Not yet resynchronized with Marcelo 2.4.22-pre1. XFS should work still but
may need some updates for the O_DIRECT changes. The exploding audio driver
problem should be fixed.
Linux 2.4.1-ac3
o Fix an hpt driver bug triggered by the new HPT (me)
BIOS
o Initial VIA and S3 DRM modules merges (VIA)
| These are marked up with some warnings and need
| a chunk of clean up work yet.
o btaudio update (Gerd Knorr)
o Backport 2.5 ipc/sem.c fix (Manfred Spraul)
o Fix scsi_register failure path for aacraid (Michael Still)
o First crack at fixing the ide reset oopses (me)
o Fix the incompatibility between via audio and (me)
esd/gnome desktops
Linux 2.4.21-ac2
o Fix an HPT37x oops case with 374 and non 33Mhz (me)
clock
o Fix some emu10k1 oopses from the ac97 updates (me)
o Fix AF_UNIX dgram select problem (Krzysztof Halasa)
o Fix Make xconfig problem (Edward Macfarlane Smith)
o Add another new eepro100 ident to eepro100.c (Tom Alsberg)
o Fix aacraid new volume crash (me)
o Fix math-emu build with gcc 3.3 hopefully (me)
o Fix wrong comparisons on some sound drivers (Adrian Bunk)
for codec detect
o Fix problem with eexpress driver (Shane Shrybman)
o Remove LNZ reference from Configure.help (Matthias Andree)
o Add support for Cirrus PCI/PCMCIA (Komuro)
o Allow both Cirrus and 82092 to exist together (me)
o Fix a remove race in the 82092 driver (me)
o Fix dnotify bug with read/writev (Zou Pengcheng)
o SiS IDE update (Lionel Bouton)
o Add PCMCIA bus ident to ethtool for pcmcia (Bill Nottingham)
devices
o Backport 2.5 updates to overcommit interface (Rik van Riel)
o Synclink driver updates (Paul Fulghum)
o Fix z85230 losing frames by waking the queue (Stefan Tamas)
too early
o O_DIRECT race fixes (Stephen Tweedie)
| Tweaked a little to merge with XFS. XFS folks should double
| check these.
o Allow use of db4 to build aicasm (J A Magallon)
o Revert problematic scsi change (fixes ppa/imm) (ChrisW)
o Further highpoint raid updates (Wilfried Weissmann)
o Fix data direction for start stop scsi command (Heiko Carstens)
o Fix a module loader memory leak (Keith Owens)
o Fix multiple ircomm bugs (Jean Tourrilhes)
o Fix irda qos handling, add tx_window setting (Jean Tourrilhes)
o Correct race in IrLMP (Jean Tourrilhes)
o Fix an RR/Poll handling error in irlap (Jean Tourrilhes)
o Update irda-usb driver (Jean Tourrilhes)
o Fix IrIAP leak (Jean Tourrilhes)
o Mask C/R bit in irlap connection address (Jean Tourrilhes)
o Fix some non modular build errors (Jean Tourrilhes)
o Remove duplicated junk from dasd_fba (Rik van Riel)
o Fix mac8390 typo (Etsushi Kato)
o Fix m68k floppy warnings (Geert Uytterhoeven)
o Fix apollo compile (Geert Uytterhoeven)
o Update mac sonic driver (Geert Uytterhoeven)
o Fix atari pamsnet driver (Geert Uytterhoeven)
o Fix mac68k keyboard warning (Geert Uytterhoeven)
o Fbcon fixes (Geert Uytterhoeven)
o M68K page handling fixes (Roman Zippel)
o Use ptrace_check_attach on m68k (Andreas Schwab)
o Remove m68k phys/virt fallback (Roman Zippel)
o Fix lmc build on some platforms (Geert Uytterhoeven)
o Update meye driver (Stelian Pop)
o Update sonypi driver (Stelian Pop)
o Fix a couple of glitches in the added 1210SA (Hugo Mills)
support for siimage
o Fix llc802.2 leak case (Oleg Drokin)
o Add more unexpected apic filtering (Randy Dunlap)
o Fix a case where hid rejected some UPS devices (Vojtech Pavlik)
o Fix via audio crash on setup (me)
Linux 2.4.21-ac1
o Further ehci fixes and updates (David Brownell)
o CPUfreq updates (Dominik Brodowski)
o Updated DMI blacklists and facility to disable (Arjan van de Ven)
APM idling
o Allow an NFS to specify if d_revalidate is a (Trond Myklebust)
stale handle event or not
| Should fix CODA and VFAT problems
o Ignore residue on autosense commands in (Tony Battersby)
sym53c8xx_2
o SiS frame buffer updates (Tomas Winischhofer)
o Add Xyratex to scsi list of large/sparse lun (David Lethe)
o MPT fusion driver updates (Pam Delaney)
o Hopefully fix the Promise SX6000 clash with (me)
PDC202xxx IDE drivers when using 20277.
o Add a framework for plug in ac97 drivers (me)
o Fix IRQ handling in speedstep-ich (Samuel Thibault)
o Redo the AC97 codec plugin interface (me)
Linux 2.4.21rc8-ac1
Sync with Marcelo 2.4.21rc8
o Fix megaraid tools breakage due to bad fix
o Add support for Dell XP 4100CX SCSI (Gregg Lebovitz)
| Possibly a contender for the longest delay in
| patch application ever (2.5 years)
o Fix force_ac97 printk in nm256 driver (Douglas Kilpatrick)
o Fix plip hang on ifdown/ifup (Stas Sergeev)
o Two more minor siimage updates (me)
o Handle PDC20275 on Promise SX6000 (me)
o Fix erroneous licensing clash (Kurt Robideau)
o DSCC driver updates/reset handling (Francois Romieu)
o Add initial support for Adaptec/SII 1210SA (Hugo Mills, me)
| Tidied up Hugo's changes so we can cleanly add
| more idents later
o Fix missing hid include (Pete Zaitcev)
o Driver for AX8817x based USB ethernet (Tivo, Dave Hollis)
| Original driver by Tivo, cleanup by Dave Hollis
| Supports Linksys USB200M, Netgear FA120, Dlink
| DUB-E100, Hawking UF200
o Remove bogus printk in AWE driver (me)
o Fix init of mp_bus_id table (Herbert Xu)
o Fix double iounmap on octagon probe fail (Francois Romieu)
o Fix several floppy driver SMP races (Jens Axboe)
o Fix a shmem deadlock (Andrea Arcangeli)
o Fix ST16654 uart overrun (Ed Vance, William King)
Linux 2.4.21rc7-ac1
Sync with Marcelo 2.4.21rc7
o PC300 updates (Henrique Gobbi,
Nigel Metheringhma)
o Fix ide-proc crash with no drivers loaded (Herbert Xu)
o Fix pc300 build with new binutils (Adrian Bunk)
o Minor | to || in fpu emulator (Margit Schubert-While)
o Fix remaining CRC32 build problem with luck (David Woodhouse)
o Use pentium3/4 targets with newer gcc (Margit Schubert-While)
o Remove DRM id string stuff as in 2.5 (Margit Schubert-While)
o Merge radeonfb updates for newer radeon (Margit Schubert-While)
o Add nokia 5510 to usb-storage (Markus Gaugusch)
o Merge hfsplus file system (Roman Zippel,
Brad Boyer, ...)
o Fix ac97 build on SMP (Adrian Bunk)
o Fix pcnet32 unload crash with multiple cards (Matt Wilson)
o Make it clear AMD IDE covers Nvidia (Paolo Ornati)
Linux 2.4.21rc6-ac2
o Add framework for pci_device_restart error (me)
recovery
o Smarter vesafb ram handling for double (Thomas Backlund)
buffering
o Fix vesafb docs (Thomas Backlund)
o Fix ppp_generic error path leak (Patrick McHardy)
o Add VIA P4M266 AGP1x/2x/4x support (Nimrod A Abing)
o Allow vga16 frame buffer on IA64 (Jeremy Katz)
o Fix & v && in sysrq.c (Margit Schubert-While)
o Remove various bits of old # if 0 in pdc202xx (Carl-Daniel Hailfinger
me)
o Further hacking on the SI680 driver (me)
o Add the Toshiba Piccolo IDE (me)
o Update ac97 layer so that it has alloc/release (me)
o Make the drivers use the ac97 changes (me)
o Fix up drivers not doing ac97 locking (me)
o Fix memory leak in forte ac97 handling (me)
o Always tune DMA on the SI680 as the BIOS (me)
firmware doesn't.
