qemu-riscv
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-riscv] [Qemu-devel] [PATCH v2] tests/boot_linux_console: add a


From: Chih-Min Chao
Subject: Re: [Qemu-riscv] [Qemu-devel] [PATCH v2] tests/boot_linux_console: add a test for riscv64 + virt
Date: Fri, 19 Jul 2019 01:10:48 +0800



On Thu, Jul 18, 2019 at 11:34 PM Alistair Francis <address@hidden> wrote:
On Thu, Jul 18, 2019 at 8:00 AM Chih-Min Chao <address@hidden> wrote:
>
>
>
> On Thu, Jul 18, 2019 at 6:47 AM Alistair Francis <address@hidden> wrote:
>>
>> On Tue, Jul 16, 2019 at 10:21 PM Chih-Min Chao <address@hidden> wrote:
>> >
>> >
>> >
>> > On Wed, Jul 17, 2019 at 6:59 AM Alistair Francis <address@hidden> wrote:
>> >>
>> >> On Tue, Jul 16, 2019 at 2:50 PM Philippe Mathieu-Daudé
>> >> <address@hidden> wrote:
>> >> >
>> >> > On 7/16/19 10:43 PM, Alistair Francis wrote:
>> >> > > On Tue, Jul 16, 2019 at 6:56 AM Chih-Min Chao <address@hidden> wrote:
>> >> > >>
>> >> > >>
>> >> > >> On Tue, Jul 16, 2019 at 12:34 AM Alistair Francis <address@hidden> wrote:
>> >> > >>>
>> >> > >>> On Mon, Jul 15, 2019 at 6:02 AM Philippe Mathieu-Daudé
>> >> > >>> <address@hidden> wrote:
>> >> > >>>>
>> >> > >>>> On 7/15/19 1:09 PM, Chih-Min Chao wrote:
>> >> > >>>>> On Mon, Jul 15, 2019 at 5:15 PM Philippe Mathieu-Daudé
>> >> > >>>>> <address@hidden <mailto:address@hidden>> wrote:
>> >> > >>>>>
>> >> > >>>>>     On 7/15/19 11:08 AM, Chih-Min Chao wrote:
>> >> > >>>>>     > Similar to the mips + malta test, it boots a Linux kernel on a virt
>> >> > >>>>>     > board and verify the serial is working.  Also, it relies on the serial
>> >> > >>>>>     > device set by the machine itself.
>> >> > >>>>>     >
>> >> > >>>>>     > If riscv64 is a target being built, "make check-acceptance" will
>> >> > >>>>>     > automatically include this test by the use of the "arch:riscv64" tags.
>> >> > >>>>>     >
>> >> > >>>>>     > Alternatively, this test can be run using:
>> >> > >>>>>     >
>> >> > >>>>>     >   $ avocado run -t arch:riscv64 tests/acceptance
>> >> > >>>>>     >
>> >> > >>>>>     > packages
>> >> > >>>>>     >   debian official
>> >> > >>>>>     >     binutils-riscv64-linux-gnu_2.32-8
>> >> > >>>>>     >     opensbi_0.4-1_all
>> >> > >>>>>     >     linux-image-4.19.0-5-riscv64 4.19.37-4
>> >> > >>>>>     >   third-party
>> >> > >>>>>     >
>> >> > >>>>>      https://github.com/groeck/linux-build-test/rootfs/riscv64/rootfs.cpio.gz
>> >> > >>>>>     >     (the repo is also used in mips target acceptance)
>> >> > >>>>>     >
>> >> > >>>>>     > Signed-off-by: Chih-Min Chao <address@hidden
>> >> > >>>>>     <mailto:address@hidden>>
>> >> > >>>>>     > ---
>> >> > >>>>>     >  .travis.yml                            |  2 +-
>> >> > >>>>>     >  tests/acceptance/boot_linux_console.py | 66
>> >> > >>>>>     ++++++++++++++++++++++++++++++++++
>> >> > >>>>>     >  2 files changed, 67 insertions(+), 1 deletion(-)
>> >> > >>>>>     >
>> >> > >>>>>     > diff --git a/.travis.yml b/.travis.yml
>> >> > >>>>>     > index 5d3d6ee..21fcead 100644
>> >> > >>>>>     > --- a/.travis.yml
>> >> > >>>>>     > +++ b/.travis.yml
>> >> > >>>>>     > @@ -232,7 +232,7 @@ matrix:
>> >> > >>>>>     >
>> >> > >>>>>     >      # Acceptance (Functional) tests
>> >> > >>>>>     >      - env:
>> >> > >>>>>     > -        - CONFIG="--python=/usr/bin/python3
>> >> > >>>>>     --target-list=x86_64-softmmu,mips-softmmu,mips64el-softmmu,aarch64-softmmu,arm-softmmu,s390x-softmmu,alpha-softmmu"
>> >> > >>>>>     > +        - CONFIG="--python=/usr/bin/python3
>> >> > >>>>>     --target-list=x86_64-softmmu,mips-softmmu,mips64el-softmmu,aarch64-softmmu,arm-softmmu,s390x-softmmu,alpha-softmmu,riscv64-softmmu"
>> >> > >>>>>     >          - TEST_CMD="make check-acceptance"
>> >> > >>>>>     >        after_failure:
>> >> > >>>>>     >          - cat tests/results/latest/job.log
>> >> > >>>>>     > diff --git a/tests/acceptance/boot_linux_console.py
>> >> > >>>>>     b/tests/acceptance/boot_linux_console.py
>> >> > >>>>>     > index 3215950..0f638bc 100644
>> >> > >>>>>     > --- a/tests/acceptance/boot_linux_console.py
>> >> > >>>>>     > +++ b/tests/acceptance/boot_linux_console.py
