Ensure PCI configuration access is within bounds
v2 cover Ensure PCI configuration access is within bounds
From: Prasad J Pandit <[email protected]>
Hello,
This patch series fixes
1. While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Add check to ensure 'addr + size' is within bounds.
2. Assert that PCI configuration access is within bounds.
Thank you.
--
Prasad J Pandit (2):
ait-vga: check address before reading configuration bytes
pci: ensure configuration access is within bounds
hw/display/ati.c | 5 ++++- hw/pci/pci.c | 2 ++ 2 files changed, 6 insertions(+), 1 deletion(-)
Signature · 1 line
2.26.2
PATCH v2 1/2 ait-vga: check address before reading configuration bytes
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Add check to ensure 'address + size' is within PCI configuration space.
--- hw/display/ati.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
Update v2: add check to avoid OOB PCI configuration space access
-> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/display/ati.c +4 -1
@@ -384,7 +384,10 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size) val = s->regs.crtc_pitch;break ; case 0xf00 ... 0xfff: - val =pci_default_read_config (&s->dev, addr -0xf00 , size); + addr = addr -0xf00 ; +if (addr + size <=0xff ) { + val =pci_default_read_config (&s->dev, addr, size); + }break ;case CUR_OFFSET : val = s->regs.cur_offset;
Signature · 1 line
2.26.2
Re: [PATCH v2 1/2] ait-vga: check address before reading configuration bytes
Quoted 23 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Add check to ensure 'address + size' is within PCI configuration space.
Reported-by Ren Ding <[email protected]>Reported-by Hanqing Zhao <[email protected]>Reported-by Yi Ren <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/display/ati.c| 5 ++++-1 file changed, 4 insertions(+), 1 deletion(-)Update v2: add check to avoid OOB PCI configuration space access
-> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.htmlhw/display/ati.c +0
-0
hw/display/ati.c static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size) · line 387
Quoted 5 lines
hw/display/ati.c +1
-1@@ -384,7 +384,10 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size) val = s->regs.crtc_pitch;break ; case 0xf00 ... 0xfff: - val =pci_default_read_config (&s->dev, addr -0xf00 , size); + addr = addr -0xf00 ;
I'd write this as addr -= 0xf00 but modifying addr here breaks the trace print out at end of function so better drop this line and do
if (addr + size <= 0xfff) {
instead. Or is that really (addr + size - 1 <= 0xfff) considering that reading the last byte with addr = 0xfff size = 1 should probably be valid.
Regards,
BALATON Zoltan
Quoted 7 lines
hw/display/ati.c +3
-0+if (addr + size <=0xff ) { + val =pci_default_read_config (&s->dev, addr, size); + }break ;case CUR_OFFSET : val = s->regs.cur_offset;
Re: [PATCH v2 1/2] ait-vga: check address before reading configuration bytes
In reply to P J PTypo: s/ait/ati/
Quoted 6 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Add check to ensure 'address + size' is within PCI configuration space.
Please include a CVE number for this security flaw if there is one.
Quoted 31 lines
Reported-by Ren Ding <[email protected]>Reported-by Hanqing Zhao <[email protected]>Reported-by Yi Ren <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>--- hw/display/ati.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)Update v2: add check to avoid OOB PCI configuration space access
-> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.htmlhw/display/ati.c +4
-1@@ -384,7 +384,10 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size) val = s->regs.crtc_pitch;break ; case 0xf00 ... 0xfff: - val =pci_default_read_config (&s->dev, addr -0xf00 , size); + addr = addr -0xf00 ; +if (addr + size <=0xff ) { + val =pci_default_read_config (&s->dev, addr, size); + }break ;case CUR_OFFSET : val = s->regs.cur_offset;Signature · 1 line
2.26.2
Regards,
Daniel
Signature · 3 lines
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
Re: [PATCH v2 1/2] ait-vga: check address before reading configuration bytes
Hello Phil,
+-- On Thu, 4 Jun 2020, Philippe Mathieu-Daudé wrote --+ | >> @@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, | >> + assert(address + len <= pci_config_size(d)); | | Yes, maybe I was not clear while reviewing v1, we need to audit the | callers and fix them first, then we can safely add the assert here.
