linux-kernel - Re: [PATCH v2] powerpc: iommu: fix build when neither PCI or IBMVIO is set

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <135d33bc-5520-d49b-dd1c-582c7d8e3bac@csgroup.eu>
Date:   Mon, 5 Apr 2021 11:04:53 +0200
From:   Christophe Leroy <christophe.leroy@...roup.eu>
To:     Randy Dunlap <rdunlap@...radead.org>, linux-kernel@...r.kernel.org
Cc:     linuxppc-dev@...ts.ozlabs.org, kernel test robot <lkp@...el.com>,
        Anton Blanchard <anton@...ba.org>
Subject: Re: [PATCH v2] powerpc: iommu: fix build when neither PCI or IBMVIO
 is set



Le 04/04/2021 à 21:26, Randy Dunlap a écrit :
> When neither CONFIG_PCI nor CONFIG_IBMVIO is set/enabled, iommu.c has a
> build error. The fault injection code is not useful in that kernel config,
> so make the FAIL_IOMMU option depend on PCI || IBMVIO.
> 
> Prevents this build error (warning escalated to error):
> ../arch/powerpc/kernel/iommu.c:178:30: error: 'fail_iommu_bus_notifier' defined but not used [-Werror=unused-variable]
>    178 | static struct notifier_block fail_iommu_bus_notifier = {
> 
> Fixes: d6b9a81b2a45 ("powerpc: IOMMU fault injection")
> Signed-off-by: Randy Dunlap <rdunlap@...radead.org>
> Reported-by: kernel test robot <lkp@...el.com>
> Suggested-by: Michael Ellerman <mpe@...erman.id.au>
> Cc: Michael Ellerman <mpe@...erman.id.au>
> Cc: linuxppc-dev@...ts.ozlabs.org
> Cc: Anton Blanchard <anton@...ba.org>
> ---
> I was supposed to update this about one month ago, but then I lost
> some email and also took a break for a few weeks, then I remembered,
> so here it is.
> 
>   arch/powerpc/Kconfig.debug |    1 +
>   1 file changed, 1 insertion(+)

Wouldn't it be cleaner to get rid of those two horid #ifdefs ?
Of course we can do both.

diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
index 216871414434..d691afa8acf8 100644
--- a/arch/powerpc/kernel/iommu.c
+++ b/arch/powerpc/kernel/iommu.c
@@ -180,12 +180,10 @@ static struct notifier_block fail_iommu_bus_notifier = {

  static int __init fail_iommu_setup(void)
  {
-#ifdef CONFIG_PCI
-	bus_register_notifier(&pci_bus_type, &fail_iommu_bus_notifier);
-#endif
-#ifdef CONFIG_IBMVIO
-	bus_register_notifier(&vio_bus_type, &fail_iommu_bus_notifier);
-#endif
+	if (IS_ENABLED(CONFIG_PCI))
+		bus_register_notifier(&pci_bus_type, &fail_iommu_bus_notifier);
+	if (IS_ENABLED(CONFIG_IBMVIO))
+		bus_register_notifier(&vio_bus_type, &fail_iommu_bus_notifier);

  	return 0;
  }
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 86c799c97b77..361f4f255911 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -964,6 +964,8 @@ enum {
  #define PCI_IRQ_MSIX		(1 << 2) /* Allow MSI-X interrupts */
  #define PCI_IRQ_AFFINITY	(1 << 3) /* Auto-assign affinity */

+extern struct bus_type pci_bus_type;
+
  /* These external functions are only available when PCI support is enabled */
  #ifdef CONFIG_PCI

@@ -986,8 +988,6 @@ enum pcie_bus_config_types {

  extern enum pcie_bus_config_types pcie_bus_config;

-extern struct bus_type pci_bus_type;
-
  /* Do NOT directly access these two variables, unless you are arch-specific PCI
   * code, or PCI core code. */
  extern struct list_head pci_root_buses;	/* List of all known PCI buses */