| Message ID | cover.1698406922.git.balaton@eik.bme.hu (mailing list archive) |
|---|---|
| Headers | show |
| Series | Add emulation of AmigaOne XE board | expand |
Am 27. Oktober 2023 11:54:48 UTC schrieb BALATON Zoltan <balaton@eik.bme.hu>: >Changes in v7: >- Increase default memory size to 512m to match pegasos2 and sam460ex >and it's a better default for AmigaOS > >Changes in v6: >- Dropped patch 1, now it's > >Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> > >([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI IDE controllers) >- Added Tested-by from Rene > >Changes in v5: >- Fixed avocado test > >Changes in v4: >- Found typo in comment in patch 1 so ended up rewording it again >trying to make it more concise. Also take the idea of using >range_covers_byte from Mark's patch >- Added RFC patch for avocado test (untested, I don't have Avocado) > >Changes in v3: >- Update values, comment and commit message in patch 1 again > >Changes in v2: >- Update comment and commit message in patch 1 (Mark) >- Fix irq mapping in patch 2 (Volker) > >Regards, >BALATON Zoltan > >BALATON Zoltan (3): > hw/pci-host: Add emulation of Mai Logic Articia S > hw/ppc: Add emulation of AmigaOne XE board > tests/avocado: Add test for amigaone board Hi Zoltan, Could you please provide some documentation on how to run Linux or some other free (as in free beer) OS on this machine? Can you provide an Avocado test booting e.g. Linux as a quality gate for the VIA south bridges? As you know I'm trying to bring these south bridges to the PC machine and relying on hard to get proprietary blobs or OSes makes this very difficult. Whenever the VIA south bridges are changed we end up having endless discussions due to this situation. We need a solution that works for everybody woking on these south bridges. Best regards, Bernhard > > MAINTAINERS | 8 + > configs/devices/ppc-softmmu/default.mak | 1 + > hw/pci-host/Kconfig | 5 + > hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ > hw/pci-host/meson.build | 2 + > hw/ppc/Kconfig | 7 + > hw/ppc/amigaone.c | 164 +++++++++++++ > hw/ppc/meson.build | 2 + > include/hw/pci-host/articia.h | 17 ++ > tests/avocado/ppc_amiga.py | 38 +++ > 10 files changed, 537 insertions(+) > create mode 100644 hw/pci-host/articia.c > create mode 100644 hw/ppc/amigaone.c > create mode 100644 include/hw/pci-host/articia.h > create mode 100644 tests/avocado/ppc_amiga.py >
On Sat, 28 Oct 2023, Bernhard Beschow wrote: > Am 27. Oktober 2023 11:54:48 UTC schrieb BALATON Zoltan <balaton@eik.bme.hu>: >> Changes in v7: >> - Increase default memory size to 512m to match pegasos2 and sam460ex >> and it's a better default for AmigaOS >> >> Changes in v6: >> - Dropped patch 1, now it's >> >> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >> >> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI IDE controllers) >> - Added Tested-by from Rene >> >> Changes in v5: >> - Fixed avocado test >> >> Changes in v4: >> - Found typo in comment in patch 1 so ended up rewording it again >> trying to make it more concise. Also take the idea of using >> range_covers_byte from Mark's patch >> - Added RFC patch for avocado test (untested, I don't have Avocado) >> >> Changes in v3: >> - Update values, comment and commit message in patch 1 again >> >> Changes in v2: >> - Update comment and commit message in patch 1 (Mark) >> - Fix irq mapping in patch 2 (Volker) >> >> Regards, >> BALATON Zoltan >> >> BALATON Zoltan (3): >> hw/pci-host: Add emulation of Mai Logic Articia S >> hw/ppc: Add emulation of AmigaOne XE board >> tests/avocado: Add test for amigaone board > > Hi Zoltan, > > Could you please provide some documentation on how to run Linux or some > other free (as in free beer) OS on this machine? There are some Linux images here that should work on amigaone: https://sourceforge.net/projects/amigaone-linux/files/debian-installer/ To boot it get the firmware as described in the comment in amigaone.c or in the avocado test then run: qemu-system-ppc -M amigaone -bios u-boot-amigaone.bin \ -cdrom "A1 Linux Net Installer.iso" -serial stdio \ -device ati-vga,model=rv100,romfile=VGABIOS-lgpl-latest.bin Then from the firmware menu select Boot sequence, Amiga Multiboot Options and set Boot device 1 to VIA CDROM then escape back to top level and from the exit menu that appears with escape on the main screen select either save settings or use settings for current boot. It takes a long time at Loading kernel... but eventually boots and you should see the installer (or select rescue from the Linux boot menu then at language selection Tab to Back button then run shell to get a prompt). > Can you provide an > Avocado test booting e.g. Linux as a quality gate for the VIA south > bridges? I don't know how to automate the above with avocado which seems to run with -display none but the Linux iso has hard coded radeonfb so I could only do a basic test with the firmware. > As you know I'm trying to bring these south bridges to the PC machine > and relying on hard to get proprietary blobs or OSes makes this very > difficult. I know that the machines and guests I work with are a bit obscure but I can't change that. These machines also run Linux and some MorphOS too which can be a test case that's available, unfortunately amigaone is not supported by MorphOS so only Linux and AmigaOS is available there. > Whenever the VIA south bridges are changed we end up having > endless discussions due to this situation. We need a solution that works > for everybody woking on these south bridges. I afree with that and I think I wasn't the one who started endless discussions about every little change to the via model so I think we can find a solution that works for all. I'd just like to keep pegasos2 and amigaone working with AmigaOS which is my main goal and also keep the device model simple and managable. Regards, BALATON Zoltan
On Fri, 27 Oct 2023, BALATON Zoltan wrote: > Changes in v7: > - Increase default memory size to 512m to match pegasos2 and sam460ex > and it's a better default for AmigaOS > > Changes in v6: > - Dropped patch 1, now it's > > Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> > > ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI IDE controllers) > - Added Tested-by from Rene > > Changes in v5: > - Fixed avocado test > > Changes in v4: > - Found typo in comment in patch 1 so ended up rewording it again > trying to make it more concise. Also take the idea of using > range_covers_byte from Mark's patch > - Added RFC patch for avocado test (untested, I don't have Avocado) > > Changes in v3: > - Update values, comment and commit message in patch 1 again > > Changes in v2: > - Update comment and commit message in patch 1 (Mark) > - Fix irq mapping in patch 2 (Volker) > > Regards, > BALATON Zoltan > > BALATON Zoltan (3): > hw/pci-host: Add emulation of Mai Logic Articia S > hw/ppc: Add emulation of AmigaOne XE board > tests/avocado: Add test for amigaone board Nick, Will you please send a pull request with this now? It's independent of the IDE fix which as a bugfix so can wait a little more but this series should be merged before the freeze starts tomorrow. (As this adds a new machine and does not touch anything else it can't break anything either.) Regards, BALATON Zoltan > MAINTAINERS | 8 + > configs/devices/ppc-softmmu/default.mak | 1 + > hw/pci-host/Kconfig | 5 + > hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ > hw/pci-host/meson.build | 2 + > hw/ppc/Kconfig | 7 + > hw/ppc/amigaone.c | 164 +++++++++++++ > hw/ppc/meson.build | 2 + > include/hw/pci-host/articia.h | 17 ++ > tests/avocado/ppc_amiga.py | 38 +++ > 10 files changed, 537 insertions(+) > create mode 100644 hw/pci-host/articia.c > create mode 100644 hw/ppc/amigaone.c > create mode 100644 include/hw/pci-host/articia.h > create mode 100644 tests/avocado/ppc_amiga.py > >
