out with no problems, so I guess its the VM hack that is buggy.
This is on an AMD K7 1200Mhz, 512MB Ram, ATA100. Nothing big was
running at the time (xchat, xmms, mozilla, gnome, x, a few xterms).
I'll do some more testing tomorrow and provide any further information
you might need.
--
Torben Mathiasen <[email protected]>
Linux ThunderLAN maintainer
http://opensource.compaq.com
On Thu, 8 Feb 2001, Torben Mathiasen wrote:
> On Thu, Feb 08 2001, Rik van Riel wrote:
> > On Thu, 8 Feb 2001, Alan Cox wrote:
> >
> > > ftp://ftp.kernel.org/pub/linux/kernel/people/alan/2.4/
> > >
> > > 2.4.1-ac7
> > > o Rebalance the 2.4.1 VM (Rik van Riel)
> > > | This should make things feel a lot faster especially
> > > | on small boxes .. feedback to Rik
> Just installed ac7 and after some 30 minutes of unpacking
> kernel-sources and diffing patches, I left my computer unattended
> for about 1 hour. When I came back the system was unusable (like it
> was frozen), and /var/log/messages just displayed messages of the
> type:
>
> Feb 8 22:54:40 fry kernel: Out of Memory: Killed process 455 (xmms).
> ...
>
> The OOM killer killed most of my apps, and finally X. I had to reboot
> in order to get the system back. I've been running ac1-ac6 since they
> came out with no problems, so I guess its the VM hack that is buggy.
Highly unlikely since the VM rebalancing patch doesn't change
any of the actual swapout mechanisms.
All it does is change how often the particular algorithms get
called by kswapd and by user programs.
As for trigerring the OOM killer, this strongly suggest a
memory leak since there's a bug in the code which makes it
very hard to trigger the OOM killer under normal situations
(I'm working on a fix for that now).
regards,
Rik
--
Linux MM bugzilla: http://linux-mm.org/bugzilla.shtml
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com/
Hi Alan,
somebody screwed up in either the IDE or the PCI code: the IDE driver
doesn't compile when CONFIG_PCI is _disabled_.
Dominik
Error message:
...
make[3]: Entering directory `/export/source/i486/linux/drivers/ide'
gcc -D__KERNEL__ -I/export/source/i486/linux/include -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strict-aliasing -pipe -mpreferred-stack-boundary=2
-march=i486 -DEXPORT_SYMTAB -c ide.c
gcc -D__KERNEL__ -I/export/source/i486/linux/include -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strict-aliasing -pipe -mpreferred-stack-boundary=2
-march=i486 -DEXPORT_SYMTAB -c ide-features.c
gcc -D__KERNEL__ -I/export/source/i486/linux/include -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strict-aliasing -pipe -mpreferred-stack-boundary=2
-march=i486 -c -o ide-proc.o ide-proc.c
ld -m elf_i386 -r -o ide-mod.o ide.o ide-features.o ide-proc.o
ide-features.o: In function `pci_match_device':
ide-features.o(.text+0x0): multiple definition of `pci_match_device'
ide.o(.text+0x0): first defined here
ide-proc.o: In function `pci_match_device':
ide-proc.o(.text+0x0): multiple definition of `pci_match_device'
ide.o(.text+0x0): first defined here
make[3]: *** [ide-mod.o] Error 1
make[3]: Leaving directory `/export/source/i486/linux/drivers/ide'
make[2]: *** [first_rule] Error 2
make[2]: Leaving directory `/export/source/i486/linux/drivers/ide'
make[1]: *** [_subdir_ide] Error 2
make[1]: Leaving directory `/export/source/i486/linux/drivers'
make: *** [_dir_drivers] Error 2
Output of ver_linux:
-- Versions installed: (if some fields are empty or look
-- unusual then possibly you have very old versions)
Linux gateway 2.4.0-ac9 #1 Tue Jan 16 23:45:51 CET 2001 i486 unknown
Kernel modules 2.4.1
Gnu C 2.95.3
Gnu Make 3.79.1
Binutils 2.10.1.0.2
Linux C Library 2.2.1
Dynamic linker ldd (GNU libc) 2.2.1
Procps 2.0.7
Mount 2.10q
Net-tools 2.05
Console-tools 0.2.3
Sh-utils 2.0.11
Modules Loaded ipt_MASQUERADE iptable_nat ip_conntrack ip_tables sb sb_lib uart401 sound soundcore ne 8390 hisax isdn slhc
Relevant .config options:
CONFIG_EXPERIMENTAL=y
...
