summaryrefslogtreecommitdiff
path: root/hw
AgeCommit message (Collapse)AuthorFilesLines
2012-04-15pseries: Consolidate hack for RTAS display-character usageDavid Gibson3-5/+3
Currently the pseries machine contains not one but two somewhat ugly hacks to allow printing of early debug messages before the guest has properly read the device tree. First, we special case H_PUT_TERM_CHAR so that a vtermno of 0 (usually invalid) will look for a suitable vty and use that. This supports Linux's early debug code which will use H_PUT_TERM_CHAR with vtermno==0 before reading the device tree. Second, we support the RTAS display-character call. This takes no vtermno so we assume the address of the default first VTY. This patch makes things more consistent by folding the second hack into the first. Now, display-character uses the existing vty_lookup() function to do the same search for a suitable VTY. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Andreas Färber <afaerber@suse.de>
2012-04-15pseries: Remove unused fields from VIOsPAPRBus structureDavid Gibson1-2/+0
The VIOsPAPRBus structure, used on the pseries machine contains some old fields which are no longer used anywhere. This patch removes them. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Andreas Färber <afaerber@suse.de>
2012-04-15pseries: Implement RTAS system-reboot callDavid Gibson1-0/+14
This patch adds the PAPR defined RTAS system-reboot call to the pseries machine emulation, providing the guest with a way to trigger a reboot. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Andreas Färber <afaerber@suse.de>
2012-04-15pseries: Fix bug with reset of VIO CRQsDavid Gibson1-8/+25
PAPR specifies a Command Response Queue (CRQ) mechanism used for virtual IO, which we implement. However, we don't correctly clean up registered CRQs when we reset the system. This patch adds a reset handler to fix this bug. While we're at it, add in some of the extra debug messages that were used to track the problem down. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> [AF: Updated hcall_dprintf()s to not duplicate the function name] Signed-off-by: Andreas Färber <afaerber@suse.de>
2012-04-15pseries: Clean up hcall_dprintf() debugging messagesDavid Gibson3-24/+15
The pseries machine code has a number of debug messages for debugging PAPR hypercalls, dependent on DEBUG_SPAPR_HCALLS. This patch cleans these messages up a bit, by adding __func__ to the hcall_dprintf() macro and simplifying up a number of the individual messages accordingly. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Andreas Färber <afaerber@suse.de>
2012-04-14Merge branch 'arm-devs.for-upstream' of ↵Blue Swirl12-169/+330
git://git.linaro.org/people/pmaydell/qemu-arm * 'arm-devs.for-upstream' of git://git.linaro.org/people/pmaydell/qemu-arm: hw/arm_gic: Remove stray hardcoded tab hw/arm_gic: gic_set_pending_private() is NVIC only hw/arm_gic: Use NVIC instead of LEGACY_INCLUDED_GIC define hw/arm_gic: Make gic_reset a sysbus reset function hw/arm11mpcore: Convert to using sysbus GIC device hw/exynos4210_gic: Convert to using sysbus GIC hw/realview_gic: switch to sysbus GIC hw/a9mpcore: Switch to using sysbus GIC hw/a15mpcore: switch to using sysbus GIC hw/arm_gic: Make the GIC its own sysbus device hw/arm_gic: Expose PPI inputs as gpio inputs hw/arm_gic: Move gic_get_current_cpu into arm_gic.c hw/arm_gic: Move NCPU definition to arm_gic.c hw/exynos4210_combiner.c: Drop excessive read/write access check. ARM: Exynos4210: Drop gic_cpu_write() after initialization. Fix bit test in Exynos4210 UART emulation to use & instead of &&
2012-04-13Merge remote-tracking branch 'stefanha/trivial-patches' into stagingAnthony Liguori1-0/+29
* stefanha/trivial-patches: configure: Insist on a Python 2, not Python 3 bsd-user: fix compile failure ps2: avoid repeated header file includes make: Always set LC_ALL=C for makeinfo configure: Fix wrong preprocessor statement configure: Remove useless uses of ARCH_CFLAGS
2012-04-13hw/arm_gic: Remove stray hardcoded tabPeter Maydell1-1/+1