On Mon, 6 Nov 2023, BALATON Zoltan wrote: > On Fri, 27 Oct 2023, BALATON Zoltan wrote: >> Changes in v7: >> - Increase default memory size to 512m to match pegasos2 and sam460ex >> and it's a better default for AmigaOS >> >> Changes in v6: >> - Dropped patch 1, now it's >> >> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >> >> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI >> IDE controllers) >> - Added Tested-by from Rene >> >> Changes in v5: >> - Fixed avocado test >> >> Changes in v4: >> - Found typo in comment in patch 1 so ended up rewording it again >> trying to make it more concise. Also take the idea of using >> range_covers_byte from Mark's patch >> - Added RFC patch for avocado test (untested, I don't have Avocado) >> >> Changes in v3: >> - Update values, comment and commit message in patch 1 again >> >> Changes in v2: >> - Update comment and commit message in patch 1 (Mark) >> - Fix irq mapping in patch 2 (Volker) >> >> Regards, >> BALATON Zoltan >> >> BALATON Zoltan (3): >> hw/pci-host: Add emulation of Mai Logic Articia S >> hw/ppc: Add emulation of AmigaOne XE board >> tests/avocado: Add test for amigaone board > > Nick, > > Will you please send a pull request with this now? It's independent of the > IDE fix which as a bugfix so can wait a little more but this series should be > merged before the freeze starts tomorrow. (As this adds a new machine and > does not touch anything else it can't break anything either.) Is there a PPC pull request in the making with this series in it? It would be really sad to miss the release not being able to merge such a simple series for weeks. As I said this is independent of any other fixes so I'd like to get this in now please. Regards, BALATON Zoltan >> MAINTAINERS | 8 + >> configs/devices/ppc-softmmu/default.mak | 1 + >> hw/pci-host/Kconfig | 5 + >> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >> hw/pci-host/meson.build | 2 + >> hw/ppc/Kconfig | 7 + >> hw/ppc/amigaone.c | 164 +++++++++++++ >> hw/ppc/meson.build | 2 + >> include/hw/pci-host/articia.h | 17 ++ >> tests/avocado/ppc_amiga.py | 38 +++ >> 10 files changed, 537 insertions(+) >> create mode 100644 hw/pci-host/articia.c >> create mode 100644 hw/ppc/amigaone.c >> create mode 100644 include/hw/pci-host/articia.h >> create mode 100644 tests/avocado/ppc_amiga.py >> >> > >
On Tue, 7 Nov 2023, BALATON Zoltan wrote: > On Mon, 6 Nov 2023, BALATON Zoltan wrote: >> On Fri, 27 Oct 2023, BALATON Zoltan wrote: >>> Changes in v7: >>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>> and it's a better default for AmigaOS >>> >>> Changes in v6: >>> - Dropped patch 1, now it's >>> >>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>> >>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI >>> IDE controllers) >>> - Added Tested-by from Rene >>> >>> Changes in v5: >>> - Fixed avocado test >>> >>> Changes in v4: >>> - Found typo in comment in patch 1 so ended up rewording it again >>> trying to make it more concise. Also take the idea of using >>> range_covers_byte from Mark's patch >>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>> >>> Changes in v3: >>> - Update values, comment and commit message in patch 1 again >>> >>> Changes in v2: >>> - Update comment and commit message in patch 1 (Mark) >>> - Fix irq mapping in patch 2 (Volker) >>> >>> Regards, >>> BALATON Zoltan >>> >>> BALATON Zoltan (3): >>> hw/pci-host: Add emulation of Mai Logic Articia S >>> hw/ppc: Add emulation of AmigaOne XE board >>> tests/avocado: Add test for amigaone board >> >> Nick, >> >> Will you please send a pull request with this now? It's independent of the >> IDE fix which as a bugfix so can wait a little more but this series should >> be merged before the freeze starts tomorrow. (As this adds a new machine >> and does not touch anything else it can't break anything either.) > > Is there a PPC pull request in the making with this series in it? It would be > really sad to miss the release not being able to merge such a simple series > for weeks. As I said this is independent of any other fixes so I'd like to > get this in now please. Nick seems to be away so please Cédric, Daniel, Phil or whoever can send a pull request step in and please take care of this. This was posted a month ago with this last version on the list for 1.5 week so I'd hate to miss the release and wait until Easter next year to get this out to users because none of the maintainers are available before and arounf the freeze. Regards, BALATON Zoltan >>> MAINTAINERS | 8 + >>> configs/devices/ppc-softmmu/default.mak | 1 + >>> hw/pci-host/Kconfig | 5 + >>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>> hw/pci-host/meson.build | 2 + >>> hw/ppc/Kconfig | 7 + >>> hw/ppc/amigaone.c | 164 +++++++++++++ >>> hw/ppc/meson.build | 2 + >>> include/hw/pci-host/articia.h | 17 ++ >>> tests/avocado/ppc_amiga.py | 38 +++ >>> 10 files changed, 537 insertions(+) >>> create mode 100644 hw/pci-host/articia.c >>> create mode 100644 hw/ppc/amigaone.c >>> create mode 100644 include/hw/pci-host/articia.h >>> create mode 100644 tests/avocado/ppc_amiga.py >>> >>> >> >> >
On 11/7/23 11:42, BALATON Zoltan wrote: > On Tue, 7 Nov 2023, BALATON Zoltan wrote: >> On Mon, 6 Nov 2023, BALATON Zoltan wrote: >>> On Fri, 27 Oct 2023, BALATON Zoltan wrote: >>>> Changes in v7: >>>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>>> and it's a better default for AmigaOS >>>> >>>> Changes in v6: >>>> - Dropped patch 1, now it's >>>> >>>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>>> >>>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI IDE controllers) >>>> - Added Tested-by from Rene >>>> >>>> Changes in v5: >>>> - Fixed avocado test >>>> >>>> Changes in v4: >>>> - Found typo in comment in patch 1 so ended up rewording it again >>>> trying to make it more concise. Also take the idea of using >>>> range_covers_byte from Mark's patch >>>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>>> >>>> Changes in v3: >>>> - Update values, comment and commit message in patch 1 again >>>> >>>> Changes in v2: >>>> - Update comment and commit message in patch 1 (Mark) >>>> - Fix irq mapping in patch 2 (Volker) >>>> >>>> Regards, >>>> BALATON Zoltan >>>> >>>> BALATON Zoltan (3): >>>> hw/pci-host: Add emulation of Mai Logic Articia S >>>> hw/ppc: Add emulation of AmigaOne XE board >>>> tests/avocado: Add test for amigaone board >>> >>> Nick, >>> >>> Will you please send a pull request with this now? It's independent of the IDE fix which as a bugfix so can wait a little more but this series should be merged before the freeze starts tomorrow. (As this adds a new machine and does not touch anything else it can't break anything either.) >> >> Is there a PPC pull request in the making with this series in it? It would be really sad to miss the release not being able to merge such a simple series for weeks. As I said this is independent of any other fixes so I'd like to get this in now please. > > Nick seems to be away so please Cédric, Daniel, Phil or whoever can send a pull request step in and please take care of this. This was posted a month ago with this last version on the list for 1.5 week so I'd hate to miss the release and wait until Easter next year to get this out to users because none of the maintainers are available before and arounf the freeze. Wait, do you think we would miss code freeze? Code freeze in QEMU is a blast! People getting mad because stuff got left behind, PRs getting remade over and over because gitlab is nagging about something, new bugs introduced, sometimes master doesn't even build the following week. We might miss any other boring date like release date or something, but code freeze? Nah, we're right here man. The reason why we didn't interfere is because it's rude to short-circuit the maintainer on duty. Nick didn't send an email asking for help, so we were just monitoring. I'll consider Nick AWOL and take this PR. And don't worry, it's still sunny in Brazil, we have plenty of time. Thanks, Daniel > > Regards, > BALATON Zoltan > >>>> MAINTAINERS | 8 + >>>> configs/devices/ppc-softmmu/default.mak | 1 + >>>> hw/pci-host/Kconfig | 5 + >>>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>>> hw/pci-host/meson.build | 2 + >>>> hw/ppc/Kconfig | 7 + >>>> hw/ppc/amigaone.c | 164 +++++++++++++ >>>> hw/ppc/meson.build | 2 + >>>> include/hw/pci-host/articia.h | 17 ++ >>>> tests/avocado/ppc_amiga.py | 38 +++ >>>> 10 files changed, 537 insertions(+) >>>> create mode 100644 hw/pci-host/articia.c >>>> create mode 100644 hw/ppc/amigaone.c >>>> create mode 100644 include/hw/pci-host/articia.h >>>> create mode 100644 tests/avocado/ppc_amiga.py >>>> >>>> >>> >>> >>
