x86/amd-iommu: enable iommu before attaching devices

commit 75f66533bc883f761a7adcab3281fe3323efbc90 upstream.

Hit another kdump problem as reported by Neil Horman.  When initializaing
the IOMMU, we attach devices to their domains before the IOMMU is
fully (re)initialized.  Attaching a device will issue some important
invalidations.  In the context of the newly kexec'd kdump kernel, the
IOMMU may have stale cached data from the original kernel.  Because we
do the attach too early, the invalidation commands are placed in the new
command buffer before the IOMMU is updated w/ that buffer.  This leaves
the stale entries in the kdump context and can renders device unusable.
Simply enable the IOMMU before we do the attach.

Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index 362ab88c73ac0bedd5f4ce4765ebe596d8c0924b..3925adfba7654c1c40da35e5b30a96141d67ea5d 100644 (file)
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -1284,6 +1284,8 @@ int __init amd_iommu_init(void)
        if (ret)
                goto free;
 
+       enable_iommus();
+
        if (iommu_pass_through)
                ret = amd_iommu_init_passthrough();
        else
@@ -1294,8 +1296,6 @@ int __init amd_iommu_init(void)
 
        amd_iommu_init_api();
 
-       enable_iommus();
-
        if (iommu_pass_through)
                goto out;
 
@@ -1314,6 +1314,8 @@ out:
        return ret;
 
 free:
+       disable_iommus();
+
        free_pages((unsigned long)amd_iommu_pd_alloc_bitmap,
                   get_order(MAX_DOMAIN_ID/8));