# CONFIG_PCI is not set
# CONFIG_EISA is not set
# CONFIG_MCA is not set
# CONFIG_HOTPLUG is not set
# CONFIG_PCMCIA is not set
...
#
# ATA/IDE/MFM/RLL support
#
CONFIG_IDE=y
#
# IDE, ATA and ATAPI Block devices
#
CONFIG_BLK_DEV_IDE=y
#
# Please see Documentation/ide.txt for help/info on IDE drives
#
# CONFIG_BLK_DEV_HD_IDE is not set
# CONFIG_BLK_DEV_HD is not set
CONFIG_BLK_DEV_IDEDISK=y
..
CONFIG_BLK_DEV_IDECD=y
...
On Thu, Feb 08, 2001 at 08:12:39PM -0200, Rik van Riel wrote:
> On Thu, 8 Feb 2001, Alan Cox wrote:
>
> > ftp://ftp.kernel.org/pub/linux/kernel/people/alan/2.4/
> >
> > 2.4.1-ac7
> > o Rebalance the 2.4.1 VM (Rik van Riel)
> > | This should make things feel a lot faster especially
> > | on small boxes .. feedback to Rik
>
> I'd really like feedback from people when it comes to this
> change. The change /should/ fix most paging performance bugs
> because it makes kswapd do the right amount of work in order
> to solve the free memory shortage every time it is run.
I just tested ac8.
If I run this test, the system gets really slow. It takes about
a second between the time I press a key, and the time it appears
on the screen. The load goes way up. Everything seems to block.
This is a box with 64 MB or RAM, and 32 MB of swap. There isn't
much running on the box while doing this, only dnetc.
It starts to get slow from the time the process starts is about
70 MB. Then you really hear the disk work. It ended up at about
75 MB, where it got killed by the OOM killer. (For once it killed
the right thing!)
I ran a vmstat 1, while doing this, and have attached the output.
It ran for serval minutes. The process itself took about 1
minutes of CPU time, and so did kswapd. It took atleast 5
minutes real time.
I once did just the same with 2.4.0, it took more like 30 minutes
then, and I ended up killing the process myself.
Kurt
On Fri, 9 Feb 2001, Kurt Roeckx wrote:
> I just tested ac8.
>
> If I run this test, the system gets really slow. It takes about
> a second between the time I press a key, and the time it appears
> on the screen. The load goes way up. Everything seems to block.
I'm sorry, but ... what test ?
And how do older kernels run the same thing ?
> It ran for serval minutes. The process itself took about 1
> minutes of CPU time, and so did kswapd. It took atleast 5
> minutes real time.
>
> I once did just the same with 2.4.0, it took more like 30
> minutes then, and I ended up killing the process myself.
So the kernel's behaviour has improved ?
regards,
Rik
--
Linux MM bugzilla: http://linux-mm.org/bugzilla.shtml
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com/
On Thu, 8 Feb 2001, Rik van Riel wrote:
> On Thu, 8 Feb 2001, Alan Cox wrote:
>
> > ftp://ftp.kernel.org/pub/linux/kernel/people/alan/2.4/
> >
> > 2.4.1-ac7
> > o Rebalance the 2.4.1 VM (Rik van Riel)
> > | This should make things feel a lot faster especially
> > | on small boxes .. feedback to Rik
>
> I'd really like feedback from people when it comes to this
> change. The change /should/ fix most paging performance bugs
> because it makes kswapd do the right amount of work in order
> to solve the free memory shortage every time it is run.
