| Age | Commit message (Collapse) | Author |
|---|
|
We always remove the old entry, and add it back if it is needed
on for the kexec'ed kernel
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Currently, we can add a lot of reservations over a small range, this
does a simple check to verify the previous entry is not the same
as the current one and skips it if so
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Currently, the device tree is passed as is. You can optionally
update the command line and specifically listed nodes but nothing
is updated automatically.
This patch updates the memreserve regions, memory node, initrd
nodes and attempts to make the device tree look as it should. Some
code is borrowed from the u-boot routines which do similiar things
Also, now if no flat device tree is passed to kexec it will attempt
to rebuild one from the /proc/device-tree file system to use for
the kexec'ed kernel for both uImage and elf formats
[ horms@verge.net.au: kernel_addr is now outside #ifdef WITH_GAMECUBE ]
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
This fixes --reuseinitrd and --ramdisk option for ppc32 on
uImage-ppc and Elf. It works for normal kexec as well as for
kdump.
When using --reuseinitrd you need to specifify retain_initrd
on the command line. Also, if you are doing kdump you need to make
sure your initrd lives in the crashdump region otherwise the
kdump kernel will not be able to access it. The --ramdisk option
should always work.
[ horms@verge.net.au: minor up-port ]
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Use current command line if none given, specifically useful for
when arguments are added causing the use of the current cmdline
to not occur
We also try to load the dtb above the kernel, this is useful for
relocatable kernels where they device tree needs to reside just
above the kernel base address
Set the kernel entry address in the relocatable purgatory code
so we jump to the correct start address if not the default. Useful
for relocatable kernels
[ horms@verge.net.au: moved kernel_addr declaration to outside #ifdef WITH_GAMECUBE ]
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Use current command line if none given, specifically useful for
when arguments are added causing the use of the current cmdline
to not occur.
We also try to load the dtb above the kernel, this is useful for
relocatable kernels where they device tree needs to reside just
above the kernel base address.
Only allocate 1 MiB extra for bss space after kernel as it appears
to be more than adequate.
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
On some actitectures the physical memory can be 64 bits, therefore
the code that reads proc entries needs to take into account it could
read either a 32 bit or 64bit value for the physical addresses.
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Booting with uImage-ppc was broken by previous work, this
patch should restore it to working order
Signed-off-by: Matthew McClintock <msm@freescale.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
The memory ranges derived from /sys/firmware/memmap may overlap, causing
the kexec command to fail to load the crash kernel. The typical failure
reports 'Overlapping memory segments at 0x...'.
The preferred remedy might be to fix the BIOS or the kernel, but given
that they may at times generate overlaps, a check in the kexec command
will prevent crash dumps from being effectively disabled.
Diffed against git.kernel.org/pub/scm/linux/kernel/git/horms/kexec-tools.git
Signed-off-by: Cliff Wickman <cpw@sgi.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
The MAX_MEMORY_RANGES of 64 is too small for a very large NUMA machine.
(A 512 processor SGI UV, for example.)
And fix a temporary workaround (hack) in load_crashdump_segments() that
assumes that 16k is sufficient for the size of the crashdump elf header.
This is too small for a machine with a large cpu count. A PT_NOTE is created
in the elf header for each cpu.
Signed-off-by: Cliff Wickman <cpw@sgi.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
The argument parsing is currently a bit broken as main()'s getopt_long()
knows nothing about either the architecture-specific options or, even
more specifically, the architecture-and-loader-specific options.
This patch introduces new #defines for all architectures,
KEXEC_ALL_OPTIONS and KEXEC_ALL_OPT_STR. These contain all possible
options for a given build, and the getopt_long() passes in main() and
arch_process_options() will now recognise arch- and loader-specific
options; these will not be re-ordered in argv[], there is no confusion
over which argv[] entry is the kernel filename, and using '--opt=foo' and
'--opt foo' both work.
All architectures have command line options (and #define OPT_BLAHs)
consolidated into their include/arch/option.h files. x86_64 builds
parts of i386/ as well, so now both share a single option.h file (with
a symlink).
Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Implement support for loading dump capture kernels for ARM architecture. ARM
doesn't need any backup memory regions so only elfcore header is allocated from
the top of the reserved memory region. Only zImages are supported.
The dump capture kernel needs to be loaded to physical address corresponding to
the memory region reserved with 'crashkernel=size@start' kernel command line
parameter.
