Re: Problem with Industry Scenario on a NUC7i7DNH


Yin, Fengwei <fengwei.yin@...>
 

On 1/13/2020 9:44 AM, Yin, Fengwei wrote:
On 1/10/2020 8:38 PM, Geoffroy Van Cutsem wrote:
I found this article that suggests Ubuntu 19.10 specifically enabled this flag for the first time by default: https://www.phoronix.com/scan.php?page=news_item&px=Ubuntu-19.10-GCC-Hardening.

I guess this is not the case in Clear Linux at this point, although I have not dug yet into where these things are set to really understand if that’s the cause of this problem.
The kernel also hit same issue and it disabled the -fcf-protection if
the -mindirect-branch is enabled (for retpoline):
https://lkml.org/lkml/2019/7/9/641
I will submit a patch accordingly.
Fixing patch was sent to acrn-dev mail list:
https://lists.projectacrn.org/g/acrn-dev/topic/patch_makefile_disable/69659519?p=,,,20,0,0,0::recentpostdate%2Fsticky,,,20,2,0,69659519

Regards
Yin, Fengwei

Regards
Yin, Fengwei


Cheers,

Geoffroy

*From:* acrn-users@... <acrn-users@...> *On Behalf Of *Steve Osselton
*Sent:* Friday, January 10, 2020 1:27 PM
*To:* acrn-users@...
*Subject:* Re: [acrn-users] Problem with Industry Scenario on a NUC7i7DNH

Thanks Geoffroy. Odd that the same version of the compiler works with Clear Linux ?

Cheers Steve.

On Fri, 10 Jan 2020 at 12:23, Geoffroy Van Cutsem <geoffroy.vancutsem@... <mailto:geoffroy.vancutsem@...>> wrote:

    Thanks Steve, I can confirm that I see the problem too using Ubuntu
    19.10 (gcc 9.2.1). Ubuntu 19.04 compiles ACRN correctly (uses gcc
    8.3.0). I have filed an issue to track this:
    https://github.com/projectacrn/acrn-hypervisor/issues/4358.

    Geoffroy

    *From:* acrn-users@...
    <mailto:acrn-users@...>
    <acrn-users@...
    <mailto:acrn-users@...>> *On Behalf Of *Steve Osselton
    *Sent:* Friday, January 10, 2020 11:33 AM
    *To:* acrn-users@...
    <mailto:acrn-users@...>
    *Subject:* Re: [acrn-users] Problem with Industry Scenario on a
    NUC7i7DNH

    Hi Geoffroy,

    Build from a clean checkout on 19.10 (not 19.04) with:

    make all BOARD=nuc7i7dnb  SCENARIO=industry RELEASE=0

    Default gcc is 9.2.1

    Cheers Steve.

    On Fri, 10 Jan 2020 at 10:24, Geoffroy Van Cutsem
    <geoffroy.vancutsem@... <mailto:geoffroy.vancutsem@...>>
    wrote:

        We use a gcc 9.2.1 in Clear Linux to build ACRN these days, I’ll
        see if I can spin up an Ubuntu 19.04 machine to reproduce the
        failure.

        Geoffroy

        *From:* acrn-users@...
        <mailto:acrn-users@...>
        <acrn-users@...
        <mailto:acrn-users@...>> *On Behalf Of *Steve
        Osselton
        *Sent:* Friday, January 10, 2020 11:03 AM
        *To:* acrn-users@...
        <mailto:acrn-users@...>
        *Subject:* Re: [acrn-users] Problem with Industry Scenario on a
        NUC7i7DNH

        Hi Liu,

        Followed exactly these instructions, except on Ubuntu 19.10. To
        me looks like a gcc 9 or config issue as clearly

        doesn't like ‘-mindirect-branch’ and ‘-fcf-protection’ used
        together. Am building a Clear Linux VM now ....

        Cheers Steve

        On Fri, 10 Jan 2020 at 09:45, Liu, Fuzhong
        <fuzhong.liu@... <mailto:fuzhong.liu@...>> wrote:

            Hi Steve,

            We don’t met this issue with Ubuntu 18.04.

            Do you install following package? As mentioned on:
https://projectacrn.github.io/1.5/getting-started/building-from-source.html

            BR.

