On Tue, 2014-09-23 at 03:34 +0000, Fan, Jeff wrote:
> Andrew,
>
> gBS->Stall () implementation has no bug.
>
> Actually, gBS->Stall () will consume EFI Metronome Arch Protocol that is
> published by Metronome module. But Metronome module is dispatched after
> CpuDxe module dispatched on Ovmf platform.
>
> Before Metronome module dispatched, gBS->Stall () is not available.
>
> EFI_STATUS
> EFIAPI
> CoreStall (
> IN UINTN Microseconds
> )
> {
> if (gMetronome == NULL) {
> return EFI_NOT_AVAILABLE_YET;
> }
> }
>
> Since CpuDxe also needs to publish CPU Arch protocol, it cannot assume
> gBS->Stall() is ready always.
>
> I suggest to use TimerLib Service instead of gBS->Stall in CpuDxe here.
I will update this suggestion to my github patches firstly.
Thanks,
Chen
>
> Another long term option, producing CPU MP protocol in another Cpu MP driver
> instead of in CpuDxe driver that is produce CPU ARCH protocol only.
>
> Thanks!
> Jeff
>
> -----Original Message-----
> From: Andrew Fish [mailto:[email protected]]
> Sent: Monday, September 22, 2014 11:58 PM
> To: Fan, Jeff
> Cc: Chen, Fan; [email protected]; Jordan Justen
> Subject: Re: [RFC v3 00/19] Introduce Mp Service protocol to UefiCpuPkg
>
>
> On Sep 22, 2014, at 12:23 AM, Fan, Jeff <[email protected]> wrote:
>
> > Yes. Timeout is root reason here.
> >
> > It seemed that gBS->Stall() does not delay enough time as expected.
> > I used MicroSecondDelay (100 * 1000) to replace gBS->Stall() in
> > StartApsStackless() to delay 100ms.
> >
> > The image ovmf.fd built by GCC could work on Windows Qemu now.
> > C:\qemu\qemu-system-x86_64.exe -pflash OVMF64.fd -smp 4 -serial COM8
> >
>
> gBS->Stall() should always work. We should fix the bug in the gBS->Stall().
>
> Thanks,
>
> Andrew Fish
>
> > I will do some verification more. Thanks!
> >
> > Jeff
> > -----Original Message-----
> > From: Chen, Fan [mailto:[email protected]]
> > Sent: Monday, September 22, 2014 1:46 PM
> > To: Fan, Jeff
> > Cc: [email protected]; Jordan Justen; Andrew Fish
> > Subject: Re: [RFC v3 00/19] Introduce Mp Service protocol to
> > UefiCpuPkg
> >
> > Hi Jeff,
> >
> > On Mon, 2014-09-22 at 05:12 +0000, Fan, Jeff wrote:
> >> Chen,
> >>
> >> Thanks. I used the real platform to verify Cpu MP. It seemed that one AP
> >> is good but 3 APs still failed. Do you know what else need to consider?
> >> Or could you tell me your test environment?
> >>
> >> 1. -smp 4 faile, please see the attached log and console output $
> >> sudo
> >> qemu-system-x86_64 -bios OVMF64.fd -smp 4 --enable-kvm -serial
> >> file:ovmf64_4.log
> >>
> >
> > Thanks for your testing, from the log file I can think of is the more CPU
> > would need more time to execute the start-up code, but MP just waited for
> > 100ms, I encountered the same issue when I set the number of cpu more than
> > 4. was the detect count of CPUs always two each time?
> >
> > I used OVMF to test:
> > command line:
> > ./OvmfPkg/build.sh qemu --enable-kvm -m 1024 -hda
> > /home/data/ovmf-rhel6.3.img -debugcon file:debug.log -global
> > isa-debugcon.iobase=0x402 -smp 4
> >
> > # /bin/qemu-system-x86_64 --version
> > QEMU emulator version 2.1.50
> >
> > Thanks,
> > Chen
> >
> >
> >> 2. -smp 2 succeed, please see the attached log.
> >> $ sudo qemu-system-x86_64 -bios OVMF64.fd -smp 2 --enable-kvm -serial
> >> file:ovmf64_2.log
> >>
> >> For IA32 arch, I got the exact result.
> >>
> >> Thanks!
