Hi,
so I rebuild with your patch applied but it is still very slow with the bsd.mp
kernel.
Thanks for you help
Morgan
18 octobre 2023 10:43 "Stuart Henderson" <stu.li...@spacehopper.org> a écrit:
> On 2023-10-17, Comète <com...@geekandfree.org> wrote:
>
>> Hi,
>>
>> Wow ! you're absolutely right ! If I unplug, no lagg anymore.
>> So the solution should be to apply your patch and rebuild the kernel ?
>
> It's certainly worth trying. If you do, please report back here.
>
>> Thanks a lot !
>>
>> Morgan
>>
>> 17 octobre 2023 14:24 "Stuart Henderson" <stu.li...@spacehopper.org> a écrit:
>>
>>> On 2023-10-16, Comète <com...@geekandfree.org> wrote:
>>
>> Hello,
>>
>> I'm experiencing big slowdowns on a LENOVO Thinkpad T14 Gen3 when using MP
>> kernel (on 7.3 and 7.4)
>> but strangely not on GENERIC.
>> For example, starting LibreOffice on GENERIC takes 7 seconds but 35 seconds
>> on MP kernel. It's even
>> lagging when typing some text in an editor or a mail.
>> Switching to GENERIC and all is working as expected...
>>
>> Thanks for your help !
>>
>> Morgan
>>
>> This is my dmesg on both kernels:
>>
>> OpenBSD 7.4 (GENERIC) #1336: Tue Oct 10 08:52:22 MDT 2023
>> dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC
>> real mem = 34026549248 (32450MB)
>> avail mem = 32975671296 (31448MB)
>> random: good seed from bootblocks
>> mpath0 at root
>> scsibus0 at mpath0: 256 targets
>> mainbus0 at root
>> bios0 at mainbus0: SMBIOS rev. 3.4 @ 0x8f8a3000 (81 entries)
>> bios0: vendor LENOVO version "N3MET16W (1.15 )" date 06/25/2023
>>> No problem with MP here, but I have an older BIOS -
>>>
>>> bios0 at mainbus0: SMBIOS rev. 3.4 @ 0x8d8a3000 (81 entries)
>>> bios0: vendor LENOVO version "N3MET12W (1.11 )" date 02/09/2023
>>>
>>> (grumble stupid US date format)
>>
>> bios0: LENOVO 21AHCTO1WW
>> efi0 at bios0: UEFI 2.7
>> efi0: Lenovo rev 0x1150
>> acpi0 at bios0: ACPI 6.3
>> acpi0: sleep states S0 S3 S4 S5
>> acpi0: tables DSDT FACP SSDT SSDT SSDT SSDT SSDT TPM2 HPET APIC MCFG ECDT
>> SSDT SSDT SSDT SSDT SSDT
>> SSDT LPIT WSMT SSDT DBGP DBG2 NHLT MSDM SSDT BATB DMAR SSDT SSDT SSDT BGRT
>> PHAT UEFI FPDT
>> acpi0: wakeup devices PEG0(S4) PEGP(S4) PEGP(S4) PEG2(S4) PEGP(S4) GLAN(S4)
>> XHCI(S3) XDCI(S4)
>> HDAS(S4) CNVW(S4) RP01(S4) PXSX(S4) RP02(S4) PXSX(S4) RP03(S4) PXSX(S4) [...]
>> acpitimer0 at acpi0: 3579545 Hz, 24 bits
>> acpihpet0 at acpi0: 19200000 Hz
>> acpimadt0 at acpi0 addr 0xfee00000: PC-AT compat
>> cpu0 at mainbus0: apid 0 (boot processor)
>> cpu0: 12th Gen Intel(R) Core(TM) i7-1260P, 2151.34 MHz, 06-9a-03, patch
>> 0000042c
>>> and different cpu:
>>>
>>> cpu0: 12th Gen Intel(R) Core(TM) i5-1245U, 1568.55 MHz, 06-9a-04, patch
>>> 0000042c
>>>
>>> FWIW I can definitely get mine to throttle when it's busy. And your
>>> CPU uses a fair bit more power than mine (I specifically looked for a
>>> U rather than a P cpu for exactly this reason) so I'd guess might be
>>> easier to hit the throttle.
>>>
>>> The OpenBSD kernel tries to set cpu clock speed high when on mains
>>> power, so it might be worth trying unplugged to see if there's any
>>> difference, or disable that thing with this
>>>
>>> Index: sched_bsd.c
>>> ===================================================================
>>> RCS file: /cvs/src/sys/kern/sched_bsd.c,v
>>> retrieving revision 1.88
>>> diff -u -p -r1.88 sched_bsd.c
>>> --- sched_bsd.c 11 Oct 2023 15:42:44 -0000 1.88
>>> +++ sched_bsd.c 17 Oct 2023 12:10:41 -0000
>>> @@ -605,7 +605,7 @@ setperf_auto(void *v)
>>> if (cpu_setperf == NULL)
>>> return;
>>>
>>> - if (hw_power) {
>>> + if (0 && hw_power) {
>>> speedup = 1;
>>> goto faster;
>>> }
