On Wed, Oct 08, 2014 at 02:36:49PM -0700, H. Peter Anvin wrote: > On 10/08/2014 12:49 PM, Chris Wilson wrote: > > > > Indeed, this appears to be the explanation. (And here I thought PAT > > superseded mtrrs - i915.ko stopped trying to use assign an mtrr for its > > GTT quite a while ago.) > > > > Replacing the stop_machine there with on_each_cpu does the trick: > > > > It should, but there seem to be quite a few drivers which still muck > with MTRRs. However, i915 is not one of them, it calls > io_mapping_create_wc() followed by arch_phys_wc_add(), so I'm wondering > what the heck is going on here.
This system also have a radeon GPU. Disabling it (not building in the module) makes no difference to the wc speed. > > Naively I would say that we lost the wc on our ioremap. > > /sys/kernel/debug/x86/pat_memtype_list remained the same across repeated > > runs. > > Could you tell me what the above looks like? # cat /sys/kernel/debug/x86/pat_memtype_list PAT memtype list: write-back @ 0x8cf34000-0x8cf43000 write-back @ 0x8cf4d000-0x8cf4e000 write-back @ 0x8cf4d000-0x8cf50000 write-back @ 0x8cf50000-0x8cf51000 write-back @ 0x8cf51000-0x8cf52000 write-back @ 0x8cf52000-0x8cf53000 write-back @ 0x8cf53000-0x8cf55000 write-back @ 0x8cf55000-0x8cf56000 write-back @ 0x8cf9d000-0x8cf9e000 write-back @ 0x8cf9f000-0x8cfa0000 write-back @ 0x8cffc000-0x8cffd000 uncached-minus @ 0x8fc00000-0x8fe00000 write-combining @ 0x8fe00000-0x90000000 uncached-minus @ 0x90220000-0x90240000 uncached-minus @ 0x90300000-0x90320000 uncached-minus @ 0x90340000-0x90341000 uncached-minus @ 0x90380000-0x90381000 write-combining @ 0xa0000000-0xc0000000 write-combining @ 0xa0139000-0xa0159000 write-combining @ 0xa0159000-0xa0179000 write-combining @ 0xa0179000-0xa0199000 write-combining @ 0xc0040000-0xc025e000 write-combining @ 0xc025e000-0xc045e000 write-combining @ 0xc045e000-0xc045f000 write-combining @ 0xc045f000-0xc075f000 uncached-minus @ 0xf8000000-0xfc000000 uncached-minus @ 0xfed00000-0xfed01000 uncached-minus @ 0xfed10000-0xfed16000 uncached-minus @ 0xfed1f000-0xfed20000 (identical for good/bad runs) # cat /proc/mtrr reg00: base=0x000000000 ( 0MB), size= 2048MB, count=1: write-back reg01: base=0x080000000 ( 2048MB), size= 256MB, count=1: write-back reg02: base=0x08e000000 ( 2272MB), size= 32MB, count=1: uncachable reg03: base=0x08d000000 ( 2256MB), size= 16MB, count=1: uncachable reg04: base=0x100000000 ( 4096MB), size= 2048MB, count=1: write-back reg05: base=0x170000000 ( 5888MB), size= 256MB, count=1: uncachable reg06: base=0x16f000000 ( 5872MB), size= 16MB, count=1: uncachable reg07: base=0x16e800000 ( 5864MB), size= 8MB, count=1: uncachable reg08: base=0x16e600000 ( 5862MB), size= 2MB, count=1: uncachable # cat /proc/iomem: 00000000-00000fff : reserved 00001000-0009bbff : System RAM 0009bc00-0009ffff : reserved 000a0000-000bffff : PCI Bus 0000:00 000c0000-000cdfff : Video ROM 000d0000-000d3fff : PCI Bus 0000:00 000d4000-000d7fff : PCI Bus 0000:00 000d8000-000dbfff : PCI Bus 0000:00 000dc000-000dffff : PCI Bus 0000:00 000e0000-000fffff : reserved 000e0000-000e3fff : PCI Bus 0000:00 000e4000-000e7fff : PCI Bus 0000:00 000f0000-000fffff : System ROM 00100000-1fffffff : System RAM 01000000-0161981b : Kernel code 0161981c-01ca20ff : Kernel data 01dac000-01e2dfff : Kernel bss 20000000-201fffff : reserved 20000000-201fffff : pnp 00:05 20200000-3fffffff : System RAM 40000000-401fffff : reserved 40000000-401fffff : pnp 00:05 40200000-8ccd2fff : System RAM 8ccd3000-8cd66fff : reserved 8cd67000-8cfe6fff : ACPI Non-volatile Storage 8cfe7000-8cffefff : ACPI Tables 8cfff000-8cffffff : System RAM 8d000000-8f9fffff : reserved 8da00000-8f9fffff : Graphics Stolen Memory 8fa00000-feafffff : PCI Bus 0000:00 8fa00000-8fa00fff : pnp 00:03 8fc00000-8fffffff : 0000:00:02.0 90000000-900fffff : PCI Bus 0000:04 90000000-900fffff : PCI Bus 0000:05 90000000-90003fff : 0000:05:00.0 90010000-900107ff : 0000:05:00.0 90100000-901fffff : PCI Bus 0000:03 90100000-90101fff : 0000:03:00.0 90200000-902fffff : PCI Bus 0000:01 90200000-9021ffff : 0000:01:00.0 90220000-9023ffff : 0000:01:00.0 90240000-90243fff : 0000:01:00.1 90300000-9031ffff : 0000:00:19.0 90300000-9031ffff : e1000e 90330000-903300ff : 0000:00:1f.3 90340000-903407ff : 0000:00:1f.2 90340000-903407ff : ahci 90350000-903503ff : 0000:00:1d.0 90360000-90363fff : 0000:00:1b.0 90370000-903703ff : 0000:00:1a.0 90380000-90380fff : 0000:00:19.0 90380000-90380fff : e1000e 90390000-90390fff : 0000:00:16.3 903a0000-903a000f : 0000:00:16.0 a0000000-bfffffff : 0000:00:02.0 c0000000-cfffffff : PCI Bus 0000:01 c0000000-cfffffff : 0000:01:00.0 f8000000-fbffffff : PCI MMCONFIG 0000 [bus 00-3f] f8000000-fbffffff : reserved f8000000-fbffffff : pnp 00:03 fec00000-fec00fff : reserved fec00000-fec003ff : IOAPIC 0 fed00000-fed003ff : HPET 0 fed00000-fed003ff : PNP0103:00 fed10000-fed13fff : reserved fed18000-fed19fff : reserved fed18000-fed18fff : pnp 00:03 fed19000-fed19fff : pnp 00:03 fed1c000-fed1ffff : reserved fed1c000-fed1ffff : pnp 00:03 fed20000-fed3ffff : pnp 00:03 fed40000-fed44fff : PCI Bus 0000:00 fed45000-fed8ffff : pnp 00:03 fed90000-fed93fff : pnp 00:03 fee00000-fee00fff : Local APIC fee00000-fee00fff : reserved ff000000-ffffffff : INT0800:00 ff980000-ffbfffff : reserved ffd80000-ffffffff : reserved 100000000-16e5fffff : System RAM 16e600000-16fffffff : RAM buffer # lspci -vv -s 0:0:2 00:02.0 VGA compatible controller: Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller (rev 09) (prog-if 00 [VGA controller]) Subsystem: Intel Corporation Device 2210 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Latency: 0 Interrupt: pin A routed to IRQ 26 Region 0: Memory at 8fc00000 (64-bit, non-prefetchable) [size=4M] Region 2: Memory at a0000000 (64-bit, prefetchable) [size=512M] Region 4: I/O ports at 3000 [size=64] Expansion ROM at <unassigned> [disabled] Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit- Address: fee0f00c Data: 41b1 Capabilities: [d0] Power Management version 2 Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-) Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME- Capabilities: [a4] PCI Advanced Features AFCap: TP+ FLR+ AFCtrl: FLR- AFStatus: TP- Kernel driver in use: i915 -- Chris Wilson, Intel Open Source Technology Centre -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/