>> >> > >>>>>     > @@ -354,3 +354,69 @@ class BootLinuxConsole(Test):
>> >> > >>>>>     >          self.vm.launch()
>> >> > >>>>>     >          console_pattern = 'Kernel command line: %s' %
>> >> > >>>>>     kernel_command_line
>> >> > >>>>>     >          self.wait_for_console_pattern(console_pattern)
>> >> > >>>>>     > +
>> >> > >>>>>     > +    def test_riscv64_virt(self):
>> >> > >>>>>     > +        """
>> >> > >>>>>     > +        :avocado: tags=arch:riscv64
>> >> > >>>>>     > +        :avocado: tags=machine:virt
>> >> > >>>>>     > +        """
>> >> > >>>>>     > +        deb_url = ('https://snapshot.debian.org/archive/debian/'
>> >> > >>>>>     > +                         '20190424T171759Z/pool/main/b/binutils/'
>> >> > >>>>>     > +
>> >> > >>>>>      'binutils-riscv64-linux-gnu_2.32-8_amd64.deb')
>> >> > >>>>>     > +        deb_hash = ('7fe376fd4452696c03acd508d6d613ca553ea15e')
>> >> > >>>>>     > +        deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash)
>> >> > >>>>>     > +        objcopy_path = '/usr/bin/riscv64-linux-gnu-objcopy'
>> >> > >>>>>     > +        objcopy_path = self.extract_from_deb(deb_path, objcopy_path)
>> >> > >>>>>     > +        libbfd_path =
>> >> > >>>>>     '/usr/lib/x86_64-linux-gnu/libbfd-2.32-riscv64.so
>> >> > >>>>>     <http://libbfd-2.32-riscv64.so>'
>> >> > >>>>>     > +        libbfd_path = self.extract_from_deb(deb_path, libbfd_path)
>> >> > >>>>>     > +        process.run('ls -al %s' % (objcopy_path))
>> >> > >>>>>     > +
>> >> > >>>>>     > +        deb_url = ('https://snapshot.debian.org/archive/debian/'
>> >> > >>>>>     > +                   '20190708T032337Z/pool/main/o/opensbi/'
>> >> > >>>>>     > +                   'opensbi_0.4-1_all.deb')
>> >> > >>>>>     > +        deb_hash = ('2319dcd702958291d323acf5649fd98a11d90112')
>> >> > >>>>>     > +        deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash)
>> >> > >>>>>     > +        opensbi_path = ('/usr/lib/riscv64-linux-gnu/opensbi/'
>> >> > >>>>>     > +                        'qemu/virt/fw_jump.elf')
>> >> > >>>>>     > +        opensbi_path = self.extract_from_deb(deb_path, opensbi_path)
>> >> > >>>>>     > +
>> >> > >>>>>     > +        deb_url =
>> >> > >>>>>     ('https://snapshot.debian.org/archive/debian-ports/'
>> >> > >>>>>     > +                   '20190620T095935Z/pool-riscv64/main/l/linux/'
>> >> > >>>>>     > +
>> >> > >>>>>      'linux-image-4.19.0-5-riscv64_4.19.37-4_riscv64.deb')
>> >> > >>>>>     > +        deb_hash = ('bf5b5680c41d92134d22caef4fbd277c5217e1f0')
>> >> > >>>>>     > +        deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash)
>> >> > >>>>>     > +        kernel_path = '/boot/vmlinux-4.19.0-5-riscv64'
>> >> > >>>>>     > +        kernel_path = self.extract_from_deb(deb_path, kernel_path)
>> >> > >>>>>     > +        kimage_path = self.workdir + "/Image"
>> >> > >>>>>     > +        env = os.environ
>> >> > >>>>>     > +        env['LD_LIBRARY_PATH'] = ('%s:' %
>> >> > >>>>>     (os.path.dirname(libbfd_path)) +
>> >> > >>>>>     > +                                 env.get('LD_LIBRARY_PATH', ''))
>> >> > >>>>>     > +        process.run(('%s -O binary -O binary -R'
>> >> > >>>>>     > +                     '.note -R .note.gnu.build-id -R .comment -S
>> >> > >>>>>     %s %s') %
>> >> > >>>>>     > +                     (objcopy_path, kernel_path, kimage_path))
>> >> > >>>>>
>> >> > >>>>>     Please explain why you need to do that...
>> >> > >>>>>
>> >> > >>>>>     Also note these tests are run on other host architecture than x86, at
>> >> > >>>>>     least ppc64/s390x (so this won't work there).
>> >> > >>>>>
>> >> > >>>>> it is because riscv64 vmlinux doesn't have physical loading address
>> >> > >>>>> information and
>> >> > >>>>> depends on certain stage bootloader to move kernel raw data to specific
>> >> > >>>>> physical address (0x8020_0000)
>> >> > >>>>>
>> >> > >>>>> The vmlinux program headers are
>> >> > >>>>>   Type             Offset       VirtAddr                  PhysAddr
>> >> > >>>>>     FileSiz  MemSiz   Flg Align
>> >> > >>>>>   LOAD           0x001000 0xffffffe000000000 0x0000000000000000 0x0303a6
>> >> > >>>>> 0x0303a6 R E 0x1000
