All of lore.kernel.org
 help / color / mirror / Atom feed
From: Hans de Goede <hdegoede@redhat.com>
To: Daniel Kiper <daniel.kiper@oracle.com>
Cc: grub-devel@gnu.org, Colin Watson <cjwatson@ubuntu.com>
Subject: Re: [PATCH 3/4] Optionally print less messages at boot
Date: Thu, 5 Apr 2018 19:53:23 +0200	[thread overview]
Message-ID: <de7f1bde-bf63-a695-7270-ea53fe2657ce@redhat.com> (raw)
In-Reply-To: <20180405121257.GT26100@olila.local.net-space.pl>

Hi,

On 05-04-18 14:12, Daniel Kiper wrote:
> On Wed, Mar 28, 2018 at 04:50:27PM +0200, Hans de Goede wrote:
>> The patch optionally makes grub not show any text (be fully quiet) when
>> timeout_style=hidden is set and the user does not interrupt the boot.
>>
>> Combined with a later patch in this series which makes grub not touch
>> the EFI console unless it actually has some text to print, this will keep
>> the vendor logo which EFI put on the display in place until the kernel
>> touches the display. Leading to a more smooth / seamless boot experience.
>>
>> At least Fedora/RHEL/CentOS and Ubuntu have been carrying patches for this
>> for a long time now (since 2013). There have been several attempts to
>> upstream these patches in the past already, which have been rejected
>> because not everyone likes the quiet behavior.
> 
> May I ask you to provide links to the relevant conversations?

Sure you may ask, but I cannot really help with that because I was not
involved in grub development back then.

I've done a quick search of the mailinglist archives, but
I'm afraid I could not find anything that way.

>> This patch makes the quiet behavior optional and defaults to off, so
>> unless grub is compiled with the new --enable-quiet-boot configure option
>> this patch changes nothing.
> 
> I am not sure why this should be build time option. I would see this as
> a runtime option, e.g. boot_quiet shell variable or even timeout_style=quiet.
> Hmmm... Latter is probably preferred.

The problem is that grub already prints various things before reading
its environment or config file.

What I've understood from the history of this patches, removing those
earlier prints (see e.g. the grub-core/boot/i386/pc/boot.S change in
this patch) was objected against because that provides info that at
least the first stage (for a classic PC BIOS boot setup) has successfully
loaded.

The menu code itself is already quiet when using timeout_style=hidden,
the problem is the messages printed before (and after) the menu code.

If you're happy with killing the messages shown before the config file
is loaded (as most distros already do), then the rest could be made runtime
configurable.

Regards,

Hans


  reply	other threads:[~2018-04-05 17:53 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-28 14:50 [PATCH 0/4] Make hidden menu really hidden Hans de Goede
2018-03-28 14:50 ` [PATCH 1/4] Add new "version" command Hans de Goede
2018-03-29  7:53   ` Olaf Hering
2018-03-29  8:37     ` Thomas Schmitt
2018-03-29 14:57     ` Hans de Goede
2018-03-29 15:02       ` Olaf Hering
2018-04-05 11:41   ` Daniel Kiper
2018-03-28 14:50 ` [PATCH 2/4] Accept Both ESC and F8 as user interrupt keys Hans de Goede
2018-03-28 14:56   ` Lennart Sorensen
2018-03-28 15:06     ` Hans de Goede
2018-03-28 15:11       ` Lennart Sorensen
2018-03-28 15:58         ` Hans de Goede
2018-04-05 11:47           ` Daniel Kiper
2018-03-28 14:50 ` [PATCH 3/4] Optionally print less messages at boot Hans de Goede
2018-04-05 12:12   ` Daniel Kiper
2018-04-05 17:53     ` Hans de Goede [this message]
2018-04-06 11:41       ` Daniel Kiper
2018-03-28 14:50 ` [PATCH 4/4] EFI: Do not set text-mode until we actually need it Hans de Goede
2018-04-05 12:34   ` Daniel Kiper
2018-04-05 18:05     ` Hans de Goede
2018-04-06 11:44       ` Daniel Kiper
2018-03-28 15:04 ` [PATCH 0/4] Make hidden menu really hidden Daniel Kiper

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=de7f1bde-bf63-a695-7270-ea53fe2657ce@redhat.com \
    --to=hdegoede@redhat.com \
    --cc=cjwatson@ubuntu.com \
    --cc=daniel.kiper@oracle.com \
    --cc=grub-devel@gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.