On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: > On 11/7/23 11:42, BALATON Zoltan wrote: >> On Tue, 7 Nov 2023, BALATON Zoltan wrote: >>> On Mon, 6 Nov 2023, BALATON Zoltan wrote: >>>> On Fri, 27 Oct 2023, BALATON Zoltan wrote: >>>>> Changes in v7: >>>>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>>>> and it's a better default for AmigaOS >>>>> >>>>> Changes in v6: >>>>> - Dropped patch 1, now it's >>>>> >>>>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>>>> >>>>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for >>>>> PCI IDE controllers) >>>>> - Added Tested-by from Rene >>>>> >>>>> Changes in v5: >>>>> - Fixed avocado test >>>>> >>>>> Changes in v4: >>>>> - Found typo in comment in patch 1 so ended up rewording it again >>>>> trying to make it more concise. Also take the idea of using >>>>> range_covers_byte from Mark's patch >>>>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>>>> >>>>> Changes in v3: >>>>> - Update values, comment and commit message in patch 1 again >>>>> >>>>> Changes in v2: >>>>> - Update comment and commit message in patch 1 (Mark) >>>>> - Fix irq mapping in patch 2 (Volker) >>>>> >>>>> Regards, >>>>> BALATON Zoltan >>>>> >>>>> BALATON Zoltan (3): >>>>> hw/pci-host: Add emulation of Mai Logic Articia S >>>>> hw/ppc: Add emulation of AmigaOne XE board >>>>> tests/avocado: Add test for amigaone board >>>> >>>> Nick, >>>> >>>> Will you please send a pull request with this now? It's independent of >>>> the IDE fix which as a bugfix so can wait a little more but this series >>>> should be merged before the freeze starts tomorrow. (As this adds a new >>>> machine and does not touch anything else it can't break anything either.) >>> >>> Is there a PPC pull request in the making with this series in it? It would >>> be really sad to miss the release not being able to merge such a simple >>> series for weeks. As I said this is independent of any other fixes so I'd >>> like to get this in now please. >> >> Nick seems to be away so please Cédric, Daniel, Phil or whoever can send a >> pull request step in and please take care of this. This was posted a month >> ago with this last version on the list for 1.5 week so I'd hate to miss the >> release and wait until Easter next year to get this out to users because >> none of the maintainers are available before and arounf the freeze. > > Wait, do you think we would miss code freeze? > > Code freeze in QEMU is a blast! People getting mad because stuff got > left behind, PRs getting remade over and over because gitlab is nagging > about something, new bugs introduced, sometimes master doesn't even > build the following week. We might miss any other boring date like > release date or something, but code freeze? Nah, we're right here man. OK, cool, thanks a lot. I was just worried that there was no answer that somebody is working on this and wanted to make sure it won't be missed. If it's being taken care of then sorry to bug you but had to make sure it won't get lost in the noise with all the things happening in last minute. > The reason why we didn't interfere is because it's rude to short-circuit > the maintainer on duty. Nick didn't send an email asking for help, so we > were just monitoring. > > I'll consider Nick AWOL and take this PR. And don't worry, it's still > sunny in Brazil, we have plenty of time. Good. It's dark and cold here and Stefan already sent a message hours ago when it was evening for him but glad to hear its bright in Brazil, I was still hoping for that. :-) Regards, BALATON Zoltan > > > Thanks, > > Daniel > >> >> Regards, >> BALATON Zoltan >> >>>>> MAINTAINERS | 8 + >>>>> configs/devices/ppc-softmmu/default.mak | 1 + >>>>> hw/pci-host/Kconfig | 5 + >>>>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>>>> hw/pci-host/meson.build | 2 + >>>>> hw/ppc/Kconfig | 7 + >>>>> hw/ppc/amigaone.c | 164 +++++++++++++ >>>>> hw/ppc/meson.build | 2 + >>>>> include/hw/pci-host/articia.h | 17 ++ >>>>> tests/avocado/ppc_amiga.py | 38 +++ >>>>> 10 files changed, 537 insertions(+) >>>>> create mode 100644 hw/pci-host/articia.c >>>>> create mode 100644 hw/ppc/amigaone.c >>>>> create mode 100644 include/hw/pci-host/articia.h >>>>> create mode 100644 tests/avocado/ppc_amiga.py >>>>> >>>>> >>>> >>>> >>> > >
Zoltan, Gitlab is complaining about a missing file in one of the tests: 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp ERROR 0.22s killed by signal 6 SIGABRT 4324>>> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k 4325――――――――――――――――――――――――――――――――――――― ✀ ――――――――――――――――――――――――――――――――――――― 4326stderr: 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' 4328Broken pipe 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU process but encountered exit status 1 (expected 0) 4330(test program exited with status code -6) 4331TAP parsing error: Too few tests run (expected 13, got 0) You can reproduce it like this: $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 ./build/tests/qtest/test-hmp I ended up amending in-tree (downloaded the firmware, put it under pc-bios, updated pc-bios/meson.build). My manual test now passes, but not sure if gitlab will nag about it. Let's wait and see. I told you: code freeze is a blast! Let's see if it's still sunny for the AmigaOne XE board emulation. Thanks, Daniel On 10/27/23 08:54, BALATON Zoltan wrote: > Changes in v7: > - Increase default memory size to 512m to match pegasos2 and sam460ex > and it's a better default for AmigaOS > > Changes in v6: > - Dropped patch 1, now it's > > Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> > > ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI IDE controllers) > - Added Tested-by from Rene > > Changes in v5: > - Fixed avocado test > > Changes in v4: > - Found typo in comment in patch 1 so ended up rewording it again > trying to make it more concise. Also take the idea of using > range_covers_byte from Mark's patch > - Added RFC patch for avocado test (untested, I don't have Avocado) > > Changes in v3: > - Update values, comment and commit message in patch 1 again > > Changes in v2: > - Update comment and commit message in patch 1 (Mark) > - Fix irq mapping in patch 2 (Volker) > > Regards, > BALATON Zoltan > > BALATON Zoltan (3): > hw/pci-host: Add emulation of Mai Logic Articia S > hw/ppc: Add emulation of AmigaOne XE board > tests/avocado: Add test for amigaone board > > MAINTAINERS | 8 + > configs/devices/ppc-softmmu/default.mak | 1 + > hw/pci-host/Kconfig | 5 + > hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ > hw/pci-host/meson.build | 2 + > hw/ppc/Kconfig | 7 + > hw/ppc/amigaone.c | 164 +++++++++++++ > hw/ppc/meson.build | 2 + > include/hw/pci-host/articia.h | 17 ++ > tests/avocado/ppc_amiga.py | 38 +++ > 10 files changed, 537 insertions(+) > create mode 100644 hw/pci-host/articia.c > create mode 100644 hw/ppc/amigaone.c > create mode 100644 include/hw/pci-host/articia.h > create mode 100644 tests/avocado/ppc_amiga.py >
