crash (7.2.1-1) unstable; urgency=medium
* New upstream (closes: #890394)
* Fix for the "runq" command on Linux 4.14 and later kernels that contain
commit cd9e61ed1eebbcd5dfad59475d41ec58d9b64b6a, titled "rbtree: cache
leftmost node internally". Without the patch, the command fails with the
error message "runq: invalid structure member offset: cfs_rq_rb_leftmost".
* Fix to prevent a useless message during session inialization. Without the
patch, if the highest possible node bit in the node_states[N_ONLINE]
multi-word bitmask is set, then a message such as "crash:
next_online_node: 256 is too large!" will be displayed.
* Additional fixes for the ARM64 "bt" command for Linux 4.14 kernels. The
patch corrects the contents of in-kernel exception frame register dumps,
and properly transitions the backtrace from the IRQ stack to the process
stack.
* Implemented a new "search -T" option, which is identical to the "search
-t" option, except that the search is restricted to the kernel stacks of
active tasks.
* Removal of the ARM64 "bt -o" option for Linux 4.14 and later kernels,
along with several cleanups/readability improvements.
* Fix for support of KASLR enabled kernels captured by the SADUMP dumpfile
facility. SADUMP dumpfile headers do not contain phys_base or VMCOREINFO
notes, so without this patch, the crash session fails during
initialization with the message "crash: seek error: kernel virtual
address: <address> type: "page_offset_base". This patch calculates the
phys_base value and the KASLR offset using the IDTR and CR3 registers from
the dumpfile header.
* Implemented a new "ps -y policy" option to filter the task display by
scheduling policy. Applicable to both standalone ps invocation as well as
via foreach.
* Fix for the "kmem -[sS]" options on Linux 4.14 and later kernels that
contain commit 2482ddec670fb83717d129012bc558777cb159f7, titled "mm: add
SLUB free list pointer obfuscation". Without the patch, there will
numerous error messages of the type "kmem: <cache name> slab: <address>
invalid freepointer: <obfuscated address>" if the kernel is configured
with CONFIG_SLAB_FREELIST_HARDENED.
* Fix for the validation of the bits located in the least significant bits of
mem_section.section_mem_map pointers. Without the patch, the validation
functions always returned valid, due to a coding error found by clang.
However, it was never really a problem because it is extremely unlikely
that an existing mem_section would ever be invalid.
* Fix for the x86_64 kernel virtual address to physical address translation
mechanism. Without the patch, when verifying that the PAGE_PRESENT bit is
set in the top-level page table, it would always test positively, and the
translation would continue parsing the remainder of the page tables. This
would virtually never be a problem in practice because if the top-level
page table entry existed, its PAGE_PRESENT bit would be set.
* Removed a check for a negative block_size value which is always a
non-negative unsigned value in the SADUMP header parsing function.
* Removed a check for an impossible negative value when calculating the
beginning address when applying the context value specified by the "search
-x <count>" option.
* Implemented a new "timer -C <cpu-specifier>" option that restricts the
timer or hrtimer output to the timer queue data associated with one or
more cpus. For multiple cpus, the cpu-specifier uses the standard comma
or dash separated list format.
* Fix for a "ps -l" regression introduced by the new "ps -y" option
introduced above. Without the patch, the -l option generates a
segmentation violation if not accompanied by a -C cpu specifier option.
* Fix for the "kmem -i" and "kmem -V" options in Linux 4.8 and later kernels
containing commit 75ef7184053989118d3814c558a9af62e7376a58, titled "mm,
vmstat: add infrastructure for per-node vmstats". Without the patch, the
CACHED line of "kmem -i" shows 0, and the VM_STAT section of "kmem -V" is
missing entirely.
* Fix for Linux 4.11 and later kernels that contain kernel commit
4b3ef9daa4fc0bba742a79faecb17fdaaead083b, titled "mm/swap: split swap
cache into 64MB trunks". Without the patch, the CACHED line of "kmem -i"
may show nonsensical data.
* Implemented a new "dev -D" option that is the same as "dev -d", but
filters out the display of disks that have no I/O in progress.
