[SOLVED] New Kernel in Slackware-current

cwizardone · 05-06-2017, 05:38 PM

Quote:

Originally Posted by ppencho

VirtualBox-5.1.14-112924-Linux_amd64.run and kernel 4.9.26 work fine together...

I usually uninstall VirtualBox before installing a new kernel, but this time forgot all about it. When I rebooted, VirtualBox automatically rebuilt its module for the "new" 4.9.26 kernel and has been running perfectly ever since.

frankbell · 05-06-2017, 07:25 PM

The update broke my Terminator. Something about Python bindings for libvte.

I'll install anew sometime this week.

willysr · 05-06-2017, 09:01 PM

Here's on my system:

Code:

vboxdrv.sh: Stopping VirtualBox services.
vboxdrv.sh: Building VirtualBox kernel modules.
vboxdrv.sh: Starting VirtualBox services.
vboxdrv.sh: Building VirtualBox kernel modules.
vboxdrv.sh: failed: modprobe vboxdrv failed. Please use 'dmesg' to find out why.

content of dmesg:

Code:

vboxdrv: Unknown symbol __udivmoddi4 (err 0)

dugan · 05-06-2017, 09:20 PM

willlysr: that's what I get too

willysr · 05-06-2017, 09:23 PM

i wonder whether this is only happening in x86 as people with x86_64 doesn't have this problem?

hitest · 05-06-2017, 09:56 PM

Quote:

Originally Posted by willysr

i wonder whether this is only happening in x86 as people with x86_64 doesn't have this problem?

Good question. My -current box is 64 bit. Vbox works well here.

ppencho · 05-07-2017, 12:13 AM

Quote:

Originally Posted by willysr

i wonder whether this is only happening in x86 as people with x86_64 doesn't have this problem?

It seems that __udivmoddi4 is part of libgcc. I have multilib and I still use gcc_5.4_multilib, maybe this makes the difference?

Code:

/usr/lib/libgcc_s.so
/usr/lib/libgcc_s.so.1
/usr/lib64/libgcc_s.so
/usr/lib64/libgcc_s.so.1
/usr/lib64/libgccpp.la
/usr/lib64/libgccpp.so
/usr/lib64/libgccpp.so.1
/usr/lib64/libgccpp.so.1.0.3
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/libgcc.a
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/libgcc_eh.a
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/32/libgcc.a
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/32/libgcc_eh.a
/usr/src/linux-4.9.26/arch/h8300/lib/libgcc.h
/usr/src/linux-4.9.26/arch/m32r/lib/libgcc.h
/usr/src/linux-4.9.26/arch/microblaze/lib/libgcc.h
/usr/src/linux-4.9.26/arch/mips/lib/libgcc.h
/usr/src/linux-4.9.26/arch/score/lib/libgcc.h
/usr/src/linux-4.9.26/arch/sh/lib/libgcc.h
/usr/src/linux-4.9.26/arch/sparc/lib/libgcc.h

rdsherman · 05-07-2017, 01:09 AM

I installed the latest -current kernel, glibc, gcc, ... & found a pleasant surprise. For the first time ever, I'm getting USB 3 flash drive write speeds close to what manufacturers claim. My Corsair Voyager II had been hitting no more than 20 MB/s; now it's 50 MB/s, as it should be. Anybody else noticed this radical driver improvement?

willysr · 05-07-2017, 02:08 AM

Quote:

Originally Posted by ppencho

It seems that __udivmoddi4 is part of libgcc. I have multilib and I still use gcc_5.4_multilib, maybe this makes the difference?

Code:

/usr/lib/libgcc_s.so
/usr/lib/libgcc_s.so.1
/usr/lib64/libgcc_s.so
/usr/lib64/libgcc_s.so.1
/usr/lib64/libgccpp.la
/usr/lib64/libgccpp.so
/usr/lib64/libgccpp.so.1
/usr/lib64/libgccpp.so.1.0.3
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/libgcc.a
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/libgcc_eh.a
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/32/libgcc.a
/usr/lib64/gcc/x86_64-slackware-linux/5.4.0/32/libgcc_eh.a
/usr/src/linux-4.9.26/arch/h8300/lib/libgcc.h
/usr/src/linux-4.9.26/arch/m32r/lib/libgcc.h
/usr/src/linux-4.9.26/arch/microblaze/lib/libgcc.h
/usr/src/linux-4.9.26/arch/mips/lib/libgcc.h
/usr/src/linux-4.9.26/arch/score/lib/libgcc.h
/usr/src/linux-4.9.26/arch/sh/lib/libgcc.h
/usr/src/linux-4.9.26/arch/sparc/lib/libgcc.h