On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: > Zoltan, > > Gitlab is complaining about a missing file in one of the tests: > > > 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp > ERROR 0.22s killed by signal 6 SIGABRT > 4324>>> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh > QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 > PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 > /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k > 4325――――――――――――――――――――――――――――――――――――― ✀ > ――――――――――――――――――――――――――――――――――――― > 4326stderr: > 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' > 4328Broken pipe > 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU > process but encountered exit status 1 (expected 0) > 4330(test program exited with status code -6) > 4331TAP parsing error: Too few tests run (expected 13, got 0) > > > You can reproduce it like this: > > $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 > ./build/tests/qtest/test-hmp > > I ended up amending in-tree (downloaded the firmware, put it under pc-bios, > updated pc-bios/meson.build). > My manual test now passes, but not sure if gitlab will nag about it. Let's > wait and see. This is handled in the avocado test and it should download the file from the URL there. When tested locally it worked and downloaded the file and extracted the firmware bin from it. Can the gitlab CI download stuff or does it expect it to be in local cache already where you need to put it somehow beforehand? I think Philippe said something about that before but I did not quite get it as I don't know neither avocado nor gitlab. Hope Philippe is reading it and can chime in. But the test is not required to run the machine so as a last resort you could just drop the avocado patch and then we can add it later if we can't figure this out now. Regards, BALATON Zoltan > I told you: code freeze is a blast! Let's see if it's still sunny for the > AmigaOne XE board emulation. > > > > Thanks, > > > Daniel > > > > > > On 10/27/23 08:54, BALATON Zoltan wrote: >> Changes in v7: >> - Increase default memory size to 512m to match pegasos2 and sam460ex >> and it's a better default for AmigaOS >> >> Changes in v6: >> - Dropped patch 1, now it's >> >> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >> >> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI >> IDE controllers) >> - Added Tested-by from Rene >> >> Changes in v5: >> - Fixed avocado test >> >> Changes in v4: >> - Found typo in comment in patch 1 so ended up rewording it again >> trying to make it more concise. Also take the idea of using >> range_covers_byte from Mark's patch >> - Added RFC patch for avocado test (untested, I don't have Avocado) >> >> Changes in v3: >> - Update values, comment and commit message in patch 1 again >> >> Changes in v2: >> - Update comment and commit message in patch 1 (Mark) >> - Fix irq mapping in patch 2 (Volker) >> >> Regards, >> BALATON Zoltan >> >> BALATON Zoltan (3): >> hw/pci-host: Add emulation of Mai Logic Articia S >> hw/ppc: Add emulation of AmigaOne XE board >> tests/avocado: Add test for amigaone board >> >> MAINTAINERS | 8 + >> configs/devices/ppc-softmmu/default.mak | 1 + >> hw/pci-host/Kconfig | 5 + >> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >> hw/pci-host/meson.build | 2 + >> hw/ppc/Kconfig | 7 + >> hw/ppc/amigaone.c | 164 +++++++++++++ >> hw/ppc/meson.build | 2 + >> include/hw/pci-host/articia.h | 17 ++ >> tests/avocado/ppc_amiga.py | 38 +++ >> 10 files changed, 537 insertions(+) >> create mode 100644 hw/pci-host/articia.c >> create mode 100644 hw/ppc/amigaone.c >> create mode 100644 include/hw/pci-host/articia.h >> create mode 100644 tests/avocado/ppc_amiga.py >> > >
On 11/7/23 14:33, BALATON Zoltan wrote: > On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >> Zoltan, >> >> Gitlab is complaining about a missing file in one of the tests: >> >> >> 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp ERROR 0.22s killed by signal 6 SIGABRT >> 4324>>> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k >> 4325――――――――――――――――――――――――――――――――――――― ✀ ――――――――――――――――――――――――――――――――――――― >> 4326stderr: >> 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' >> 4328Broken pipe >> 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU process but encountered exit status 1 (expected 0) >> 4330(test program exited with status code -6) >> 4331TAP parsing error: Too few tests run (expected 13, got 0) >> >> >> You can reproduce it like this: >> >> $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 ./build/tests/qtest/test-hmp >> >> I ended up amending in-tree (downloaded the firmware, put it under pc-bios, updated pc-bios/meson.build). >> My manual test now passes, but not sure if gitlab will nag about it. Let's wait and see. > > This is handled in the avocado test and it should download the file from the URL there. When tested locally it worked and downloaded the file and extracted the firmware bin from it. Can the gitlab CI download stuff or does it expect it to be in local cache already where you need to put it somehow beforehand? I think Philippe said something about that before but I did not quite get it as I don't know neither avocado nor gitlab. Hope Philippe is reading it and can chime in. Gitlab CI can download stuff. But note that the error above is not avocado, it's hmp. It expects a firmware file to be available, and I'm not entirely sure it'll make any effort (e.g. downloading it) aside from checking if the file exists. > > But the test is not required to run the machine so as a last resort you could just drop the avocado patch and then we can add it later if we can't figure this out now. Not sure if dropping the avocado test would remediate the situation, but noted. For now let's push stuff upstream. We have the freeze window to make smaller adjustments if needed. Thanks, Daniel > > Regards, > BALATON Zoltan > >> I told you: code freeze is a blast! Let's see if it's still sunny for the >> AmigaOne XE board emulation. >> >> >> >> Thanks, >> >> >> Daniel >> >> >> >> >> >> On 10/27/23 08:54, BALATON Zoltan wrote: >>> Changes in v7: >>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>> and it's a better default for AmigaOS >>> >>> Changes in v6: >>> - Dropped patch 1, now it's >>> >>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>> >>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI IDE controllers) >>> - Added Tested-by from Rene >>> >>> Changes in v5: >>> - Fixed avocado test >>> >>> Changes in v4: >>> - Found typo in comment in patch 1 so ended up rewording it again >>> trying to make it more concise. Also take the idea of using >>> range_covers_byte from Mark's patch >>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>> >>> Changes in v3: >>> - Update values, comment and commit message in patch 1 again >>> >>> Changes in v2: >>> - Update comment and commit message in patch 1 (Mark) >>> - Fix irq mapping in patch 2 (Volker) >>> >>> Regards, >>> BALATON Zoltan >>> >>> BALATON Zoltan (3): >>> hw/pci-host: Add emulation of Mai Logic Articia S >>> hw/ppc: Add emulation of AmigaOne XE board >>> tests/avocado: Add test for amigaone board >>> >>> MAINTAINERS | 8 + >>> configs/devices/ppc-softmmu/default.mak | 1 + >>> hw/pci-host/Kconfig | 5 + >>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>> hw/pci-host/meson.build | 2 + >>> hw/ppc/Kconfig | 7 + >>> hw/ppc/amigaone.c | 164 +++++++++++++ >>> hw/ppc/meson.build | 2 + >>> include/hw/pci-host/articia.h | 17 ++ >>> tests/avocado/ppc_amiga.py | 38 +++ >>> 10 files changed, 537 insertions(+) >>> create mode 100644 hw/pci-host/articia.c >>> create mode 100644 hw/ppc/amigaone.c >>> create mode 100644 include/hw/pci-host/articia.h >>> create mode 100644 tests/avocado/ppc_amiga.py >>> >> >>