Hi Rik,
This change makes my box swap madly under load. It appears to be
keeping more cache around than is really needed, and therefore
having to resort to swap instead. The result is MUCH more I/O than
previous kernels while doing the same exact job.
My test load is make -jN bzImage. Previous kernels kept cache at
an average of ~20ish mb at a job level N at which level I had nearly
zero measurable throughput loss compared to single task compile.
>From that, I surmise that the cachable component of this job must
fit in that roughly 20ish mb of space. (for otherwise, I would be
suffering throughput loss). With this vm change, cache is nearly
three times as large as usual. Where 30 tasks will run with only
modest throughput loss in ac5, ac8 throughput tapers off rapidly
at half of that.
-Mike
On Sat, 10 Feb 2001, Mike Galbraith wrote:
> Hi Rik,
>
> This change makes my box swap madly under load. It appears to be
> keeping more cache around than is really needed, and therefore
> having to resort to swap instead. The result is MUCH more I/O than
> previous kernels while doing the same exact job.
>
> My test load is make -jN bzImage. Previous kernels kept cache at
> an average of ~20ish mb at a job level N at which level I had nearly
> zero measurable throughput loss compared to single task compile.
>
> >>From that, I surmise that the cachable component of this job must
> fit in that roughly 20ish mb of space. (for otherwise, I would be
> suffering throughput loss). With this vm change, cache is nearly
> three times as large as usual. Where 30 tasks will run with only
> modest throughput loss in ac5, ac8 throughput tapers off rapidly
> at half of that.
Swapped out pages were not being counted in the flushing limitation.
Could you try the following patch?
Thanks
--- linux.orig/mm/vmscan.c Sat Feb 10 08:26:17 2001
+++ linux/mm/vmscan.c Sat Feb 10 09:34:20 2001
@@ -515,6 +515,7 @@
writepage(page);
flushed_pages++;
+ max_launder--;
page_cache_release(page);
/* And re-start the thing.. */
On Sat, 10 Feb 2001, Marcelo Tosatti wrote:
> On Sat, 10 Feb 2001, Mike Galbraith wrote:
>
> > This change makes my box swap madly under load.
>
> Swapped out pages were not being counted in the flushing limitation.
>
> Could you try the following patch?
Marcelo's patch should do the trick wrt. to making page_launder()
well-behaved again. It should fix the problems some people have
seen with bursty swap behaviour.
> --- linux.orig/mm/vmscan.c Sat Feb 10 08:26:17 2001
> +++ linux/mm/vmscan.c Sat Feb 10 09:34:20 2001
> @@ -515,6 +515,7 @@
>
> writepage(page);
> flushed_pages++;
> + max_launder--;
> page_cache_release(page);
>
> /* And re-start the thing.. */
Rik
--
Linux MM bugzilla: http://linux-mm.org/bugzilla.shtml
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com/
On Sat, 10 Feb 2001, Rik van Riel wrote:
> On Sat, 10 Feb 2001, Marcelo Tosatti wrote:
> > On Sat, 10 Feb 2001, Mike Galbraith wrote:
> >
> > > This change makes my box swap madly under load.
> >
> > Swapped out pages were not being counted in the flushing limitation.
> >
> > Could you try the following patch?
>
> Marcelo's patch should do the trick wrt. to making page_launder()
> well-behaved again. It should fix the problems some people have
> seen with bursty swap behaviour.
It's still reluctant to shrink cache. I'm hitting I/O saturation
at 20 jobs vs 30 with ac5. (difference seems to be the delta in
space taken by cache.. ~same space shows as additional swap volume).
-Mike
On Sat, 10 Feb 2001, Mike Galbraith wrote:
> On Sat, 10 Feb 2001, Rik van Riel wrote:
> > On Sat, 10 Feb 2001, Marcelo Tosatti wrote:
> > > On Sat, 10 Feb 2001, Mike Galbraith wrote:
> > >
> > > > This change makes my box swap madly under load.
