Linux 2.4.19-pre10-ac2

Linux 2.4.19-pre10-ac2

Post by Alan Co » Fri, 07 Jun 2002 03:20:23



[+ indicates stuff that went to Marcelo, o stuff that has not,
 * indicates stuff that is merged in mainstream now, X stuff that proved
   bad and was dropped out]

This should hopefully fix up a few of the outstanding buglets, especially
for Promise IDE. It also merges speakup - support for blind console use.
The speakup code would benefit from a chunk of kernel janitoring I think.

Linux 2.4.19pre10-ac2
o       Merge speakup support for blind users
o       CSB6 cable detect for Dell                      (Matt Domsch)
o       Update pci ids for Intel i8xx                   (Wim Van Sebroeck)
o       Add AMD766 PCI irq router support               (Wayne Whitney)
o       ACARD scsi update                               (Matthew Chang)
o       Fix idle-period bug in APM parser               (Stephen Rothwell)
o       Printk levels for 3c501 ethernet                (Felipe Damasio)
o       AMD768 TCO watchdog driver - * needs testing *  (Zwane Mwaikambo)
o       Fix IDE port offset for pdc202xx                (Hang Yang)
        | should fix LBA48 drives on primary channel
o       Fix incorrect speedstep multiplier detect       (Dominik Brodowski)
o       Add support for Aptiva with Bose subwoofer      (Toshio Spoor,
                                                         John Rood)
o       Autodetect SiS 745 AGP                          (Carsten Rietzschel)
o       More scsi sparselun entries                     (Arjan van de Ven)
o       Fix possible crash on shutdown with AF_ROSE     (Jean-Paul Roubelat)
o       Intel 845G IDE support                          (Andre Hedrick)
o       Further CPiA driver updates                     (Duncan Haldane)
o       Fix DAC960 diff that went astray                (Juan Quintela)
o       Add HP arrays to the sparselun list             (Andrew Patterson)

Linux 2.4.19pre10-ac1
o       Merge with Linux 2.4.19-pre10

Linux 2.4.19pre9-ac3
o       Cpufreq updates                 (Dominik Brodowski, Dave Jones0
        | Now includes some reverse engineered speedstep support
o       JFS updates                     (David Kleikamp, Christoph Hellwig)
o       CPiA updates/Intel microscope support           (Duncan Haldane)
o       Fix vm86 locking errors on SMP                  (Ben LaHaise)
o       Remove dead vm86mode field                      (Ben LaHaise)
o       Fix make clean for cl2llc                       (Keith Owens)
o       Fix loop errors with highmem                    (Ben LaHaise)
*       Fix ipc/sem.c SuS/LSB compliance                (Christopher Yeoh)
o       Update swsuspend maintainer info                (Pavel Machek)
o       Add another drive quirk for the promise         (Hank Yang)
        drivers
o       Merge external journal support for jfs          (David Kleikamp)
o       Add documentation about O(1) scheduler          (Robert Love)
o       O(1) scheduler tidy ups                         (Robert Love)
o       Fix remaining extern inline users               (Christoph Hellwig)
o       Cache alignment cleanups for SMP apic timers    (Ravikiran Thirumalai)
o       Ext3 file system updates                        (Stephen Tweedie)
o       Fix 'dump corrupts live fs bug'                 (Stephen Tweedie)
o       Add DAC960 devices to init table                (Oliver Pitzeier)
        | Lilo doesn't care but grub does ..

