All of lore.kernel.org
 help / color / mirror / Atom feed
From: Boszormenyi Zoltan <zboszor@pr.hu>
To: DRI Development <dri-devel@lists.freedesktop.org>
Subject: radeon.pcie_gen2=1 unusable with AMD 990X chipset
Date: Mon, 05 Mar 2012 11:58:10 +0100	[thread overview]
Message-ID: <4F549C42.1020003@pr.hu> (raw)

Hi,

last week I bought a new ASUS M5A99X EVO AM3+ mainboard.
It supports PCIe 2.0. I also have a Radeon 6570 videocard (PCIe 2.0)
and a 3ware 9650SE 8LPML RAID card (PCIe 1.0).

There are 3 PCIe x16 slots on the mainboard, two of them share x16
wiring so single card x16 or dual card x8/x8 is possible. The 3rd slot
is PCIe x16 slot with dedicated x4 wiring, no sharing with the first 2 slots.
The RAID card is sitting in the 3rd slot.

I am running Fedora 16 with the latest 3.2.7-1.fc16.x86_64 kernel.

I have set radeon.pcie_gen2=1 in grub2 config to test PCIe 2.0 speed with
the videocard. It reliably locks up on boot. As a result, the mainboard
gets confused so much that it needs CMOS clear, otherwise even after
hard reset it cannot boot up successfully. I tried it a couple of times,
the symptoms (without CMOS clear) are mysteriously different. From the
few attempts, I experienced these:

- the mainboard reports faulty videocard, no image on display
- the mainboard passes the videocard test but reports faulty boot drive
  (I guess it cannot initialize the RAID card as it's the primary boot drive)
  and still no image on display
- the first few screens appear on the display but e.g. the onboard SATA
  controller cannot find the DVD drive and the BIOS initialization sequence
  stops somewhere, it doesn't reach the GRUB menu

The motherboard has an AMD 990X chipset.

Without radeon.pcie_gen2=1, the system works and is stable.

I don't know what to blame. The chipset, BIOS bugs, or the mix of
PCIe 1.0 and 2.0 devices. Anyway, I thought it would be an interesting
detail for Radeon developers.

Best regards,
Zoltán Böszörményi

                 reply	other threads:[~2012-03-05 12:18 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=4F549C42.1020003@pr.hu \
    --to=zboszor@pr.hu \
    --cc=dri-devel@lists.freedesktop.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.