This doesn't even seem to want to build. Config attached.
make[2]: Entering directory `/usr/src/linux/mm'
gcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer -fno-strict-aliasing -fno-common -pipe -march=i586 -DKBUILD_BASENAME=filemap -DEXPORT_SYMTAB -c filemap.c
filemap.c: In function `__find_page_nolock':
filemap.c:404: structure has no member named `next_hash'
make[2]: *** [filemap.o] Error 1
make[2]: Leaving directory `/usr/src/linux/mm'
make[1]: *** [first_rule] Error 2
make[1]: Leaving directory `/usr/src/linux/mm'
make: *** [_dir_mm] Error 2
--
bill davidsen <[email protected]> CTO, TMR Associates, Inc
Programming without software engineering is like sculpting with a chain
saw. The very talented can produce a work of art, the mediocre wind up with
a misshapen lump in a pile of rubble, and in neither case does the end
result have more than a passing resemblance to the original intent.
On Mon, 2002-02-18 at 13:25, [email protected] wrote:
> filemap.c: In function `__find_page_nolock':
> filemap.c:404: structure has no member named `next_hash'
> make[2]: *** [filemap.o] Error 1
> make[2]: Leaving directory `/usr/src/linux/mm'
> make[1]: *** [first_rule] Error 2
> make[1]: Leaving directory `/usr/src/linux/mm'
> make: *** [_dir_mm] Error 2
There was a fix posted to lkml earlier, though I'd say that likely if
you can't fix this, you don't need to run this kernel. try looking at
lkml. Next version will have it fixed, however. =)
------
Michael Cohen
OhDarn.net
On 18 Feb 2002, Michael Cohen wrote:
> On Mon, 2002-02-18 at 13:25, [email protected] wrote:
> > filemap.c: In function `__find_page_nolock':
> > filemap.c:404: structure has no member named `next_hash'
> > make[2]: *** [filemap.o] Error 1
> > make[2]: Leaving directory `/usr/src/linux/mm'
> > make[1]: *** [first_rule] Error 2
> > make[1]: Leaving directory `/usr/src/linux/mm'
> > make: *** [_dir_mm] Error 2
>
> There was a fix posted to lkml earlier, though I'd say that likely if
> you can't fix this, you don't need to run this kernel. try looking at
> lkml. Next version will have it fixed, however. =)
I didn't ask for a fix, I was offering a problem report, from a machine
with no lkml or anything else (don't let the name and IP fool you, I was
testing). I was trying to add it to my summary of recent kernels which run
well on small slow machines which might be available for the asking.
I included the config because I thought the P5 might be triggering a
problem others hadn't tested, but it looks as if everyone had the problem.
I thought these were like -ac patches which has already been tested to see
if they compile.
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
On Mon, 2002-02-18 at 16:50, Bill Davidsen wrote:
> On 18 Feb 2002, Michael Cohen wrote:
>
> > On Mon, 2002-02-18 at 13:25, [email protected] wrote:
> > > filemap.c: In function `__find_page_nolock':
> > > filemap.c:404: structure has no member named `next_hash'
> > > make[2]: *** [filemap.o] Error 1
> > > make[2]: Leaving directory `/usr/src/linux/mm'
> > > make[1]: *** [first_rule] Error 2
> > > make[1]: Leaving directory `/usr/src/linux/mm'
> > > make: *** [_dir_mm] Error 2
> >
> > There was a fix posted to lkml earlier, though I'd say that likely if
> > you can't fix this, you don't need to run this kernel. try looking at
> > lkml. Next version will have it fixed, however. =)
>
> I didn't ask for a fix, I was offering a problem report, from a machine
> with no lkml or anything else (don't let the name and IP fool you, I was
> testing). I was trying to add it to my summary of recent kernels which run
> well on small slow machines which might be available for the asking.
>
> I included the config because I thought the P5 might be triggering a
> problem others hadn't tested, but it looks as if everyone had the problem.
> I thought these were like -ac patches which has already been tested to see
> if they compile.
Sorry, I wasn't in too great of a mood. Didn't mean to sound *that*
rough. Anyways, I had just started using bitkeeper, and the change
didn't get committed properly, so it fell out of my bk tree when I made
the patch. It had been tested mostly though. I started using a much
more inclusive test .config though; and I'll have a working next
release. Take a look at how well 2.5 currently compiles though =)
Fix is to remove the function __find_page_nolock from mm/filemap.c.
------
Michael Cohen
OhDarn.net
On 18 Feb 2002, Michael Cohen wrote:
> Sorry, I wasn't in too great of a mood. Didn't mean to sound *that*
> rough. Anyways, I had just started using bitkeeper, and the change
> didn't get committed properly, so it fell out of my bk tree when I made
> the patch. It had been tested mostly though. I started using a much
> more inclusive test .config though; and I'll have a working next
> release. Take a look at how well 2.5 currently compiles though =)
>
> Fix is to remove the function __find_page_nolock from mm/filemap.c.
And after it compiles and depmod lists all the symbols which aren't
exported you rewite ksyms.c with:
/* all the crap missing from mjc2 */
EXPORT_SYMBOL(blkdev_varyio);
EXPORT_SYMBOL(pidhash_bits);
EXPORT_SYMBOL(pidhash_size);
Some of which is cruft you inhereted which will be fixed in 2.4.28-rc77, I
think one you created yourself.
Since the test machine take ~100min to build a kernel, I have just
finished the build and will boot Friday, assuming that rc3-mjc1 or so
isn't out.
for the record, 2.4.18-pre9-mjc1 did not run as well on the small machine
test as just rmap-12e+K3. I was hoping the preempt in mjc1 would make it
more responsive. I have a new test for thread handling, I want to run
that, and maybe Thursday I'll get to try all of these on SMP and large
memory systems.
Soon I have to write all this stuff up...
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.