From: Gerd Hoffmann <kraxel@redhat.com>
To: alexey.korolev@endace.com
Cc: mst@redhat.com, sfd@endace.com, seabios@seabios.org,
qemu-devel@nongnu.org, kevin@koconnor.net, avi@redhat.com
Subject: Re: [Qemu-devel] [SeaBIOS] [Seabios] [PATCH 0/6] 64bit PCI BARs allocations (take 2)
Date: Thu, 01 Mar 2012 10:05:54 +0100 [thread overview]
Message-ID: <4F4F3BF2.7040003@redhat.com> (raw)
In-Reply-To: <1330581043.29508.61.camel@nzhmlwks0057.ad.endace.com>
Hi,
> a) Provide 64bit bar support for PCI BARs and bridges with 64bit memory
> window.
Bridge support seems to be completely untested. /me has a test setup
using mst's bridge patches which looks like this:
[root@fedora ~]# lspci -tv
-[0000:00]-+-00.0 Intel Corporation 440FX - 82441FX PMC [Natoma]
+-01.0 Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II]
+-01.1 Intel Corporation 82371SB PIIX3 IDE [Natoma/Triton II]
+-01.2 Intel Corporation 82371SB PIIX3 USB [Natoma/Triton II]
+-01.3 Intel Corporation 82371AB/EB/MB PIIX4 ACPI
+-02.0 Red Hat, Inc. Device 0100
+-03.0 Red Hat, Inc Virtio network device
+-05.0 Red Hat, Inc Virtio console
+-06.0 Intel Corporation 82801AA AC'97 Audio Controller
+-0a.0 Red Hat, Inc Device 1009
+-0b.0 Red Hat, Inc Device 1009
+-10.0-[01]----01.0 Red Hat, Inc Device 1110
+-10.1-[02]----01.0 Red Hat, Inc Virtio memory balloon
+-10.2-[03]--
+-10.3-[04]--
+-11.0-[05-07]--+-01.0-[06]--+-01.0 Red Hat, Inc Device 1110
| | \-02.0 Red Hat, Inc Device 1110
| \-02.0-[07]--+-01.0 Red Hat, Inc Device 1110
| \-02.0 Red Hat, Inc Device 1110
+-1d.0 Intel Corporation 82801I (ICH9 Family) USB UHCI
Controller #1
+-1d.1 Intel Corporation 82801I (ICH9 Family) USB UHCI
Controller #2
+-1d.2 Intel Corporation 82801I (ICH9 Family) USB UHCI
Controller #3
+-1d.7 Intel Corporation 82801I (ICH9 Family) USB2 EHCI
Controller #1
\-1f.0 Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port
SATA AHCI Controller
The linux kernel completely redoes the bridge resource assignment. The
seabios log looks sane for the root bus devices but not for devices
behind bridges:
=== PCI new allocation pass #2 ===
PCI region entry: bdf=00:06.0 BAR 0 base=0x000000000000c000
size=0x0000000000000400 [io, 32bits]
PCI region entry: bdf=00:06.0 BAR 1 base=0x000000000000c400
size=0x0000000000000100 [io, 32bits]
PCI region entry: bdf=00:0b.0 BAR 0 base=0x000000000000c500
size=0x0000000000000040 [io, 32bits]
PCI region entry: bdf=00:0a.0 BAR 0 base=0x000000000000c540
size=0x0000000000000040 [io, 32bits]
PCI region entry: bdf=00:1f.0 BAR 4 base=0x000000000000c580
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:1d.2 BAR 4 base=0x000000000000c5a0
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:1d.1 BAR 4 base=0x000000000000c5c0
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:1d.0 BAR 4 base=0x000000000000c5e0
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:05.0 BAR 0 base=0x000000000000c600
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:03.0 BAR 0 base=0x000000000000c620
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:02.0 BAR 3 base=0x000000000000c640
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:01.2 BAR 4 base=0x000000000000c660
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=00:01.1 BAR 4 base=0x000000000000c680
size=0x0000000000000010 [io, 32bits]
PCI region entry: bdf=00:02.0 BAR 0 base=0x00000000f8000000
size=0x0000000004000000 [mem, 32bits]
PCI region entry: bdf=00:02.0 BAR 1 base=0x00000000fc000000
size=0x0000000001000000 [mem, 32bits]
PCI region entry: bdf=00:03.0 BAR 6 base=0x00000000fd000000
size=0x0000000000010000 [mem, 32bits]
PCI region entry: bdf=00:02.0 BAR 6 base=0x00000000fd010000