> > >
> > > Swapped out pages were not being counted in the flushing limitation.
> > >
> > > Could you try the following patch?
> >
> > Marcelo's patch should do the trick wrt. to making page_launder()
> > well-behaved again. It should fix the problems some people have
> > seen with bursty swap behaviour.
>
> It's still reluctant to shrink cache. I'm hitting I/O saturation
> at 20 jobs vs 30 with ac5. (difference seems to be the delta in
> space taken by cache.. ~same space shows as additional swap volume).
Indeed, to "fix" that we'll need to work at refill_inactive().
However, I am very much against tuning the VM for one particular
workload. If you can show me that this problem also happens under
other workloads we can work at changing it, but I don't think it's
right to optimise the VM for a specific workload...
regards,
Rik
--
Linux MM bugzilla: http://linux-mm.org/bugzilla.shtml
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com/
Hello Rik , As an aside to the below conversation .
Is there a URL/doc/... that gives basic tuning examples
for various types workloads ? Tia , JimL
On Sat, 10 Feb 2001, Rik van Riel wrote:
...snip...
> > It's still reluctant to shrink cache. I'm hitting I/O saturation
> > at 20 jobs vs 30 with ac5. (difference seems to be the delta in
> > space taken by cache.. ~same space shows as additional swap volume).
>
> Indeed, to "fix" that we'll need to work at refill_inactive().
>
> However, I am very much against tuning the VM for one particular
> workload. If you can show me that this problem also happens under
> other workloads we can work at changing it, but I don't think it's
...snip...
+----------------------------------------------------------------+
| James W. Laferriere | System Techniques | Give me VMS |
| Network Engineer | 25416 22nd So | Give me Linux |
| [email protected] | DesMoines WA 98198 | only on AXP |
+----------------------------------------------------------------+
On Sat, 10 Feb 2001, Mike Galbraith wrote:
> > > o Rebalance the 2.4.1 VM (Rik van Riel)
> This change makes my box swap madly under load. It appears to be
> keeping more cache around than is really needed, and therefore
> having to resort to swap instead. The result is MUCH more I/O than
> previous kernels while doing the same exact job.
I concur this, I watched a DVD tonight, and actually it got so bad
I had to reboot at one point as the it became too jerky to watch.
free output looked like this at this point...
total used free shared buffers cached
Mem: 254960 253252 1708 0 24116 174500
-/+ buffers/cache: 54636 200324
Swap: 248996 20848 228148
It appeared, that rather than free the cached buffers and reuse the
memory, it preferred to hit swap space. Streaming I/O performance seems
to have taken a hit lately.
(This was 2.4.1-ac9 btw)
regards,
Dave.
--
| Dave Jones. http://www.suse.de/~davej
| SuSE Labs
On Sat, 10 Feb 2001, Rik van Riel wrote:
> On Sat, 10 Feb 2001, Mike Galbraith wrote:
> > On Sat, 10 Feb 2001, Rik van Riel wrote:
> > > On Sat, 10 Feb 2001, Marcelo Tosatti wrote:
> > > > On Sat, 10 Feb 2001, Mike Galbraith wrote:
> > > >
> > > > > This change makes my box swap madly under load.
> > > >
> > > > Swapped out pages were not being counted in the flushing limitation.
> > > >
> > > > Could you try the following patch?
> > >
> > > Marcelo's patch should do the trick wrt. to making page_launder()
> > > well-behaved again. It should fix the problems some people have
> > > seen with bursty swap behaviour.
> >
> > It's still reluctant to shrink cache. I'm hitting I/O saturation
> > at 20 jobs vs 30 with ac5. (difference seems to be the delta in
> > space taken by cache.. ~same space shows as additional swap volume).
>
> Indeed, to "fix" that we'll need to work at refill_inactive().