That's an elaborate task. Could we please make that into another patch series?
+-- On Thu, 4 Jun 2020, Daniel P. Berrangé wrote --+ | On Thu, Jun 04, 2020 at 01:52:50AM +0530, P J P wrote: | > While reading PCI configuration bytes, a guest may send an | > address towards the end of the configuration space. It may lead | > to an OOB access issue. Add check to ensure 'address + size' is | > within PCI configuration space. | | Please include a CVE number for this security flaw if there is | one.
Yes. For now I'll send a revised patch to fix this ati-vga OOB access issue.
Thank you.
--
Prasad J Pandit / Red Hat Product Security Team
8685 545E B54C 486B C6EB 271E E285 8B5A F050 DE8D
Re: [PATCH v2 1/2] ait-vga: check address before reading configuration bytes
Quoted 10 lines · P J P
Hello Phil,
+-- On Thu, 4 Jun 2020, Philippe Mathieu-Daudé wrote --+ | >> @@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, | >> +   assert(address + len <= pci_config_size(d)); | | Yes, maybe I was not clear while reviewing v1, we need to audit the | callers and fix them first, then we can safely add the assert here.That's an elaborate task. Could we please make that into another patch series?
So let's make this a separate patch as defence in depth in case there are more bugs somewhere. Patch 1 is a CVE fix. Patch 2 is not.
Quoted 16 lines
+-- On Thu, 4 Jun 2020, Daniel P. Berrangé wrote --+ | On Thu, Jun 04, 2020 at 01:52:50AM +0530, P J P wrote: | > While reading PCI configuration bytes, a guest may send an | > address towards the end of the configuration space. It may lead | > to an OOB access issue. Add check to ensure 'address + size' is | > within PCI configuration space. | | Please include a CVE number for this security flaw if there is | one.Yes. For now I'll send a revised patch to fix this ati-vga OOB access issue.
Thank you.
--
Prasad J Pandit / Red Hat Product Security Team
8685 545E B54C 486B C6EB 271E E285 8B5A F050 DE8D
PATCH v2 2/2 pci: ensure configuration access is within bounds
In reply to P J PFrom: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
--- hw/pci/pci.c | 2 ++ 1 file changed, 2 insertions(+)
Update v2: assert PCI configuration access is within bounds
-> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +2 -0
@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {uint32_t val =0 ; +assert (address + len <=pci_config_size (d)); +if (pci_is_express_downstream_port (d) &&ranges_overlap (address, len, d->exp.exp_cap +PCI_EXP_LNKSTA ,2 )) {pcie_sync_bridge_lnk (d);
Signature · 1 line
2.26.2
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds
-> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.htmlhw/pci/pci.c +0
-0
hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {uint32_t val =0 ; +assert (address + len <=pci_config_size (d));
Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.c
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Regards,
BALATON Zoltan
Quoted 5 lines
hw/pci/pci.c +1
-0+if (pci_is_express_downstream_port (d) &&ranges_overlap (address, len, d->exp.exp_cap +PCI_EXP_LNKSTA ,2 )) {pcie_sync_bridge_lnk (d);
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Hi,
Quoted 3 lines
Quoted 1 line
+
assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU?
Looks like it does (didn't actually try though).
Quoted 3 lines
I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest.
Correct. We do have guest-triggerable asserts in the code base. They are not the end of the world as the guest will only hurt itself. But in general we try to get rid of them instead of adding new ones ...
Often you can just ignore the illegal guest action (bonus points for logging GUEST_ERROR as debugging aid). Sometimes it is more difficult to deal with it (in case the hardware is expected to throw an error irq for example).
take care,
Gerd
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 16 lines · Gerd Hoffmann
Hi,
Quoted 3 lines
Quoted 1 line
+
assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU?
