qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Philippe Mathieu-Daudé" <f4bug@amsat.org>
To: qemu-devel@nongnu.org
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
	"Philippe Mathieu-Daudé" <f4bug@amsat.org>,
	"Peter Xu" <peterx@redhat.com>, "Greg Kurz" <groug@kaod.org>,
	qemu-arm@nongnu.org, qemu-ppc@nongnu.org,
	"Cédric Le Goater" <clg@kaod.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Hervé Poussineau" <hpoussin@reactos.org>,
	"David Gibson" <david@gibson.dropbear.id.au>
Subject: [PATCH v2 06/11] hw/pci-host/raven: Assert PCI I/O AddressSpace is based at 0x80000000
Date: Sat, 17 Apr 2021 12:30:23 +0200	[thread overview]
Message-ID: <20210417103028.601124-7-f4bug@amsat.org> (raw)
In-Reply-To: <20210417103028.601124-1-f4bug@amsat.org>

Commit 1ae1dc5ba24 ("raven: Set a correct PCI I/O memory region")
abused an AddressSpace API weakness which allows set non-zero base
address to AddressSpace root region. We will fix that in the next
commit. First add an assertion to ensure no regression is introduced.
As raven_io_address() is called by both MemoryRegionOps handlers, it
is a good place for such assert call.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
---
 hw/pci-host/raven.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/hw/pci-host/raven.c b/hw/pci-host/raven.c
index 730f31a8931..36652122424 100644
--- a/hw/pci-host/raven.c
+++ b/hw/pci-host/raven.c
@@ -141,6 +141,17 @@ static const MemoryRegionOps raven_intack_ops = {
 static inline hwaddr raven_io_address(PREPPCIState *s,
                                       hwaddr addr)
 {
+    /*
+     * We shouldn't access AddressSpace internals. However this assert
+     * is temporarily introduced to prove a subtle inconsistency from
+     * commit 1ae1dc5ba24 ("raven: Set a correct PCI I/O memory region"):
+     * AddressSpace root region must be zero-based, but the Raven use is not.
+     *
+     * Assert the root region is based on physical address 0x80000000
+     * until the issue is fixed.
+     */
+    assert(s->pci_io_as.root->addr == PCI_IO_BASE_ADDR);
+
     if (s->contiguous_map == 0) {
         /* 64 KB contiguous space for IOs */
         addr &= 0xFFFF;
-- 
2.26.3



  parent reply	other threads:[~2021-04-17 10:36 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-17 10:30 [PATCH v2 00/11] memory: Forbid mapping AddressSpace root MemoryRegion Philippe Mathieu-Daudé
2021-04-17 10:30 ` [PATCH v2 01/11] hw/arm/aspeed: Do not directly map ram container onto main address bus Philippe Mathieu-Daudé
2021-04-20 18:28   ` Peter Xu
2021-04-21  5:53     ` Cédric Le Goater
2021-04-21 13:02       ` Philippe Mathieu-Daudé
2021-04-17 10:30 ` [PATCH v2 02/11] hw/aspeed/smc: Use the RAM memory region for DMAs Philippe Mathieu-Daudé
2021-04-17 10:30 ` [PATCH v2 03/11] hw/arm/aspeed: Do not sysbus-map mmio flash region directly, use alias Philippe Mathieu-Daudé
2021-04-17 10:30 ` [PATCH v2 04/11] hw/pci-host: Rename Raven ASIC PCI bridge as raven.c Philippe Mathieu-Daudé
2021-04-19  0:46   ` David Gibson
2021-04-17 10:30 ` [PATCH v2 05/11] hw/pci-host/raven: Add PCI_IO_BASE_ADDR definition Philippe Mathieu-Daudé
2021-04-19  0:47   ` David Gibson
2021-04-17 10:30 ` Philippe Mathieu-Daudé [this message]
2021-04-19  1:00   ` [PATCH v2 06/11] hw/pci-host/raven: Assert PCI I/O AddressSpace is based at 0x80000000 David Gibson
2021-04-17 10:30 ` [PATCH v2 07/11] hw/pci-host/raven: Use MR alias for AS root, not sysbus mapped MR Philippe Mathieu-Daudé
2021-04-17 10:30 ` [RFC PATCH v2 08/11] hw/pci-host/raven: Remove pointless alias mapping onto system bus Philippe Mathieu-Daudé
2021-04-17 10:30 ` [PATCH v2 09/11] hw/pci-host/prep: Do not directly map bus-master region onto main bus Philippe Mathieu-Daudé
2021-04-17 10:30 ` [PATCH v2 10/11] memory: Make sure root MR won't be added as subregion Philippe Mathieu-Daudé
2021-04-17 10:30 ` [PATCH v2 11/11] hw/pci-host/raven: Remove temporary assertion 'root MR is zero-based' Philippe Mathieu-Daudé
2021-04-19  7:17 ` [PATCH v2 00/11] memory: Forbid mapping AddressSpace root MemoryRegion Cédric Le Goater
2021-04-19  9:48   ` Philippe Mathieu-Daudé
2021-04-20 19:07 ` Peter Xu

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=20210417103028.601124-7-f4bug@amsat.org \
    --to=f4bug@amsat.org \
    --cc=clg@kaod.org \
    --cc=david@gibson.dropbear.id.au \
    --cc=groug@kaod.org \
    --cc=hpoussin@reactos.org \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=peterx@redhat.com \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@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 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).