* If a line number request for a module text address initially fails, force
the embedded gdb module to complete its two-stage strategy used for
reading debuginfo symbol tables from module object files, and then retry
the line number extraction. This automatically does what the "mod -r" or
"crash --readnow" options accomplish.
* Update for support of Linux 4.12 and later PPC64 kernels where the hash
page table geometry accommodates a larger virtual address range. Without
the patch, the virtual-to-physical translation of user space virtual
addresses by "vm -p", "vtop", and "rd -u" may generate an invalid
translation or otherwise fail.
* Implemented a new "runq -T" option that displays the time lag of each CPU
relative to the most recent runqueue timestamp.
* Fix to support Linux 4.15 and later kernels that contain kernel commit
e8cfbc245e24887e3c30235f71e9e9405e0cfc39, titled "pid: remove pidhash".
The kernel's traditional usage of a pid_hash[] array to store PIDs has
been replaced by an IDR radix tree, requiring a new crash plug-in function
to gather the system's task set. Without the patch, the crash session
fails during initialization with the error message "crash: cannot resolve
init_task_union".
* Fix for the "net" command when the network device listing has an unusually
large number of IP addresses. In that case, without the patch, the
command may generate a segmentation violation.
* Fix for Linux 4.15 and later kernels that are configured with
CONFIG_SPARSEMEM_EXTREME, and that contain kernel commit
83e3c48729d9ebb7af5a31a504f3fd6aff0348c4, titled "mm/sparsemem: Allocate
mem_section at runtime for CONFIG_SPARSEMEM_EXTREME=y". Without the
patch, kernels configured with SPARSEMEM_EXTREME have changed the data
type of "mem_section" from an array to a pointer, leading to errors in
commands such as "kmem -p", "kmem -n", "kmem -s", and any other command
that translates a physical address to its page struct address.
* With the latest PPC64 NMI IPI changes, crash_ipi_callback is found
multiple times on the stack of active non-panic tasks. Ensure that the
symbol reference relates to an actual backtrace stack frame.
* Update the starting virtual address of vmalloc space for kernels
configured with CONFIG_X86_5LEVEL.
* Update the X86_64 VSYSCALL_END address to reflect that it only contains 1
page.
* Prevent the X86_64 FILL_PML() macro from updating the internal
machdep->machspec->last_pml4_read address every time a vmalloc'd kernel
virtual address is translated.
* Fix for the "bt" command in x86_64 kernels that contain, or have backports
of, kernel commit 4950d6d48a0c43cc61d0bbb76fb10e0214b79c66, titled
"x86/dumpstack: Remove 64-byte gap at end of irq stack". Without the
patch, backtraces fail to transition from the IRQ stack back to the
process stack, showing an error message such as "bt: cannot transition
exception stack to IRQ stack to current process stack".
* Initial pass for support of kernel page table isolation. The x86_64 "bt"
command may indicate "bt: cannot transition from exception stack to
current process stack" if the crash callback NMI occurred while an active
task was running on the new entry trampoline stack. This has only been
tested on the RHEL7 backport of the upstream patch because as of this
commit, crash does not run on 4.15-rc kernels. Further changes may be
required for upstream kernels, and distributions that implement the kernel
changes differently than upstream.
* Fix for the "bt" command and the "ps -s" option for zombie tasks whose
kernel stacks have been freed/detached. Without the patch, the "bt"
command indicates "bt: invalid kernel virtual address: 0 type: stack
contents" and "bt: read of stack at 0 failed"; it will be changed to
display "(no stack)". The "ps -s" option would fail prematurely upon
reaching such a task, indicating "ps: invalid kernel virtual address: 0
type: stack contents" and "ps: read of stack at 0 failed".
* Fix for running on live systems on 4.15-rc2 and later kernels that are
configured with CONFIG_RANDOMIZE_BASE and contain kernel commit
668533dc0764b30c9dd2baf3ca800156f688326b, titled "kallsyms: take advantage
of the new '%px' format". Without the patch, a live crash session does
not show the "WARNING: kernel relocated ..." message expected with KASLR,
and then displays the message "crash: cannot set context for pid: <pid>"
prior to generating a SIGSEGV.