Remove the single instance of a hardcoded tab from hw/arm_gic.c. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/arm_gic: gic_set_pending_private() is NVIC onlyPeter Maydell1-2/+3
The function gic_set_pending_private() is now used by the NVIC only (for the GIC we now set PPI interrupts via gpio lines and gic_set_irq()). So make it #ifdef NVIC and remove the 'attribute unused' annotation. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/arm_gic: Use NVIC instead of LEGACY_INCLUDED_GIC definePeter Maydell2-4/+10
Now all the A profile cores have been switched to use the standalone sysbus GIC, the only remaining code which #includes arm_gic.c is the v7M NVIC. The coupling is much closer here so it's not so easily disentangled. For now, add a comment about how arm_gic.c is compiled, and assume that the NVIC always includes arm_gic.c and the non-NVIC GIC is always compiled standalone. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/arm_gic: Make gic_reset a sysbus reset functionPeter Maydell2-2/+19
Make gic_reset a sysbus reset function, so we actually reset the GIC on system reset rather than only at init. For the NVIC this requires us also to implement reset of the SysTick. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/arm11mpcore: Convert to using sysbus GIC devicePeter Maydell1-19/+30
Convert arm11mpcore to using the standalone sysbus GIC device. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/exynos4210_gic: Convert to using sysbus GICPeter Maydell1-8/+24
Convert the Exynos GIC code to use the standalone sysbus GIC device. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Evgeny Voevodin <e.voevodin@samsung.com>
2012-04-13hw/realview_gic: switch to sysbus GICPeter Maydell1-12/+26
Switch the realview_gic device to the standalone sysbus GIC. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/a9mpcore: Switch to using sysbus GICPeter Maydell1-25/+35
Switch the a9mpcore to using the sysbus GIC device rather than having the a9mp private memory region device subclass the GIC. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/a15mpcore: switch to using sysbus GICPeter Maydell1-9/+26
Switch the a15mpcore private peripheral region to using the standalone sysbus GIC device. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/arm_gic: Make the GIC its own sysbus devicePeter Maydell7-1/+57
Compile arm_gic.c as a standalone C file to produce a self contained sysbus GIC device. Support the legacy usage by #include of the .c file by making those users #define LEGACY_INCLUDED_GIC, so we can convert them one by one. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Evgeny Voevodin <e.voevodin@samsung.com>
2012-04-13hw/arm_gic: Expose PPI inputs as gpio inputsPeter Maydell1-9/+40
Expose the Private Peripheral Interrupt inputs as GPIO inputs. The layout of the GPIO array is thus: [0..N-1] SPIs [N..N+31] PPIs for CPU 0 [N+32..N+63] PPIs for CPU 1 ... Treating PPIs as being another kind of input line is in line with the GIC architecture specification, where they are clearly described that way. The 11MPCore TRM is a bit more ambiguous, but there is no practical difference between "set PPI X as pending" and "0->1 transition on a PPI input line configured as edge triggered", and PPIs are always edge triggered, so this change won't affect behaviour. Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13hw/arm_gic: Move gic_get_current_cpu into arm_gic.cPeter Maydell7-48/+15
Move the gic_get_current_cpu() function into arm_gic.c. There are only two implementations: (1) "get the index of the currently executing CPU", used by all multicore GICs, and (2) "always 0", used by all GICs instantiated with a single CPU interface (the Realview board GIC and the v7M NVIC). So we can move this into the main GIC source file. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Evgeny Voevodin <e.voevodin@samsung.com>
2012-04-13hw/arm_gic: Move NCPU definition to arm_gic.cPeter Maydell7-26/+19