            Fuzhong

            *From:* acrn-users@...
            <mailto:acrn-users@...>
            <acrn-users@...
            <mailto:acrn-users@...>> *On Behalf Of
            *Steve Osselton
            *Sent:* Friday, January 10, 2020 5:33 PM
            *To:* acrn-users@...
            <mailto:acrn-users@...>
            *Subject:* Re: [acrn-users] Problem with Industry Scenario
            on a NUC7i7DNH

            Hi Yin,

            Will validate whether suggested quick fix works. Looks like
            not happy building with gcc 9 (am using Ubuntu 19.10):

            cc  -Iinclude  -Iinclude/lib  -Iinclude/lib/crypto
              -Iinclude/common  -Iinclude/arch/x86
              -Iinclude/arch/x86/boot  -Iinclude/arch/x86/guest
              -Iinclude/arch/x86/lib  -Iinclude/debug  -Iinclude/public
              -Iinclude/dm  -Iinclude/hw  -Iboot/include
              -Iboot/include/guest
              -I/home/steve/git/acrn-hypervisor/build/hypervisor/include
              -Iarch/x86/configs/nuc7i7dnb  -Iscenarios/industry -I. -c
              -include
/home/steve/git/acrn-hypervisor/build/hypervisor/include/config.h
            -Wall -W -ffunction-sections -fdata-sections -fshort-wchar
            -ffreestanding -fsigned-char -m64 -mno-mmx -mno-sse
            -mno-sse2 -mno-80387 -mno-fp-ret-in-387 -mno-red-zone
            -mpopcnt -nostdinc -nostdlib -fno-common -Werror -O2 -fpie
            -fstack-protector-strong -DSTACK_PROTECTOR
            -mindirect-branch=thunk-extern -mindirect-branch-register
            -DCONFIG_RETPOLINE -DHV_DEBUG -DPROFILING_ON
            -fno-omit-frame-pointer  -gdwarf-2 pre_build/static_checks.c
            -o
/home/steve/git/acrn-hypervisor/build/hypervisor/pre_build/static_checks.o
            -MMD -MT
/home/steve/git/acrn-hypervisor/build/hypervisor/pre_build/static_checks.o
            In file included from pre_build/static_checks.c:7:
            include/lib/util.h: In function ‘mem_aligned_check’:
            include/lib/util.h:30:1: error: ‘-mindirect-branch’ and
            ‘-fcf-protection’ are not compatible

            Should I submit this as a bug report? Will try building on
            Clear Linux ...

            Cheers Steve.

            On Fri, 10 Jan 2020 at 09:18, Yin, Fengwei
            <fengwei.yin@... <mailto:fengwei.yin@...>> wrote:

                Hi Steve,

                On 2020/1/10 下午5:12, Steve Osselton wrote:
                 > Hi Yin,
                 >
                 > What is the preferred way? Is there any chance of
                having the default
                 > value increased as I guess it would be nice
                 > to support Samsung SSDs "out of the box", without
                having to rebuild.
                The default way is change it through "make menuconfig"
                command and
                rebuild the ACRN. Changing Kconfig is kind of hack here.

                For default value, we need trade off like whether there
                are many such
                devices.

                Could you please try the hack first so we could know
                whether this is
                the root cause of your issue? Thanks.