>> >> > >>>>>   LOAD           0x032000 0xffffffe000031000 0x0000000000031000 0x828f0c
>> >> > >>>>> 0x828f0c RWE 0x1000
>> >> > >>>>>   NOTE           0x85aec0 0xffffffe000859ec0 0x0000000000859ec0 0x00004c
>> >> > >>>>> 0x00004c R   0x4
>> >> > >>>>>
>> >> > >>>>> For legacy bootloader, bbl, It can handle payload in ELF format.
>> >> > >>>>> However, for the newer OpenSBI, it can only handle raw data and
>> >> > >>>>> debian doesn't provide linux image in raw data format.
>> >> > >>>
>> >> > >>> This doesn't sound right. You should just be able to say -kernel
>> >> > >>> <anything> and it should work. It has worked for me with vmlinux and
>> >> > >>> Image files when using master (not the 4.0 release but master/4.1).
>> >> > >>> Although all of my testing was with the 5.1 kernel, so maybe there is
>> >> > >>> a difference there?
>> >> > >>>
>> >> > >>> What isn't working for you? Can you include the errors and output of -d in_asm?
>> >> > >>>
>> >> > >>>
>> >> > >>> Alistair
>> >> > >>
>> >> > >>
>> >> > >> Hi Alistair,
>> >> > >>    I have come across error before starting target simulation.  What I means is to execute
>> >> > >>             qemu-system-riscv64 -M virt -m 256M -nographic -bios ./opensbi/build/platform/qemu/virt/firmware/fw_jump.elf -kernel vmlinux
>> >> > >>    then get the error message
>> >> > >>             rom: requested regions overlap (rom mrom.reset. free=0x000000000001eb7c, addr=0x0000000000001000)
>> >> > >>             qemu-system-riscv64: rom check and register reset failed
>> >> > >
>> >> > > Hmmm.... I don't remember seeing this problem in my testing, but it
>> >> > > does seem to make sense.
>> >> > >
>> >> > > We have two options here:
>> >> > > 1. Use the Image file from Linux 5.0+ (my preferred option as 4.19 is
>> >> > > getting old)
>> >
>> >  I choose 4.19 version rather than 5.0+  because 4.19 is stable branch maintained by Greg.
>> >  But it is ok to use 5.0 trunk, the latest version provided by debian.
>> >
>> >>
>> >> > > 2. Use the device loader (documented here:
>> >> > > https://github.com/riscv/opensbi/blob/master/docs/platform/qemu_virt.md)
>> >
>> >
>> > The document describes two user cases
>> >   1. fw_payload.elf
>> >       a. include the Image (raw data file, not elf)  into data section
>> >       b. -kernel load the each part to the correct address
>> >
>> >      the program header of  fw_payload.elf are
>> >      Type           Offset   VirtAddr           PhysAddr           FileSiz  MemSiz   Flg Align
>> >     LOAD           0x0000b0 0x0000000080000000 0x0000000080000000 0x00a008 0x00b0b8 RWE 0x10               <-  fw_payload  main program
>> >     LOAD           0x00a0b8 0x0000000080200000 0x0000000080200000 0x8dbe20 0x8dbe20 R E 0x1                    <-  arch/riscv/boot/Image   or u-boot.bin
>> >
>> >   2. fw_jump.elf
>> >       a. use  "-kernel <raw image> or -device  loader,file=<raw Image>,addr=0x80200000" to load correct address
>> >       b. fw_jump.elf just jump to the entry address 0x80200000
>> >
>> >
>> >  The issues are
>> >    a. debian linux-image package only provides vmlinux.
>>
>> This can be fixed by upgrading the kernel version. Fedora should have
>> a 5.2 kernel that you can use instead of Debian.
>>
>> >        so I use objcopy in acceptance test
>> >    b. riscv64 vmlinux doesn't have correct PhysAddr and -kernel can't move data to expected address
>> >        explicit load the Image to specific
>> >            -kernel raw_image   (if file is not elf or uImage,  riscv_kerner_loader treat it is as raw image and place it at 0x8020_0000)
>> >            -device loader,file=<raw_image>,0x80200000  (do the same thing)
>>
>> Does the device loader not work?
>>
>> Alistair
>
>
>   The device loader works for "Image" case (raw data)
>   If the data is raw image,  for riscv_load_kernel implementation,   "-kernel <raw_image>" and  -device loader,file=<raw_image>,0x80200000  are the same.
>   They finally call "load_image_targphys_as"
>   The only difference is that target address is hardcoded or feed from command line option.
>
>   I don't prefe one than the other.  Both of them are Ok to me.  I could refine that if one of them is more preferable.

