All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrea Arcangeli <aarcange@redhat.com>
To: Gerd Hoffmann <kraxel@redhat.com>
Cc: qemu-devel@nongnu.org, Anthony Liguori <aliguori@amazon.com>
Subject: Re: [Qemu-devel] [PATCH 1/2] pc: add etc/e820 fw_cfg file
Date: Tue, 5 Nov 2013 18:48:01 +0100	[thread overview]
Message-ID: <20131105174801.GG3835@redhat.com> (raw)
In-Reply-To: <1383567431-13540-2-git-send-email-kraxel@redhat.com>

Hi,

On Mon, Nov 04, 2013 at 01:17:10PM +0100, Gerd Hoffmann wrote:
> Unlike the existing FW_CFG_E820_TABLE entry which carries reservations
> only the new etc/e820 file also has entries for RAM.

Acked, it looks the best the way to go if the objective is to keep
backwards compatibility with older seabios protocol.

I have to trust you on why we need to stay backwards compatible at all
times and why we can't commit an updated bios.bin before a new seabios
stable release happened. Otherwise we could have just fixed
FW_CFG_E820_TABLE breaking backwards compatibility without introducing
a partially overlapping fw_cfg.

About the file, I wonder what happens if too many people starts to use
files and we'll run out of FW_CFG_FILE_SLOTS at runtime (assert(index
< FW_CFG_FILE_SLOTS);). To me seeing the list of all fw_cfg IDs in a
header file to define all possible paravirt APIs seabios need to know
about, looked not so bad, but then grepping for fw_cfg_add_file is
equivalent. The main issue is that if we use files from now, it would
be nicer not to limit those to FW_CFG_FILE_SLOTS but to allocate them
a bit more dynamically without asserts but this is offtopic (I just
happened to read how files are created to review this patch).

Probably one patch could be added to
s/FW_CFG_E820_TABLE/FW_CFG_E820_TABLE_OLD/, otherwise if somebody read
fw_cfg.h, it won't be apparent that the grepping shouldn't stop there
to reach the real e820 map.

Thanks!
Andrea

  reply	other threads:[~2013-11-05 17:48 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-04 12:17 [Qemu-devel] [PULL for-1.7 0/2] pc: e820 fw_cfg fixup Gerd Hoffmann
2013-11-04 12:17 ` [Qemu-devel] [PATCH 1/2] pc: add etc/e820 fw_cfg file Gerd Hoffmann
2013-11-05 17:48   ` Andrea Arcangeli [this message]
2013-11-06  7:54     ` Gerd Hoffmann
2013-11-04 12:17 ` [Qemu-devel] [PATCH 2/2] pc: register e820 entries for ram Gerd Hoffmann
  -- strict thread matches above, loose matches on Subject: below --
2013-10-22 14:05 [Qemu-devel] [PATCH v2 0/2] fw_cfg: add etc/e820 Gerd Hoffmann
2013-10-22 14:05 ` [Qemu-devel] [PATCH 1/2] pc: add etc/e820 fw_cfg file Gerd Hoffmann

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=20131105174801.GG3835@redhat.com \
    --to=aarcange@redhat.com \
    --cc=aliguori@amazon.com \
    --cc=kraxel@redhat.com \
    --cc=qemu-devel@nongnu.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.