2018-09-12 17:49:13 |
bugproxy |
bug |
|
|
added bug |
2018-09-12 17:59:31 |
bugproxy |
tags |
|
architecture-ppc64le bugnameltc-171273 severity-high targetmilestone-inin1804 |
|
2018-09-13 00:18:57 |
Ubuntu Foundations Team Bug Bot |
tags |
architecture-ppc64le bugnameltc-171273 severity-high targetmilestone-inin1804 |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 |
|
2018-09-13 15:08:08 |
Brian Murray |
affects |
ubuntu |
linux (Ubuntu) |
|
2018-09-17 08:16:48 |
Joseph Salisbury |
linux (Ubuntu): importance |
Undecided |
Medium |
|
2018-09-17 08:16:52 |
Joseph Salisbury |
linux (Ubuntu): status |
New |
Incomplete |
|
2018-09-26 17:06:42 |
Manoj Iyer |
bug task added |
|
ubuntu-power-systems |
|
2018-09-26 17:07:02 |
Manoj Iyer |
ubuntu-power-systems: assignee |
|
Canonical Kernel Team (canonical-kernel-team) |
|
2018-09-26 17:07:17 |
Manoj Iyer |
linux (Ubuntu): assignee |
|
Canonical Kernel Team (canonical-kernel-team) |
|
2018-09-26 17:07:21 |
Manoj Iyer |
ubuntu-power-systems: importance |
Undecided |
High |
|
2018-09-26 18:21:01 |
Joseph Salisbury |
linux (Ubuntu): status |
Incomplete |
Triaged |
|
2018-09-26 18:22:15 |
Joseph Salisbury |
linux (Ubuntu): assignee |
Canonical Kernel Team (canonical-kernel-team) |
Joseph Salisbury (jsalisbury) |
|
2018-10-01 13:38:38 |
Manoj Iyer |
ubuntu-power-systems: status |
New |
Triaged |
|
2018-10-01 18:13:23 |
Joseph Salisbury |
linux (Ubuntu): status |
Triaged |
In Progress |
|
2018-10-01 18:15:47 |
Joseph Salisbury |
nominated for series |
|
Ubuntu Bionic |
|
2018-10-01 18:15:47 |
Joseph Salisbury |
bug task added |
|
linux (Ubuntu Bionic) |
|
2018-10-01 18:15:55 |
Joseph Salisbury |
linux (Ubuntu Bionic): status |
New |
In Progress |
|
2018-10-01 18:15:59 |
Joseph Salisbury |
linux (Ubuntu Bionic): importance |
Undecided |
Medium |
|
2018-10-01 18:16:03 |
Joseph Salisbury |
linux (Ubuntu Bionic): assignee |
|
Joseph Salisbury (jsalisbury) |
|
2018-10-08 14:32:39 |
Frank Heimes |
ubuntu-power-systems: status |
Triaged |
Incomplete |
|
2018-10-10 07:39:48 |
bugproxy |
attachment added |
|
backported dd1 plus patch https://bugs.launchpad.net/bugs/1792195/+attachment/5199401/+files/lp1792195-removedd1-fix-dd2.2.patch |
|
2018-10-10 09:26:45 |
Andrew Cloke |
ubuntu-power-systems: status |
Incomplete |
In Progress |
|
2018-10-10 13:38:55 |
Andrew Cloke |
ubuntu-power-systems: importance |
High |
Critical |
|
2018-10-10 13:38:57 |
Manoj Iyer |
linux (Ubuntu): importance |
Medium |
Critical |
|
2018-10-10 13:38:58 |
Manoj Iyer |
linux (Ubuntu Bionic): importance |
Medium |
Critical |
|
2018-10-11 16:21:20 |
Joseph Salisbury |
nominated for series |
|
Ubuntu Cosmic |
|
2018-10-11 16:21:20 |
Joseph Salisbury |
bug task added |
|
linux (Ubuntu Cosmic) |
|
2018-10-11 16:21:56 |
Joseph Salisbury |
description |
-- Problem Description --
GPFS mmfsd daemon is mapping shared tracing buffer(allocated from kernel driver using vmalloc) and then writing trace records from user space threads in parallel. While the SIGBUS happened, the access virtual memory address is in the mapped range, no overflow on access.
Worked with Benjamin Herrenschmidt on GPFS tracing kernel driver code and he made a suggestion as workaround on the driver code to bypass the problem, and it works....
the workaround code change as below:
- rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, PAGE_SHARED);
+ rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, __pgprot(pgprot_val(PAGE_SHARED)|_PAGE_DIRTY);
As Benjamin mentioned, this is a Linux kernel bug and this is just a workaround. He will give the details about the kernel bug and why this workaround works....