I'm a little confused with what you mean, so if you have something
that will work just send a new version and we can discuss it there.

PS: I'm still in favour of just using a newer kernel and a Image file.

Alistair

 ok, I will use the latest kernel available in Debian package and refine the command in next version.
 
chihmin.chao
 
>
>
>  chihmin.chao
>>
>>
>> >
>> >   ELF should work only if it has correct loading information.
>> >
>> >   chihmin.chao
>> >>
>> >> >
>> >> > Nice doc, this example is exactly what I had in mind:
>> >> >
>> >> >   qemu-system-riscv64 -M virt -m 256M -nographic \
>> >> >         -kernel build/platform/qemu/virt/firmware/fw_jump.elf \
>> >> >         -device
>> >> > loader,file=<linux_build_directory>/arch/riscv/boot/Image,addr=0x80200000 \
>> >> >         -drive file=<path_to_linux_rootfs>,format=raw,id=hd0 \
>> >> >         -device virtio-blk-device,drive=hd0 \
>> >> >         -append "root=/dev/vda rw console=ttyS0"
>> >> >
>> >> > If the Image is an ELF, I assume we don't need to specify the address,
>> >> > because the device loader knows how to parse this format. Again,
>> >> > Alistair is the expert ;)
>> >>
>> >> Yep, an elf should just work.
>> >>
>> >  describe the issue above
>> >
>> >>
>> >> >
>> >> > If for some reason it only works with Linux 5.0+, let's use these!
>> >>
>> >> Only 5.0+ builds an elf.
>> >>
>> >> Alistair
>> >>
>> >> >
>> >> > Regards,
>> >> >
>> >> > Phil.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]