Looks like it does (didn't actually try though).
Quoted 3 lines
I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest.
Correct. We do have guest-triggerable asserts in the code base. They are not the end of the world as the guest will only hurt itself. But in general we try to get rid of them instead of adding new ones ...
Quoted 7 lines
Often you can just ignore the illegal guest action (bonus points for logging
GUEST_ERRORas debugging aid). Sometimes it is more difficult to deal with it (in case the hardware is expected to throw an error irq for example).take care,
Gerd
In this case it's not supposed to be guest triggerable, so I'm inlined to merge this, but as a separate patch from patch 1,
and commit log need to be clearer that it's defence in depth not a bugfix.
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
In reply to BALATON Zoltan+-- On Thu, 4 Jun 2020, BALATON Zoltan wrote --+ | On Thu, 4 Jun 2020, P J P wrote: | > + assert(address + len <= pci_config_size(d)); | | Does this allow guest now to crash QEMU?
Yes, possible. Such crash (assert failure) can be a regular bug, as reading PCI configuration is likely a privileged operation inside guest.
| If this is considered to be an error now to call this function with wrong | parameters did you check other callers?
No, I haven't checked all other cases.
| Would it be better to not crash just log invalid access and either fix up | parameters or return some garbage like 0?
* Earlier patch v1 did the same, returned 0.
* Assert(3) may help to fix current and future incorrect usage of the call.
@mst ...?
Thank you.
--
Prasad J Pandit / Red Hat Product Security Team
8685 545E B54C 486B C6EB 271E E285 8B5A F050 DE8D
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
In reply to BALATON ZoltanQuoted 33 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, { uint32_t val = 0; + assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
Quoted 7 lines
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.cand maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
Quoted 10 lines
Regards,
BALATON ZoltanQuoted 6 lines
hw/pci/pci.c +1
-0+if (
pci_is_express_downstream_port(d)&&ranges_overlap(address, len, d->exp.exp_cap + PCI_EXP_LNKSTA, 2)) {
pcie_sync_bridge_lnk(d);
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 39 lines · Philippe Mathieu-Daudé
Quoted 38 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds  -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, { Â Â Â uint32_t val = 0; +Â Â Â assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.c
Quoted 5 lines
Quoted 2 lines
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
We can add assert here regardless of auditing callers. Doing that will also make fuzzying easier. But the assert is unrelated to CVE imho.
Quoted 10 lines
Quoted 10 lines
Regards,
BALATON ZoltanQuoted 6 lines
hw/pci/pci.c +1
-0+Â Â Â if (
pci_is_express_downstream_port(d)&&
      Âranges_overlap(address, len, d->exp.exp_cap + PCI_EXP_LNKSTA, 2)) {
      Âpcie_sync_bridge_lnk(d);
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 40 lines · Michael S. Tsirkin
Quoted 39 lines · Philippe Mathieu-Daudé
Quoted 38 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds  -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, { Â Â Â uint32_t val = 0; +Â Â Â assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.c
Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
We can add assert here regardless of auditing callers. Doing that will also make fuzzying easier. But the assert is unrelated to CVE imho.
I wonder why isn't the check added to pci_default_read_config() right away? If we have an assert there the overhead is the same and adding the check there would make it unnecessary to patch all callers so it's just one patch instead of a whole series.
Regards,
BALATON Zoltan
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 50 lines · BALATON Zoltan
Quoted 40 lines · Michael S. Tsirkin
Quoted 39 lines · Philippe Mathieu-Daudé
Quoted 38 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds  -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {    uint32_t val = 0; +   assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.cQuoted 8 lines
Quoted 5 lines
Quoted 2 lines
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
We can add assert here regardless of auditing callers. Doing that will also make fuzzying easier. But the assert is unrelated to CVE imho.