Move the NCPU definition to arm_gic.c: the maximum number of CPU interfaces is defined by the GIC architecture specification to be 8, so we don't need to have this #define in each of the sources files which currently includes arm_gic.c. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Evgeny Voevodin <e.voevodin@samsung.com>
2012-04-13hw/exynos4210_combiner.c: Drop excessive read/write access check.Evgeny Voevodin1-10/+0
Access to reserved area at offset higher than 0x3c is allowed in External Combiner. Samsung Galaxy Kernel implements this. So, drop excessive checks in read/write functions. Signed-off-by: Evgeny Voevodin <e.voevodin@samsung.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13ARM: Exynos4210: Drop gic_cpu_write() after initialization.Evgeny Voevodin4-2/+34
Remove gic_cpu_write() call after initialization that was emulating functionality of earliest SOC bootloader which enables external GIC CPU1 interface. Instead introduce Exynos4210-specific secondary CPU bootloader, which enables both Internal and External GIC CPU1 interfaces. Signed-off-by: Evgeny Voevodin <e.voevodin@samsung.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-13Fix bit test in Exynos4210 UART emulation to use & instead of &&Daniel P. Berrange1-3/+3
* hw/exynos4210_uart.c: s/&&/&/ Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2012-04-12kvmclock: guest stop notificationEric B Munson1-0/+16
Often when a guest is stopped from the qemu console, it will report spurious soft lockup warnings on resume. There are kernel patches being discussed that will give the host the ability to tell the guest that it is being stopped and should ignore the soft lockup warning that generates. This patch uses the qemu Notifier system to tell the guest it is about to be stopped. Signed-off-by: Eric B Munson <emunson@mgebm.net> Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> Reviewed-by: Andreas Färber <afaerber@suse.de> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2012-04-12kvm: allow arbitrarily sized mmio ioeventfdMichael S. Tsirkin1-4/+4
We use a 2 byte ioeventfd for virtio memory, add support for this. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Amos Kong <akong@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2012-04-12ps2: avoid repeated header file includesWanpeng Li1-0/+29
Signed-off-by: Wanpeng Li <liwp@linux.vnet.ibm.com> Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
2012-04-11rtl8139: do not assume TxStatus[] and TxAddr[] are adjacentStefan Hajnoczi1-11/+15
Commit afe0a595356192d5f79703cf6462fcc112df007c ("rtl8139: support byte read to TxStatus registers") reused rtl8139_TxStatus_read() for reading TxAddr registers. It relies on the fact that TxStatus[] and TxAddr[] are adjacent. This causes a gcc warning because the compiler can detect that array access is out-of-bounds: hw/rtl8139.c:2501:27: error: array subscript is above array bounds [-Werror=array-bounds] This patch refactors the function so that we don't rely on out-of-bounds accesses. Cc: Jason Wang <jasonwang@redhat.com> Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-04-10Merge commit 'ff71f2e8cacefae99179993204172bc65e4303df' into stagingAnthony Liguori15-149/+1048
* commit 'ff71f2e8cacefae99179993204172bc65e4303df': (21 commits) rtl8139: do the network/host communication only in normal operating mode rtl8139: correctly check the opmode net: move compute_mcast_idx() to net.h rtl8139: support byte read to TxStatus registers rtl8139: remove unused marco rtl8139: limit transmission buffer size in c+ mode pci_regs: Add PCI_EXP_TYPE_PCIE_BRIDGE virtio-net: add DATA_VALID flag pci_bridge: upper 32 bit are long registers pci: fix bridge IO/BASE pcie: drop functionality moved to core pci: set memory type for memory behind the bridge pci: add standard bridge device slotid: add slot id capability shpc: standard hot plug controller pci_bridge: user-friendly default bus name pci: make another unused extern function static pci: don't export an internal function pci_regs: Fix value of PCI_EXP_TYPE_RC_EC. pci: Do not check if a bus exist in pci_parse_devaddr. ...