On Tue, 7 Nov 2023, BALATON Zoltan wrote: > On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >> Zoltan, >> >> Gitlab is complaining about a missing file in one of the tests: >> >> >> 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp ERROR 0.22s >> killed by signal 6 SIGABRT >> 4324>>> >> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh >> QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 >> PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 >> /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k >> 4325――――――――――――――――――――――――――――――――――――― ✀ >> ――――――――――――――――――――――――――――――――――――― >> 4326stderr: >> 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' >> 4328Broken pipe >> 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU >> process but encountered exit status 1 (expected 0) >> 4330(test program exited with status code -6) >> 4331TAP parsing error: Too few tests run (expected 13, got 0) >> >> >> You can reproduce it like this: >> >> $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 >> ./build/tests/qtest/test-hmp >> >> I ended up amending in-tree (downloaded the firmware, put it under pc-bios, >> updated pc-bios/meson.build). >> My manual test now passes, but not sure if gitlab will nag about it. Let's >> wait and see. > > This is handled in the avocado test and it should download the file from the > URL there. When tested locally it worked and downloaded the file and > extracted the firmware bin from it. Can the gitlab CI download stuff or does > it expect it to be in local cache already where you need to put it somehow > beforehand? I think Philippe said something about that before but I did not > quite get it as I don't know neither avocado nor gitlab. Hope Philippe is > reading it and can chime in. $ make check-avocado AVOCADO_TAGS=machine:amigaone [...] Fetching asset from tests/avocado/ppc_amiga.py:AmigaOneMachine.test_ppc_amigaone JOB ID : d638ea21275304aee911d63fdb9dc9f11a0f14e3 JOB LOG : [...]tests/results/job-2023-11-07T18.38-d638ea2/job.log (1/1) tests/avocado/ppc_amiga.py:AmigaOneMachine.test_ppc_amigaone: PASS (0.11 s) RESULTS : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0 JOB TIME : 1.87 s This creates $HOME/avocado where it downloads the zip. Maybe you need to have this dir on your CI runner if it can't download it itself. Looking at the job.log might tell what happened. I'm afraid I don't know anything else about this. Regards, BALATON Zoltan > But the test is not required to run the machine so as a last resort you could > just drop the avocado patch and then we can add it later if we can't figure > this out now. > > Regards, > BALATON Zoltan > >> I told you: code freeze is a blast! Let's see if it's still sunny for the >> AmigaOne XE board emulation. >> >> >> >> Thanks, >> >> >> Daniel >> >> >> >> >> >> On 10/27/23 08:54, BALATON Zoltan wrote: >>> Changes in v7: >>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>> and it's a better default for AmigaOS >>> >>> Changes in v6: >>> - Dropped patch 1, now it's >>> >>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>> >>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI >>> IDE controllers) >>> - Added Tested-by from Rene >>> >>> Changes in v5: >>> - Fixed avocado test >>> >>> Changes in v4: >>> - Found typo in comment in patch 1 so ended up rewording it again >>> trying to make it more concise. Also take the idea of using >>> range_covers_byte from Mark's patch >>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>> >>> Changes in v3: >>> - Update values, comment and commit message in patch 1 again >>> >>> Changes in v2: >>> - Update comment and commit message in patch 1 (Mark) >>> - Fix irq mapping in patch 2 (Volker) >>> >>> Regards, >>> BALATON Zoltan >>> >>> BALATON Zoltan (3): >>> hw/pci-host: Add emulation of Mai Logic Articia S >>> hw/ppc: Add emulation of AmigaOne XE board >>> tests/avocado: Add test for amigaone board >>> >>> MAINTAINERS | 8 + >>> configs/devices/ppc-softmmu/default.mak | 1 + >>> hw/pci-host/Kconfig | 5 + >>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>> hw/pci-host/meson.build | 2 + >>> hw/ppc/Kconfig | 7 + >>> hw/ppc/amigaone.c | 164 +++++++++++++ >>> hw/ppc/meson.build | 2 + >>> include/hw/pci-host/articia.h | 17 ++ >>> tests/avocado/ppc_amiga.py | 38 +++ >>> 10 files changed, 537 insertions(+) >>> create mode 100644 hw/pci-host/articia.c >>> create mode 100644 hw/ppc/amigaone.c >>> create mode 100644 include/hw/pci-host/articia.h >>> create mode 100644 tests/avocado/ppc_amiga.py >>> >> >
On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: > On 11/7/23 14:33, BALATON Zoltan wrote: >> On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >>> Zoltan, >>> >>> Gitlab is complaining about a missing file in one of the tests: >>> >>> >>> 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp ERROR 0.22s >>> killed by signal 6 SIGABRT >>> 4324>>> >>> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh >>> QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 >>> PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 >>> /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k >>> 4325――――――――――――――――――――――――――――――――――――― ✀ >>> ――――――――――――――――――――――――――――――――――――― >>> 4326stderr: >>> 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' >>> 4328Broken pipe >>> 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU >>> process but encountered exit status 1 (expected 0) >>> 4330(test program exited with status code -6) >>> 4331TAP parsing error: Too few tests run (expected 13, got 0) >>> >>> >>> You can reproduce it like this: >>> >>> $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 >>> ./build/tests/qtest/test-hmp >>> >>> I ended up amending in-tree (downloaded the firmware, put it under >>> pc-bios, updated pc-bios/meson.build). >>> My manual test now passes, but not sure if gitlab will nag about it. Let's >>> wait and see. >> >> This is handled in the avocado test and it should download the file from >> the URL there. When tested locally it worked and downloaded the file and >> extracted the firmware bin from it. Can the gitlab CI download stuff or >> does it expect it to be in local cache already where you need to put it >> somehow beforehand? I think Philippe said something about that before but I >> did not quite get it as I don't know neither avocado nor gitlab. Hope >> Philippe is reading it and can chime in. > > Gitlab CI can download stuff. But note that the error above is not > avocado, it's hmp. It expects a firmware file to be available, and I'm > not entirely sure it'll make any effort (e.g. downloading it) aside from > checking if the file exists. Hmm, that's strange because the extracted firmware is passed with -bios in the avocado test. But maybe it's then something else tries to run the machine with some other parameters but it needs a firmware. I was told we can't add the firmware because while it's GPL the sources were lost and not available any more so we only have the binary. Then the question is which test is trying to run the machine without -bios? What's test-hmp? Regards, BALATON Zoltan >> But the test is not required to run the machine so as a last resort you >> could just drop the avocado patch and then we can add it later if we can't >> figure this out now. > > Not sure if dropping the avocado test would remediate the situation, but > noted. > > For now let's push stuff upstream. We have the freeze window to make smaller > adjustments if needed. > > > Thanks, > > Daniel > >> >> Regards, >> BALATON Zoltan >> >>> I told you: code freeze is a blast! Let's see if it's still sunny