if you read the file /etc/rc.d/init.d/microcode_ctl
you see that microcode_ctl is the software that mange to update the microcode. for this it uses "microcode driver as a component)
then i assume it is not deprecated
see
*****************************************************************************************************
if [ $minor -ge 29 ]; then
gprintf "Loading AMD microcode update module\n"
/sbin/modprobe microcode
action "" /bin/true
touch /var/lock/subsys/microcode_ctl
/sbin/rmmod microcode
exit 0
fi
else
action "" /bin/true
touch /var/lock/subsys/microcode_ctl
exit 0
fi
if [ ! -e $DATAFILE ]; then
action "CPU microcode data file not present (%s)" "$DATAFILE" /bin/false
exit 1
fi
/sbin/modprobe microcode
lt=0
while [ ! -c /dev/cpu/microcode ]; do
*****************************************************************************************************
more
it seems microcode_ctl is strangely only for amd family 16 and above
see
********************************************************************************************************
start|reload|restart|force-reload)
# Intel 686 and above, AMD family 16 and above
********************************************************************************************************
then
i understand why my cpu is not updated
Asus M2N, AMD Athlon 64 X2 5000+, 8 GB, Asus Nvidia 210 silent, Linksys WMP600N, Plugable USB-BT4LE Bluetooth 4.0 Adapter, lecteur multicarte Addonics AEIDDSAU, HP LaserJet 1220
promeneur wrote:if you read the file /etc/rc.d/init.d/microcode_ctl
you see that microcode_ctl is the software that mange to update the microcode. for this it uses "microcode driver as a component)
[****]
then i assume it is not deprecated
[***]
it seems microcode_ctl is strangely only for amd family 16 and above
see
********************************************************************************************************
start|reload|restart|force-reload)
# Intel 686 and above, AMD family 16 and above
********************************************************************************************************
then
i understand why my cpu is not updated
Well, I still did not look at it, as you did, but I'm on a AMD fam15h CPU as I showed previously and in my system the microcode update does not need
at all the use of microcode_ctl, because it happens automagically at the boot time...
so I guess the above sw is needed only when it cannot happen (as, ie, with a CPU > fam15h: hence the limitation in the script and the *Intel* CPU).
Perhaps we will need to back-port and adjust also the microcode_ctl package...
Sembra tutto OK nel Intel, mentre sembra che c'è qualcosa che non va (non carica il module) in AMD...
ninarello19
=========
RosaDesktop.Fresh 2012.1 R1 - arch. x86_64 - KDE 4.10.4 MIB - Kernel 3.10.15-nrjQL-desktop-69rosa #1 SMP PREEMPT
O.S. alternativi su VMware workstation 10.0.0>>> Ubuntu, Windows XP Pro, Windows Seven e Mac come Macchine VIrtuali
per AMD, il modulo NON l'ha ancora caricato, ma hai provato almeno a riavviare???
riavvia, e fammi sapere!
.
--- Professional experience ---
Kernel designer, engineer, maintainer and tester for ROSA Desktop and OpenMandriva Lx O.S.
--- currently I'm playing with ---
LTS Kernels > Linux 4.1.12-nrjQL <<< Linux 3.18.17-nrjQL <<< Linux 3.14.46-nrjQL
EOL Kernels > Linux 3.19.8-nrjQL <<< Linux 3.17.8-nrjQL <<< Linux 3.15.10-nrjQL
ninarello19
=========
RosaDesktop.Fresh 2012.1 R1 - arch. x86_64 - KDE 4.10.4 MIB - Kernel 3.10.15-nrjQL-desktop-69rosa #1 SMP PREEMPT
O.S. alternativi su VMware workstation 10.0.0>>> Ubuntu, Windows XP Pro, Windows Seven e Mac come Macchine VIrtuali
Sembra dunque che per INTEL non occorra riavvio, mentre per AMD occorra sempre almeno un riavvio...
.
--- Professional experience ---
Kernel designer, engineer, maintainer and tester for ROSA Desktop and OpenMandriva Lx O.S.
--- currently I'm playing with ---
LTS Kernels > Linux 4.1.12-nrjQL <<< Linux 3.18.17-nrjQL <<< Linux 3.14.46-nrjQL
EOL Kernels > Linux 3.19.8-nrjQL <<< Linux 3.17.8-nrjQL <<< Linux 3.15.10-nrjQL
su AMD X64 x 2 4200 non c'è verso i messaggi sono quelli (invalid argument per modprobe microcode e CPU fam 15 not supported)
se riavvio dmesg, non da niente segno che non carica il modulo
a me non funziona
In teoria dovrebbero essere supportate tutte le cpu AMD appartenenti alle famiglie da 10 a 15,
ma con la famiglia 15 qualcosa non va, o non si scompatta o ci sono problemi di path, dobbiamo indagare...
.
--- Professional experience ---
Kernel designer, engineer, maintainer and tester for ROSA Desktop and OpenMandriva Lx O.S.
--- currently I'm playing with ---
LTS Kernels > Linux 4.1.12-nrjQL <<< Linux 3.18.17-nrjQL <<< Linux 3.14.46-nrjQL
EOL Kernels > Linux 3.19.8-nrjQL <<< Linux 3.17.8-nrjQL <<< Linux 3.15.10-nrjQL
vmalav wrote:su AMD X64 x 2 4200 non c'è verso i messaggi sono quelli (invalid argument per modprobe microcode e CPU fam 15 not supported)
se riavvio dmesg, non da niente segno che non carica il modulo
a me non funziona
Cosa ti dice il buon cpuid nella sezione sulla version info (1/eax slot dell'output)?
A me, con un Phenom II X4 810 (una fam 15h, quindi) funziona perfettamente (v. il mio precedente messaggio a promeneur)...
I would verify that:
1) the microcode_amd*.bin files are in their right place (/lib/firmware/amd-ucode)
2) my CPU is supported: what does say cpuid in the version information (1/eax) slot?
promeneur wrote:
2.
there is no "version information(1/eax)
Yes, you looked directly to /proc/cpuinfo, while I was using /usr/bin/cpuid:
same result, however...
the CPU should be supported out-of-the-box from a recent kernel (I tried 3.4.x and 3.5.x).
Perhaps are you on something other kernel version?