Signed-off-by: Mika Westerberg <ext-mika.1.westerberg@nokia.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
This fixes the kexec-build on ppc32 when
the --game-cube option is supplied to ./configure.
It seems to have bit-rotted a little.
Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Cc: Maxim Uvarov <muvarov@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
add_dyn_reconf_usable_mem_property() iterates over memory spans
in /ibm,dynamic-reconfiguration-memory/ibm,dynamic-memory and intersects
these with usablemem_rgns ranges. In doing so it used an unchecked
fixed-size array which will overrun on machines with lots of LMBs.
This patch removes the fixed-sized arrays from
add_dyn_reconf_usable_mem_property() and add_usable_mem_property(), in lieu of
malloc/realloc/free.
Signed-off-by: Matt Evans <matt@ozlabs.org>
Michael Neuling <mikey@neuling.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Maxim Uvarov <muvarov@gmail.com>
Signed-off-by: Maxim Uvarov <muvarov@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
mips, ppc)
These architectures don't have any architecture-specific options
(note: distinct from loader options, which are more specific), yet
their arch_process_options() functions contain getopts_long() calls
followed by no argument processing. The code doesn't do anything,
so this patch removes it.
Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Revert this portion of recent MIPS changes as it actually effects the
CFLAGS for all architectures. Moreover I'm somewhat dubious about
defaulting to -Werror for any architecture.
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
The original fixed address of 0x800000 may be overwritten after the
kernel is decompressed. Assume a maximum decompression ratio of 4
and place initrd right after that, just to be safer.
Signed-off-by: Eric Miao <eric.miao@canonical.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
ppc64's fs2dt used to use a fixed-size array into which the device tree
was parsed. There was no bounds checking, so with a large device tree other
heap data ended up getting stomped -- SIGSEGV time.
This patch adds a function, 'dt_reserve', to check whether there's enough space
left prior to writing data to the array. If not, the array is realloced.
Signed-off-by: Matt Evans <matt@ozlabs.org>
Ack-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
6adc05c6e3fdbc8b9f5d915af78ca05d0a09cb17 "some kexec MIPS
improvements" broke pp64 as it turned on -Werror for all archs.
This fixes the warning and hence ppc64 building.
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
|
|
The now generic probe function is more complete than the currently used.
It seems that ARM's and SH's uImage are always uncompressed so it might
be good to check for this.
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Marc Andre Tanner <mat@brain-dump.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
|
|
The check and uncompress code could be split and recycled by other arch.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
|
|
Signed-off-by: Ameya Palande <ameya.palande@nokia.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Generate the correct crashdump header for x86_64 kernels
when a 32bit kernel is used.
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
- using simple mips* ) in configure.ac to make it compilable on mips2
and mips64
- remove kexec/arch/mips/mips-setup-simple.S which prepares cmdline for
new kernel, it is better to move this work to kernel code. BTW this code was
compilable only on o32 because of t4 is not defined on 64-64 or n32 MIPS ABIs.
- simple put cmdline as string, kernel code should catch cmdline like this
int board_kexec_prepare(struct kimage *image)
{
int i;
char *bootloader = "kexec";
board_boot_desc_ptr->argc = 0;
for(i=0;i<image->nr_segments;i++)
{
printk("segment %d
if (!strncmp(bootloader, (char*)image->segment[i].buf,
strlen(bootloader)))
{
/*
* convert command line string to array
* of parameters (as bootloader does).
*/
int argc = 0, offt;
char *str = (char *)image->segment[i].buf;
char *ptr = strchr(str, ' ');
while (ptr && (ARGV_MAX_ARGS > argc)) {
*ptr = '\0';
if (ptr[1] != ' ') {
offt = (int)(ptr - str + 1);
boot_desc_ptr->argv[argc] =
image->segment[i].mem + offt;
argc++;
}
ptr = strchr(ptr + 1, ' ');
}
boot_desc_ptr->argc = argc;
break;
}
}
Keep it as string make code simple and more readable.
- add crashdump support
- do not redefine syscalls numbers if they defined in system
remove fixups for /proc/iomem. If your board provides wrong /proc/iomem please
fix kernel, or at least you local version of kexec. No need to support it in
main line. At least add option --fake-iomem
- some minor fixes
Signed-off-by: Maxim Uvarov <muvarov@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
On PowerPC the uImage usually contains the compressed "final" kernel and
not a tiny wrapper which relocates itself und uncomprosses the final
kernel to its final position. Instead we uncompress the gzip image and
put it the its final position.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
|
|
A few device nodes are dynamically determined by the bootloader (MAC address,
memory size, ..) and are not part of the device tree. The kernel command line
is also read from dtb and usually not part of the device tree.