Linux 2.4.19pre9-ac2
o       Clean up after SIGURG properly                  (David Weinehall)
        | Needed to match the other SuS compliance fix for it
+       Fixed wrong elf section in neofb                (Thomas Mirlacher,
                                                         Andrey Panin)
o       Don't write to reserved bits on 815 gart        (Nicolas Aspert)
o       Make fcntl locking POSIX 2001 compliant         (Andries Brouwer)
+       Fix an mmap corner case                         (Ral)
o       Merge 3c59x vlan support                        (Paul Komkoff)
+       Update URLS for LDP documentation               (John Kacur)
o       Fix rmem setting for low memory                 (J A Magallon)
o       Reparent scsi error thread to init              (J A Magallon)
o       Backport FPU init fixes                         (J A Magallon)
+       Fix AGPgart crash on I830M/I845G when using
        8Mb/8Mb split                                   (Jeff Hartmann)
*       Fix phy masking on 8139too                      (Jeff Garzik)
*       Fix link state reporting on generic phy code    (Jeff Garzik)
*       Tulip phy handling fix                          (Jeff Garzik)
*       Update 8139too docs                             (Jeff Garzik)
*       cs89x0 update                                   (Jeff Garzik)
*       VIA rhine fixes                                 (Jeff Garzik)
+       Hamachi quick fixup for 2.4.19                  (Keith Underwood)
o       Revert escaped procfs debug code                (Todd Eigenschink)
*       Merge the 2.5 additions to ethtool              (Jeff Garzik)
*       Update dl2k driver                              (Jeff Garzik)
*       Fix kernel api docs to reflect fb changes       (Juan Quintela)
*       Fix problems with pcnet32 workaround for x250   (Go Taniguchi)
*       De4x5 cleanups                                  (Jeff Garzik)

Linux 2.4.19pre9-ac1
o       Merge with 2.4.19pre9
o       Fix SuS violation on readv/writev               (me)
        | I believe this one is correct, please double check

Linux 2.4.19pre8-ac5
*       Fix various audio copy*user                     (Rusty Russell)
o       Update to rmap 13               (Rik van Riel, Christoph Hellwig)
*       Fix joystick copy_user bugs                     (Robert Johnson)
*       Document the i2o_pci module                     (me)
*       Switch i2o_block back to direct pointers        (me)
        to avoid promise firmware bugs
*       Remove cache error paths from i2o_block         (me)
        | new code doesnt trip that bug
*       Reduce the i2o queue depth per device           (me)
        | pending tuning - might need more yet
*       Set i2o default limit at 48K a write            (me)
        | more firmware bug stuff
*       Clean up i2o cache strategy, add tuning ioctl   (me)
*       Allow users to force dpt cards to use base i2o  (me)
        | tested i2o_block on DPT with my cards
*       Remove duplicate ac97_codec inclusion           (Keith Owens)
o       Tidy up patch for swsuspend                     (Pavel Machek)
+       Fix wrong __init in 3c509                       (Kasper Dupont)
o       Fix mm/bootmem.c build on cris                  (Johan Adolfsson)
*       Remove config tools for 8253x from kernel tree  (Keith Owens)
*       Rename files in aacraid ready for merge         (me)
        of updates
*       Merge bridge specific changes in aac code       (Deanna Bonds)
*       Merge most of the fixups/cleanups for aacraid   (Deanna Bonds)
o       Set PCI masks for the 64 and 32bit aacraids     (me)
*       Don't program up the ali secondary codec for    (me)
        6 channel if you don't have one fitted
*       Block layer copy*user fixups            (Arnaldo Carvalho de Melo)
*       Fix missing intermezzo include          (Marc-Christian Petersen)
o       Slab cache for iobufs           (Andrea Arcangeli, Chuck Lever,
                                                Christoph Hellwig)
*       Fix intermezzo copy*user                (Arnaldo Carvalho de Melo)
o       down_trylock                                    (Christoph Hellwig)
*       Fix video compile for split module              (Michal Jaegermann)
        and compiled in
o       Kill 3c59x debug bits                           (Andrew Morton)
*       Char fixes for copy*user                (Arnaldo Carvalho de Melo)
*       Fix a few errors in the janitor copy* fixes     (me)

Linux 2.4.19pre8-ac4
o       Fix warnings in pc_keyb.c                       (Christoph Hellwig)
*       Fix undefined C in rivafb                       (Christoph Hellwig)
*       Fix dnotify warnings                            (Christoph Hellwig)
o       Remove unused nfs label                         (Christoph Hellwig)
o       Fix vm_validate_enough prototypes               (Christoph Hellwig)
*       Fix wrong comment in agpgart                    (Nicolas Aspert)
*       JFFS2 fixes                                     (David Woodhouse)
o       Hopefully fix zisofs breakage                   (David Woodhouse)
*       Remove a defunct soc_probe call                 (Christoph Hellwig)
o       Update initrd documentation                     (Mark Post)
o       Fix SMP build                                   (Robert Love)
o       Numa-Q apic timer update                        (Martin Bligh)