Quoted 7 lines
I wonder why isn't the check added to
pci_default_read_config()right away? If we have an assert there the overhead is the same and adding the check there would make it unnecessary to patch all callers so it's just one patch instead of a whole series.Regards,
BALATON Zoltan
We need to return something, and we can't be sure that callers will handle returning random stuff correctly. Callers know what to do on errors, we don't.
Signature · 1 line
MST
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 56 lines · Michael S. Tsirkin
Quoted 55 lines · BALATON Zoltan
Quoted 40 lines · Michael S. Tsirkin
Quoted 39 lines · Philippe Mathieu-Daudé
Quoted 38 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds  -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {    uint32_t val = 0; +   assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.cQuoted 8 lines
Quoted 5 lines
Quoted 2 lines
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
We can add assert here regardless of auditing callers. Doing that will also make fuzzying easier. But the assert is unrelated to CVE imho.
I wonder why isn't the check added to
pci_default_read_config()right away? If we have an assert there the overhead is the same and adding the check there would make it unnecessary to patch all callers so it's just one patch instead of a whole series.
Quoted 6 lines
Quoted 2 lines
Regards,
BALATON ZoltanWe need to return something, and we can't be sure that callers will handle returning random stuff correctly. Callers know what to do on errors, we don't.
This is an invalid case where behaviour will be undefined anyway so returning anything such as 0 or -1 is probably OK (what do most hardware return in this case?). If callers need better error handling they can do a check before calling the function but for other (most) callers which will just return the same random value you would return from pci_default_read_config() having an assert instead makes it necessary to modify all of them one by one and doubles the check overhead by unnecessarily double checking. So I think having a default check and error handling in pci_default_read_config() would be better so callers who don't care would work and those few who might care could check before calling or actually implement their own callback (which I expect they already do as this is just the default implementation of this callback).
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 60 lines · BALATON Zoltan
Quoted 59 lines · Michael S. Tsirkin
Quoted 57 lines · BALATON Zoltan
Quoted 40 lines · Michael S. Tsirkin
Quoted 39 lines · Philippe Mathieu-Daudé
Quoted 38 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds  -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {    uint32_t val = 0; +   assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.cQuoted 8 lines
Quoted 5 lines
Quoted 2 lines
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
We can add assert here regardless of auditing callers. Doing that will also make fuzzying easier. But the assert is unrelated to CVE imho.
I wonder why isn't the check added to
pci_default_read_config()right away? If we have an assert there the overhead is the same and adding the check there would make it unnecessary to patch all callers so it's just one patch instead of a whole series.Regards,
BALATON Zoltan
Quoted 7 lines
Quoted 3 lines
We need to return something, and we can't be sure that callers will handle returning random stuff correctly. Callers know what to do on errors, we don't.
This is an invalid case where behaviour will be undefined anyway so returning anything such as 0 or -1 is probably OK (what do most hardware return in this case?).
This is an internal detail of the API. It's not about what hardware returns. Look at the ati as an example.
Quoted 10 lines
If callers need better error handling they can do a
check before calling the function but for other (most) callers which will just return the same random value you would return frompci_default_read_config()having an assert instead makes it necessary to modify all of them one by one and doubles the check overhead by unnecessarily double checking. So I think having a default check and error handling inpci_default_read_config()would be better so callers who don't care would work and those few who might care could check before calling or actually implement their own callback (which I expect they already do as this is just the default implementation of this callback).
Basically if you look at the specific example, you will see that it triggers because of a misaligned access which device code never expected. Which memory core should not allow at all.
It will likely trigger other bugs, some of them could be security related. assert is a reasonable way to help us catch them in fuzzying.
Signature · 1 line
MST
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 65 lines · Michael S. Tsirkin
Quoted 64 lines · BALATON Zoltan
Quoted 62 lines · Michael S. Tsirkin
Quoted 50 lines · BALATON Zoltan
Quoted 40 lines · Michael S. Tsirkin
Quoted 39 lines · Philippe Mathieu-Daudé
Quoted 38 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds  -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {    uint32_t val = 0; +   assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.cQuoted 8 lines
Quoted 5 lines
Quoted 2 lines
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
We can add assert here regardless of auditing callers. Doing that will also make fuzzying easier. But the assert is unrelated to CVE imho.