for the >>> AmigaOne XE board emulation. >>> >>> >>> >>> Thanks, >>> >>> >>> Daniel >>> >>> >>> >>> >>> >>> On 10/27/23 08:54, BALATON Zoltan wrote: >>>> Changes in v7: >>>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>>> and it's a better default for AmigaOS >>>> >>>> Changes in v6: >>>> - Dropped patch 1, now it's >>>> >>>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>>> >>>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for >>>> PCI IDE controllers) >>>> - Added Tested-by from Rene >>>> >>>> Changes in v5: >>>> - Fixed avocado test >>>> >>>> Changes in v4: >>>> - Found typo in comment in patch 1 so ended up rewording it again >>>> trying to make it more concise. Also take the idea of using >>>> range_covers_byte from Mark's patch >>>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>>> >>>> Changes in v3: >>>> - Update values, comment and commit message in patch 1 again >>>> >>>> Changes in v2: >>>> - Update comment and commit message in patch 1 (Mark) >>>> - Fix irq mapping in patch 2 (Volker) >>>> >>>> Regards, >>>> BALATON Zoltan >>>> >>>> BALATON Zoltan (3): >>>> hw/pci-host: Add emulation of Mai Logic Articia S >>>> hw/ppc: Add emulation of AmigaOne XE board >>>> tests/avocado: Add test for amigaone board >>>> >>>> MAINTAINERS | 8 + >>>> configs/devices/ppc-softmmu/default.mak | 1 + >>>> hw/pci-host/Kconfig | 5 + >>>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>>> hw/pci-host/meson.build | 2 + >>>> hw/ppc/Kconfig | 7 + >>>> hw/ppc/amigaone.c | 164 +++++++++++++ >>>> hw/ppc/meson.build | 2 + >>>> include/hw/pci-host/articia.h | 17 ++ >>>> tests/avocado/ppc_amiga.py | 38 +++ >>>> 10 files changed, 537 insertions(+) >>>> create mode 100644 hw/pci-host/articia.c >>>> create mode 100644 hw/ppc/amigaone.c >>>> create mode 100644 include/hw/pci-host/articia.h >>>> create mode 100644 tests/avocado/ppc_amiga.py >>>> >>> >>> > >
On Tue, 7 Nov 2023, BALATON Zoltan wrote: > On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >> On 11/7/23 14:33, BALATON Zoltan wrote: >>> On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >>>> Zoltan, >>>> >>>> Gitlab is complaining about a missing file in one of the tests: >>>> >>>> >>>> 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp ERROR 0.22s >>>> killed by signal 6 SIGABRT >>>> 4324>>> >>>> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh >>>> QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 >>>> PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 >>>> /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k >>>> 4325――――――――――――――――――――――――――――――――――――― ✀ >>>> ――――――――――――――――――――――――――――――――――――― >>>> 4326stderr: >>>> 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' >>>> 4328Broken pipe >>>> 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU >>>> process but encountered exit status 1 (expected 0) >>>> 4330(test program exited with status code -6) >>>> 4331TAP parsing error: Too few tests run (expected 13, got 0) >>>> >>>> >>>> You can reproduce it like this: >>>> >>>> $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 >>>> ./build/tests/qtest/test-hmp >>>> >>>> I ended up amending in-tree (downloaded the firmware, put it under >>>> pc-bios, updated pc-bios/meson.build). >>>> My manual test now passes, but not sure if gitlab will nag about it. >>>> Let's wait and see. >>> >>> This is handled in the avocado test and it should download the file from >>> the URL there. When tested locally it worked and downloaded the file and >>> extracted the firmware bin from it. Can the gitlab CI download stuff or >>> does it expect it to be in local cache already where you need to put it >>> somehow beforehand? I think Philippe said something about that before but >>> I did not quite get it as I don't know neither avocado nor gitlab. Hope >>> Philippe is reading it and can chime in. >> >> Gitlab CI can download stuff. But note that the error above is not avocado, >> it's hmp. It expects a firmware file to be available, and I'm not entirely >> sure it'll make any effort (e.g. downloading it) aside from checking if the >> file exists. > > Hmm, that's strange because the extracted firmware is passed with -bios in > the avocado test. But maybe it's then something else tries to run the machine > with some other parameters but it needs a firmware. I was told we can't add > the firmware because while it's GPL the sources were lost and not available > any more so we only have the binary. Then the question is which test is > trying to run the machine without -bios? What's test-hmp? So if it's tests/qtest/test-hmp.c that seems to try to run a bunch of command for each machine AFAIU. This machine needs a firmware image but this test seems to run it with -S and never starts the machine so could take any file with -bios for that, e.g. -bios vof.bin (thought about -bios /dev/null but that does not work but any other file should probably be OK). I don't know how to tell test-hmp to use that thogh or to skip this machine. At some point pegasos2 also needed a firmware but this wasn't a problem back then. Either this test is new or we could just ignore this error? Regards, BALATON Zoltan >>> But the test is not required to run the machine so as a last resort you >>> could just drop the avocado patch and then we can add it later if we can't >>> figure this out now. >> >> Not sure if dropping the avocado test would remediate the situation, but >> noted. >> >> For now let's push stuff upstream. We have the freeze window to make >> smaller >> adjustments if needed. >> >> >> Thanks, >> >> Daniel >> >>> >>> Regards, >>> BALATON Zoltan >>> >>>> I told you: code freeze is a blast! Let's see if it's still sunny for the >>>> AmigaOne XE board emulation. >>>> >>>> >>>> >>>> Thanks, >>>> >>>> >>>> Daniel >>>> >>>> >>>> >>>> >>>> >>>> On 10/27/23 08:54, BALATON Zoltan wrote: >>>>> Changes in v7: >>>>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>>>> and it's a better default for AmigaOS >>>>> >>>>> Changes in v6: >>>>> - Dropped patch 1, now it's >>>>> >>>>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>>>> >>>>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for >>>>> PCI IDE controllers) >>>>> - Added Tested-by from Rene >>>>> >>>>> Changes in v5: >>>>> - Fixed avocado test >>>>> >>>>> Changes in v4: >>>>> - Found typo in comment in patch 1 so ended up rewording it again >>>>> trying to make it more concise. Also take the idea of using >>>>> range_covers_byte from Mark's patch >>>>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>>>> >>>>> Changes in v3: >>>>> - Update values, comment and commit message in patch 1 again >>>>> >>>>> Changes in v2: >>>>> - Update comment and commit message in patch 1 (Mark) >>>>> - Fix irq mapping in patch 2 (Volker) >>>>> >>>>> Regards, >>>>> BALATON Zoltan >>>>> >>>>> BALATON Zoltan (3): >>>>> hw/pci-host: Add emulation of Mai Logic Articia S >>>>> hw/ppc: Add emulation of AmigaOne XE board >>>>> tests/avocado: Add test for amigaone board >>>>> >>>>> MAINTAINERS | 8 + >>>>> configs/devices/ppc-softmmu/default.mak | 1 + >>>>> hw/pci-host/Kconfig | 5 + >>>>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>>>> hw/pci-host/meson.build | 2 + >>>>> hw/ppc/Kconfig | 7 + >>>>> hw/ppc/amigaone.c | 164 +++++++++++++ >>>>> hw/ppc/meson.build | 2 + >>>>> include/hw/pci-host/articia.h | 17 ++ >>>>> tests/avocado/ppc_amiga.py | 38 +++ >>>>> 10 files changed, 537 insertions(+) >>>>> create mode 100644 hw/pci-host/articia.c >>>>> create mode 100644 hw/ppc/amigaone.c >>>>> create mode 100644 include/hw/pci-host/articia.h >>>>> create mode 100644 tests/avocado/ppc_amiga.py >>>>> >>>> >>>> >> >