Linux 2.4.21rc6-ac1
Merge with Marcelo 2.4.21-rc6
o Finish off the codec->digital operations (me)
o Move i810 code into generic digital ops (me)
o Make i810_audio use ac97_codec ops properly (me)
o Add per card quirk handling to aacraid (me)
o Resolve pipe writing to r/o file systems (Stephen Tweedie)
o Add missing hidden device check (Mark Salyzyn)
o SG fixes (Douglas Gilbert)
o Update ACPI to next Intel release (Bernhard Rosenkraenzer)
o Fix pcibios code on SH3 for core changes (Paul Mundt)
o Resync tlan with Jeff (Jeff Garzik)
o Fix solutionengine build (Paul Mundt)
o SH build issues fixes (Paul Mundt)
o First cut at making modular IDE happy again (me)
Linux 2.4.21rc5-ac2
o Backport tlan update and add 64bit support (me)
o Make SiS APIC work (me)
o Fix incorrect strncmp in sclp driver (Pete Zaitcev)
o Fix further cmpci copy/user bugs (Hollis Blanchard)
o Update ACPI to newer intel base code (Bero Rosenkraenzer)
o Fix a bogus ; in the ACPI code (Pavel Machek)
o IDE I/O and DMA state machine fixes (Alexander Atanasov)
o Fix gcc 3.3 build of ma600 (Eduardo Pereira Habkost)
Linux 2.4.21rc5-ac1
Resync with Marcelo 2.4.21-rc5
o IBM S/390 architecture bug fixes (Martin Schwidefsky)
o IBM S/390 DASD driver fixes (Martin Schwidefsky)
o IBM S/390 sclp fixes (Martin Schwidefsky)
o IBM tape fixes (Martin Schwidefsky)
o AGPgart support for SiS651 (Thomas Winischhofer)
o Fix emulation bug in the mips utsname handling (Ben Collins)
o Fix module list race (Keith Owens)
o Fixes to CRC32 library and optimisations (David Woodhouse,
Joakim Tjernlund)
o I/O pause waitqueue fix (Jens Axboe)
o Fix typo in ide Config.in (Carl-Daniel Hailfinger)
o Fix ac97 to init volume mute bit on volume (me)
control lacking codecs
o Correct vm86 virtualisation of pushf iopl (Stas Sergeev)
| Fixes some problems with dos4gw
o Nvidia AGP support (someone in Nvidia)
o Fix gcc 3.3 build of sdla_chdlc (Eduardo Pereira Habkost)
o Fix gcc 3.3 build of olympic (Eduardo Pereira Habkost)
o Fix gcc 3.3 build of DECnet (Eduardo Pereira Habkost)
o Fix gcc 3.3 build of cs46xx (Eduardo Pereira Habkost)
o Report -EFAULT back on /proc/misc (Daniele Bellucci)
o Merge S/390 ctrlchar handler fix (Pete Zaitcev)
o Add oneshot support to UHCI USB as well (Pete Zaitcev)
o Fix type errors in kcopy irq save (Guy Streeter)
o Major merge of aacraid updates (Mark Salyzn, Deanna Bonds)
| 64bit stuff, raid mode, SATA, other new idents
o Rewrite the aacraid link list code to (me)
be 64bit safe
| This aacraid is likely to need a bit more debugging yet
| and also there is work to do on error recovery handling
o Fix CRC library problem with ksyms (me)
Linux 2.4.21rc4-ac1
Resync with Marcelo 2.4.21-rc4
o Allow setuid core dumps by a sysctl (me)
| First test release. Note that you probably want to set
| the core dumping path to somewhere like /cores/ via sysctl
| when using this feature. Setuid cores are made root only
| and owner rw only.
Linux 2.4.21rc2-ac3
o Possible fix for IDE lost IRQ problem (Alexander Atanasov)
o Add VIA KL/KM266 AGP (Dietrich Radel)
o Handle older Compaq ACPI * prefix to HID (Andrew Grover)
o Add mmio==2 support to the ide-dma layer (me)
o Clean up and document the CMD680 driver (me)
o Fix the CMD680 resource handling bugs (me)
o VIA 8327 IDE support (Vojtech Pavlik)
o Update SiS IDE for 655/630SET and oddments (Lionel Bouton)
o Fix missing exports for modular XFS (Bero Rosenkraenzer)
o Update worst case quota block count for ext3 (Jan Kara)
o Add CRC32 libraries backport (David Woodhouse, Duncan Sands)
o HDLC doc fixes (Krzysztof Halasa)
o Use tail -n for the posixly afflicted (Olaf Hering)
o Fix wrong use of strstr in reiserfs (Sam Ravnborg)
o Correct AMD enable bits (Vojtech Pavlik)
| Should fix IDE boot timeout when probing empty AMD device slots
o Fix netlink compile with gcc 3.3 (Andrew Church)
o Frame buffer is in bits so fix vesafb (Adam Mercer)
o Make ksoftirqd cpuid match 2.5 (and allow for (Martin Hicks)
100 cpus without overflow..)