> >> Jeff
> >> -----Original Message-----
> >> From: Chen, Fan [mailto:[email protected]]
> >> Sent: Friday, September 19, 2014 5:24 PM
> >> To: Fan, Jeff
> >> Cc: [email protected]; Jordan Justen; Andrew Fish
> >> Subject: Re: [RFC v3 00/19] Introduce Mp Service protocol to
> >> UefiCpuPkg
> >>
> >> On Fri, 2014-09-19 at 07:07 +0000, Fan, Jeff wrote:
> >>> Chen & Jordan,
> >>>
> >>> I encountered on exception as blew on Ubuntu12.04 (on virtual machine by
> >>> Virtual Box, VT-x enabled) with GCC46. Log is attached. Do you have any
> >>> idea on it?
> >>>
> >>> $ qemu-system-i386 -smp 4 -serial file:ovmf.log
> >>>
> >>> Could not access KVM kernel module: No such file or directory failed
> >>> to initialize KVM: No such file or directory
> >>
> >> I think you probably did not load the KVM modules. and used the qemu
> >> command with --enable-kvm.
> >>
> >> Thanks,
> >> Chen
> >>
> >>
> >>> Back to tcg accelerator.
> >>> qemu: fatal: Trying to execute code outside RAM or ROM at 0x000a0000
> >>>
> >>> EAX=00000000 EBX=00000000 ECX=00000000 EDX=00000633
> >>> ESI=00000000 EDI=00001ff8 EBP=00000000 ESP=00000000 EIP=00000ffc
> >>> EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0000
> >>> 00000000 0000ffff 00009300 CS =9f00 0009f000 0000ffff 00009b00 SS
> >>> =0000
> >>> 00000000 0000ffff 00009300 DS =0000 00000000 0000ffff 00009300 FS
> >>> =0000 00000000 0000ffff 00009300 GS =0000 00000000 0000ffff 00009300
> >>> LDT=0000 00000000 0000ffff 00008200
> >>> TR =0000 00000000 0000ffff 00008b00
> >>> GDT= 00000000 0000ffff
> >>> IDT= 00000000 0000ffff
> >>> CR0=60000010 CR2=00000000 CR3=00000000 CR4=00000000
> >>> DR0=00000000 DR1=00000000 DR2=00000000 DR3=00000000
> >>> DR6=ffff0ff0 DR7=00000400
> >>> CCS=00000000 CCD=00000000 CCO=SUBW
> >>> EFER=0000000000000000
> >>> FCW=037f FSW=0000 [ST=0] FTW=00 MXCSR=00001f80
> >>> FPR0=0000000000000000 0000 FPR1=0000000000000000 0000
> >>> FPR2=0000000000000000 0000 FPR3=0000000000000000 0000
> >>> FPR4=0000000000000000 0000 FPR5=0000000000000000 0000
> >>> FPR6=0000000000000000 0000 FPR7=0000000000000000 0000
> >>> XMM00=00000000000000000000000000000000
> >>> XMM01=00000000000000000000000000000000
> >>> XMM02=00000000000000000000000000000000
> >>> XMM03=00000000000000000000000000000000
> >>> XMM04=00000000000000000000000000000000
> >>> XMM05=00000000000000000000000000000000
> >>> XMM06=00000000000000000000000000000000
> >>> XMM07=00000000000000000000000000000000
> >>> Aborted (core dumped)
> >>>
> >>> Thanks!
> >>> Jeff
> >>> -----Original Message-----
> >>> From: Chen Fan [mailto:[email protected]]
> >>> Sent: Thursday, September 18, 2014 11:12 AM
> >>> To: [email protected]
> >>> Cc: Jordan Justen; Fan, Jeff; Andrew Fish
> >>> Subject: [RFC v3 00/19] Introduce Mp Service protocol to UefiCpuPkg
> >>>
> >>> This series patchset try to implement Mp Service protocol in UefiCpuPkg,
> >>> Jordan had implemented the startup APs code, and I try to add more
> >>> initialization code to let all APs work up, this Mp Service protocol's
> >>> implementation used EmulatorPkg/MpService for reference.
> >>> this patches works on my github:
> >>> https://github.com/ChenFanFnst/edk2/tree/cpu-mp-service
> >>>
> >>> V2-V3:
> >>> 1. rebase codes due to Jordan'tree updated:
> >>> https://github.com/jljusten/edk2/tree/ap-startup-example
> >>> 2. add supported on Ia32 arch
> >>> 3. add a new Lock to replace present SpinLock mechanisms in mutilple
> >>> processors, maybe the SpinLock mechanisms is not MP safe.
> >>> 4. add function header
> >>> 5. add StartupAllAPs() supported
> >>> 6. add SwitchBSP() function, which is unsupported.
> >>>
> >>> V1-V2:
> >>> 1. do not call anything EFI API from APs.
> >>> 2. add AP busy-wait for task assignment from BSP and get rid of
> >>> IPI sent mechanism.