Linux 2.4.19pre8-ac3
o       Kbuild fixes                                    (Keith Owens)
o       Fix eepro100 bug/typo                           (Michael Rozhavsky)
*       Intel 845G GART support                         (Graeme Fisher)
*       Fix tasklet disable/kill in pppoatm             (Luca Barbier)
*       Add another PCI ident to the acenic driver      (Eric Smith)
o       Major IDE updates                               (Andre Hedrick)

Linux 2.4.19pre8-ac2
*       Fix more compile problems                       (me)
*       Fix a possible hang on shutdown in 3270 tty     (Martin Schwidefsky)
*       Make "make rpm" sane for non x86              (Cesar Cardoso)
*       Two new AC97 codec entries                      (Lei Hu)
*       Thread exit race fix                            (Dave McCracken)
*       Further sg buffer clearing fix                  (Douglas Gilbert)
*       Fix do_mounts printk                            (Al Viro)
*       Umembp fixups                                   (Neil Brown)
*       Umembp shift bug fixup                          (me)
o       Kbuild fixes and improvements                   (Keith Owens)
*       Add a new tulip clone pci ident entry           (Ohta Kyuma)
*       Fix url on via pci fixups                       (Erich Schubert)
*       koi8-ru handling fixes                          (Petr Vandrovec)
o       Clean up remaining code to use yield            (Robert Love)
o       Clean up migration_init as per 2.5              (Erich Focht)
o       Clean up maximum real time priorities           (Robert Love)
*       Kill unused variable in bpck6                   (Adrian Bunk)
*       Fix dnotify/process exit handling               (Stephen Rothwell)
*       Add another vaio bios to the table              (Yves Lafon)
*       Allow users to disable hyperthreading           (Hugh Dickins)

Linux 2.4.19pre8-ac1
o       Merge with Linux 2.4.19pre8
        -       Fix some compile problems

Linux 2.4.19pre7-ac4
*       Test AMD768 IRQ router support                  (me)
o       Fix ext2 build error
o       Improve i810 audio documentation                (Johannes Feigl)
o       Ensure UTS data is in C locale                  (Martin Dalecki)
*       Add the Intel ICH4 to the i810 audio driver     (Wang Jun)
*       Fix qlogicfc crash under load                   (Dave Miller)
*       Fix snprintf return values in some cases        (Ben LaHaise)
*       Fix a bug that got into the iph5526 code when   (Vineet Abraham)
        networking
*       Add more scanners that respond to all LUNs      (Frank Zago)
+       Synclink PCMCIA wan driver                      (Paul Fulghum)
o       Fix sparc64/ppc64 bluetooth ioctl build         (Martin Eriksson)
*       Change 5/6bit codec resolution detect for       (Wan Tat Chee)
        AC97
o       Fix v4l compile bug in one option case          (Iain Stevenson)
o       Clean up powernow initcalls                     ("CaT")
o       Add PIO mode support for the Pacific Digital    (Mark Lord)
        ADMA-100i card

Linux 2.4.19pre7-ac3
o       Back merge some documentation fixes             (Daniel Dickman)
*       Update sisfb driver                             (Thomas Winischhofer)
o       Remove sync wakeups now O(1) handles it         (Robert Love)
o       Abstract away need_resched                      (Robert Love)
o       Fix scheduler deadlock during switch_mm         (Dave Miller)
        on sparc etc
o       Optimise sched_yield                            (Robert Love)
o       Handle tasks becoming runnable during           (Robert Love)
        schedule
o       Clean up assumptions about MAX_RT_PRIO          (Robert Love)
o       Backport of migration fixes/irq off             (Robert Love
        fixes and migration_init                         William Irwin)
o       Cleanups from 2.5->2.4 O(1) backport         (Robert Love)
        | The entire O(1) block above is a backport
        | of all the fixes from Ingo, Robert and others
o       Swsuspend fix crash on boot add cleanups        (Pavel Machek)
*       Scsi generic buffer tidy up                     (Douglas Gilbert)
*       Correct kd.h definitions                        (Andrej Lajovic)
o       Fix missing include for swsuspend               (Mauricio Zambrano)
o       Configure.help typo fixes               (Arnaldo Carvalho de Melo)
o       Identify PIV Xeon in mptable                    (James Bourne)
o       Fix "skip_ioapc_setup" compile problem        (Mikael Pettersson)
o       Additional ext2/ext3 sanity checker             (Andreas Dilger)
*      
...