o Fix a pcmcia without ISA crash (Pavel Roskin)
o Fix cs89x0 set_mac_address handling (Stefano Fedrigo)
o Update the ide tags I forgot to do before (me)
o Fix a delayed block/xfs problem from a missed (Christoph Hellwig)
merge
Linux 2.4.21rc2-ac2
o Use pid_t for pid in aacraid (Walter Harms)
o Prevent user compiling both megaraid drivers in (Adriank Bunk)
o Fix missing wolfson codec bits (me)
o Add SIS648 AGP (Grzesiek Wilk)
o Fix devexit in kahlua audio (Adrian Bunk)
o SunRPC timer missing HZ scaling (Steve Dickson)
o Fix a bogus kfree in iphase ATM (Patrick McHardy)
o Don't clobber SMI APIC routing (John Stultz)
o Fix big endian rtl8150 (Geert Uytterhoeven)
o Fix binfmt_elf leak on error paths (Oleg Drokin)
o Handle thread create errors in aacraid (Walter Harms)
o Allow a user to mark a disk as for scsi at (Matan Ziv-Av)
boot even with ide-scsi is modular
Linux 2.4.21rc2-ac1
Merge with Marcelo 2.4.21rc2
o Ensure we do not enable DMA on early SLC82105 (Ben Herrenschmidt)
o Fix sclp compile (Rik van Riel)
o Fix C7000 compile (Rik van Riel)
o Fix CMD680/SII clocking problems (me)
o C7000 header clean up (me)
o Fix IDE setup for old PC202xx raid (Bartlomiej Zolnierkiewicz)
o Update Quadrics PCI idents with vendor data (Daniel Blueman)
o Update XFS to current XFS (Christoph Hellwig)
o Update wolfson codec support with vendor (Liam Girdwood)
supplied changes
o Allow user to override vesa video ram (Thomas Backlund)
o Fix ibm hotplug build (Geller Sandor,
Andreas Haumer)
o Fix ipmi build with ACPI (Andreas Haumer)
Linux 2.4.21rc1-ac4
o Fix vesafb over allocation of I/O memory (Adam Mercer)
| VESA reports ram on card but that may be banked
| and is more than we need to map. On 128/256Mb
| cards we really don't want to have this happen
o Relaxed AML checking (needed for Toshiba (Rick Richardson)
laptops and other stuff built with some old
buggy AML tools)
o Fix v110 fill on hisax PCI (David Woodhouse)
o Ensure cable id pin setup is correct on hpt ide (Duncan Laurie)
o Quirkhandler for serverworks CSB5 IRQ (Duncan Laurie)
o Add NFSv3 pathconf/fsinfo support (Steve Dickson)
o Fix NFS close/open consistency setting bug (Shantanu Goel)
o Fix the mess in ibmphp_ebda (me)
o Add Vivitar Vivicam to unusual_devs (Thomas Rabe)
o Clean up lots of code that could use mod_timer (Vinay Nallamothu)
so that it does (sch_csz, sch_htb, synclink, sdla_x25,
sdla_fr, sdla_ppp, sdla_chdlc, sch_cbq, mptctl)
o Allow IDE drivers to reserve interfaces for
found hotplug controller with no disk (Ben Herrenschmidt)
o Intel ICH5 basic SATA support (Andre Hedrick)
o Merge some of Greg's ibmphp cleanups (Greg Kroah-Hartmann)
o vsprintf fix (Al Viro)
o Initial ICH5 audio support (Martin Schlemmer)
| Please report any problems
o Merge new AMI Megaraid driver as megaraid2 (Atul Mukker)
o Add ALC100P codec (Ni Wei)
Linux 2.4.21rc1-ac3
o Fix copy/user handling errors in mpu401, (me)
mdc800, eicon, vicam
| From Stanford checker
o Fix an i810 error path bug that showed up (John Stultz)
in new Macromedia flash player
o parisc arch code resync (Joel Soete)
o Merge big endian sstfb updates (Joel Soete)
o Fix compile with no quota again (without a (me)
typo this time)
o Fix missing fc_type_trans (Andreas Haumer)
o Fix DRM 4.0 build (Xosé Vázquez Pérez)
o Fix SiS746 AGP merge (Volker Armin Hemmann)
o Merge XFS core code (Steve Lord, Christoph Hellwig, and a load
more people)
o Merge current Intel ACPI
| Except the mem= bits which neec bootloader resyncs
| This breaks ipmi but that shouldnt be too hard to clean up
| and should end up a lot nicer
Linux 2.4.21rc1-ac2
o Add hwif->rw_disk callout (me)
| This allows us to remove the PDC4030 special case
| and also allows for the 372N clock switch stuff.
o Add HPT372N clock switcher (needs testing carefully)
o TDFX framebuffer improvements/fixes (Jakub Bogusz)
o Hopefully fix legacy hd only build (me)
|Reported by Jerome Chantelauze but different fix
o Orinoco updates (David Gibson)
o AGP update for Intel 852/855 (David Dawes)
o Fix leak in rio firmware handler (Oleg Drokin)
o Fix leak on aironet4500 error path (Oleg Drokin)
o Fix leak in roadrunner exit path (Oleg Drokin)
o Use the FAT free cluster hints in Linux (Björn Stenberg)
o Update Intermezzo contacts (Jörn Engel)
o Add DMI handling for broken PnPBIOS (me)
o Fix build without quota support (Pavel Roskin)
o Backport 2.5 slab poison improvements (Faik Uygur)
o Initial SiS 746 AGP (not for 8x yet) (Volker Hemmann)
o CCISS updates (support for 6404/256, cross (Mike M)
platform fixes, 64bit DMA
o Fix the ide unregister deadlock bug (me)
o Generic XAPIC support (8 way HT etc) (Venkatesh Pallipadi,
Ingo Molnar)
o A collection of NFS fixes (Steve Dickson)
o Fix IDE makefile a bit further (Christoph Hellwig)
o Use new ieee1394 code (Ben Collins)
o Minimal S/390 fixes to get -ac running ok (Martin Schwidefsky)
o Update S/390 cio layer (Martin Schwidefsky)
o Update S/390 DASD drivers (Martin Schwidefsky)
o Update S/390 31bit emulation (Martin Schwidefsky)
o S/390 documentation updates (Martin Schwidefsky)
o 3215 driver updates (Martin Schwidefsky)
o Update S/390 ctc layer (Martin Schwidefsky)
o S/390 iucv updates (Martin Schwidefsky)
o Replace hwc with backport of 2.5 sclp (Martin Schwidefsky)
o Do the same with the 2.4/2.5 S/390 tape (Martin Schwidefsky)
o Updates to Serverworks IDE (Duncan Lane, me)
Linux 2.4.21rc1-ac1
Merge Marcelo 2.4.21-rc1
- Drop broken m68k ide change
o Fix PPC build (Olaf Hering)
o Fix up d_path handling (Christoph Hellwig)
o Update IPMI (Corey Minyard)
o Fix ext3 orphan race (Ernie Petrides)
o Update seq_file to match 2.5 (Randy Dunlap)
o Remove experimental runtime scsi switch for IDE (me)
| Fixing it requires major ide register rewriting
o Fix a deadlock on ide_unregister_subdriver (Ben Herrenschmidt, me)
o Fix an ext3 quota deadlock (Jan Kara)
o Fix ohci single shot interrupt out (Frode Isaken)
o Update summit idents (James Cleverdon)
o Clear sense buffer before retrying command (Alan Stern)
o Fix 82092 on a PCI bus with no ISA bridge (David Woodhouse)
o Fix duplicate pid corner case (Takayoshi Kochi)
o Add VIA phy to SiS900 driver (Pedro A Gracia Fajorda)
Linux 2.4.21pre7-ac2
o HPT raid support for disk-spanning/initial bits (Wilfried Weissmann)
of raid1
o Cyclades PC300 driver initial merge (Henrique Gobbi)
o Fix bigendian use of pegasus driver (Paul Mackerras)
o Fix copy/user bugs in zoran drivers (me)
|From Stanford checker
o Fix copy/user bugs in sisfb (me)
|From Stanford checker
o Fix copy/user bugs in intermezzo (me)
|From Stanford checker
o Fix copy/user bug in cmi8330 driver (me)
|From Stanford checker
o Fix copy/user bug in awe sound (me)
|From Stanford checker
o Merge GPL version of UTS Global CLAW driver (Bob Scardapane)
o Make cardbus fall back to PCI irq routing if (Pavel Roskin)
needed
o Fix sign bug in decnet (Oleg Drokin)
o Add AZT1008 PnP identifiers to ad1848 (Zwane Mwaikambo)
Linux 2.4.21pre7-ac1
Merge with Marcelo 2.4
o Merge memory barrier bits (Zwane Mwaikambo)
o Fix ip_conntrack merge after free (Martin Josefsson)
o Stop failing sethostname from clearing entire (Stephan Maciej)
field
o Fix Config.in syntax for ADMA-100 (Mark Lord)
o Remove IDE_DEBUG macro from 2.4 as well (Alexander Atanasov)