The root cause is that for PTEs created by a driver at mmap time (ie, that aren't created dynamically at fault time), it's not legit for ptep_set_access_flags() to make them invalid even temporarily. A concurrent access while they are invalid will be unable to service the page fault and will cause as SIGBUS.
Thankfully such PTEs shouldn't normally be the subject of a RO->RW privilege escalation.
What happens is that the GPFS driver creates the PTEs using remap_pfn_range(...,PAGE_SHARED).
PAGE_SHARED has _PAGE_ACCESSED (R) but not _PAGE_DIRTY (C) set.
Thus on the first write, we try set C and while doing so, hit the above workaround, which causes the problem described earlier.
The proposed patch will ensure we only do the Nest MMU hack when changing _PAGE_RW and not for normal R/C updates.
The workaround tested by the GPFS team consists of adding _PAGE_DIRTY to the mapping created by remap_pfn_range() to avoid the RC update fault completely.
This is fixed by these:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=bd0dbb73e01306a1060e56f81e5fe287be936477
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=f08d08f3db55452d31ba4a37c702da6245876b96
Since DD1 support is still in (ie, 2bf1071a8d50928a4ae366bb3108833166c2b70c is not applied) the second doesn't apply cleanly. Did you want that attached? |
== SRU Justification ==
IBM is requesting these commits in bionic and cosmic. These commits
also rely on commit 7acf50e4efa6, which was SRU'd in bug 1792102. The
first patch, commit 2bf1071a8d50, was backported by IBM themself.
Description of bug:
GPFS mmfsd daemon is mapping shared tracing buffer(allocated from kernel
driver using vmalloc) and then writing trace records from user space threads
in parallel. While the SIGBUS happened, the access virtual memory address
is in the mapped range, no overflow on access.
The root cause is that for PTEs created by a driver at mmap time (ie, that
aren't created dynamically at fault time), it's not legit for ptep_set_access_flags()
to make them invalid even temporarily. A concurrent access while they are
invalid will be unable to service the page fault and will cause as SIGBUS.
== Fixes ==
2bf1071a8d50 ("powerpc/64s: Remove POWER9 DD1 support")
bd0dbb73e013 ("powerpc/mm/books3s: Add new pte bit to mark pte temporarily invalid.")
f08d08f3db55 ("powerpc/mm/radix: Only need the Nest MMU workaround for R -> RW transition")
== Regression Potential ==
Low. Limited to powerpc.
== Test Case ==
A test kernel was built with these patches and tested by IBM.
IBM states the test kernel resolved the bug.
-- Problem Description --
GPFS mmfsd daemon is mapping shared tracing buffer(allocated from kernel driver using vmalloc) and then writing trace records from user space threads in parallel. While the SIGBUS happened, the access virtual memory address is in the mapped range, no overflow on access.
Worked with Benjamin Herrenschmidt on GPFS tracing kernel driver code and he made a suggestion as workaround on the driver code to bypass the problem, and it works....
the workaround code change as below:
- rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, PAGE_SHARED);
+ rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, __pgprot(pgprot_val(PAGE_SHARED)|_PAGE_DIRTY);
As Benjamin mentioned, this is a Linux kernel bug and this is just a workaround. He will give the details about the kernel bug and why this workaround works....
The root cause is that for PTEs created by a driver at mmap time (ie, that aren't created dynamically at fault time), it's not legit for ptep_set_access_flags() to make them invalid even temporarily. A concurrent access while they are invalid will be unable to service the page fault and will cause as SIGBUS.
Thankfully such PTEs shouldn't normally be the subject of a RO->RW privilege escalation.
What happens is that the GPFS driver creates the PTEs using remap_pfn_range(...,PAGE_SHARED).
PAGE_SHARED has _PAGE_ACCESSED (R) but not _PAGE_DIRTY (C) set.
Thus on the first write, we try set C and while doing so, hit the above workaround, which causes the problem described earlier.
The proposed patch will ensure we only do the Nest MMU hack when changing _PAGE_RW and not for normal R/C updates.
The workaround tested by the GPFS team consists of adding _PAGE_DIRTY to the mapping created by remap_pfn_range() to avoid the RC update fault completely.
This is fixed by these:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=bd0dbb73e01306a1060e56f81e5fe287be936477
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=f08d08f3db55452d31ba4a37c702da6245876b96
Since DD1 support is still in (ie, 2bf1071a8d50928a4ae366bb3108833166c2b70c is not applied) the second doesn't apply cleanly. Did you want that attached? |
|
2018-10-16 14:47:40 |
Joseph Salisbury |
description |
== SRU Justification ==
IBM is requesting these commits in bionic and cosmic. These commits
also rely on commit 7acf50e4efa6, which was SRU'd in bug 1792102. The
first patch, commit 2bf1071a8d50, was backported by IBM themself.