read more »

 
 
 

Linux 2.4.19-pre10-ac2

Post by Arnaldo Carvalho de Mel » Fri, 07 Jun 2002 03:20:26


Em Wed, Jun 05, 2002 at 02:04:48PM -0400, Alan Cox escreveu:

Quote:> The speakup code would benefit from a chunk of kernel janitoring I think.

me downloading... I have to have something to do on the taxy on my way to the
office... 8)

- Arnaldo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Tomas Szep » Fri, 07 Jun 2002 03:30:16


Quote:> o  Fix IDE port offset for pdc202xx                (Hang Yang)
>    | should fix LBA48 drives on primary channel

Btw, the big IDE update seems to have written off the ServerWorks
driver -- DMA mode on OSB4 cannot be enabled no matter what.

Tomas
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by dean gaude » Fri, 07 Jun 2002 04:20:10


so i haven't had a chance to dig into this further, but i think there may
be some .PRECIOUS foo missing.  i had hit ^C a few times to cancel out a
"make -j3 dep", and a "make -j3 bzImage" while i tweaked other things...
and somehow in the process include/linux/sunrpc/svcsock.h disappeared.

i notice two files where svcsock.h appears on the LHS of a make rule --
.hdepend and tmp_include_depends.  .hdepend has a .PRECIOUS entry, but
tmp_include_depends doesn't.

maybe one of those files wasn't generated properly when i hit ^C during
the "make -j3 dep"?  after the ^C i re-configured and re-did make dep...
but i recall after the make menuconfig it didn't tell me i needed to do
the make dep... and i know it didn't finish the first make dep.

anyhow, sorry i haven't tried to reproduce it, just thought i'd mention it
though.

-dean

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Stephen Rothwel » Fri, 07 Jun 2002 10:30:06



Quote:

> o  Fix idle-period bug in APM parser               (Stephen Rothwell)


the patch on.

--
Cheers,

http://www.canb.auug.org.au/~sfr/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Zwane Mwaikamb » Fri, 07 Jun 2002 15:50:05



Quote:> Em Wed, Jun 05, 2002 at 02:04:48PM -0400, Alan Cox escreveu:
> > The speakup code would benefit from a chunk of kernel janitoring I think.

> me downloading... I have to have something to do on the taxy on my way to the
> office... 8)

Thats either a damn long drive, or one helluva fast laptop ;)

--
http://function.linuxpower.ca

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Matt Bernstei » Fri, 07 Jun 2002 19:20:07


Since when was it OK to do a parallel make dep?


Quote:>so i haven't had a chance to dig into this further, but i think there may
>be some .PRECIOUS foo missing.  i had hit ^C a few times to cancel out a
>"make -j3 dep", and a "make -j3 bzImage" while i tweaked other things...
>and somehow in the process include/linux/sunrpc/svcsock.h disappeared.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
 
 
 

Linux 2.4.19-pre10-ac2

Post by David S. Mille » Fri, 07 Jun 2002 19:20:08



   Date: Thu, 6 Jun 2002 11:11:09 +0100 (BST)

   Since when was it OK to do a parallel make dep?

I've been doing that successfully for a long time now.
It has always worked as far as I can remember..
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Keith Owen » Fri, 07 Jun 2002 22:00:12


On Thu, 6 Jun 2002 11:11:09 +0100 (BST),


>Since when was it OK to do a parallel make dep?

Arch dependent.  Parallel make dep will generate incomplete output on
some architectures, mainly those that generate files at make dep time.
mkdep.c only adds .h files to .[h]depend if the file exists.  With
parallel make dep the scanning of .c files can occur before the .h
files have been generated, resulting in an incomplete dependency tree.
Later changes may not rebuild everything that should be rebuilt.

Not a problem for kbuild 2.5 of course.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Arnaldo Carvalho de Mel » Sat, 08 Jun 2002 00:30:12


Em Thu, Jun 06, 2002 at 08:09:05AM +0200, Zwane Mwaikambo escreveu:


> > Em Wed, Jun 05, 2002 at 02:04:48PM -0400, Alan Cox escreveu:
> > > The speakup code would benefit from a chunk of kernel janitoring I think.