o In some situatiosn rq->buffer changes under (Stephan von Krawcyznski)
us in scsi. Store the idescsi_pc in ->special
where it probably belongs anyway
o Arcnet oops fixes (Herbert Xu)
o Pmac IDE update (Ben Herrenschmidt)
o Jbd compile warnings fixes (Stephen Tweedie)
o Dquot lock fix (Oleg Drokin)
o I2c fixups (Greg Kroah-Hartmann)
o Scsi tape updates (Kai Makisara)
o Update DAC960 and Qlogic drivers for Alpha (Jay Estabrook)
o Fix non pci build (Stephane Oullette)
o Fix non DMA ide build (Andries Brouwer)
o Fix PIO boot serverworks IDE problem (Robert Hentosh,
me)
o Small nfs dentry/dir fix (Steve Dickson)
o Allow longer for diagnostic commands in scsi (Douglas Gilbert)
o Sunrpc locking fix (Steve Dickson)
o Make tty->count atomic (Jes Sorensen)
o Update ipmi (Corey Minyard)
o Fix multiplex syscall wrong return code (Ulrich Drepper)
o M68K IDE updates (Geert Uytterhoeven)
o Small quota compatibility fix (Jan Kara)
o FPU copy fix (Ingo Molnar)
o MPT Fusion update (Pam Delaney)
o SonyPi update (Stelian Pop)
o Reiserfs journal fixup (Oleg Drokin)
| Sanity test fail on old fs's
o Fix X.25 crash on unknown facilities (Tiaan Wessels)
o Fix iphase module on new binutils (Adrian Bunk)
o Fix ad1889 module on new binutils (Adrian Bunk)
o Ditto for nsp32, ips, rtl8169
o SiS frame buffer updates (Thomas Winischhofer)
o ndelay for m68k systems (Geert Uytterhoeven)
o m68k raw I/O updates (Geert Uytterhoeven)
o Fix IDE completion race (Jens Axboe, Andrew Morton)
o m68k needs WANT_PAGE_VIRTUAL except sun (Richard Zidlicky)
o Remove duplicate copy of PROC_CONSOLE (Geert Uytterhoeven)
o Fix swapoff crash (Szabolcs Berecz)
o Fix is_dumpable on zombies (Marc-Christian Petersen)
o Add vicicam to unusual storage devices
o Update sony unusual device entries (Hanno Böck)
Linux 2.4.21pre5-ac4 (not released generally)
o Add initial test support for HPT372N (me)
o Fall back to PIO if the BIOS got mmio setup (me)
wrong for an SI3112/CMD680
| Still doesnt explain some problems
o Update amiga floppy driver (Geert Uytterhoeven)
o AmigaFB wrong IRQ fix (Geert Uytterhoeven)
o Amiga RTC updates (Kars de Jong)
o Amiga PCMCIA ethernet cleanups (Kars de Jong)
o Fix Amiga isa space mapping (Kars de Jong)
o Update apollo MMIO and pseudio MMIO (Geert Uytterhoeven)
o Fix bitop abuse in 5380 drivers for m68k (Geert Uytterhoeven)
o Fix m68k with recent binutils (Andreas Schwab)
o m68k prototype fix (Geert Uytterhoeven)
o m68k heartbeat config fix (Geert Uytterhoeven)
o Convert m68k cache macros to be inline (Geert Uytterhoeven)
o Update m68k VIA stuff (Ray Knight)
o Make m68k page size to fix warnings (Geert Uytterhoeven)
o Allow mac68k to build with no fb (Geert Uytterhoeven)
o Fix m68k network driver warnings (Geert Uytterhoeven)
o Backport m68k page_to_phys from 2.5 (Richard Zidlicky)
o Move m68k low level iomap defines around (Richard Zidlicky)
o Update sun3 contact info (Geert Uytterhoeven)
o m68k warning fixes for scsi (Geert Uytterhoeven)
o Optimised stack check for m68k (Roman Zippel)
o M68K spelling fixes (Steven Cole)
o Make all sun3 pages as zone 0 (Sam Creasey)
o Add ioremap for sun3 and use it in drivers (Sam Creasey)
o Sun3/3x updates and cleanups (Sam Creasey)
o Fix page calculation for first virtual page (Sam Creasey)
on sun3
o Rename sbus structs for sparc compatibility (Sam Creasey)
o Update sun3 vectored interrupts (Sam Creasey)
o Dont update rtc from clock eveyr 11 mins (Geert Uytterhoeven)
o Add Sun3 VME support (Sam Creasey)
o ISDN ppp locking fix (Patrick McHardy)
o Semtimedop backport (Mark Fasheh)
o Fix missing cli in isdn_net (Patrick McHardy)
o Handle radeonfb mobility cards reporting (Hanno Bock)
no memory
o Add another broken APM bios (Arjan van de Ven)
o Add Centrino IDE support (Dean Gaudet)
o Fix ibm hotplug memory leaks (Oleg Drokin)
o Fix xjack memory leaks (Oleg Drokin)
o I2O memory leak fix (Oleg Drokin)
o Emu10K memory leak fix (Oleg Drokin)
o cpqfc memory leak fix (Oleg Drokin)
o dpt_i2o memory leak notes (Oleg Drokin)
o Fix -ac build on alpha (Ivan Kokshaysky)
o Fix fd leak in initrd (Pete Zaitcev)
o Megaraid cleanup/check fix (Oleg Drokin)
o sx memory leak fix (Oleg Drokin)
o Kobil USB memory leak fix (Oleg Drokin)
o USB memory leak fix on hub (Oleg Drokin)
o Fix iphase driver null cells bug (Eric Leblond)
o Fix non zero offset reads on /proc/cmdline (Dick Streefland)
o Fix pdcraid ioctl pass through (Jens Axboe)
o Make hdparm report error on cable refusal (Jens Axboe)
o Reiserfs warning fix (Maciej Soltysiak)
o Fix warning in make_configs (Maciej Soltysiak)
o Remove unused variable in ide-proc (Maciej Soltysiak)
o CMD640 locking bug fixups (Alexander Atanasov)
o General IDE driver resync
o Add another datafab kecf to the dev list (Chris Clayton)
o Fix wrong type for timer in aha152x (Christoph Hellwig)
o Avoid IDE hang on SMP when doing DMA->PIO (Petr Vandrovec)
changedown on error
o Fix ide_wait_50ms fencepost error (Alexander Atanasov)
Linux 2.4.21pre5-ac3
o Add cpuid for SiS processors (SiS SiS SiS) (me)
o Fix basic ADMA100 driver support (Mark Lord)
o Fix memory leak on UFS error path (Oleg Drokin)
o Fix eepro100 ethtool hang (Jason Lunz)
o Fix procfs memory leak (Kazuto Miyoshi)
o Forte media driver update (Martin Petersen)
o WIN_SET_MAX crashes some old Samsung disks so (Jens Axboe)
dont issue it on disks < 32Gb in size
o Compaq MS1000 may have sparse lun (Tom Coughlan)
o Add SiS FB idents for newer chipsets (Thomas Winischhofer)
o Fix vsscanf in hex mode (Kevin Corry)
o Fix 64bit jiffy cleanness in sis900, shaper, (Dave Miller)
dgrs, qlogicfc and tty layer
o Reiserfs journal overflow fix (Hans Reiser)
o PCMCIA oops fix with HostAP (Pavel Roskin)
o Handle more panasonic compact USB CD-ROMs (Go Taniguchi)
o Extend USB hotplug to handle multi interface (Go Taniguchi)
HID devices (eg IBM BladeCenter)
o Update ALi PCI ident data (TH Chou)
o Fix memory leak in ldm error path (Oleg Drokin)
o NCPFs ioctl passed wrong parameter (Oleg Drokin)
o Fix leak in ircomm core error path (Oleg Drokin)
o Make xconfig syntax error fixes (Andreas Gruenbacher)
o Fix memory leak in vlanproc exit path (Oleg Drokin)
o Fix iphase misaligned skb (I hope) (me, based on stuff by
Eric Leblond)
o Fix a couple of printk levels in IDE (Alan Cox)
Linux 2.4.21pre5-ac2
o Add PCI idents for ALi 1563 to dmfe (Clear Zhang)
o Busproc operations now error if unsupported (me)
o Make busproc handler return a status
o Fix IDE reset locking. We don't want an IRQ (me)
poking around during a reset while the iface
state is undefined
o Remove half baked request clean up code (me)
from ide_do_reset. We require the caller
cleans up first
o Add ide_abort functions to abort due to (me)
host not target triggered events
o Remove a pile of surplus hwgroup checks (me)
o Fix the reset ioctl paths to use (me)
ide_abort
o Fix PCI posting on ide resets (me)
o Call the dma_check routine when trying to (me)
enable DMA via hdparm
o Add per driver abort handlers and use them (me)
o Forward port 8.0 ALi driver updates from (me)
Clear Zhang at ALi
Linux 2.4.21pre5-ac1
o Merge with 2.4.21pre5
o Do the final hatchet work on drive->id (me)
| IDE drive->id is now always valid so people
| can no longer get that one wrong.