If this reluctance to munch cache can be relaxed a little, I think
we'll see the end of a long standing problem. I often see a scenario
wherein we flush everything flushable, then steal the entire cache
before doing any paging. The result (we hit a wall) is a mondo swapout
followed immediately by swapping it all right back in. We seem to have
done a complete turnaround wrt paging vs flush/cache reap preference,
and that does effectively cure this scenario.. but methinks optimal
(-ENOENT?) lies somewhere in between.
> However, I am very much against tuning the VM for one particular
> workload. If you can show me that this problem also happens under
> other workloads we can work at changing it, but I don't think it's
> right to optimise the VM for a specific workload...
I'll watch behavior under other loads. (I don't have enough network
capacity to do anything stressful there, and whatever load I pick
has to be compute bound as to not end up benchmarking my modest I/O
capacity.. suggestions welcome. I use make -j primarily because it
doesn't need much I/O bandwidth for itself, but does allocate quite
a bit.. that leaves most I/O capacity free for vm usage)
Something else I see while watching it run: MUCH more swapout than
swapin. Does that mean we're sending pages to swap only to find out
that we never need them again?
-Mike
On Sun, 11 Feb 2001, Mike Galbraith wrote:
> Something else I see while watching it run: MUCH more swapout than
> swapin. Does that mean we're sending pages to swap only to find out
> that we never need them again?
(numbers might be more descriptive)
user : 0:07:21.70 54.3% page in : 142613
nice : 0:00:00.00 0.0% page out: 155454
system: 0:03:40.63 27.1% swap in : 56334
idle : 0:02:30.50 18.5% swap out: 149872
uptime: 0:13:32.83 context : 519726
On Sun, 11 Feb 2001 [email protected] wrote:
> On Sat, 10 Feb 2001, Mike Galbraith wrote:
>
> > > > o Rebalance the 2.4.1 VM (Rik van Riel)
> > This change makes my box swap madly under load. It appears to be
> > keeping more cache around than is really needed, and therefore
> > having to resort to swap instead. The result is MUCH more I/O than
> > previous kernels while doing the same exact job.
>
> I concur this,
> It appeared, that rather than free the cached buffers and reuse
> the memory, it preferred to hit swap space. Streaming I/O
> performance seems to have taken a hit lately.
OK, so we're back to the old VM magic number game again ;(
In short, we have to be more agressive towards unmapped
cache pages than towards mapped pages in processes, except
that this horribly breaks down when somebody does streaming
IO using mmap while somebody else is at the same time re-using
data from cached files (say, .h files)...
Now the question is ... WHY do we need to change this behaviour
and HOW exactly should it be changed ?
I don't really feel comfortable just tweaking stuff until we
get a half-dozen benchmarks right, I think we need to understand
what is happening and change things accordingly.
It's fine with me to put some temporary thing in place to get
at least -ac5 behaviour back, but I don't think we should have
this as a long-term thing.
regards,
Rik
--
Linux MM bugzilla: http://linux-mm.org/bugzilla.shtml
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com/
On Sun, 11 Feb 2001, Mike Galbraith wrote:
> On Sun, 11 Feb 2001, Mike Galbraith wrote:
>
> > Something else I see while watching it run: MUCH more swapout than
> > swapin. Does that mean we're sending pages to swap only to find out
> > that we never need them again?
>
> (numbers might be more descriptive)
>
> user : 0:07:21.70 54.3% page in : 142613
> nice : 0:00:00.00 0.0% page out: 155454
> system: 0:03:40.63 27.1% swap in : 56334
> idle : 0:02:30.50 18.5% swap out: 149872
> uptime: 0:13:32.83 context : 519726
Indeed, in this case we send a lot more pages to swap
than we read back in from swap, this means that the
data is still sitting in swap space and was never needed
again.
regards,
Rik
--
Linux MM bugzilla: http://linux-mm.org/bugzilla.shtml
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com/
On Sun, 11 Feb 2001, Rik van Riel wrote:
> On Sun, 11 Feb 2001, Mike Galbraith wrote:
> > On Sun, 11 Feb 2001, Mike Galbraith wrote:
> >
> > > Something else I see while watching it run: MUCH more swapout than
> > > swapin. Does that mean we're sending pages to swap only to find out
> > > that we never need them again?