On Tue, 7 Nov 2023 at 18:05, BALATON Zoltan <balaton@eik.bme.hu> wrote: > So if it's tests/qtest/test-hmp.c that seems to try to run a bunch of > command for each machine AFAIU. This machine needs a firmware image but > this test seems to run it with -S and never starts the machine so could > take any file with -bios for that, e.g. -bios vof.bin (thought about -bios > /dev/null but that does not work but any other file should probably be > OK). I don't know how to tell test-hmp to use that thogh or to skip this > machine. > > At some point pegasos2 also needed a firmware but this wasn't a problem > back then. Either this test is new or we could just ignore this error? The test has been around for a long time, and no, you can't ignore 'make check' failures, they will break the CI. It's one of the tests we have that test properties that must be true for all board types (or all devices, in the case of some other tests). One of the properties is "it ought to start up cleanly without extra arguments". pegasos2 works because there's a vof.bin for it in pc-bios/. thanks -- PMM
On Tue, 7 Nov 2023, Peter Maydell wrote: > On Tue, 7 Nov 2023 at 18:05, BALATON Zoltan <balaton@eik.bme.hu> wrote: >> So if it's tests/qtest/test-hmp.c that seems to try to run a bunch of >> command for each machine AFAIU. This machine needs a firmware image but >> this test seems to run it with -S and never starts the machine so could >> take any file with -bios for that, e.g. -bios vof.bin (thought about -bios >> /dev/null but that does not work but any other file should probably be >> OK). I don't know how to tell test-hmp to use that thogh or to skip this >> machine. >> >> At some point pegasos2 also needed a firmware but this wasn't a problem >> back then. Either this test is new or we could just ignore this error? > > The test has been around for a long time, and no, you can't ignore > 'make check' failures, they will break the CI. It's one of > the tests we have that test properties that must be true for > all board types (or all devices, in the case of some other > tests). One of the properties is "it ought to start up cleanly > without extra arguments". I'll send a patch in a minute which hopefully fixes that skipping the exit when using qtest and called without -bios. > pegasos2 works because there's a vof.bin for it in pc-bios/. Now it has vof.bin but it did not have that previously and it still worked when it needed -bios too. But that was a while ago so don't remember the details. Regards, BALATON Zoltan
On Tue, 7 Nov 2023 at 18:18, BALATON Zoltan <balaton@eik.bme.hu> wrote: > > On Tue, 7 Nov 2023, Peter Maydell wrote: > > On Tue, 7 Nov 2023 at 18:05, BALATON Zoltan <balaton@eik.bme.hu> wrote: > >> So if it's tests/qtest/test-hmp.c that seems to try to run a bunch of > >> command for each machine AFAIU. This machine needs a firmware image but > >> this test seems to run it with -S and never starts the machine so could > >> take any file with -bios for that, e.g. -bios vof.bin (thought about -bios > >> /dev/null but that does not work but any other file should probably be > >> OK). I don't know how to tell test-hmp to use that thogh or to skip this > >> machine. > >> > >> At some point pegasos2 also needed a firmware but this wasn't a problem > >> back then. Either this test is new or we could just ignore this error? > > > > The test has been around for a long time, and no, you can't ignore > > 'make check' failures, they will break the CI. It's one of > > the tests we have that test properties that must be true for > > all board types (or all devices, in the case of some other > > tests). One of the properties is "it ought to start up cleanly > > without extra arguments". > > I'll send a patch in a minute which hopefully fixes that skipping the exit > when using qtest and called without -bios. Yeah, looking at eg hw/mips/malta.c that's what we do there: we check qtest_enabled() before printing the error about not being able to load a BIOS blob. thanks -- PMM
On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: > On 11/7/23 14:33, BALATON Zoltan wrote: >> On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >>> Zoltan, >>> >>> Gitlab is complaining about a missing file in one of the tests: >>> >>> >>> 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp ERROR 0.22s >>> killed by signal 6 SIGABRT >>> 4324>>> >>> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh >>> QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 >>> PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 >>> /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k >>> 4325――――――――――――――――――――――――――――――――――――― ✀ >>> ――――――――――――――――――――――――――――――――――――― >>> 4326stderr: >>> 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' >>> 4328Broken pipe >>> 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU >>> process but encountered exit status 1 (expected 0) >>> 4330(test program exited with status code -6) >>> 4331TAP parsing error: Too few tests run (expected 13, got 0) OK, how about with v8 of patch2 I've just sent? The other pathches are unchanged so did not resend those. It should allow ronning without -bios for qtest and as this test just runs monitor commands without starting the machine that should be OK. Hopefully this fixes it. Regards, BALATON Zoltan >>> You can reproduce it like this: >>> >>> $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 >>> ./build/tests/qtest/test-hmp >>> >>> I ended up amending in-tree (downloaded the firmware, put it under >>> pc-bios, updated pc-bios/meson.build). >>> My manual test now passes, but not sure if gitlab will nag about it. Let's >>> wait and see. >> >> This is handled in the avocado test and it should download the file from >> the URL there. When tested locally it worked and downloaded the file and >> extracted the firmware bin from it. Can the gitlab CI download stuff or >> does it expect it to be in local cache already where you need to put it >> somehow beforehand? I think Philippe said something about that before but I >> did not quite get it as I don't know neither avocado nor gitlab. Hope >> Philippe is reading it and can chime in. > > Gitlab CI can download stuff. But note that the error above is not avocado, > it's hmp. > It expects a firmware file to be available, and I'm not entirely sure it'll > make any > effort (e.g. downloading it) aside from checking if the file exists. >> >> But the test is not required to run the machine so as a last resort you >> could just drop the avocado patch and then we can add it later if we can't >> figure this out now. > > Not sure if dropping the avocado test would remediate the situation, but > noted. > > For now let's push stuff upstream. We have the freeze window to make smaller > adjustments if needed. > > > Thanks, > > Daniel > >> >> Regards, >> BALATON Zoltan >> >>> I told you: code freeze is a blast! Let's see if it's still sunny for the >>> AmigaOne XE board emulation. >>> >>> >>> >>> Thanks, >>> >>> >>> Daniel >>> >>> >>> >>> >>> >>> On 10/27/23 08:54, BALATON Zoltan wrote: >>>> Changes in v7: >>>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>>> and it's a better default for AmigaOS >>>> >>>> Changes in v6: >>>> - Dropped patch 1, now it's >>>> >>>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>>> >>>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for >>>> PCI IDE controllers) >>>> - Added Tested-by from Rene >>>> >>>> Changes in v5: >>>> - Fixed avocado test >>>> >>>> Changes in v4: >>>> - Found typo in comment in patch 1 so ended up rewording it again >>>> trying to make it more concise. Also take the idea of using >>>> range_covers_byte from Mark's patch >>>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>>> >>>> Changes in v3: >>>> - Update values, comment and commit message in patch 1 again >>>> >>>> Changes in v2: >>>> - Update comment and commit message in patch 1 (Mark) >>>> - Fix irq mapping in patch 2 (Volker) >>>> >>>> Regards, >>>> BALATON Zoltan >>>> >>>> BALATON Zoltan (3): >>>> hw/pci-host: Add emulation of Mai Logic Articia S >>>> hw/ppc: Add emulation of AmigaOne XE board >>>> tests/avocado: Add test for amigaone board >>>> >>>> MAINTAINERS | 8 + >>>> configs/devices/ppc-softmmu/default.mak | 1 + >>>> hw/pci-host/Kconfig | 5 + >>>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>>> hw/pci-host/meson.build | 2 + >>>> hw/ppc/Kconfig | 7 + >>>> hw/ppc/amigaone.c | 164 +++++++++++++ >>>> hw/ppc/meson.build | 2 + >>>> include/hw/pci-host/articia.h | 17 ++ >>>> tests/avocado/ppc_amiga.py | 38 +++ >>>> 10 files changed, 537 insertions(+) >>>> create mode 100644 hw/pci-host/articia.c >>>> create mode 100644 hw/ppc/amigaone.c >>>> create mode 100644 include/hw/pci-host/articia.h >>>> create mode 100644 tests/avocado/ppc_amiga.py >>>> >>> >>> > >