o DRIVER(drive) in IDE != NULL always now (me)
| A dummy driver removes a ton of conditions
| and a load of bugs
o Move modem awareness into ac97_codec.c (me)
| Fixes CXT66 support I hope
o Minimal cmedia codec setup/bug stuff (me)
| Note these codecs dont yet support AC3 and also
| don't support volume control. May fix some sis7012
| laptop setups with luck.
o Fix mkdep bug causing devlist.h problem with (Pavel Roskin)
some versions of make
o Fix missing mtd Makefile entry (Adrian Bunk)
o APIC initialisation fix (Mikael Pettersson)
o CCISS update (Stephen Cameron)
o USB transport size handling fix (Alan Stern)
o Add AGP entry for the VIA EPIA (John Eckerdal)
o Add Laneed idents to pegasus usb ethernet (Go Taniguchi)
o Add HID workaround for OKI USB keyboard (Go Taniguchi)
o Add idents for MTT_TE MN128 USB ethernet (Go Taniguchi)
o Add USB quirks for another memorystick (Go Taniguchi)
o Some minor typo fixes to keep 2.4/2.5 easier (Steven Cole)
to diff
o Fix several operator and precdence problems (Norbert Kiesel)
o cciss error handling unregister fix (Herbert Xu)
o Kerneldoc for user access functions (Jon Foster)
o Further ALi IDE fixes (Ivan Kokshaysky)
o Improved 440GX bios workarounds (Arjan van de Ven)
| Thanks to the guys at Intel for hints on this
o AMD74xx cable detect fixes (Zoltan Hidvegi,
Vojtech Pavlik)
o io/irq in mpu401 must not be initdata (Daniel Ritz)
o Handle shared irq on pcmcia qlogicfas (Komuro)
Linux 2.4.21pre4-ac7
o Next chunk of DRM merge towards 4.3 codebase
o Fix ide-scsi deadlock on reset with SMP (me)
o Add some sun arrays to the scsi quirks list (Joel Buckley)
| They want multilun scanning always
o Fix skbuff abuse in atm lec (Chas Williams)
o Update the ips driver (Jack Hammer)
o Fix intelfb compile on SMP (Arjan van de Ven)
o One shot elevator contention fixing cache (Stephen Tweedie)
o Support swapoff from initrd (Stephen Tweedie)
o Add another transparent bridge quirk (Arjan van de Ven)
o ieee1394 sleep fixes (Arjan van de Ven)
o Use 0xff for cpu target (Arjan van de Ven)
o kmap leak fix for nfs symlink (Arjan van de Ven)
o Fix incorrect kernel/user address handling (me)
crash in swapoff (root only)
o kiovec accelerator (??)
o Export symbol needed by ipmi (Andreas Haumer)
o Add another 3c59x pci identifier (Daniel Kopko)
o Alpha build fix (Elliot Lee)
o Add new chips to e100 (Matt Wilson)
Linux 2.4.21pre4-ac6
o Update IPMI to v18 (Corey Minyard)
o More intel PIIX identifiers (Bill Nottingham)
o Update e100 for new identifiers (Jeff Garzik)
o Update Athlon SSE enabler (Dave Jones)
o Update auerswald USB isdn driver (Wolfgang)
o USB storage updates (Matthew Dharm)
o Add tripp idents to the pl2303 usb serial (John Moses)
o Add a new ftdi_sio ident (Philipp G?hring)
o Remove unused ohci driver field (Johannes Erdfelt)
o Fix EHCI abuse of SLAB_KERNEL in interrupt (Oliver Neukum)
o Fix dhcp on kaweth (Oliver Neukum)
o Fix some wrong idents in the pegasus driver (Petko Manolov)
o Fix ipaq name in usbnet (Carsten)
o USB macro cleanup (Joern Engel)
o Remove proc files in uhci that get stuck
o Remove wrong comment in ohci/uhci drivers (Johannes Erdfelt)
o Roland SC8820 USB midi support (Andrew Wood)
o Fix USB naming bug (Johannes Edrfelt)
o Add ontrack to the hid ignore list (Greg Kroah Hartmann)
o Add tangtop to the hid blacklist (Greg Kroah Hartmann)
o USB scanner updates (Henning Meier-Geinitz)
o Fix an oom handling bug in sis drm
o DRM updates for Radeon
| Flightgear now takes > 2hrs to hang on my R9000
o Fix various abusers of GFP_KERNEL in USB (Arjan van de Ven)
o Fix aic7xxx updates eaten by exclude file (Sergio Visinoni)
o Use check_gcc on crusoe (Stelian Pop)
o Update sonypi and meye drivers (Stelian Pop)
o Make input layer accept jogdial as valid (Stelian Pop)
o Intel i8xx framebuffer driver (David Dawes)
Linux 2.4.21pre4-ac5
o Fix the AMD ide bug() on boot up
o Pass device to outbsync so that we can whack (Ben Herrenschmidt)
the bridge on weird platforms
o Default sl82c05 second channel to PIO0 (Ben Herrenschmidt)
o EHCI speed up fixes (David Brownell)
o Assorted cpia fixes (Duncan Haldane)
o SSE enable for later Athlon (Daniel Egger)
o 3com 3c990 driver (David Dillow)
o Fix config syntax error in DRM config (Andrzej Krzysztofowicz)
o Update pci-skeleton to fix pad bug in example (me)
| Noted by Roger Luethi
o Supress popping when audio starts on via82cxxx (Jorg Schuler)
o Fix reiserfs direct I/O crash (Oleg Drokin)
o Allow cramfs initrd (Christoph Hellwig)
o Fix error path on dscc wan driver (me)
o Fix sign mishandling in epca driver (me)
o Fix sign mishandling in mwave driver (Oleg Drokin)
o Fix sign mishandling in mpt fusion (Oleg Drokin)
o Fix sign mishandling in aacraid (Oleg Drokin)
o Fix sign mishandling in tun (Oleg Drokin, me)
Linux 2.4.21pre4-ac4
o Attach a fake id struct to old/unprobed drives (me)
| Fixes a ton of special casing some of which was
| buggy.
