Re: [PATCH 4/5] automation: improve checking for MSI/MSI-X in PCI passthrough tests

["Marek Marczykowski-Górecki" <marmarek@invisiblethingslab.com>]

[-- Attachment #1: Type: text/plain, Size: 1709 bytes --]

On Thu, Oct 19, 2023 at 12:30:22AM +0100, Andrew Cooper wrote:
> On 06/10/2023 3:05 am, Marek Marczykowski-Górecki wrote:
> > diff --git a/automation/scripts/qubes-x86-64.sh b/automation/scripts/qubes-x86-64.sh
> > index 5464d10fc343..842e6fae7204 100755
> > --- a/automation/scripts/qubes-x86-64.sh
> > +++ b/automation/scripts/qubes-x86-64.sh
> > @@ -90,23 +90,18 @@ on_reboot = "destroy"
> >  
> >      domU_check="
> >  set -x -e
> > -ip link set eth0 up
> > -timeout 30s udhcpc -i eth0
> > +interface=eth0
> > +ip link set \"\$interface\" up
> > +timeout 30s udhcpc -i \"\$interface\"
> >  pingip=\$(ip -o -4 r show default|cut -f 3 -d ' ')
> >  ping -c 10 \"\$pingip\"
> >  echo domU started
> > -cat /proc/interrupts
> > +pcidevice=\$(basename \$(readlink /sys/class/net/\$interface/device))
> > +lspci -vs \$pcidevice
> 
> Actually, I know I said I had no more comments, and maybe this is one
> for further cleanup rather than for now, but wouldn't we be much better
> using a heredoc?
> 
> read -r -d '' domU_check <<"EOF"
> 
> lorem ipsum, no \ escaping " or $
> 
> EOF
> 
> If nothing else it would make the innards of the more readable as a
> script fragment, and less likely go to wrong with variable expansion in
> the wrong context.

Some of those snippets have intentionally unescaped $ (as in - have it
expanded in place), and IMO having different method depending whether
you want to expand variables inside or not will be more error-prone.
Syntax highlighting in an editor makes it rather easy to spot unescaped
$ or such (and then decide whether it was intentional or not).

-- 
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]