On Tue, 7 Nov 2023, BALATON Zoltan wrote: > On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >> On 11/7/23 14:33, BALATON Zoltan wrote: >>> On Tue, 7 Nov 2023, Daniel Henrique Barboza wrote: >>>> Zoltan, >>>> >>>> Gitlab is complaining about a missing file in one of the tests: >>>> >>>> >>>> 8/259 qemu:qtest+qtest-ppc / qtest-ppc/test-hmp ERROR 0.22s >>>> killed by signal 6 SIGABRT >>>> 4324>>> >>>> G_TEST_DBUS_DAEMON=/builds/danielhb/qemu/tests/dbus-vmstate-daemon.sh >>>> QTEST_QEMU_BINARY=./qemu-system-ppc MALLOC_PERTURB_=87 >>>> PYTHON=/builds/danielhb/qemu/build/pyvenv/bin/python3 >>>> /builds/danielhb/qemu/build/tests/qtest/test-hmp --tap -k >>>> 4325――――――――――――――――――――――――――――――――――――― ✀ >>>> ――――――――――――――――――――――――――――――――――――― >>>> 4326stderr: >>>> 4327qemu-system-ppc: Could not find firmware 'u-boot-amigaone.bin' >>>> 4328Broken pipe >>>> 4329../tests/qtest/libqtest.c:195: kill_qemu() tried to terminate QEMU >>>> process but encountered exit status 1 (expected 0) >>>> 4330(test program exited with status code -6) >>>> 4331TAP parsing error: Too few tests run (expected 13, got 0) > > OK, how about with v8 of patch2 I've just sent? The other pathches are OK, sent v9 then. This shuold be it now. > unchanged so did not resend those. It should allow ronning without -bios for > qtest and as this test just runs monitor commands without starting the > machine that should be OK. Hopefully this fixes it. > > Regards, > BALATON Zoltan > >>>> You can reproduce it like this: >>>> >>>> $ make -j -C build && QTEST_QEMU_BINARY=./build/qemu-system-ppc64 >>>> ./build/tests/qtest/test-hmp >>>> >>>> I ended up amending in-tree (downloaded the firmware, put it under >>>> pc-bios, updated pc-bios/meson.build). >>>> My manual test now passes, but not sure if gitlab will nag about it. >>>> Let's wait and see. >>> >>> This is handled in the avocado test and it should download the file from >>> the URL there. When tested locally it worked and downloaded the file and >>> extracted the firmware bin from it. Can the gitlab CI download stuff or >>> does it expect it to be in local cache already where you need to put it >>> somehow beforehand? I think Philippe said something about that before but >>> I did not quite get it as I don't know neither avocado nor gitlab. Hope >>> Philippe is reading it and can chime in. >> >> Gitlab CI can download stuff. But note that the error above is not avocado, >> it's hmp. >> It expects a firmware file to be available, and I'm not entirely sure it'll >> make any >> effort (e.g. downloading it) aside from checking if the file exists. >>> >>> But the test is not required to run the machine so as a last resort you >>> could just drop the avocado patch and then we can add it later if we can't >>> figure this out now. >> >> Not sure if dropping the avocado test would remediate the situation, but >> noted. >> >> For now let's push stuff upstream. We have the freeze window to make >> smaller >> adjustments if needed. >> >> >> Thanks, >> >> Daniel >> >>> >>> Regards, >>> BALATON Zoltan >>> >>>> I told you: code freeze is a blast! Let's see if it's still sunny for the >>>> AmigaOne XE board emulation. >>>> >>>> >>>> >>>> Thanks, >>>> >>>> >>>> Daniel >>>> >>>> >>>> >>>> >>>> >>>> On 10/27/23 08:54, BALATON Zoltan wrote: >>>>> Changes in v7: >>>>> - Increase default memory size to 512m to match pegasos2 and sam460ex >>>>> and it's a better default for AmigaOS >>>>> >>>>> Changes in v6: >>>>> - Dropped patch 1, now it's >>>>> >>>>> Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> >>>>> >>>>> ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for >>>>> PCI IDE controllers) >>>>> - Added Tested-by from Rene >>>>> >>>>> Changes in v5: >>>>> - Fixed avocado test >>>>> >>>>> Changes in v4: >>>>> - Found typo in comment in patch 1 so ended up rewording it again >>>>> trying to make it more concise. Also take the idea of using >>>>> range_covers_byte from Mark's patch >>>>> - Added RFC patch for avocado test (untested, I don't have Avocado) >>>>> >>>>> Changes in v3: >>>>> - Update values, comment and commit message in patch 1 again >>>>> >>>>> Changes in v2: >>>>> - Update comment and commit message in patch 1 (Mark) >>>>> - Fix irq mapping in patch 2 (Volker) >>>>> >>>>> Regards, >>>>> BALATON Zoltan >>>>> >>>>> BALATON Zoltan (3): >>>>> hw/pci-host: Add emulation of Mai Logic Articia S >>>>> hw/ppc: Add emulation of AmigaOne XE board >>>>> tests/avocado: Add test for amigaone board >>>>> >>>>> MAINTAINERS | 8 + >>>>> configs/devices/ppc-softmmu/default.mak | 1 + >>>>> hw/pci-host/Kconfig | 5 + >>>>> hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ >>>>> hw/pci-host/meson.build | 2 + >>>>> hw/ppc/Kconfig | 7 + >>>>> hw/ppc/amigaone.c | 164 +++++++++++++ >>>>> hw/ppc/meson.build | 2 + >>>>> include/hw/pci-host/articia.h | 17 ++ >>>>> tests/avocado/ppc_amiga.py | 38 +++ >>>>> 10 files changed, 537 insertions(+) >>>>> create mode 100644 hw/pci-host/articia.c >>>>> create mode 100644 hw/ppc/amigaone.c >>>>> create mode 100644 include/hw/pci-host/articia.h >>>>> create mode 100644 tests/avocado/ppc_amiga.py >>>>> >>>> >>>> >> >
Patches 1 and 3 from v7, and 2 from v9, queued in ppc-next. Thanks, Daniel On 10/27/23 08:54, BALATON Zoltan wrote: > Changes in v7: > - Increase default memory size to 512m to match pegasos2 and sam460ex > and it's a better default for AmigaOS > > Changes in v6: > - Dropped patch 1, now it's > > Based-on: <20231024224056.842607-1-mark.cave-ayland@ilande.co.uk> > > ([PATCH v2 0/3] ide: implement simple legacy/native mode switching for PCI IDE controllers) > - Added Tested-by from Rene > > Changes in v5: > - Fixed avocado test > > Changes in v4: > - Found typo in comment in patch 1 so ended up rewording it again > trying to make it more concise. Also take the idea of using > range_covers_byte from Mark's patch > - Added RFC patch for avocado test (untested, I don't have Avocado) > > Changes in v3: > - Update values, comment and commit message in patch 1 again > > Changes in v2: > - Update comment and commit message in patch 1 (Mark) > - Fix irq mapping in patch 2 (Volker) > > Regards, > BALATON Zoltan > > BALATON Zoltan (3): > hw/pci-host: Add emulation of Mai Logic Articia S > hw/ppc: Add emulation of AmigaOne XE board > tests/avocado: Add test for amigaone board > > MAINTAINERS | 8 + > configs/devices/ppc-softmmu/default.mak | 1 + > hw/pci-host/Kconfig | 5 + > hw/pci-host/articia.c | 293 ++++++++++++++++++++++++ > hw/pci-host/meson.build | 2 + > hw/ppc/Kconfig | 7 + > hw/ppc/amigaone.c | 164 +++++++++++++ > hw/ppc/meson.build | 2 + > include/hw/pci-host/articia.h | 17 ++ > tests/avocado/ppc_amiga.py | 38 +++ > 10 files changed, 537 insertions(+) > create mode 100644 hw/pci-host/articia.c > create mode 100644 hw/ppc/amigaone.c > create mode 100644 include/hw/pci-host/articia.h > create mode 100644 tests/avocado/ppc_amiga.py >