> >
> > (numbers might be more descriptive)
> >
> > user : 0:07:21.70 54.3% page in : 142613
> > nice : 0:00:00.00 0.0% page out: 155454
> > system: 0:03:40.63 27.1% swap in : 56334
> > idle : 0:02:30.50 18.5% swap out: 149872
> > uptime: 0:13:32.83 context : 519726
>
> Indeed, in this case we send a lot more pages to swap
> than we read back in from swap, this means that the
> data is still sitting in swap space and was never needed
> again.
But it looks and feels (box is I/O hyper-saturated) like
it wrote it all to disk.
(btw, ac5 does more disk read.. ie the reduced cache size
of earlier kernels under heavy pressure does have it's price
with this workload.. quite visible in agregates. looks to
be much cheaper than swap though.. for this workload)
-Mike
On Sun, 11 Feb 2001, Mike Galbraith wrote:
> On Sun, 11 Feb 2001, Rik van Riel wrote:
>
> > On Sun, 11 Feb 2001, Mike Galbraith wrote:
> > > On Sun, 11 Feb 2001, Mike Galbraith wrote:
> > >
> > > > Something else I see while watching it run: MUCH more swapout than
> > > > swapin. Does that mean we're sending pages to swap only to find out
> > > > that we never need them again?
> > >
> > > (numbers might be more descriptive)
> > >
> > > user : 0:07:21.70 54.3% page in : 142613
> > > nice : 0:00:00.00 0.0% page out: 155454
> > > system: 0:03:40.63 27.1% swap in : 56334
> > > idle : 0:02:30.50 18.5% swap out: 149872
> > > uptime: 0:13:32.83 context : 519726
> >
> > Indeed, in this case we send a lot more pages to swap
> > than we read back in from swap, this means that the
> > data is still sitting in swap space and was never needed
> > again.
>
> But it looks and feels (box is I/O hyper-saturated) like
> it wrote it all to disk.
>
> (btw, ac5 does more disk read.. ie the reduced cache size
> of earlier kernels under heavy pressure does have it's price
> with this workload.. quite visible in agregates. looks to
> be much cheaper than swap though.. for this workload)
Mike,
Could you please try the attached patch on top of latest Rik's patch?
Thanks!
--- linux.orig/mm/vmscan.c Sun Feb 11 07:56:29 2001
+++ linux/mm/vmscan.c Sun Feb 11 11:05:30 2001
@@ -563,7 +566,8 @@
/* The buffers were not freed. */
if (!clearedbuf) {
add_page_to_inactive_dirty_list(page);
- flushed_pages++;
+ if (wait)
+ flushed_pages++;
/* The page was only in the buffer cache. */
} else if (!page->mapping) {
On Mon, 12 Feb 2001, Marcelo Tosatti wrote:
> On Sun, 11 Feb 2001, Mike Galbraith wrote:
>
> > On Sun, 11 Feb 2001, Rik van Riel wrote:
> >
> > > On Sun, 11 Feb 2001, Mike Galbraith wrote:
> > > > On Sun, 11 Feb 2001, Mike Galbraith wrote:
> > > >
> > > > > Something else I see while watching it run: MUCH more swapout than
> > > > > swapin. Does that mean we're sending pages to swap only to find out
> > > > > that we never need them again?
> > > >
> > > > (numbers might be more descriptive)
> > > >
> > > > user : 0:07:21.70 54.3% page in : 142613
> > > > nice : 0:00:00.00 0.0% page out: 155454
> > > > system: 0:03:40.63 27.1% swap in : 56334
> > > > idle : 0:02:30.50 18.5% swap out: 149872
> > > > uptime: 0:13:32.83 context : 519726
> > >
> > > Indeed, in this case we send a lot more pages to swap
> > > than we read back in from swap, this means that the
> > > data is still sitting in swap space and was never needed
> > > again.