o Fix incorrect sign handling in setup-pci noted (me)
by Oleg Drokin
o Fix bogon error returns from init_chipset noted (me)
by Oleg Drokin
| Fixes hpt366 crash on 66Mhz bus
o Fix mishandling of flash/disk combinations (me)
o Fix handling of /proc/ide/*/identify with (me)
no driver loaded (band aid for now)
o Fix IDE hang on rmmod and on poweroff (me)
o Fix IDE printk <6> bug (Henning Schmiedehausen)
o Radeon no longer needs AGPgart (James McClain)
o REPORTING-BUGS typo fix (Faik Uygur)
o ndelay() for PPC (Ben Herrenschmidt)
o PPC ioflush handling (Ben Herrenschmidt)
o PowerMac IDE updates (Ben Herrenschmidt)
o 8169 missing includes for Alpha build (Geoffrey Lee)
o Fix sisfb build on boxes with no MTRR (Geoffrey Lee)
o Fix cpqfc build on Alpha (Geoffrey Lee)
o Fix forte build on Alpha (Geoffrey Lee)
o Add eth_io_copy_and_sum for Alpha (Geoffrey Lee)
o Fix bogus semicolon in 8253xtty (Oleg Drokin)
o Fix incorrect if in megaraid driver (Oleg Drokin)
o Fix sign warning in radio_cadet driver found (me)
by Oleg Drokin
Linux 2.4.21pre4-ac3
o ALi FIFO setup channel fix (Al Viro)
| This needs careful testing. Treat -ac3 with a lot of care
| on ALi platforms and report how it goes
o Fix the dma waiting overflow (Ben Herrenschmidt)
o Fix ATAPI devices on VIA8235 (Vojtech Pavlik)
o Add ndelay for Alpha (Ivan kokshaysky)
o Give ndelay sensible argument names (Geert Uytterhoeven)
o Fix pcnet32 big endian filtering (Marcus Meissner)
o Fix ordering problem with PCI radeon causing (Chris Ison)
DRI hangs
o Fix C3 gcc compiler flags for newer gcc (Jeff Garzik)
o Replace nvidia and amd IDE drivers with new (Vojtech Pavlik)
driver
o Fix missing ; in aicasm_gram.y (Thibaut VARENE)
o NCR5380 trivial fix (Geert Uytterhoeven)
o Make constants in maxiradio static (Arnd Bergmann)
o Fix typos of 'available' (Alfredo Sanjuan)
o Fix wrong checks in bttv ioctl code (Alexandre Pereira Nunes)
o Fix i2c_ack cris extra ";"
o Fix JSIOCSBTNMAP extra ";"
o Fix VIDIOCGTUNER on w9966
o Fix amd8111e_read_regs
o Fix smctr_load_node_addr
o Fix sym53c8xxx extra ";"
o Fix sym53c8xxx_2 extra ";"
o Fix cs46xx download area clear
o Fix hysdn bootup error handling
o Fix mtd mount error checks
o Fix dpt_i2o reset error paths
o Fix a jffs error path handler
o Fix es1371 error path on register
o Fix sscape operator precedence
o Fix copy counting in vrc5477 audio
o Fix cdu31a oops with data cd (Mauricio Martinez)
o Fix ide taskfile if ";" errors (Oleg Drokin)
o Add 3com 3c460 to kaweth (Oliver Neukum)
o Kaweth length/dhcp fix (Oliver Neukum)
o ISD-200 requires IDE (Olaf Hering)
Linux 2.4.21pre4-ac2
o Turn on use of ide_execute_command everywhere (Ross Biro, me)
o First cut at settings locking for IDE (me)
o Add driver for CS5530 Kahlua audio (me)
o Fix wrong semicolons in system.h (Mikael Pettersson)
o Support root=nbd (Ben LaHaise)
o x86 byte order swapping optimisations (Andi Kleen)
o PMAC ide updates (Ben Herrenschmidt)
o Fix mishandling of nfsroot port= option (Eric Lammerts)
o Fix ALi audio on systems with > 2Gb RAM (Ivan Kokshaysky)
o Enable generic rtc on PPC boxes (Geert Uytterhoeven)
o Fix ide build with gcc 3.3 snapshot (Olaf Hering)
o Merge EHCI updates (qh state machine fix etc) (David Brownell)
o Fix radio-cadet SMP build (Adrian Bunk)
o Starfire updates (Ion Badulescu)
o Backport seq_file fix to 2.4 (Eric Sandeen)
o Fix ext3 crash deleting a single non sparse (Stephen Tweedie)
file exceeding 1Tb
Linux 2.4.21pre4-ac1
o Restore the mmap corner case fix (Raul)
o Add sendfile64 to 2.4.x (Christoph Hellwig)
o NLM garbage collection hang fix (Daniel Forrest)
o Enable kernel side pcigart for radeon (Michael Danzer)
| Requires recent XFree and ForcePCIMode
o Don't bash legacy floppy on x86_64 bootup (Mikael Petersson)
o Forward sony joygdial input to input layer (Stelian Pop)
o TCP session stall fix (Alexey Kuznetsov)
o Ian Nelson has moved (Ian Nelson)
o Add unplugged iops ready for hotplug IDE support(me)
o Add an OUTBSYNC iop for the IDE layer (Ben Herrenschmidt)
o Finish the ide_execute_command code (me)
o Switch ide-cd to ide_execute_command (me)
| Always good to test stuff on read only devices first 8)
o Fix IDE masking logic error (Ross Biro)
o Fix IDE mishandling of IRQ 0 devices (me)
o Fix printk levels on promise drivers (me)
o Clean up duplicate mmio ops/printk in siimage (me)
o Always set interrupt line with VIA northbridge (me)
| Should fix apic mode problems with USB/audio/net on VIA boards
o Add Diamond technology dt0893 codec (Thomas Davis)
o Add IBM 'Ruthless' platform string to summit
o Don't warn about IRQ when enabling a pure (me)
legacy mode IDE class device
o Clean up radio_cadet locking and other bugs (me)
o Fix jiffies mishandling in eata drivers (Tim Schmielau)
o Quieten confusing DMA disabled messages (Tomas Szepe)
o i830 DRM update port over (Arjan van de Ven)
Linux 2.4.21pre3-ac5
o Fix erratic oopsing on 2.4.21pre3-ac* (Hugh Dickins)
o Fix an incorrect check in raw.c (Artur Frycze)
o Fix highmem IDE DMA (Jens Axboe)
o Fix the size of the EDD area (Kevin Lawton)
o Remove incorrect ACPI blacklist entry (Pavel Machek)
o SCSI memory leak fix (Justin Gibbs)
o Fix mmap of vmalloc area in kmem giving wrong (Tony Dziedzic)
results
o Fix date in the microcode driver (Jonah Sherman)
o Fix incorrect smc9194 handling of skb_padto (David McCullough)
o Fix use of old check_regio function in umc8672 (William Stinson)
o Remove unused variable in sc1200 (Bob Miller)
o Perform ide_cs unregister in task context (Paul Mackerras)
| This doesn't fix all the bugs yet...