size=0x0000000000010000 [mem, 32bits]
PCI region entry: bdf=00:02.0 BAR 2 base=0x00000000fd020000
size=0x0000000000002000 [mem, 32bits]
PCI region entry: bdf=00:1f.0 BAR 5 base=0x00000000fd022000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=00:1d.7 BAR 0 base=0x00000000fd023000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=00:0b.0 BAR 1 base=0x00000000fd024000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=00:0a.0 BAR 1 base=0x00000000fd025000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=00:05.0 BAR 1 base=0x00000000fd026000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=00:03.0 BAR 1 base=0x00000000fd027000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=00:02.0 BAR 4 base=0x00000000e0000000
size=0x0000000010000000 [prefmem, 64bits]
PCI region entry: bdf=01:01.0 BAR 0 base=0x0000000000000000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=01:01.0 BAR 2 base=0x0000000000000000
size=0x0000000040000000 [prefmem, 64bits]
PCI region entry: bdf=02:01.0 BAR 0 base=0x0000000000000000
size=0x0000000000000020 [io, 32bits]
PCI region entry: bdf=06:02.0 BAR 0 base=0x0000000000000000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=06:01.0 BAR 0 base=0x0000000000001000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=06:02.0 BAR 2 base=0x0000000000000000
size=0x0000000002000000 [prefmem, 64bits]
PCI region entry: bdf=06:01.0 BAR 2 base=0x0000000002000000
size=0x0000000002000000 [prefmem, 64bits]
PCI region entry: bdf=07:02.0 BAR 0 base=0x0000000000000000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=07:01.0 BAR 0 base=0x0000000000001000
size=0x0000000000001000 [mem, 32bits]
PCI region entry: bdf=07:02.0 BAR 2 base=0x0000000000000000
size=0x0000000002000000 [prefmem, 64bits]
PCI region entry: bdf=07:01.0 BAR 2 base=0x0000000002000000
size=0x0000000002000000 [prefmem, 64bits]
cheers,
Gerd
next prev parent reply other threads:[~2012-03-01 9:06 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-01 5:50 [Qemu-devel] [Seabios] [PATCH 0/6] 64bit PCI BARs allocations (take 2) Alexey Korolev
2012-03-01 6:05 ` [Qemu-devel] [Seabios] [PATCH 1/6] Adding new structures Alexey Korolev
2012-03-01 6:15 ` [Qemu-devel] [PATCH 2/6] New service functions and ported old functions to 64bit Alexey Korolev
2012-03-01 6:40 ` [Qemu-devel] [PATCH 3/6] Fill PCI regions with etnries Alexey Korolev
2012-03-01 6:57 ` [Qemu-devel] [PATCH 4/6] Mapping of BARs and Bridge regions Alexey Korolev
2012-03-01 9:22 ` [Qemu-devel] [SeaBIOS] " Gerd Hoffmann
2012-03-01 22:01 ` Alexey Korolev
2012-03-02 7:21 ` Gerd Hoffmann
2012-03-05 5:31 ` Alexey Korolev
2012-03-01 7:02 ` [Qemu-devel] [PATCH 5/6] Delete old code Alexey Korolev
2012-03-01 7:11 ` [Qemu-devel] [PATCH 6/6] 64bit PCI range in _CRS table Alexey Korolev
2012-03-01 9:05 ` Gerd Hoffmann [this message]
2012-03-01 21:48 ` [Qemu-devel] [SeaBIOS] [Seabios] [PATCH 0/6] 64bit PCI BARs allocations (take 2) Alexey Korolev
2012-03-02 7:08 ` Gerd Hoffmann
2012-03-05 5:34 ` Alexey Korolev
2012-03-05 10:12 ` Gerd Hoffmann
2012-03-06 4:28 ` Alexey Korolev
2012-03-04 19:40 ` [Qemu-devel] " Kevin O'Connor
2012-03-05 6:03 ` Alexey Korolev
2012-03-05 13:16 ` Kevin O'Connor
2012-03-05 9:53 ` [Qemu-devel] [SeaBIOS] " Gerd Hoffmann
2012-03-05 13:49 ` Kevin O'Connor
2012-03-06 4:44 ` Alexey Korolev
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=4F4F3BF2.7040003@redhat.com \
--to=kraxel@redhat.com \
--cc=alexey.korolev@endace.com \
--cc=avi@redhat.com \
--cc=kevin@koconnor.net \
--cc=mst@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=seabios@seabios.org \
--cc=sfd@endace.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).