                Regards
                Yin, Fengwei

                 >
                 > Cheers Steve.
                 >
                 > On Fri, 10 Jan 2020 at 09:06, Yin, Fengwei
                <fengwei.yin@... <mailto:fengwei.yin@...>
                 > <mailto:fengwei.yin@...
                <mailto:fengwei.yin@...>>> wrote:
                 >
                 >     Hi Steve,
                 >
                 >     On 2020/1/10 下午5:02, Steve Osselton wrote:
                 >      > Hi Yin,
                 >      >
                 >      > I have a Samsung PCI SSD installed. Looking at
                it's capabilities
                 >     I get:
                 >      >
                 >      > Capabilities: [b0] MSI-X: Enable+ Count=33 Masked-
                 >      >
                 >      > So looks like this is the problem. It it easy
                to up the number of
                 >      > supported MSIX tables?
                 >     The easiest way to try is change the file:
                 >          hypervisor/arch/x86/Kconfig
                 >     which has:
                 >          config MAX_MSIX_TABLE_NUM
                 >               int "Maximum number of MSI-X tables per
                device"
                 >               range 1 2048
                 >               default 16  ---> 64
                 >
                 >     You could try it as quick check. It's not the
                preferred way...
                 >
                 >
                 >     Regards
                 >     Yin, Fengwei
                 >
                 >      >
                 >      > Cheers Steve.
                 >      >
                 >      > On Fri, 10 Jan 2020 at 05:00, Yin, Fengwei
                <fengwei.yin@... <mailto:fengwei.yin@...>
                 >     <mailto:fengwei.yin@...
                <mailto:fengwei.yin@...>>
                 >      > <mailto:fengwei.yin@...
                <mailto:fengwei.yin@...>
                <mailto:fengwei.yin@...
                <mailto:fengwei.yin@...>>>> wrote:
                 >      >
                 >      >     Hi Osselton,
                 >      >     It looks like there is one PCI device in
                your system which
                 >     has more than
                 >      >     16 MSIX table.
                 >      >
                 >      >     Can you boot to native linux and dump the
                MSI table of each
                 >     PCI devices
                 >      >     on your board?
                 >      >
                 >      >
                 >      >     Regards
                 >      >     Yin, Fengwei
                 >      >
                 >      >     On 2020/1/9 下午6:30, Steve Osselton wrote:
                 >      >      > Hi Liz,
                 >      >      >
                 >      >      > Was just following the instructions for
                example industry
                 >     scenario
                 >      >     for
                 >      >      > the 1.4 release, didn't build anything
                myself.
                 >      >      > Have tried again from scratch using
                latest updates for 1.5:
                 >      >      >
                 >      >      > 1 -
                Installed clear-32030-live-server.iso onto hard disk.
                 >      >      > 2 - Boot:
                 >      >      >
                 >      >      > # wget
                 >      >      >
                 >      >
                 >
https://raw.githubusercontent.com/projectacrn/acrn-hypervisor/master/doc/getting-started/acrn_quick_setup.sh
                 >      >      > # chmod +x acrn_quick_setup.sh
                 >      >      > # ./acrn_quick_setup.sh -s 32030 -d -i
                 >      >      > # efibootmgr -c -l '\EFI\acrn\acrn.efi'
                -d /dev/sda -p 1
                 >     -L ACRN -u
                 >      >      > "uart=port@0x3f8"
                 >      >      >
                 >      >      > 3 - Reboot:
                 >      >      >
                 >      >      > Console output:
                 >      >      >
                 >      >      > ACRN Hypervisor
                 >      >      > calibrate_tsc, tsc_khz=2100000
                 >      >      >
                [11187656us][cpu=0][(null)][sev=2][seq=1]:HV version
                 >      >      > 1.5-unstable-2020-01-05 11:33:11- DBG
                (daily tag:) build by
                 >      >     mockbuild,
                 >      >      > start time 11171478us
                 >      >      >
                [11201829us][cpu=0][(null)][sev=2][seq=2]:API version 1.0
                 >      >      >
                [11207570us][cpu=0][(null)][sev=2][seq=3]:Detect
                 >     processor: Intel(R)
                 >      >      > Core(TM) i7-8650U CPU @ 1.90GHz
                 >      >      >
                [11218392us][cpu=0][(null)][sev=2][seq=4]:Assertion failed
                 >     in file
                 >      >      > hw/pci.c,line 448 : fatal error
                 >      >      >
                 >      >      > Still failing ...
                 >      >      >
                 >      >      > Cheers Steve.
                 >      >      >
                 >      >      >
                 >      >      > On Thu, 9 Jan 2020 at 02:40, Liu, Fuzhong
                 >     <fuzhong.liu@...
                <mailto:fuzhong.liu@...>
                <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>
                 >      >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>
                <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>>
                 >      >      > <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>
                 >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>
                <mailto:fuzhong.liu@... <mailto:fuzhong.liu@...>
                 >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>>>> wrote:
                 >      >      >
                 >      >      >     Hi Steve____
                 >      >      >
                 >      >      >     Seem like, you are using very old
                codebase, can you
                 >     try with v1.4
                 >      >      >     codebase first?____
                 >      >      >
                 >      >      >     If not; please share the commit id.____
                 >      >      >
                 >      >      >     Thanks!____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     BR.____
                 >      >      >
                 >      >      >     Fuzhong____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     *From:*
                acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>
                 >      >      >  <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>>
                 >      >      >     <acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>
                 >      >      >  <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>>> *On Behalf
                Of *Steve
                 >      >     Osselton
                 >      >      >     *Sent:* Wednesday, January 8, 2020
                11:34 PM
                 >      >      >     *To:*
                acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>
                 >      >      >  <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>>
                 >      >      >     *Subject:* Re: [acrn-users] Problem
                with Industry
                 >     Scenario on a
                 >      >      >     NUC7i7DNH____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     Hi Liu,____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     On inspection was running an old
                BIOS version, have
                 >     upgraded to
                 >      >      >     latest version (DNi70067) and
                disabled hyper