Description of bug:
GPFS mmfsd daemon is mapping shared tracing buffer(allocated from kernel
driver using vmalloc) and then writing trace records from user space threads
in parallel. While the SIGBUS happened, the access virtual memory address
is in the mapped range, no overflow on access.
The root cause is that for PTEs created by a driver at mmap time (ie, that
aren't created dynamically at fault time), it's not legit for ptep_set_access_flags()
to make them invalid even temporarily. A concurrent access while they are
invalid will be unable to service the page fault and will cause as SIGBUS.
== Fixes ==
2bf1071a8d50 ("powerpc/64s: Remove POWER9 DD1 support")
bd0dbb73e013 ("powerpc/mm/books3s: Add new pte bit to mark pte temporarily invalid.")
f08d08f3db55 ("powerpc/mm/radix: Only need the Nest MMU workaround for R -> RW transition")
== Regression Potential ==
Low. Limited to powerpc.
== Test Case ==
A test kernel was built with these patches and tested by IBM.
IBM states the test kernel resolved the bug.
-- Problem Description --
GPFS mmfsd daemon is mapping shared tracing buffer(allocated from kernel driver using vmalloc) and then writing trace records from user space threads in parallel. While the SIGBUS happened, the access virtual memory address is in the mapped range, no overflow on access.
Worked with Benjamin Herrenschmidt on GPFS tracing kernel driver code and he made a suggestion as workaround on the driver code to bypass the problem, and it works....
the workaround code change as below:
- rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, PAGE_SHARED);
+ rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, __pgprot(pgprot_val(PAGE_SHARED)|_PAGE_DIRTY);
As Benjamin mentioned, this is a Linux kernel bug and this is just a workaround. He will give the details about the kernel bug and why this workaround works....
The root cause is that for PTEs created by a driver at mmap time (ie, that aren't created dynamically at fault time), it's not legit for ptep_set_access_flags() to make them invalid even temporarily. A concurrent access while they are invalid will be unable to service the page fault and will cause as SIGBUS.
Thankfully such PTEs shouldn't normally be the subject of a RO->RW privilege escalation.
What happens is that the GPFS driver creates the PTEs using remap_pfn_range(...,PAGE_SHARED).
PAGE_SHARED has _PAGE_ACCESSED (R) but not _PAGE_DIRTY (C) set.
Thus on the first write, we try set C and while doing so, hit the above workaround, which causes the problem described earlier.
The proposed patch will ensure we only do the Nest MMU hack when changing _PAGE_RW and not for normal R/C updates.
The workaround tested by the GPFS team consists of adding _PAGE_DIRTY to the mapping created by remap_pfn_range() to avoid the RC update fault completely.
This is fixed by these:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=bd0dbb73e01306a1060e56f81e5fe287be936477
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=f08d08f3db55452d31ba4a37c702da6245876b96
Since DD1 support is still in (ie, 2bf1071a8d50928a4ae366bb3108833166c2b70c is not applied) the second doesn't apply cleanly. Did you want that attached? |
== SRU Justification ==
IBM is requesting these commits in bionic and cosmic. These commits
also rely on commit 7acf50e4efa6, which was SRU'd in bug 1792102.
Description of bug:
GPFS mmfsd daemon is mapping shared tracing buffer(allocated from kernel
driver using vmalloc) and then writing trace records from user space threads
in parallel. While the SIGBUS happened, the access virtual memory address
is in the mapped range, no overflow on access.
The root cause is that for PTEs created by a driver at mmap time (ie, that
aren't created dynamically at fault time), it's not legit for ptep_set_access_flags()
to make them invalid even temporarily. A concurrent access while they are
invalid will be unable to service the page fault and will cause as SIGBUS.
== Fixes ==
bd0dbb73e013 ("powerpc/mm/books3s: Add new pte bit to mark pte temporarily invalid.")
f08d08f3db55 ("powerpc/mm/radix: Only need the Nest MMU workaround for R -> RW transition")
== Regression Potential ==
Low. Limited to powerpc.
== Test Case ==
A test kernel was built with these patches and tested by IBM.
IBM states the test kernel resolved the bug.
-- Problem Description --
GPFS mmfsd daemon is mapping shared tracing buffer(allocated from kernel driver using vmalloc) and then writing trace records from user space threads in parallel. While the SIGBUS happened, the access virtual memory address is in the mapped range, no overflow on access.