Quoted 7 lines
I wonder why isn't the check added to
pci_default_read_config()right away? If we have an assert there the overhead is the same and adding the check there would make it unnecessary to patch all callers so it's just one patch instead of a whole series.Regards,
BALATON ZoltanWe need to return something, and we can't be sure that callers will handle returning random stuff correctly. Callers know what to do on errors, we don't.
Quoted 6 lines
Quoted 3 lines
This is an invalid case where behaviour will be undefined anyway so returning anything such as 0 or -1 is probably OK (what do most hardware return in this case?).
This is an internal detail of the API. It's not about what hardware returns. Look at the ati as an example.
Considering that this function implements reading PCI config space its API should aligh with what happens on hardware normally. You could make it unrelated but that does not make much sense other than causing trouble for callers.
Quoted 12 lines
Quoted 10 lines
If callers need better error handling they can do a
check before calling the function but for other (most) callers which will just return the same random value you would return frompci_default_read_config()having an assert instead makes it necessary to modify all of them one by one and doubles the check overhead by unnecessarily double checking. So I think having a default check and error handling inpci_default_read_config()would be better so callers who don't care would work and those few who might care could check before calling or actually implement their own callback (which I expect they already do as this is just the default implementation of this callback).
Quoted 6 lines
Basically if you look at the specific example, you will see that it triggers because of a misaligned access which device code never expected. Which memory core should not allow at all.
It will likely trigger other bugs, some of them could be security related. assert is a reasonable way to help us catch them in fuzzying.
The specific example (ati-vga) does expect and should support unaligned access. Not for all regs but for most registers, there's a table in docs which says for PCI POS registers (whatever those are) unalligned access is supported. This works now, if it should not work witout .impl.unaligned or some other value set somewhere that should be patched instead.
Regards,
BALATON Zoltan
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 70 lines · BALATON Zoltan
Quoted 69 lines · Michael S. Tsirkin
Quoted 67 lines · BALATON Zoltan
Quoted 56 lines · Michael S. Tsirkin
Quoted 55 lines · BALATON Zoltan
Quoted 40 lines · Michael S. Tsirkin
Quoted 39 lines · Philippe Mathieu-Daudé
Quoted 38 lines · BALATON Zoltan
Quoted 21 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>---
hw/pci/pci.c| 2 ++1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds  -> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.html
hw/pci/pci.c +0
-0hw/pci/pci.c uint32_t pci_default_read_config(PCIDevice *d, · line 1384
Quoted 4 lines
hw/pci/pci.c +1
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {    uint32_t val = 0; +   assert(address + len <= pci_config_size(d));Does this allow guest now to crash QEMU? I think it was suggested that assert should only be used for cases that can only arise from a programming error and not from values set by the guest. If this is considered to be an error now to call this function with wrong parameters did you check other callers? I've found a few such as:
hw/scsi/esp-pci.chw/watchdog/wdt_i6300esb.chw/ide/cmd646.chw/vfio/pci.cQuoted 8 lines
Quoted 5 lines
Quoted 2 lines
and maybe others. Would it be better to not crash just log invalid access and either fix up parameters or return some garbage like 0?
Yes, maybe I was not clear while reviewing v1, we need to audit the callers and fix them first, then we can safely add the assert here.
We can add assert here regardless of auditing callers. Doing that will also make fuzzying easier. But the assert is unrelated to CVE imho.
I wonder why isn't the check added to
pci_default_read_config()right away? If we have an assert there the overhead is the same and adding the check there would make it unnecessary to patch all callers so it's just one patch instead of a whole series.Quoted 6 lines
Quoted 2 lines
Regards,
BALATON ZoltanWe need to return something, and we can't be sure that callers will handle returning random stuff correctly. Callers know what to do on errors, we don't.