> > me downloading... I have to have something to do on the taxy on my way to the
> > office... 8)

> Thats either a damn long drive, or one helluva fast laptop ;)

or the statement was a joke 8) Anyway, here is the first cleanup, the laptop
is not that fast...

Alan is it important to keep all the ugly #ifdefs for 2.2 compatibility?

- Arnaldo

--- linux-2.4.19-pre10-ac2/drivers/char/speakup/speakup.c       2002-06-05 21:20:11.000000000 -0300

 static int errno;
 char *spk_cfg[] = { DEFAULT_SPKUP_VARS };
-long spk_cfg_map = 0;          /* which ones have been re'alloc'ed */
-int synth_file_inuse = 0;
+long spk_cfg_map;              /* which ones have been re'alloc'ed */
+int synth_file_inuse;
 static struct spk_variable spk_vars[] = { SPKUP_VARS };
-static unsigned char pitch_shift = 0;
+static unsigned char pitch_shift;
 char saved_punc_level = 0x30;
-char mark_cut_flag = 0;
-unsigned short mark_x = 0;
-unsigned short mark_y = 0;
+char mark_cut_flag;
+unsigned short mark_x;
+unsigned short mark_y;
 static char synth_name[10] = CONFIG_SPEAKUP_DEFAULT;
 static struct spk_synth *synths[] = {

        "white space"
 };

-int spk_keydown = 0;
-int bell_pos = 0;
-static int spk_lastkey = 0;
+int spk_keydown;
+int bell_pos;
+static int spk_lastkey;

 struct spk_t *speakup_console[MAX_NR_CONSOLES];

 speakup_diacr (unsigned char ch, unsigned int currcons)
 {
        static unsigned char *buf = "\0\0\0\0\0";
-       static int num = 0;
+       static int num;
        int tmp = 0;


 int
 spkup_write (const char *buf, int count)
 {
-       static unsigned short rep_count = 0;
+       static unsigned short rep_count;
        static char old_ch, oldest_ch, count_buf[30], punc_buf[128];
        int in_count = count;

        return 0;
 }

-#if LINUX_VERSION_CODE >= 0x20300
 static struct file_operations synth_fops = {
-       read:speakup_file_read,
-       write:speakup_file_write,
-       ioctl:speakup_file_ioctl,
-       open:speakup_file_open,
-       release:speakup_file_release,
+       .owner   = THIS_MODULE;
+       .read    = speakup_file_read,
+       .write   = speakup_file_write,
+       .ioctl   = speakup_file_ioctl,
+       .open    = speakup_file_open,
+       .release = speakup_file_release,
 };
-#else
-static struct file_operations synth_fops = {
-       NULL,                   /* seek */
-       speakup_file_read,
-       speakup_file_write,
-       NULL,                   /* readdir */
-       NULL,                   /* poll */
-       speakup_file_ioctl,
-       NULL,                   /* mmap */
-       speakup_file_open,
-       NULL,                   /* flush */
-       speakup_file_release,
-       NULL,
-       NULL,                   /* fasync */
-};
-#endif

 void

        return ((count < begin + len - off) ? count : begin + len - off);
 }

-static volatile int chars_timer_active = 0;    // indicates when a timer is set
+static volatile int chars_timer_active;        // indicates when a timer is set
 #if (LINUX_VERSION_CODE < 0x20300)  /* is it a 2.2.x kernel? */
-static struct wait_queue *chars_sleeping_list = NULL;
+static struct wait_queue *chars_sleeping_list;
 #else                          /* nope it's 2.3.x */
 static DECLARE_WAIT_QUEUE_HEAD (chars_sleeping_list);

                               unsigned long count, void *data)
 {
        static const int max_description_len = 72;
-       static int cnt = 0, num = 0, state = 0;
+       static int cnt, num, state;
        static char desc[max_description_len + 1];
-       static unsigned long jiff_last = 0;
+       static unsigned long jiff_last;
        int i;
        char ch, *s1, *s2;

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Arnaldo Carvalho de Mel » Sat, 08 Jun 2002 00:30:14


Em Thu, Jun 06, 2002 at 03:14:17AM -0700, David S. Miller escreveu:


>    Date: Thu, 6 Jun 2002 11:11:09 +0100 (BST)

>    Since when was it OK to do a parallel make dep?

> I've been doing that successfully for a long time now.
> It has always worked as far as I can remember..

