x86: don't assume all fb devices are PCI devices
PATCH v3 1/1 x86: don't assume all fb devices are PCI devices
When booting Hyper-V Generation 2 guests KASAN reports the following out-of-bounds access:
BUG: KASAN: slab-out-of-bounds in fb_is_primary_device+0x58/0x70 at addr ffff880079cf0eb0 Read of size 8 by task swapper/0/1 ... [<ffffffff81581308>] dump_stack+0x63/0x8b [<ffffffff812e1f99>] print_trailer+0xf9/0x150 [<ffffffff812e7344>] object_err+0x34/0x40 [<ffffffff812e9630>] kasan_report_error+0x230/0x550 [<ffffffff812e9ee8>] kasan_report+0x58/0x60 [<ffffffff812e4500>] ? ___slab_alloc+0x80/0x490 [<ffffffff81878a28>] ? fb_is_primary_device+0x58/0x70 [<ffffffff812e87cd>] __asan_load8+0x5d/0x70 [<ffffffff81878a28>] fb_is_primary_device+0x58/0x70 [<ffffffff8162357a>] register_framebuffer+0xda/0x5b0 [<ffffffff816234a0>] ? remove_conflicting_framebuffers+0x50/0x50
...
The issue is caused by the to_pci_dev() call with no check that the given info->device is in fact a pci device and some fb devices (Hyper-V FB, EFI FB,...) are not. While on it, cleanup the function.
---
Changes since v2: cleanup fb_is_primary_device() [Bjorn Helgaas, Ingo Molnar].
--- arch/x86/video/fbdev.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-)
arch/x86/video/fbdev.c +8 -10
@@ -14,26 +14,24 @@int fb_is_primary_device (struct fb_info *info) {struct device *device = info->device; -struct pci_dev *pci_dev =NULL ;struct pci_dev *default_device =vga_default_device (); -struct resource *res =NULL ; +struct pci_dev *pci_dev; +struct resource *res; -if (device) - pci_dev =to_pci_dev (device); - -if (!pci_dev) +if (!device || !dev_is_pci (device))return 0 ; + pci_dev =to_pci_dev (device); +if (default_device) {if (pci_dev == default_device)return 1 ; -else -return 0 ; +return 0 ; } - res = &pci_dev->resource[PCI_ROM_RESOURCE ]; + res = pci_dev->resource +PCI_ROM_RESOURCE ; -if (res && res->flags &IORESOURCE_ROM_SHADOW ) +if (res->flags &IORESOURCE_ROM_SHADOW )return 1 ;return 0 ;
Signature · 1 line
2.5.0
Ingo Molnar
AppliedQuoted 20 lines · Vitaly Kuznetsov
When booting Hyper-V Generation 2 guests KASAN reports the following out-of-bounds access:
BUG: KASAN: slab-out-of-bounds in fb_is_primary_device+0x58/0x70 at addr ffff880079cf0eb0 Read of size 8 by task swapper/0/1 ... [<ffffffff81581308>] dump_stack+0x63/0x8b [<ffffffff812e1f99>] print_trailer+0xf9/0x150 [<ffffffff812e7344>] object_err+0x34/0x40 [<ffffffff812e9630>] kasan_report_error+0x230/0x550 [<ffffffff812e9ee8>] kasan_report+0x58/0x60 [<ffffffff812e4500>] ? ___slab_alloc+0x80/0x490 [<ffffffff81878a28>] ? fb_is_primary_device+0x58/0x70 [<ffffffff812e87cd>] __asan_load8+0x5d/0x70 [<ffffffff81878a28>] fb_is_primary_device+0x58/0x70 [<ffffffff8162357a>] register_framebuffer+0xda/0x5b0 [<ffffffff816234a0>] ? remove_conflicting_framebuffers+0x50/0x50...
Quoted 7 lines
The issue is caused by the
to_pci_dev()call with no check that the giveninfo->deviceis in fact a pci device and some fb devices (Hyper-V FB, EFI FB,...) are not. While on it, cleanup the function.Signed-off-by Vitaly Kuznetsov <[email protected]>---
Changes since v2: cleanupfb_is_primary_device()[Bjorn Helgaas, Ingo Molnar].
Applied, thanks!
I also added Bjorn's Acked-by to the commit, as the only change from the last version are the cleanups.
Thanks,
Ingo
tip-bot for Vitaly Kuznetsov
merge bot[tip:x86/urgent] x86/video: Don't assume all FB devices are PCI devices
In reply to Vitaly Kuznetsovx86/video: Don't assume all FB devices are PCI devices
Commit-ID: 743146db071c4a828159211a295d12ff4f61752f Gitweb: http://git.kernel.org/tip/743146db071c4a828159211a295d12ff4f61752f Author: Vitaly Kuznetsov <[email protected]> AuthorDate: Tue, 15 Mar 2016 09:20:33 +0100 Committer: Ingo Molnar <[email protected]> CommitDate: Tue, 15 Mar 2016 11:08:26 +0100
x86/video: Don't assume all FB devices are PCI devices
When booting Hyper-V Generation 2 guests KASAN reports the following out-of-bounds access:
BUG: KASAN: slab-out-of-bounds in fb_is_primary_device+0x58/0x70 at addr ffff880079cf0eb0 Read of size 8 by task swapper/0/1 ... [<ffffffff81581308>] dump_stack+0x63/0x8b [<ffffffff812e1f99>] print_trailer+0xf9/0x150 [<ffffffff812e7344>] object_err+0x34/0x40 [<ffffffff812e9630>] kasan_report_error+0x230/0x550 [<ffffffff812e9ee8>] kasan_report+0x58/0x60 [<ffffffff812e4500>] ? ___slab_alloc+0x80/0x490 [<ffffffff81878a28>] ? fb_is_primary_device+0x58/0x70 [<ffffffff812e87cd>] __asan_load8+0x5d/0x70 [<ffffffff81878a28>] fb_is_primary_device+0x58/0x70 [<ffffffff8162357a>] register_framebuffer+0xda/0x5b0 [<ffffffff816234a0>] ? remove_conflicting_framebuffers+0x50/0x50 ...
The issue is caused by the to_pci_dev() call with no check that the given info->device is in fact a PCI device and some FB devices (Hyper-V FB, EFI FB,...) are not.
While on it, clean up the function.
--- arch/x86/video/fbdev.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-)
arch/x86/video/fbdev.c +8 -10
@@ -14,26 +14,24 @@int fb_is_primary_device (struct fb_info *info) {struct device *device = info->device; -struct pci_dev *pci_dev =NULL ;struct pci_dev *default_device =vga_default_device (); -struct resource *res =NULL ; +struct pci_dev *pci_dev; +struct resource *res; -if (device) - pci_dev =to_pci_dev (device); - -if (!pci_dev) +if (!device || !dev_is_pci (device))return 0 ; + pci_dev =to_pci_dev (device); +if (default_device) {if (pci_dev == default_device)return 1 ; -else -return 0 ; +return 0 ; } - res = &pci_dev->resource[PCI_ROM_RESOURCE ]; + res = pci_dev->resource +PCI_ROM_RESOURCE ; -if (res && res->flags &IORESOURCE_ROM_SHADOW ) +if (res->flags &IORESOURCE_ROM_SHADOW )return 1 ;return 0 ;