2012-04-10Merge remote-tracking branch 'qmp/queue/qmp' into stagingAnthony Liguori6-15/+31
* qmp/queue/qmp: qapi: convert device_del qdev: qdev_unplug(): use error_set()
2012-04-10Merge remote-tracking branch 'kwolf/for-anthony' into stagingAnthony Liguori7-18/+109
* kwolf/for-anthony: (46 commits) qed: remove incoming live migration blocker qed: honor BDRV_O_INCOMING for incoming live migration migration: clear BDRV_O_INCOMING flags on end of incoming live migration qed: add bdrv_invalidate_cache to be called after incoming live migration blockdev: open images with BDRV_O_INCOMING on incoming live migration block: add a function to clear incoming live migration flags block: Add new BDRV_O_INCOMING flag to notice incoming live migration block stream: close unused files and update ->backing_hd qemu-iotests: Fix call syntax for qemu-io qemu-iotests: Fix call syntax for qemu-img qemu-iotests: Test unknown qcow2 header extensions qemu-iotests: qcow2.py sheepdog: fix send req helpers sheepdog: implement SD_OP_FLUSH_VDI operation block: bdrv_append() fixes qed: track dirty flag status qemu-img: add dirty flag status qed: image fragmentation statistics qemu-img: add image fragmentation statistics block: document job API ...
2012-04-10petalogix_s3adsp1800: deleted bad FIXME commentPeter A. G. Crosthwaite1-1/+0
This FIXME has already been actioned. Deleted comment. Signed-off-by: Peter A. G. Crosthwaite <peter.crosthwaite@petalogix.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2012-04-09qapi: convert device_delLuiz Capitulino1-13/+5
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
2012-04-09qdev: qdev_unplug(): use error_set()Luiz Capitulino6-11/+35
It currently uses qerror_report(), but next commit will convert the drive_del command to the QAPI and this requires using error_set(). One particularity of qerror_report() is that it knows when it's running on monitor context or command-line context and prints the error message accordingly. error_set() doesn't do this, so we have to be careful not to drop error messages. qdev_unplug() has three kinds of usages: 1. It's called when hot adding a device fails, to undo anything that has been done before hitting the error 2. It's called by function monitor functions like device_del(), to unplug a device 3. It's used by xen_platform.c in a way that doesn't _seem_ to be in monitor context Only item 2 can print an error message to the user, this commit maintains that. Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
2012-04-07Replace Qemu by QEMU in commentsStefan Weil6-11/+11
The official spelling is QEMU. Signed-off-by: Stefan Weil <sw@weilnetz.de> Reviewed-by: Andreas Färber <afaerber@suse.de> [blauwirbel@gmail.com: fixed comment style in hw/sun4m.c] Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2012-04-05Use DMADirection type for dma_bdrv_ioDavid Gibson2-2/+4
Currently dma_bdrv_io() takes a 'to_dev' boolean parameter to determine the direction of DMA it is emulating. We already have a DMADirection enum designed specifically to encode DMA directions. This patch uses it for dma_bdrv_io() as well. This involves removing the DMADirection definition from the #ifdef it was inside, but since that only existed to protect the definition of dma_addr_t from places where config.h is not included, there wasn't any reason for it to be there in the first place. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Reviewed-by: Andreas Färber <afaerber@suse.de> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2012-04-05qdev: add blocksize property typeStefan Hajnoczi2-0/+49
Storage interfaces like virtio-blk can be configured with block size information so that the guest can take advantage of efficient I/O request sizes. According to the SCSI Block Commands (SBC) standard a device's block size is "almost always greater than one byte and may be a multiple of 512 bytes". QEMU currently has a 512 byte minimum block size because the block layer functions work at that granularity. Furthermore, the block size should be a power of 2 because QEMU calculates bitmasks from the value. Introduce a "blocksize" property type so devices can enforce these constraints on block size values. If the constraints are relaxed in the future then this property can be updated. Introduce the new PropertyValueNotPowerOf2 QError so QMP clients know exactly why a block size value was rejected. Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2012-04-05ide: Adds wwn=hex qdev optionFloris Bos3-8/+29
Allow the user to specify a disk's World Wide Name. Linux guests can address disks by their unique World Wide Name number (e.g. /dev/disk/by-id/wwn-0x5001517959123522). This patch adds support for assigning a World Wide Name number to a virtual IDE disk. Cc: kwolf@redhat.com Signed-off-by: Floris Bos <dev@noc-ps.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2012-04-05ide: Change serial number strncpy() to pstrcpy()Floris Bos1-1/+1
