Re: [PATCH v2 1/4] bhyve: support parsing fbuf PCI device
On a Tuesday in 2020, Roman Bogorodskiy wrote:
> +static int
> +bhyveParsePCIFbuf(virDomainDefPtr def,
> + virDomainXMLOptionPtr xmlopt,
> + unsigned caps G_GNUC_UNUSED,
> + unsigned bus,
> + unsigned slot,
> + unsigned function,
> + const char *config)
> +{
> + /* -s slot,fbuf,wait,vga=on|io|off,rfb=<ip>:port,w=width,h=height */
> +
> + virDomainVideoDefPtr video = NULL;
> + virDomainGraphicsDefPtr graphics = NULL;
> + char **params = NULL;
> + char *param = NULL, *separator = NULL;
> + size_t nparams = 0;
> + unsigned int i = 0;
Interesting, some of the CI jobs complain that this should be "size_t"
(which I'll fix before merging), and some don't (including my local
environment). Wondering if that's caused by different sed versions or
something else.
Looking at .gitlab-ci.yml:
We have a 'codestyle' job which specifically runs syntax-check:
meson test -C build --suite syntax-check --no-rebuild ||
And it's also run by the centos-7 job (which is special due to an old
git version):
ninja -C build test
The other jobs execute:
ninja -C build dist
Which apparently does not include syntax-check.
A different sed version might be the culprit - we use it to build the
list of different checks in build-aux/meson.build:
rc = run_command(
'sed', '-n',
's/^\\(sc_[a-zA-Z0-9_-]*\\):.*/\\1/p',
meson.current_source_dir() / 'syntax-check.mk',
check: true,
)
Jano
Attachments:
- signature.asc (application/pgp-signature — 488 bytes)