/me too, doing that for ages without a single problem.

- Arnaldo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by dean gaude » Sat, 08 Jun 2002 02:50:13



> Since when was it OK to do a parallel make dep?

wow such a useful response.  can you point me where this is documented?

i've never seen "make -j3" cause *source files* to be deleted.  there's a
bug here somewhere.

-dean


> >so i haven't had a chance to dig into this further, but i think there may
> >be some .PRECIOUS foo missing.  i had hit ^C a few times to cancel out a
> >"make -j3 dep", and a "make -j3 bzImage" while i tweaked other things...
> >and somehow in the process include/linux/sunrpc/svcsock.h disappeared.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
 
 
 

Linux 2.4.19-pre10-ac2

Post by dean gaude » Sat, 08 Jun 2002 03:40:13



> On Thu, 6 Jun 2002 11:11:09 +0100 (BST),

> >Since when was it OK to do a parallel make dep?

> Arch dependent.  Parallel make dep will generate incomplete output on
> some architectures, mainly those that generate files at make dep time.
> mkdep.c only adds .h files to .[h]depend if the file exists.  With
> parallel make dep the scanning of .c files can occur before the .h
> files have been generated, resulting in an incomplete dependency tree.
> Later changes may not rebuild everything that should be rebuilt.

hmm, so i can imagine there being a race condition there somewhere... but
i just tried comparing a few "make -j3 dep" trees versus a "make dep" tree
and there weren't any differences.  (i'm building for i386).

anyhow, i think i see the problem.

in the top level Makefile there's this:

        tmp_include_depends: include/config/MARKER dummy
                $(MAKE) -r -f tmp_include_depends all

and if you look in tmp_include_depends it has all the header files on the
LHS of dependencies (it touches them all)... but there's no .PRECIOUS for
any of the header files.  so if you hit ^C while this is going on you'll
lose source files.

i believe the patch below fixes the problem... but i couldn't hit ^C at
the right time to reproduce it, so maybe there's something else going on.

btw, it doesn't appear like anything actually uses .hdepend... at least i
couldn't see a use when i grepped the tree for '\.hdepend', maybe it's
hidden somewhere.

-dean

--- linux/Makefile.orig Thu Jun  6 10:44:29 2002

 endif
        (find $(TOPDIR) \( -name .depend -o -name .hdepend \) -print | xargs $(AWK) -f scripts/include_deps) > tmp_include_depends
        sed -ne 's/^\([^ ].*\):.*/  \1 \\/p' tmp_include_depends > tmp_include_depends_1
+       (echo ""; echo ".PRECIOUS: \\"; cat tmp_include_depends_1; echo "") >> tmp_include_depends
        (echo ""; echo "all: \\"; cat tmp_include_depends_1; echo "") >> tmp_include_depends
        rm tmp_include_depends_1

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Heinz Dieh » Sat, 08 Jun 2002 06:10:10



Quote:> Linux 2.4.19pre10-ac2

[....]

After updating to pre10-ac2, the second harddisk gives the following error
output in dmesg:

      VP_IDE: not 100% native mode: will probe irqs later
      VP_IDE: VIA vt82c586b (rev 41) IDE UDMA33 controller on pci00:07.1
      ide0: BM-DMA at 0xe000-0xe007, BIOS settings: hda:DMA, hdb:DMA
      ide1: BM-DMA at 0xe008-0xe00f, BIOS settings: hdc:DMA, hdd:DMA
      hda: IBM-DHEA-36481, ATA DISK drive
      hdb: Conner Peripherals 1275MB - CFS1275A, ATA DISK drive
      hdc: CD-540E, ATAPI CD/DVD-ROM drive
      hdd: CD-W54E, ATAPI CD/DVD-ROM drive
      ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
      ide1 at 0x170-0x177,0x376 on irq 15
      hda: 12692736 sectors (6499 MB) w/472KiB Cache, CHS=790/255/63, UDMA(33)
      hdb: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
      hdb: task_no_data_intr: error=0x04 { DriveStatusError }
      hdb: 2496876 sectors (1278 MB) w/64KiB Cache, CHS=619/64/63, DMA
      Partition check:
      [....]

