Previously, copying the image would consume a lot of space and was
I/O-intensive, to the point that the marionette connection timeout of
20s could be reached when running tests like "docker-system".
* gnu/system/vm.scm (common-qemu-options): Pass 'format=' for each
'-drive' option.
(system-qemu-image/shared-store-script)[copy-image]: New variable.
[builder]: Use it when VOLATILE? is false.
This code duplicates the (gnu system image) and (gnu build image) code. Using
VM for image generation is not needed, not portable and really slow. Remove
all the VM image generation code to make sure that only the image API is used.
* gnu/build/vm.scm: Remove it. Move the qemu-command procedure to ...
* gnu/build/marionette.scm: ... here.
* gnu/local.mk (GNU_SYSTEM_MODULES): Adapt it.
* tests/modules.scm: Ditto.
* gnu/tests/install.scm: Ditto.
* gnu/system/vm.scm: Adapt it and remove expression->derivation-in-linux-vm,
qemu-img, system-qemu-image/shared-store and system-docker-image procedures.
* doc/guix.texi (G-Expressions): Adapt it.
Also add a volatile? argument to the virtual-machine record. When volatile? is
true generate a QEMU script that mounts an overlay on top of a read only
storage. When volatile? is false, use a persistent, read-write storage.
* gnu/system/vm.scm (common-qemu-options): Add a rw-image? argument to use a
persistent storage.
(system-qemu-image/shared-store-script): Add a volatile? argument and honor
it. Use the image API to build the QEMU image.
(<virtual-machine>)[volatile?]: New field.
(virtual-machine-compiler): Pass the volatile? argument to the
system-qemu-image/shared-store-script procedure.
* gnu/system/vm.scm (common-qemu-options): Remove the "-vga std" argument that
is the default since QEMU 2.2.
Suggested-by: Attila Lendvai <attila@lendvai.name>
It configures QEMU to use the tty it was started in for IO (as opposed to
opening a separate graphical window). This enables copy-pasting and
scrollback. Exit QEMU with C-a x.
* gnu/system/vm.scm (system-qemu-image/shared-store-script): Also add the
'-nographic' QEMU argument when called with #:graphic? #false.
* guix/scripts/system.scm (system-derivation-for-action): Propagate
the #:graphic? keyword arg.
(perform-action): Add and propagate the #:graphic? keyword arg.
(show-help): Add --no-graphic CLI arg.
(%options): Process the new --no-graphic CLI arg.
(process-action): Feed in the CLI arg through the #:graphic? keyword arg.
* doc/guix.texi (Invoking guix system): Add example to combine guix system and
invoking the vm start script into one command. Split the description of the
various options into separate paragraphs. Add a paragraph to describe
--no-graphic.
Signed-off-by: Mathieu Othacehe <othacehe@gnu.org>
'qemu-minimal' is enough for system tests, which is the main (only?) use
of <virtual-machine>.
* gnu/system/vm.scm (<virtual-machine>)[qemu]: Default to QEMU-MINIMAL.
Fixes <https://issues.guix.gnu.org/47225>.
* gnu/system/vm.scm (%default-msize-value): New variable.
(%linux-vm-file-systems): Use it as the value of the msize option.
(mapping->file-system): Likewise.
Reported-by: Leo Famulari <leo@famulari.name>
Remove the 'vm-image' command that has been superseded by the 'image'
command.
* gnu/system/vm.scm (system-qemu-image): Remove it.
* guix/scripts/system.scm (system-derivation-for-action): Mark 'vm-image'
command as deprecated and use the image API to produce the VM image.
(perform-action, show-help): Adapt accordingly.
* tests/guix-system.sh: Ditto.
* doc/guix.texi (Invoking guix system,
Running Guix in a VM): Ditto.
* etc/completion/fish/guix.fish: Ditto.
* etc/completion/zsh/_guix: Ditto.
Fixes <https://bugs.gnu.org/44511>.
Reported by Maxim Cournoyer <maxim.cournoyer@gmail.com>.
* gnu/system/vm.scm (system-qemu-image): Add 'bootloader' field to OS.