> >
> > But it looks and feels (box is I/O hyper-saturated) like
> > it wrote it all to disk.
> >
> > (btw, ac5 does more disk read.. ie the reduced cache size
> > of earlier kernels under heavy pressure does have it's price
> > with this workload.. quite visible in agregates. looks to
> > be much cheaper than swap though.. for this workload)
>
> Mike,
>
> Could you please try the attached patch on top of latest Rik's patch?
Sure thing.. (few minutes later) no change.
-Mike
P.S.
(said I'd try other loads...)
I tried a different workload yesterday (only one because it took
the entire day to finish). Glimpseindexing my webserver test area
played well. As it was running, all of it's growth was pushed to
swap (very slow trickle for some hours). I figured it would get
beaten up when it started actually using the dataset it was building,
but the way it used it, sending it to swap was perfect. When it
started using, it paged in a small burst and then used this data
combined with a truckload of I/O (actual database building bit).
Since there were hours between dataset generation and subsequent
use, I had free use of about ~30 mb of pages for those hours.
Doing other things (like letting lynx webcrawl over my other box's
server area while it was running) were not even visible.. ie caused
no additional paging, so huge cache/buffers space (~100mb of 128mb)
was not plugging up the mana supply in any way. What was in cache
_looked_ to be old cruft, and the system released these resources
just fine. (so why won't it give up cache to gcc? either the vm
doesn't like gcc, or I flat don't understand page aging yet. nod;)
Very boring test session. I hope the results mean more to you than
they do to me ;-)
On Tue, 13 Feb 2001, Mike Galbraith wrote:
> On Mon, 12 Feb 2001, Marcelo Tosatti wrote:
>
> > Could you please try the attached patch on top of latest Rik's patch?
>
> Sure thing.. (few minutes later) no change.
That's because your problem requires a change to the
balancing between swap_out() and refill_inactive_scan()
in refill_inactive()...
The big problem here is that no matter which magic
proportion between the two functions we use, it'll always
be wrong for a large proportion of the people out there.
This means we need to have a good way to auto-tune this
thing. I'm thinking of letting swap_out() start out way
less active than refill_inactive_scan() with extra calls
to swapout being made from refill_inactive_scan when we
think it's needed...
(... I'm writing a patch right now ...)
regards,
Rik
--
Virtual memory is like a game you can't win;
However, without VM there's truly nothing to lose...
http://www.surriel.com/
http://www.conectiva.com/ http://distro.conectiva.com.br/
On Tue, 13 Feb 2001, Rik van Riel wrote:
> On Tue, 13 Feb 2001, Mike Galbraith wrote:
> > On Mon, 12 Feb 2001, Marcelo Tosatti wrote:
> >
> > > Could you please try the attached patch on top of latest Rik's patch?
> >
> > Sure thing.. (few minutes later) no change.
>
> That's because your problem requires a change to the
> balancing between swap_out() and refill_inactive_scan()
> in refill_inactive()...
>
> The big problem here is that no matter which magic
> proportion between the two functions we use, it'll always
> be wrong for a large proportion of the people out there.
>
> This means we need to have a good way to auto-tune this
> thing. I'm thinking of letting swap_out() start out way
> less active than refill_inactive_scan() with extra calls
> to swapout being made from refill_inactive_scan when we
> think it's needed...
>
> (... I'm writing a patch right now ...)
We're using nr_async_pages to calculate the number of pages which should
be flushed, but nr_async_pages counts on flight swap _readaheads_ (each
swapin increases nr_async_pages by (1 << page_cluster)) and writes, not
only writes.
That makes the "pageout free shortage and sleep" kswapd behaviour you
wanted a bit messy.