o Fix bugs in the gx power management code (Hiroshi Miura)
o Fix the sl82c105 driver for the new IDE code (Benjamin Herrenschmidt,
Russell King)
o Remove cacheflush debug printk (me)
o Fix IDE paths in docs for new layout (Karl-Heinz Eischer)
o Generic RTC driver backport (Geert Uytterhoeven)
o HDLC driver updates (Krzysztof Halasa)
o AMD8111 random number generator support (Andi Kleen)
o Fix crashes on e2100 driver (me)
Linux 2.4.21pre3-ac4
o Finish verifying PIIX/ICH drivers versus errata (me)
o Fix handling of DMA0 MWDMA on early ICH (me)
o Fix compile in kernel for Aurora SIO16 (Adrian Bunk)
o Clean up various Configure.help bits (Adrian Bunk)
o Disallow write combining on 450NX (me)
o Ensure rev C0 450NX has restreaming off (me)
o Don't do IDE DMA on rev B0 450NX or later (me)
450NX without BIOS workarounds for the hang
o Update Configure.help for HPT IDE (Adrian Bunk)
o Fix harmless code error in sb_mixer (Jeff Garzik)
o Fix ethernet padding on via-rhine (Roger Luethi)
o Add ndelay functionality for x86 (me)
| Based on Ross Biro's code
o Add ide_execute_command (me)
| Again based on Ross Biro's changed. Not yet used
| This will be the new correct way to kick off an
| IDE command from non IRQ context
o Matroxfb compile fix for one option combination (Petr Vandrovec)
Linux 2.4.21pre3-ac3
o Address comments on wcache value/issuing (me)
cache flush requests
o Update credits entry for Stelian Pop (Stelian Pop)
o Backport some sonypi improvements from 2.5 (Kunihiko IMAI)
o Fix pdcraid/silraid symbol clash (Arjan van de Ven)
o Fix ehci build with older gcc (Greg Kroah-Hartmann)
o Fix via 8233/5 hang (me)
o Fix non SMP cpufreq build (Eyal Lebidinsky)
o Fix sbp2 build with some config options (Eyal Lebidinsky)
o Fix ATM build bugs (Francois Romieu)
o Fix an ipc/sem.c race (Bernhard Kaindl)
o Fix toshiba keyboard double release (Unknown)
o CPUFreq updaes/fixes (Dominik Brodowski)
o Natsemi Geode/Cyrix MediaGX cpufreq support (Hiroshi Miura,
Zwane Mwaikambo)
o Add frequency table helpers to CPUfreq (Dominik Brodowski)
Linux 2.4.21pre3-ac2
o Fix the dumb bug in skb_pad (Dave Miller)
o Confirm some sparc bits are wrong and drop them (Dave Miller)
o Remove a wrong additional copyright comment (Dave Miller)
o Upgrade IPMI driver to v16 (Corey Minyard)
o Fix 3c523 compile (Francois Romieu)
o Handle newer rpm where -ta is rpmbuild not rpm (me)
o Driver for Aurora Sio16 PCI adapter series (Joachim Martillo)
(SIO8000P, 16000P, and CPCI)
| Initial merge
o Backport Hammer 32bit mtrr/nmi changes (Andi Kleen)
o Add the fast IRQ path to via 8233/5 audio (me)
Linux 2.4.21pre3-ac1
+ Handle battery quirk on the Vaio Z600-RE (Paul Mitcheson)
* EHCI USB updates (David Brownell)
+ IDE Raid support for AMI/SI 'Medley' IDE Raid (Arjan van de Ven)
+ NVIDIA nForce2 IDE PCI identifiers (Johannes Deisenhofer,
Tim Krieglstein)
* CPU bitmask truncation fix (Bjorn Helgaas)
o HP100 cleanup (Pavel Machek)
o Fix initial capslock handling on USB keyboard (Pete Zaitcev)
+ Update dscc4 driver for new wan (Francois Romieu)
+ Fix boot on Chaintech 4BEA/4BEA-R and (Alexander Achenbach)
Gigabyte 9EJL by handing wacky E820 memory
reporting
o SysKonnect driver updates (Mirko Lindner)
o Fix memory leak in n_hdlc (Paul Fulghum)
o Fix missing mtd dependancy (Herbert Xu)
+ Clean up ide-tape printk stuff (Pete Zaitcev)
+ IDE tape fixes (Pete Zaitcev)
o Fix size reporting of large disks in scsi (Andries Brouwer)
+ Fix excessive stack usage in NMI handlers (Mikael Pettersson)
+ Add support for Epson 785EPX USB printer pcmcia (Khalid Aziz)
* Quirk handler to sort out IDE compatibility (Ivan Kokshaysky)
mishandling
+ Model 1 is valid for PIV in MP table (Egenera)
+ Ethernet padding fixes for various drivers (me)
o Allow trident codec setup to time out (Ian Soboroff)
This can happen with non PM codecs
o Fix broken documentation link (Henning Meier-Geinitz)
o Update video4linux docbook (William Stimson)
o Correct kmalloc check in dpt_i2o (Pablo Menichini)
o Shrink kmap area to required space only (Manfred Spraul)
o Fix irq balancing (Ben LaHaise)
o CPUfreq updates (Dominik Brodowski)
o Fix typo in pmagb fb (John Bradford)
o EDD backport (Matt Domsch)
REMOVED FOR NOW
- RMAP
REMOVED FOR GOOD
- LLC (See 2.5)
- VaryIO (Never accepted mainstream)
--
"... and for $64000 question, could you get yourself vaguely familiar with
the notion of on-topic posting?"
-- Al Viro
Hi,
Since 2.4.21-ac2 i'm experiencing problems with Postfix on Debian
Stable. Messages get corrupted while being handled by Postfix.
Vanilla 2.4.21 and 2.4.21-ac1 are fine, but 2.4.21-ac2/3 causes
problems. Going back to ac1 resolves the issue. I tried kernels on
several Debian servers, but all have the same problem. Could be it's
something in postfix that emerges with this specific patch, but it's at
least curious. I'm not too familiar with this matter unfortunately.
The only logentries i see are:
postfix/qmgr[399]: warning: active/0/4/04E5B17E3F: too many length bits,
record type 255
postfix/qmgr[399]: warning: 04E5B17E3F: envelope records out of order
postfix/qmgr[399]: warning: saving corrupt file "04E5B17E3F" from queue
"active" to queue "corrupt"
...or just...
postfix/smtp[536]: warning: corrupted queue file: active/C/7/C7AC317E3F
Tested on: Intel PPro, Intel Celeron, AMD Duron
Tested on: ext2 and ext3
All other programms seem to work fine, no other strange messages
whatsoever...
Regards,
--Thijs Welman
Delft University of Technology, the Netherlands
In article <[email protected]>,
Thijs <[email protected]> writes:
T> Hi,
T>
T> Since 2.4.21-ac2 i'm experiencing problems with Postfix on Debian
T> Stable. Messages get corrupted while being handled by Postfix.
T>
T> Vanilla 2.4.21 and 2.4.21-ac1 are fine, but 2.4.21-ac2/3 causes
T> problems. Going back to ac1 resolves the issue. I tried kernels on
T> several Debian servers, but all have the same problem. Could be it's
T> something in postfix that emerges with this specific patch, but it's at
T> least curious. I'm not too familiar with this matter unfortunately.
Same problem here observed with -ac3 (reiserfs).
Thanks for the warning.
> Not yet resynchronized with Marcelo 2.4.22-pre1.
I've had issues with 2.4.22-pre1 & 2.4.21-ac1 with PCMCIA & my orinoco
based wireless card (SMC 2632W) - as soon as I insert the card the
machine freezes. Likewise with 2.4.21 normal. However 2.4.21-ac3 seems
to have fixed this.
Compiling with gcc 3.3; I thought this might be the problem and dropped
back to gcc 2.95, but 2.4.22-pre1 still hung.
Laptop is a Compaq N200; yenta CardBus driver compiled into the kernel.
pcmcia-cs 3.1.33; the machine will boot without the wireless card
inserted but hangs as soon as it is, or if inserted while booting hangs
when cardmgr starts.
Configs at:
http://the.earth.li/~noodles/config-2.4.21-ac3
http://the.earth.li/~noodles/config-2.4.22-pre1
J.
--
Ok ramblers, let's get rambling.