This is an invalid case where behaviour will be undefined anyway so returning anything such as 0 or -1 is probably OK (what do most hardware return in this case?).
Quoted 7 lines
Quoted 2 lines
This is an internal detail of the API. It's not about what hardware returns. Look at the ati as an example.
Considering that this function implements reading PCI config space its API should aligh with what happens on hardware normally. You could make it unrelated but that does not make much sense other than causing trouble for callers.
What happens on hardware is that there's no way to send to device a transaction that is out of range: on pci offset is 8 bit so <= 0xff, and on express 12 bit so <= 4K.
So this handles something that never happens on real hardware and it happens because of a bug elsewhere in QEMU. assert seems appropriate.
Quoted 19 lines
Quoted 12 lines
Quoted 10 lines
If callers need better error handling they can do a
check before calling the function but for other (most) callers which will just return the same random value you would return frompci_default_read_config()having an assert instead makes it necessary to modify all of them one by one and doubles the check overhead by unnecessarily double checking. So I think having a default check and error handling inpci_default_read_config()would be better so callers who don't care would work and those few who might care could check before calling or actually implement their own callback (which I expect they already do as this is just the default implementation of this callback).Quoted 6 lines
Basically if you look at the specific example, you will see that it triggers because of a misaligned access which device code never expected. Which memory core should not allow at all.
It will likely trigger other bugs, some of them could be security related. assert is a reasonable way to help us catch them in fuzzying.
Quoted 2 lines
The specific example (ati-vga) does expect and should support unaligned access.
Then it should set "unaligned = true". It does not seem to do so.
Quoted 4 lines
Not for all regs but for most registers, there's a table in docs which says for PCI POS registers (whatever those are) unalligned access is supported. This works now, if it should not work witout .impl.unaligned or some other value set somewhere that should be patched instead.
Argue with the docs/devel/memory.rst about this please, that's not what it says.
Quoted 3 lines
Regards,
BALATON Zoltan
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
In reply to P J PQuoted 7 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
What does the spec say should happen when the guest does this? Does it depend on the pci controller implementation?
thanks
-- PMM
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
Quoted 10 lines · Peter Maydell
Quoted 7 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
What does the spec say should happen when the guest does this?
Spec says anything can happen *to the device*. Naturally there's an expectation that while device might crash it stays resettable and does not blow up.
Quoted 4 lines
Does it depend on the pci controller implementation?
thanks
-- PMM
Shouldn't I think.
Signature · 1 line
MST
Re: [PATCH v2 2/2] pci: ensure configuration access is within bounds
In reply to P J PQuoted 9 lines · P J P
From: Prasad J Pandit <[email protected]>
While reading PCI configuration bytes, a guest may send an address towards the end of the configuration space. It may lead to an OOB access issue. Assert that 'address + len' is within PCI configuration space.
Suggested-by Philippe Mathieu-Daudé <[email protected]>Signed-off-by Prasad J Pandit <[email protected]>
My understanding is that this can't really happen normally, this is more an assert in case some pci host devices are buggy, as is the case of alt-vga.
Right?
Pls clarify commit log so it's obvious this is defence in depth.
Quoted 22 lines
--- hw/pci/pci.c | 2 ++ 1 file changed, 2 insertions(+)Update v2: assert PCI configuration access is within bounds
-> https://lists.gnu.org/archive/html/qemu-devel/2020-06/msg00711.htmlhw/pci/pci.c +2
-0@@ -1381,6 +1381,8 @@ uint32_t pci_default_read_config(PCIDevice *d, {uint32_t val =0 ; +assert (address + len <=pci_config_size (d)); +if (pci_is_express_downstream_port (d) &&ranges_overlap (address, len, d->exp.exp_cap +PCI_EXP_LNKSTA ,2 )) {pcie_sync_bridge_lnk (d);Signature · 1 line
2.26.2