strncpy may not null-terminate the destination string. Cc: kwolf@redhat.com Signed-off-by: Floris Bos <dev@noc-ps.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2012-04-05ide: Add "model=s" qdev optionFloris Bos3-8/+29
Allow the user to override the default disk model name "QEMU HARDDISK". Some Linux distributions use the /dev/disk/by-id/scsi-SATA_name-of-disk- model_serial addressing scheme when refering to partitions in /etc/fstab and elsewhere. This causes problems when starting a disk image taken from an existing physical server under qemu, because when running under qemu name-of-disk-model is always "QEMU HARDDISK". This patch introduces a model=s option which in combination with the existing serial=s option can be used to fake the disk the operating system was previously on, allowing the OS to boot properly. Cc: kwolf@redhat.com Signed-off-by: Floris Bos <dev@noc-ps.com> Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2012-04-05ide: IDENTIFY word 86 bit 14 is reservedKevin Wolf1-1/+1
Reserved bits should be cleared to zero. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
2012-04-05aio: move BlockDriverAIOCB to qemu-aio.hPaolo Bonzini2-2/+0
And remove several block_int.h inclusions that should not be there. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2012-04-04Improve interrupt handling priorityArtyom Tarasenko1-1/+6
The vector interrupt has higher priority than interrupt_level_n. Also check only interrupt_level_n concurency when TL > 0, the traps of other types may be nested. Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2012-04-04Fix vector interrupt handlingArtyom Tarasenko1-13/+16
Don't produce stray irq 5, don't overwrite ivec_data if still busy with processing of the previous interrupt. Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2012-04-02qdev: put all devices under /machinePaolo Bonzini5-5/+18
Avoid cluttering too much the QOM root. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-04-02qdev: give all devices a canonical pathPaolo Bonzini1-3/+13
A strong limitation of QOM right now is that unconverted ports (e.g. all...) do not give a canonical path to devices that are part of the board. This in turn makes it impossible to replace PROP_PTR with a QOM link for example. Reviewed-by: Anthony Liguori <aliguori@us.ibm.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-04-02qdev: add children before qdev_initPaolo Bonzini4-24/+7
We want the composition tree to to be in order by the time we call qdev_init, so that a single set of the toplevel realize property can propagate all the way down the composition tree. This is not the case so far. Unfortunately, this is incompatible with calling qdev_init in the constructor wrappers for devices, so for now we need to unattach some devices that are created through those wrappers. This will be fixed by removing qdev_init and instead setting the toplevel realize property after machine init. Reviewed-by: Anthony Liguori <aliguori@us.ibm.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-04-02qom: add container_getPaolo Bonzini1-6/+2
This is QOM "mkdir -p". It is useful when referring to container objects such as "/machine". Reviewed-by: Anthony Liguori <aliguori@us.ibm.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-04-02serial: clear LSR.TEMT when populating the TSRAnthony Liguori1-0/+1
We never actually clear the TEMT (transmit sending register empty) flag when populating the TSR. We set the flag, but since it's never cleared, setting it is sort of pointless.. I found this with a unit test case. Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-04-02serial: fix retry logicAnthony Liguori1-1/+3
I'm not sure if the retry logic has ever worked when not using FIFO mode. I found this while writing a test case although code inspection confirms it is definitely broken. The TSR retry logic will never actually happen because it is guarded by an 'if (s->tsr_rety > 0)' but this is the only place that can ever make the variable greater than zero. That effectively makes the retry logic an 'if (0)'. I believe this is a typo and the intention was >= 0. Once this is fixed though, I see double transmits with my test case. This is because in the non FIFO case, serial_xmit may get invoked while LSR.THRE is still high because the character was processed but the retransmit timer was still active. We can handle this by simply checking for LSR.THRE and returning early. It's possible that the FIFO paths also need some attention. Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-04-01milkymist-vgafb: add missing registerMichael Walle1-1/+4
This bug existed since the first commit. Fortunately, the affected registers have no functionality in qemu. This will only prevent the following warning: milkymist_vgafb: write access to unknown register 0x00000034 Signed-off-by: Michael Walle <michael@walle.cc>
generated by cgit v1.2.1 (git 2.18.0) at 2024-07-09 22:52:05 +0000