Thijs on Wed 25/06 22:29 +0200:
> Since 2.4.21-ac2 i'm experiencing problems with Postfix on
> Debian Stable. Messages get corrupted while being handled
> by Postfix.
I thought I was crazy, I'm seeing this also. I had to
switch my kernel back on our mail hub because postfix kept
doing this every time I boot the new kernel. I tried to
figure out why but could not; the system *appears* to run
normally otherwise. I was moving from 2.4.21-pre4-ac1
+gibbs-aic7xxx patches from that time, to 2.4.21-ac2
+gibbs-aic7xxx from 20030603, and I thought that might be it
(the AIC patches); are you running AIC adapters on your
system with the gibbs patches by any chance?
here's what I'm seeing on my mail hub getting a test message
from one of the exchangers:
Jun 24 05:07:54 lujuria postfix/smtpd[997]: connect from mx1.mydomain.tld[1.2.3.4]
Jun 24 05:07:54 lujuria postfix/nqmgr[983]: 7017833812: from=<[email protected]>, size=347, nrcpt=1 (queue active)
Jun 24 05:07:54 lujuria postfix/nqmgr[983]: warning: active/7/0/7017833812: too many length bits, record type 255
Jun 24 05:07:54 lujuria postfix/nqmgr[983]: warning: 7017833812: envelope records out of order
Jun 24 05:07:54 lujuria postfix/nqmgr[983]: warning: saving corrupt file "7017833812" from queue "active" to queue "corrupt"
Jun 24 05:07:54 lujuria postfix/smtpd[997]: A6CF3339AA: client=mx1.mydomain.tld[1.2.3.4]
Jun 24 05:07:54 lujuria postfix/cleanup[991]: A6CF3339AA: message-id=<[email protected]>
Jun 24 05:07:54 lujuria postfix/smtpd[997]: disconnect from mx1.mydomain.tld[1.2.3.4]
this is system is RH8+updates with the mentioned kernels.
Very bizarre.
> All other programms seem to work fine, no other strange
> messages whatsoever...
yeah that's the weird part that had me scratching my head as
well. The other poster said it happens for him on reiserfs,
and I'm on ext3, so that would seem to rule out
filesystems...
hmm looking at -ac2 log, looks like the following might be
possible:
- Fix AF_UNIX dgram select problem
+ postfix uses lots of these, this sounds like the
most likely candidate for sure
- Revert problematic scsi change (fixes ppa/imm)
+ does it touch generic code?
- Fix data direction for start stop scsi command
+ have no idea what this is
- Add more unexpected apic filtering
+ I'm on SMP...don't know if this is related??
I don't see how it could be any of the other changes.
In article <[email protected]>,
Scott McDermott <[email protected]> writes:
SM> Thijs on Wed 25/06 22:29 +0200:
>> Since 2.4.21-ac2 i'm experiencing problems with Postfix on
>> Debian Stable. Messages get corrupted while being handled
>> by Postfix.
SM>
SM> I thought I was crazy, I'm seeing this also. I had to
SM> switch my kernel back on our mail hub because postfix kept
SM> doing this every time I boot the new kernel. I tried to
SM> figure out why but could not; the system *appears* to run
SM> normally otherwise. I was moving from 2.4.21-pre4-ac1
SM> +gibbs-aic7xxx patches from that time, to 2.4.21-ac2
SM> +gibbs-aic7xxx from 20030603, and I thought that might be it
SM> (the AIC patches); are you running AIC adapters on your
SM> system with the gibbs patches by any chance?
SM>
No AIC or any kind here. Bring on the next suspect.
Jonathan Hudson on Wed 25/06 23:34 +0100:
> No AIC or any kind here. Bring on the next suspect.
does changing the unix_dgram_ops `poll' op from `dgram_poll'
to `datagram_poll' in net/unix/af_unix.c change anything for
you? I can't test this myself until later this week. Also I
don't know what other bug the unix_peer_get() addition is
supposed to fix, so...
On Wed, 25 Jun 2003, Thijs wrote:
> Since 2.4.21-ac2 i'm experiencing problems with Postfix on Debian
> Stable. Messages get corrupted while being handled by Postfix.
>
> Vanilla 2.4.21 and 2.4.21-ac1 are fine, but 2.4.21-ac2/3 causes
> problems. Going back to ac1 resolves the issue. I tried kernels on
> several Debian servers, but all have the same problem. Could be it's
> something in postfix that emerges with this specific patch, but it's at
> least curious. I'm not too familiar with this matter unfortunately.
>
> The only logentries i see are:
>
> postfix/qmgr[399]: warning: active/0/4/04E5B17E3F: too many length bits,
> record type 255
> postfix/qmgr[399]: warning: 04E5B17E3F: envelope records out of order
> postfix/qmgr[399]: warning: saving corrupt file "04E5B17E3F" from queue
> "active" to queue "corrupt"
> ...or just...
> postfix/smtp[536]: warning: corrupted queue file: active/C/7/C7AC317E3F
>
> Tested on: Intel PPro, Intel Celeron, AMD Duron
> Tested on: ext2 and ext3
I can confirm this problem. I use postfix 2.0.12 on a Debian woody
installation, all mails get corrupted. Postfix queue in on reiserfs, so
I think the bug is filesystem independent.
Geller Sandor <[email protected]>
On Thu, 2003-06-26 at 00:19, Scott McDermott wrote:
> Jonathan Hudson on Wed 25/06 23:34 +0100:
> > No AIC or any kind here. Bring on the next suspect.
>
> does changing the unix_dgram_ops `poll' op from `dgram_poll'
> to `datagram_poll' in net/unix/af_unix.c change anything for
> you? I can't test this myself until later this week. Also I
> don't know what other bug the unix_peer_get() addition is
> supposed to fix, so...
I'll see if I can test this out, once I've build -ac3 on a box where I
don't drop real messages.
-jonathan
Scott McDermott wrote:
>
> does changing the unix_dgram_ops `poll' op from `dgram_poll'
> to `datagram_poll' in net/unix/af_unix.c change anything for
> you? I can't test this myself until later this week. Also I
> don't know what other bug the unix_peer_get() addition is
> supposed to fix, so...
Tried it, but doesn't fix it, unfortunately...
guess it's time for the next suspect ;)
regards,
--Thijs Welman
On Wed, 2003-06-25 at 19:36, Alan Cox wrote:
> Linux 2.4.1-ac3
> o Fix an hpt driver bug triggered by the new HPT (me)
> BIOS
> o Initial VIA and S3 DRM modules merges (VIA)
> | These are marked up with some warnings and need
> | a chunk of clean up work yet.
> o btaudio update (Gerd Knorr)
> o Backport 2.5 ipc/sem.c fix (Manfred Spraul)
> o Fix scsi_register failure path for aacraid (Michael Still)
> o First crack at fixing the ide reset oopses (me)
> o Fix the incompatibility between via audio and (me)
> esd/gnome desktops
Looks like ICH5-SATA IDE was also disabled in favor of a SCSI ATA
driver. Caused me a momentary headache before I figured out why my disk
drives went away. :)
MikaL
Alan,
W.R.T the VIA CLE266 DRM merge, do you have any kernel side MPEG2 decoder
bits to add in (along with the V4L part presumably)? And do you have any
information about the user-space library (libddmpeg.so????) that would be
necessary to go along with?
I read when you wrote "The CLE266 has some fairly complicated Linux kernel
entanglements to resolve for the hardware mpeg engine stuff," but didn't
know how to interpret.
Thanks,
David
--
/==============================\
| David Mansfield |
| [email protected] |
\==============================/