Worked with Benjamin Herrenschmidt on GPFS tracing kernel driver code and he made a suggestion as workaround on the driver code to bypass the problem, and it works....
the workaround code change as below:
- rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, PAGE_SHARED);
+ rc = remap_pfn_range(vma, start, pfn, PAGE_SIZE, __pgprot(pgprot_val(PAGE_SHARED)|_PAGE_DIRTY);
As Benjamin mentioned, this is a Linux kernel bug and this is just a workaround. He will give the details about the kernel bug and why this workaround works....
The root cause is that for PTEs created by a driver at mmap time (ie, that aren't created dynamically at fault time), it's not legit for ptep_set_access_flags() to make them invalid even temporarily. A concurrent access while they are invalid will be unable to service the page fault and will cause as SIGBUS.
Thankfully such PTEs shouldn't normally be the subject of a RO->RW privilege escalation.
What happens is that the GPFS driver creates the PTEs using remap_pfn_range(...,PAGE_SHARED).
PAGE_SHARED has _PAGE_ACCESSED (R) but not _PAGE_DIRTY (C) set.
Thus on the first write, we try set C and while doing so, hit the above workaround, which causes the problem described earlier.
The proposed patch will ensure we only do the Nest MMU hack when changing _PAGE_RW and not for normal R/C updates.
The workaround tested by the GPFS team consists of adding _PAGE_DIRTY to the mapping created by remap_pfn_range() to avoid the RC update fault completely.
This is fixed by these:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=bd0dbb73e01306a1060e56f81e5fe287be936477
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=f08d08f3db55452d31ba4a37c702da6245876b96
Since DD1 support is still in (ie, 2bf1071a8d50928a4ae366bb3108833166c2b70c is not applied) the second doesn't apply cleanly. Did you want that attached? |
|
2018-10-23 14:46:32 |
Kleber Sacilotto de Souza |
linux (Ubuntu Bionic): status |
In Progress |
Fix Committed |
|
2018-10-24 13:35:28 |
Brad Figg |
tags |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 verification-needed-bionic |
|
2018-10-24 14:50:01 |
Brad Figg |
tags |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 verification-needed-bionic |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 verification-needed-bionic verification-needed-cosmic |
|
2018-10-25 19:28:40 |
Mike Ranweiler |
tags |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 verification-needed-bionic verification-needed-cosmic |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 verification-done-bionic verification-needed-cosmic |
|
2018-11-08 13:43:23 |
Kleber Sacilotto de Souza |
linux (Ubuntu Cosmic): status |
In Progress |
Fix Committed |
|
2018-11-12 14:43:10 |
Andrew Cloke |
ubuntu-power-systems: status |
In Progress |
Fix Committed |
|
2018-11-12 19:26:24 |
Terry Rudd |
bug |
|
|
added subscriber Terry Rudd |
2018-11-13 18:51:26 |
Launchpad Janitor |
linux (Ubuntu Bionic): status |
Fix Committed |
Fix Released |
|
2018-11-13 18:51:26 |
Launchpad Janitor |
cve linked |
|
2017-13168 |
|
2018-11-13 18:51:26 |
Launchpad Janitor |
cve linked |
|
2018-15471 |
|
2018-11-13 18:51:26 |
Launchpad Janitor |
cve linked |
|
2018-16658 |
|
2018-11-13 18:51:26 |
Launchpad Janitor |
cve linked |
|
2018-9363 |
|
2018-11-13 19:09:36 |
Launchpad Janitor |
linux (Ubuntu Cosmic): status |
Fix Committed |
Fix Released |
|
2018-11-13 19:09:37 |
Launchpad Janitor |
linux (Ubuntu Cosmic): status |
Fix Committed |
Fix Released |
|
2018-11-13 19:15:48 |
Andrew Cloke |
ubuntu-power-systems: status |
Fix Committed |
In Progress |
|
2018-11-14 16:01:57 |
Joseph Salisbury |
linux (Ubuntu): status |
In Progress |
Fix Released |
|
2018-11-14 16:33:36 |
Andrew Cloke |
ubuntu-power-systems: status |
In Progress |
Fix Released |
|
2019-07-24 20:53:58 |
Brad Figg |
tags |
architecture-ppc64le bot-comment bugnameltc-171273 severity-high targetmilestone-inin1804 verification-done-bionic verification-needed-cosmic |
architecture-ppc64le bot-comment bugnameltc-171273 cscc severity-high targetmilestone-inin1804 verification-done-bionic verification-needed-cosmic |
|