The drive itself works without any problems and is not defective.
All partitions are ext3. Here are the IDE related options from .config:

CONFIG_IDE=y
CONFIG_BLK_DEV_IDE=y
CONFIG_BLK_DEV_IDEDISK=y
CONFIG_BLK_DEV_IDESCSI=y
CONFIG_IDE_TASK_IOCTL=y
CONFIG_IDE_TASKFILE_IO=y

hdparm -i says:

/dev/hdb:

Model=Conner Peripherals 1275MB - CFS1275A, FwRev=8.32, SerialNo=EV9228R
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=2477/16/63, TrkSize=40887, SectSize=649, ECCbytes=4
BuffType=DualPortCache, BuffSize=64kB, MaxMultSect=16, MultSect=8
CurCHS=2477/16/63, CurSects=2496816, LBA=yes, LBAsects=2496876
IORDY=on/off, tPIO={min:270,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio3 pio4
DMA modes: mdma0 mdma1 *mdma2
Kernel Drive Geometry LogicalCHS=619/64/63 PhysicalCHS=2477/16/63

Does anybody know how to fix this/ what's the cause?

             Heinz
--
# Heinz Diehl, 68259 Mannheim, Germany
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

Linux 2.4.19-pre10-ac2

Post by Matt Bernstei » Sat, 08 Jun 2002 17:50:08




>> Since when was it OK to do a parallel make dep?

>wow such a useful response.  can you point me where this is documented?

No--I read it a few times on lkml, admittedly a year or three ago.
Googling reveals others must've seen it too, eg:
http://www.ecst.csuchico.edu/~dranch/LINUX/TrinityOS/cHTML/TrinityOS-...

Quote:>i've never seen "make -j3" cause *source files* to be deleted.  there's a
>bug here somewhere.

OK so my response was rather tangential! What unlink()s does strace
show? What is your fs?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

1. 2.4.19-pre10-ac2, compile warnings/failures

cpqphp.h: In function `cpq_get_latch_status':
cpqphp.h:698: warning: concatenation of string literals with
__FUNCTION__ is deprecated
cpqphp.h: In function `wait_for_ctrl_irq':
cpqphp.h:736: warning: concatenation of string literals with
__FUNCTION__ is deprecated
cpqphp.h:746: warning: concatenation of string literals with
__FUNCTION__ is deprecated

cpqphp_nvram.c:163: warning: concatenation of string literals with
__FUNCTION__ is deprecated

cpqphp_nvram.c:179:17: missing terminating " character
cpqphp_nvram.c: In function `access_EV':
cpqphp_nvram.c:180: parse error before "xorl"
cpqphp_nvram.c:184:28: missing terminating " character

People, please don't do things like:

   spin_lock_irqsave(&int15_lock, flags);
   __asm__ (
      "xorl   %%ebx,%%ebx
      xorl    %%edx,%%edx
      pushf
      push    %%cs
      cli
      call    *%6"

Patches keep going in to fix this.

Please do something like:

   spin_lock_irqsave(&int15_lock, flags);
   __asm__ (
      "xorl   %%ebx,%%ebx  \n"
      "xorl    %%edx,%%edx \n"
      "pushf               \n"
      "push    %%cs        \n"
      "cli                 \n"
      "call    *%6         \n"

i2o_core.c:3393:75: missing terminating " character
         printk(KERN_WARNING "i2o: Could not quiesce %s."  "
            Verify setup on next system power up.\n", c->name);

Please write it like:

         printk(KERN_WARNING "i2o: Could not quiesce %s."  
            "Verify setup on next system power up.\n", c->name);

-d

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

2. Wanted: recommendations for 2.5-inch portable disk drives

3. Problem with 2.4.19-pre10-ac2 and i2o SX6000 controller

4. Two questions about Linux

5. PROBLEM: 2.4.19-pre10-ac2 bug in page_alloc.c:131

6. non-blocking gethostbyname

7. CMD649 on SMP 2.4.19-pre10-ac2: still not working: please help

8. SoundStudio problems

9. Oddities with HighPoint HPT374, 2.4.19-pre10-ac2

10. kbuild-2.5 on 2.4.19-pre10-ac2 build error when modular support turned off

11. Kernel 2.4.19-pre10-ac2 Oops in mke2fs on HPT374 RAID Array

12. 2.4.19-pre10-ac2 tulip.o

13. 2.4.19-pre10-ac2: APM & ACPI