                 >     threading.____
                 >      >      >
                 >      >      >     Things are improved in that I now
                get diagnostics from
                 >     the serial
                 >      >      >     port:____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     ACRN Hypervisor
                 >      >      >     calibrate_tsc, tsc_khz=2100000
                 >      >      >  [11041475us][cpu=0][sev=2][seq=1]:HV version
                 >      >     1.4-unstable-2019-11-18
                 >      >      >     09:13:55- DBG (daily tag:) build by
                mockbuild, start time
                 >      >     11028016us
                 >      >      >  [11054921us][cpu=0][sev=2][seq=2]:API version 1.0
                 >      >      >  [11059914us][cpu=0][sev=2][seq=3]:Detect processor:
                 >     Intel(R)
                 >      >      >     Core(TM) i7-8650U CPU @ 1.90GHz
                 >      >      >  [11069846us][cpu=0][sev=2][seq=4]:Assertion failed in file
                 >      >      >     hw/pci.c,line 250 : fatal error____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     So still failing.____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     Also as a comment, the BIOS
                settings descriptions, are
                 >     badly
                 >      >     out of
                 >      >      >     step with what you see from the
                latest visual BIOS.____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     Cheers Steve.____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     On Wed, 8 Jan 2020 at 13:46, Liu,
                Fuzhong
                 >      >     <fuzhong.liu@...
                <mailto:fuzhong.liu@...>
                <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>
                 >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>
                <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>>
                 >      >      >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>
                 >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>
                 >      >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>
                 >     <mailto:fuzhong.liu@...
                <mailto:fuzhong.liu@...>>>>> wrote:____
                 >      >      >
                 >      >      >         Hi Steve,____
                 >      >      >
                 >      >      >         Please disable Hyper thread in
                BIOS for
                 >     workaround.____
                 >      >      >
                 >      >      >         This issue is fixed in  v1.5;
                will release soon.____
                 >      >      >
                 >      >      >         ____
                 >      >      >
                 >      >      >         BR.____
                 >      >      >
                 >      >      >         Fuzhong____
                 >      >      >
                 >      >      >         ____
                 >      >      >
                 >      >      >         *From:*
                acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>
                 >      >      >  <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>>
                 >      >      >  <acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>
                 >      >      >  <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>>> *On Behalf
                Of *Steve
                 >      >      >         Osselton
                 >      >      >         *Sent:* Wednesday, January 8,
                2020 8:32 PM
                 >      >      >         *To:*
                acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>
                 >      >      >  <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>
                 >      >     <mailto:acrn-users@...
                <mailto:acrn-users@...>
                 >     <mailto:acrn-users@...
                <mailto:acrn-users@...>>>>
                 >      >      >         *Subject:* [acrn-users] Problem
                with Industry
                 >     Scenario on a
                 >      >      >         NUC7i7DNH____
                 >      >      >
                 >      >      >         ____
                 >      >      >
                 >      >      >         Hi,
                 >      >      >
                 >      >      >         Am essentially following the
                deployment
                 >     instructions for the
                 >      >      >         ACRN industry scenario
                 >      >      >         using a NUC7i7DNH with a serial
                port connected.
                 >      >      >
                 >      >      >
                 >
https://projectacrn.github.io/latest/getting-started/rt_industry.html
                 >      >      >
                 >      >      >  Flashing clear-31670-live-server.iso worked fine. Then
                 >      >     followed
                 >      >      >         the "Use the pre-installed
                industry ACRN hypervisor"
                 >      >      >         section and installed serial
                port support. Again
                 >     this all
                 >      >     worked
                 >      >      >         fine and looked good.
                 >      >      >
                 >      >      >         I then rebooted the NUC. At
                HDMI console get
                 >     "Intel NUC"
                 >      >     banner
                 >      >      >         with the following at top LHS
                of screen:
                 >      >      >
                 >      >      >         detected 8 processes, 8 enabled
                 >      >      >         current on process 0
                 >      >      >
                 >      >      >         And on serial port:
                 >      >      >
                 >      >      >         E0S0 S0
                 >      >      >
                 >      >      >         Not sure what has happened
                here. Am I missing
                 >     something
                 >      >     obvious ?
                 >      >      >
                 >      >      >         Thanks Steve ____
                 >      >      >
                 >      >      >
                 >      >      >     ____
                 >      >      >
                 >      >      >     __ __
                 >      >      >
                 >      >      >     -- ____
                 >      >      >
                 >      >      >     Engineering Director____
                 >      >      >
                 >      >      >     IOTech Systems Ltd.____
                 >      >      >
                 >      >      >     __
                 >      >      >
                 >      >      >
                 >      >      >
                 >      >      > --
                 >      >      > Engineering Director
                 >      >      > IOTech Systems Ltd.
                 >      >      >
                 >      >
                 >      >
                 >      >
                 >      >
                 >      >
                 >      > --
                 >      > Engineering Director
                 >      > IOTech Systems Ltd.
                 >      >
                 >
                 >
                 >
                 >
                 >
                 > --
                 > Engineering Director
                 > IOTech Systems Ltd.
                 >


            --
            Engineering Director

            IOTech Systems Ltd.


        --
        Engineering Director

        IOTech Systems Ltd.


    --
    Engineering Director

    IOTech Systems Ltd.


--

Engineering Director

IOTech Systems Ltd.

Join acrn-users@lists.projectacrn.org to automatically receive all group messages.