* Fix for 4.15-rc5 and later x86_64 kernels that contain kernel commit
c482feefe1aeb150156248ba0fd3e029bc886605, titled "x86/entry/64: Make
cpu_entry_area.tss read-only". Without the patch, the addresses and sizes
of the x86_64 exception stacks cannot be determined; therefore if a
backtrace starts on one of the exception stacks, then the "bt" command
will fail.
* Additional fix for support of KASLR enabled kernels captured by the SADUMP
dumpfile facility, where this patch fixes a problem when Page Table
Isolation(PTI) is enabled. When PTI is enabled, bit 12 of CR3 register is
used to split user space and kernel space. Also bit 11:0 is used for
Process Context IDentifiers(PCID). To open an SADUMP dumpfile, the value
of CR3 is used to calculate KASLR offset and phys_base; this patch masks
the CR3 register value correctly for a PTI enabled kernel.
* Second phase of future support for x86_64 5-level page tables. This patch
is a cleanup/collaboration of the original logic used by the various vtop
functions, where several new common functions have been added for
extracting page table entries from PGD, P4D, PUD, PMD and PTE pages. The
usage of the former PML4 and UPML pages have been replaced with the use of
the common PGD page, and use the PUD page in 4-level page table
translation. Support for 5-level page tables has been incorporated into
the the existing x86_64_kvtop() and x86_64_uvtop_level4() functions.
Backwards compatibility for older legacy kernels has been maintained. The
third phase of support will automatically detect whether the kernel
proper, and whether an individual user task, is utilizing 5-level page
tables. This patch enables support for kernel-only 5-level page tables by
entering the command line option "--machdep vm=5level".
* Xen commit 615588563e99a23aaf37037c3fee0c413b051f4d (Xen 4.0.0.) extended
the direct mapping to 5 TB. This area was previously reserved for future
use, so it is OK to simply change the upper bound unconditionally.
* Add a new "foreach gleader" qualifier option, restricting the output to
user-space tasks that are thread group leaders.
* Since Xen commit 666aca08175b ("sched: use the auto-generated list of
schedulers") crash cannot open Xen vmcores because the "schedulers" symbol
no longer exists. Xen 4.7 implemented schedulers as its own section in
"xen/arch/x86/xen.lds.S", delimited by the two symbols
"__start_schedulers_array" and "__end_schedulers_array". Without the
patch, the crash session fails during initialization with the error
message "crash: cannot resolve schedulers"
* Fix the sample crash.ko memory driver to prevent an s390X kernel
addressing exception. Legitimate pages of RAM that successfully pass the
page_is_ram() and pfn_valid() verifier functions may not be provided by
the s390x hypervisor, and the memcpy() from the non-existent memory to the
bounce buffer panics the kernel. The patch replaces the the memcpy() call
with probe_kernel_read().
* Fix for the ARM64 "bt" command running against Linux 4.14 and later
kernels. Without the patch, the backtraces of the active tasks in a
kdump-generated dumpfile are truncated. Without the patch, the panic task
will just show the "crash_kexec" frame and the kernel-entry user-space
exception frame; the non-panic tasks will show their backtraces starting
from the stackframe addresses captured in the per-cpu NT_PRSTATUS notes,
and will not display the exception frame generated by the NMI callback,
nor any stackframes on the IRQ stack.
* Fix for the ARM64 "bt" command in kernels that contain commit
30d88c0e3ace625a92eead9ca0ad94093a8f59fe, titled "arm64: entry: Apply BP
hardening for suspicious interrupts from EL0". Without the patch, there
may be invalid kernel kernel exception frames displayed on an active
task's kernel stack, often below a stackframe of the
"do_el0_ia_bp_hardening" function; the address translation of the PC and
LR values in the the bogus exception frame will display "[unknown or
invalid address]".
-- Troy Heber <email address hidden> Fri, 16 Feb 2018 10:47:33 -0700
