tag:blogger.com,1999:blog-1016709139053396535.post6217502661884297377..comments2024-03-18T17:34:35.079+00:00Comments on A Smackerel of Opinion: Making sense of PCIe ASPMColin Ian Kinghttp://www.blogger.com/profile/06458723239721015750noreply@blogger.comBlogger7125tag:blogger.com,1999:blog-1016709139053396535.post-55338580523203111882017-02-15T07:03:37.675+00:002017-02-15T07:03:37.675+00:00I'm having trouble with ASPM on Skylake and no...I'm having trouble with ASPM on Skylake and noticed the root port says "ASPM not supported" does that mean UEFI/Linux disabled it or that the hardware itself does not support it?<br /><br />00:1b.0 0604: 8086:a167 (rev f1) (prog-if 00 [Normal decode])<br /> Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+<br /> Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- SERR- TAbort- Reset- FastB2B-<br /> PriDiscTmr- SecDiscTmr- DiscTmrStat- DiscTmrSERREn-<br /> Capabilities: [40] Express (v2) Root Port (Slot+), MSI 00<br /> DevCap: MaxPayload 256 bytes, PhantFunc 0<br /> ExtTag- RBE+<br /> DevCtl: Report errors: Correctable+ Non-Fatal+ Fatal+ Unsupported+<br /> RlxdOrd- ExtTag- PhantFunc- AuxPwr- NoSnoop-<br /> MaxPayload 128 bytes, MaxReadReq 128 bytes<br /> DevSta: CorrErr- UncorrErr- FatalErr- UnsuppReq- AuxPwr+ TransPend-<br /> LnkCap: Port #17, Speed 8GT/s, Width x4, ASPM not supported, Exit Latency L0s <1us, L1 <16us<br /> ClockPM- Surprise- LLActRep+ BwNot+ ASPMOptComp+<br /> LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- CommClk+<br /> ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-<br /> LnkSta: Speed 8GT/s, Width x4, TrErr- Train- SlotClk+ DLActive+ BWMgmt+ ABWMgmt-<br /><br />(was too long for comment) - Linux 4.10calchttps://www.blogger.com/profile/00761979115677332112noreply@blogger.comtag:blogger.com,1999:blog-1016709139053396535.post-79203202117983490472011-12-01T23:33:26.103+00:002011-12-01T23:33:26.103+00:00@Alex, many thanks for spotting my mistakes. I'...@Alex, many thanks for spotting my mistakes. I've re-worked the article.Colin Ian Kinghttps://www.blogger.com/profile/06458723239721015750noreply@blogger.comtag:blogger.com,1999:blog-1016709139053396535.post-85834952572822373242011-12-01T13:46:28.570+00:002011-12-01T13:46:28.570+00:00@Colin,
"PCIe ASPM Controls" does not i...@Colin,<br /><br />"PCIe ASPM Controls" does not indicate "BIOS can indicate that ASPM should be disabled". <br /><br />Instead, it means that BIOS tells OSPM, i.e. Linux or Windows, not to control ASPM settings ("If set, indicates to OSPM that it must not enable OSPM ASPM control on this platform" from ACPI 4.0a)<br /><br />I suspect the primary purpose of this bit is to workaround some faulty PCIE devices that causes the instability. BIOS can force to disable L0s and/or L1 and OSPM will not enable them. I saw many such cases before myself.<br /><br />~Alex HungAlex Hunghttps://www.blogger.com/profile/17630352059792385002noreply@blogger.comtag:blogger.com,1999:blog-1016709139053396535.post-25843395831801800712011-11-20T01:09:02.930+00:002011-11-20T01:09:02.930+00:00Ran across this post after I rolled my own version...Ran across this post after I rolled my own version of 3.2-rc2 with MG's patches. Booting with pcie_aspm=powersave didn't do a lot for me, however, changing /sys/module/pcie_aspm/parameters/policy to powersave shaved approx 2-3 watts off. Measured with a wattmeter on the mains plug.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1016709139053396535.post-80041481539355126532011-10-11T09:07:45.918+01:002011-10-11T09:07:45.918+01:00@Rogelio, I'm not a hardware engineer, but I s...@Rogelio, I'm not a hardware engineer, but I suspect the latencies are down to powering up the PLLs and getting reference clocks re-synchronized and other link related housekeeping.Colin Ian Kinghttps://www.blogger.com/profile/06458723239721015750noreply@blogger.comtag:blogger.com,1999:blog-1016709139053396535.post-26905617312358971202011-10-11T04:46:55.758+01:002011-10-11T04:46:55.758+01:00why the latency? is there a way to design devices ...why the latency? is there a way to design devices with a much faster poweron time?Rogeliohttps://www.blogger.com/profile/14269444824192058618noreply@blogger.comtag:blogger.com,1999:blog-1016709139053396535.post-76032472097330433742011-04-12T08:02:01.701+01:002011-04-12T08:02:01.701+01:00I am curious: Do you think the PCI spec can be use...I am curious: Do you think the PCI spec can be used to turn off a device completely? I am asking because, I'd like to add a power-hungry GFX to my Desktop that I'd like to turn off occasionally and use the on-board ones.Unknownhttps://www.blogger.com/profile/03302860043986900644noreply@blogger.com