Re: not working /dev/console

From: Warner Losh <imp_at_bsdimp.com>
Date: Fri, 24 Jan 2025 06:50:26 UTC
On Thu, Jan 23, 2025, 2:19 PM Garrett Wollman <wollman@bimajority.org>
wrote:

> Resurrecting an old thread, because I find myself in a position where
> I have to get a serial console working on a UEFI-only Dell R360 (with
> iDRAC Express, so no remote video console that I can look at).
>

What version?

>
> <<On Wed, 27 Nov 2024 09:42:24 -0700, Warner Losh <imp@bsdimp.com> said:
>
> > But, the tl;dr is likely just adding 'boot_serial=YES' and the
> > "hw.uart.console"
> > line above to /loader.conf and deleting all the comconsole lines (since
> they
> > will result in doubled output if the console really is redirected at the
> > UEFI
> > level, see below for more details on all this :_
>
> So this does not seem to be working for me.  The server boots and the
> loader is accessible over SOL so I am pretty confident that it's
> configured correctly on the UEFI side, but when booting the kernel,
> the only thing it displays is the initial probe message where it
> identifies where the EFI framebuffer is -- nothing else goes to the
> serial port.
>
> OK boot -s
> staging 0x1a800000 (not copying) tramp 0x284f1000 PT4 0x1a6e5000
> Start @ 0xffffffff803081e0 ...
> EFI framebuffer information:
> addr, size     0x40000000, 0x300000
> dimensions     1024 x 768
> stride         1024
> masks          0x00ff0000, 0x0000ff00, 0x000000ff, 0xff000000
> [then nothing]
>
> (The system doesn't have a network connection yet, and won't have SSH
> enabled when it's finally installed, so I actually need the serial
> console to work.  I have some other UEFI systems in a remote DC that
> hav broken consoles too, but fixing them requires a disruptive outage,
> so I'm trying to get it fixed first on a new machine.)
>
> Running "show" in the loader tells me the followin things are set:
>
> beastie_disable=YES
> boot_multicons=YES
> boot_serial=YES
> comconsole_pcidev=
> comconsole_port=760
> comconsole_speed=115200
> console=efi
> efi-version=2.70
> efi_8250_uid=0
> efi_com_speed=115195
> efi_max_resolution=1x1
> hint.uart.1.at=acpi
> hint.uart.1.flags=16
> hint.uart.1.port=0x2F8
> hw.uart.console=io:0x2f8,br:115200
> smbios.system.product=PowerEdge R360
> [a bunch of irrelevant stuff elided]
>
> The loader reports the following in `efi-show` (which if it is
> documented at all I do not know where):
>
> OK efi-show -v ConOut
> global NV,BS,RS ConOut =
> PciRoot(0x0)/Pci(0x1C,0x4)/Pci(0x0,0x0)/Pci(0x0,0x0)/AcpiAdr(0x80010100),/PciRoot(0x0)/Pci(0x1F,0x0)/Serial(0x0)/Uart(115195,8,N,1)/VenVt100()
>

Odd band rate. Makes me suspect serial clock rate. But this is from the
bios..

OK
> OK efi-show -v ConIn
> global NV,BS,RS ConIn =
> UsbHID(0xFFFF,0xFFFF,0x1,0x1),/PciRoot(0x0)/Pci(0x1F,0x0)/Acpi(PNP0303,0x0),/PciRoot(0x0)/Pci(0x1F,0x0)/Serial(0x0)/Uart(115195,8,N,1)/VenVt100()
>
> ...ao I'm guessing that this is taking precedence over `boot_serial`?
> Is it safe to just swap these around?
>

Boot -s should do that. If this is 14 can you see if 15 works? I've fixed a
few things.

Warner

>
> -GAWollman
>
>