> >>>
> >>> TODO:
> >>> 1) StartupThisAP/StartAllAps: when Timeout expires before AP returns
> >>> from Procedure, constrainedly terminate the executed Procedure.
> >>>
> >>> Any Comment is welcome.
> >>>
> >>> Chen Fan (13):
> >>> UefiCpuPkg/CpuDxe: Introduce AsmApDoneWithCommonStack
> >>> UefiCpuPkg/CpuDxe: Waiting for Aps initialization done
> >>> UefiCpuPkg/CpuDxe: Switch Ap CommonStack with NewStack
> >>> UefiCpuPkg/CpuDxe: install Mp Service protocol
> >>> UefiCpuPkg/CpuDxe: implement Mp Protocol: GetNumberOfProcessors()
> >>> UefiCpuPkg/CpuDxe: implement Mp Protocol: WhoAmI()
> >>> UefiCpuPkg/CpuDxe: implement Mp Services: GetProcessorInfo()
> >>> UefiCpuPkg/CpuDxe: implement Mp Protocol: EnableDisableAP()
> >>> UefiCpuPkg/CpuDxe: implement Mp Protocol: StartupThisAP()
> >>> UefiCpuPkg/CpuDxe: introduce CPU bus lock for sync data
> >>> UefiCpuPkg/CpuDxe: implement Mp Services: StartupAllAPs()
> >>> UefiCpuPkg/CpuDxe: implement Mp Services: SwitchBSP()
> >>> UefiCpuPkg/CpuDxe: add Mp Service TestCase
> >>>
> >>> Jordan Justen (6):
> >>> UefiCpuPkg/CpuDxe: Add no-op InitializeMpSupport
> >>> UefiCpuPkg/CpuDxe: Add ApEntryPointInC
> >>> UefiCpuPkg/CpuDxe: Add stackless assembly AP entry points
> >>> UefiCpuPkg/CpuDxe: Move GDT structures into CpuGdt.h
> >>> UefiCpuPkg/CpuDxe: Add StartApsStackless routine
> >>> UefiCpuPkg/CpuDxe: Startup APs
> >>>
> >>> UefiCpuPkg/CpuDxe/ApStartup.asm | 111 +++
> >>> UefiCpuPkg/CpuDxe/ApStartup.c | 209 ++++++
> >>> UefiCpuPkg/CpuDxe/CpuDxe.c | 5 +
> >>> UefiCpuPkg/CpuDxe/CpuDxe.inf | 8 +
> >>> UefiCpuPkg/CpuDxe/CpuGdt.c | 52 +-
> >>> UefiCpuPkg/CpuDxe/CpuGdt.h | 72 ++
> >>> UefiCpuPkg/CpuDxe/CpuMp.c | 1349
> >>> +++++++++++++++++++++++++++++++++++++
> >>> UefiCpuPkg/CpuDxe/CpuMp.h | 643 ++++++++++++++++++
> >>> UefiCpuPkg/CpuDxe/CpuMpTest.c | 76 +++
> >>> UefiCpuPkg/CpuDxe/Ia32/MpAsm.S | 109 +++
> >>> UefiCpuPkg/CpuDxe/Ia32/MpAsm.nasm | 144 ++++
> >>> UefiCpuPkg/CpuDxe/X64/MpAsm.S | 111 +++
> >>> UefiCpuPkg/CpuDxe/X64/MpAsm.nasm | 142 ++++
> >>> 13 files changed, 2980 insertions(+), 51 deletions(-) create mode
> >>> 100644 UefiCpuPkg/CpuDxe/ApStartup.asm create mode 100644
> >>> UefiCpuPkg/CpuDxe/ApStartup.c create mode 100644
> >>> UefiCpuPkg/CpuDxe/CpuGdt.h create mode 100644
> >>> UefiCpuPkg/CpuDxe/CpuMp.c create mode 100644
> >>> UefiCpuPkg/CpuDxe/CpuMp.h create mode 100644
> >>> UefiCpuPkg/CpuDxe/CpuMpTest.c create mode 100644
> >>> UefiCpuPkg/CpuDxe/Ia32/MpAsm.S create mode 100644
> >>> UefiCpuPkg/CpuDxe/Ia32/MpAsm.nasm create mode 100644
> >>> UefiCpuPkg/CpuDxe/X64/MpAsm.S create mode 100644
> >>> UefiCpuPkg/CpuDxe/X64/MpAsm.nasm
> >>>
> >>> --
> >>> 1.9.3
> >>>
> >>
> >
>
------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
_______________________________________________
edk2-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/edk2-devel