That could be it, but i'm not really sure
i have GCC 7.1.0 here and i tested with the latest update per May 6 and it still persists

willysr · 05-07-2017, 02:09 AM

Quote:

Originally Posted by rdsherman

I installed the latest -current kernel, glibc, gcc, ... & found a pleasant surprise. For the first time ever, I'm getting USB 3 flash drive write speeds close to what manufacturers claim. My Corsair Voyager II had been hitting no more than 20 MB/s; now it's 50 MB/s, as it should be. Anybody else noticed this radical driver improvement?

most likely due to improvements in the kernel side

arahmadi · 05-07-2017, 04:23 AM

Hi,

I am new in this community.

I would like to report that perl is not happy with the recent upgrade of Slackware-current, I've got various error with *.c when trying to upgrade via cpan.

i.e.

LibIDN.c: loadable library and perl binaries are mismatched (got handshake key 0xdb00080, needed 0xdb80080)
Zlib.c: loadable library and perl binaries are mismatched (got handshake key 0xdb00080, needed 0xdb80080)

anyhelp?

I am trying to recompile the Perl now using
https://slackbuilds.org/mirror/slack...source/d/perl/

willysr · 05-07-2017, 05:45 AM

Quote:

Originally Posted by arahmadi

Hi,

I am new in this community.

I would like to report that perl is not happy with the recent upgrade of Slackware-current, I've got various error with *.c when trying to upgrade via cpan.

i.e.

LibIDN.c: loadable library and perl binaries are mismatched (got handshake key 0xdb00080, needed 0xdb80080)
Zlib.c: loadable library and perl binaries are mismatched (got handshake key 0xdb00080, needed 0xdb80080)

anyhelp?

I am trying to recompile the Perl now using
https://slackbuilds.org/mirror/slack...source/d/perl/

Please don't mix between perl packages built via CPAN and SlackBuilds. Just stick with one method and you will be safe.

There are several threads here on LQ (here, here, and here) that discussed about this issue.

arahmadi · 05-07-2017, 06:14 AM

Thanks for the clue Willy. I got more clue from here.

Well, I managed to remove old cpan modules (built while in slackware64-14.2).
I've tried to rebuild Zlib, libidn, but the same errors are still there.

Now trying to downgrade the perl from 5.24.0 to 5.22.3, first. Then, summon the cpan console to uninstall things.
It's been a while not playing with -current.

arahmadi · 05-07-2017, 09:47 AM

All good now.
This is my quick and dirty solution:

I move all related modules that was compiled with Perl 5.22.
In this case I back them up in different folders

Something like:

VERSION=_5.22
mv /usr/local/lib64/perl5 /usr/local/lib64/perl5$VERSION
mv /usr/local/shared/perl5 /usr/local/shared/perl5$VERSION
mv /root/.cpan /root/.cpan$VERSION
mv /usr/lib64/perl5/vendor_perl /usr/lib64/perl5/vendor_perl$VERSION
mv /usr/share/perl5/vendor_perl /usr/share/perl5/vendor_perl$VERSION
mv /usr/lib64/perl5 /usr/lib64/perl5$VERSION
mv /usr/share/perl5 /usr/share/perl5$VERSION

Then I recompiled Perl 5.24.1 with the script supplied. cpan console is back.

Everything seems OK for now.

Hence I report successful upgrade to the latest -current on:
Intel DG31PR, Intel Xeon x5450 (mod) @3.0 GHz, DDR2 4GB, Radeon HD6400 1GB, Panasonic (TV) 1920x1080, Microsoft Wireless Mouse (series: 2000) and Keyboard (series: 1000).

Will do more test next weekend.

Good job Pat & Slackware team.

dugan · 05-07-2017, 12:18 PM

willysr: my VM is also 32-bit.