* gnu/system/vm.scm (system-qemu-image)[file-systems-to-keep]: Check
whether SOURCE is a string before calling 'string-prefix?'. Remove
UUIDs and file system labels as well.
* gnu/build/vm.scm (load-in-linux-vm): Rename the RNG.
* gnu/system/vm.scm (common-qemu-options): Likewise.
(system-docker-image): Rename the ROOT-DIRECTORY.
* gnu/packages/crypto.scm (eschalot)[arguments]: Use a different
arbitrary string.
* gnu/packages/wicd.scm (wicd)[arguments]: Remove unused configure flag.
* gnu/packages/xorg.scm (xorg-server): Set a more accurate OS vendor.
This allows users to specify network interface settings with 'guix system vm'
without having to create a new NIC. Fixes <https://bugs.gnu.org/42252>.
Reported by Christopher Lemmer Webber <cwebber@dustycloud.org>.
This reverts commit 5379392731.
* gnu/system/vm.scm (qemu-image)[preserve-target, inputs*]: New variables.
In gexp, use INPUTS* instead of INPUTS. Wrap OS and BOOTCFG-DRV in
'preserve-target'. Pass INPUTS* instead of INPUTS as the #:references-graphs.
* gnu/system/vm.scm (expression->derivation-in-linux-vm): Remove #:target.
[builder]: Use #+. Don't pass #:target-arm32? and #:target-aarch64? to
'load-in-linux-vm'.
Pass #:target #f to 'gexp->derivation'.
(qemu-image): Adjust accordingly.
* gnu/build/vm.scm (load-in-linux-vm): Remove #:target-aarch64?
and #:target-arm32?. Define them as local variables.
* gnu/system/vm.scm (expression->derivation-in-linux-vm)[check]: New macro.
[builder]: Use 'let-system' and 'check' instead of referencing
'%current-system' and '%current-target-system'.
* gnu/system/vm.scm (system-docker-image): Add #:shared-network? and
pass it to 'containerized-operating-system'.
(qemu-image):
* guix/scripts/system.scm (system-derivation-for-action): Pass
#:shared-network? to 'system-docker-image'.
* doc/guix.texi (Invoking guix system): Document it.
* gnu/build/vm.scm (install-efi, make-iso9660-image): Remove those procedures
that are now implemented in (gnu build image) module,
(initialize-hard-disk): remove efi support.
* gnu/system/vm.scm (iso9660-image): Remove it,
(qemu-image): adapt it to remove ISO9660 support.
Raw disk-images and ISO9660 images are created in a Qemu virtual machine. This
is quite fragile, very slow, and almost unusable without KVM.
For all these reasons, add support for host image generation. This implies the
use new image generation mechanisms.
- Raw disk images: images of partitions are created using tools such as mke2fs
and mkdosfs depending on the partition file-system type. The partition
images are then assembled into a final image using genimage.
- ISO9660 images: the ISO root directory is populated within the store. GNU
xorriso is then called on that directory, in the exact same way as this is
done in (gnu build vm) module.
Those mechanisms are built upon the new (gnu image) module.
* gnu/image.scm: New file.
* gnu/system/image.scm: New file.
* gnu/build/image: New file.
* gnu/local.mk: Add them.
* gnu/system/vm.scm (system-disk-image): Rename to system-disk-image-in-vm.
* gnu/ci.scm (qemu-jobs): Adapt to new API.
* gnu/tests/install.scm (run-install): Ditto.
* guix/scripts/system.scm (system-derivation-for-action): Ditto.
* gnu/build/vm.scm (root-partition-initializer): Add #:make-device-nodes
parameter and use it.
* gnu/system/vm.scm (qemu-image): Add #:device-node parameter. Pass
#:make-device-nodes to 'root-partition-initializer'.
* gnu/build/vm.scm (<partition>)[file-system-options]: New field.
(create-ext-file-system, create-fat-file-system)
(format-partition): Add #:options and honor it.
(initialize-partition): Pass #:options to 'format-partition'.
* gnu/system/vm.scm (qemu-image): Add #:file-system-options and use it
for the root partition.