With the libfdt it is now easy to add/replace nodes in the device tree.
The user may specify "--reuse-node=/memory/reg" to update the memory/reg
node to what ever is now. This requires the kernel to export the device
tree via the procfs in /proc/device-tree.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
|
|
This is v1.2.0 of libfdt from the dtc project which is available at
git://www.jdl.com/software/dtc.git
The other files
- include/page.h
- include/types.h
- libfdt-wrapper.c
- ops.h
are part of the glue code which is used the powerpc boot wrapper code
is comming from the Linux kernel v2.6.27-rc6 and has been modified a
little to fit.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
|
|
Some code dtb scanning & filling has been borrowed from ppc64.
The old behavior is still available if compiled with GameCube,
other PowerPC platform use the can purgatory and specify a new
dtb.
Booting a self contained elf image (incl. dtb / without the need
for a bd sturct or the like) can be booted. The dtb support is currently
optional. That means if the elf image does not contain a dtb file then
the user has to supply a complete dtb (including mem size, command line,
bus freq., mac addr, ...)
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
|
|
The RMO region can only exist in '/memory@0' node in the device tree.
While adding support for '/ibm,dynamic-reconfiguration-memory' in
kexec-tools, I incorrectly extended the code of '/memory@' nodes to
drconf memory with the assumption that what applies to '/memory@'
nodes would also apply to drconf memory. This is incorrect and
hence removing the code from kexec-tools now.
With this code in place, kexec on kexec fails. The first kexec
works because '/memory@0' node is read after drconf memory and
that correctly sets rmo_base and rmo_top from '/memory@0' node.
While in the kexec'ed kernel, '/ibm,dynamic-reconfiguration-memory'
is read after '/memory@xxx' nodes and this incorrectly sets both
rmo_base and rmo_top from drconf memory.
Sorry for introducing this bug.
Signed-off-by: Chandru Siddalingappa <chandru@linux.vnet.ibm.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
This fixes a bug when using gdb with vmcore
as explained by Dave Anderson:
The kexec/arch/x86_64/crashdump-x86_64.h file contains a
stale PAGE_OFFSET value. In 2.6.27 it was changed from
0xffff810000000000UL to 0xffff880000000000UL. This is
only a problem when using gdb with the vmlinux/vmcore
pair, because gdb relies upon the PT_LOAD segment's p_vaddr
values in the ELF header to be correct.
Cc: Dave Anderson <anderson@redhat.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
# crisv32-axis-linux-gnu-gcc --version
crisv32-axis-linux-gnu-gcc (GCC) 3.2.1 Axis release R64/1.64
Copyright (C) 2002 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# make
kexec/arch/cris/kexec-elf-cris.c: In function `elf_cris_load':
kexec/arch/cris/kexec-elf-cris.c:134: warning: assignment makes pointer from integer without a cast
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Although the following compiler warning is bogus
it seems harmless to initialise initrd_start to NULL.
# arm-unknown-linux-gnu-gcc --version
arm-unknown-linux-gnu-gcc (GCC) 4.1.1
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# make
kexec/arch/arm/kexec-zImage-arm.c: In function 'zImage_arm_load':
kexec/arch/arm/kexec-zImage-arm.c:135: warning: 'initrd_start' may be used
uninitialized in this function
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
# sh4-unknown-linux-gnu-gcc --version
sh4-unknown-linux-gnu-gcc (GCC) 4.1.1
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# make
kexec/arch/sh/kexec-netbsd-sh.c:29: warning: 'extern' is not at beginning of declaration
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
This avoids comparisons between signed and unsigned integers.
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Apparently this never happens, but the current code seems excessively loose.
1) die if it does happen
2) initialise offset to make gcc happy.
# ia64-unknown-linux-gnu-gcc --version
ia64-unknown-linux-gnu-gcc (GCC) 3.4.5
Copyright (C) 2004 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# make
kexec/arch/ia64/kexec-elf-ia64.c: In function `move_loaded_segments':
kexec/arch/ia64/kexec-elf-ia64.c:102: warning: 'offset' might be used
uninitialized in this function
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
The two-stage cast is because we really do want a 32bit integer value
from a 64bit pointer, and gcc assumes that we're making a mistake
if there is just one cast.
# powerpc64-unknown-linux-gnu-gcc --version
powerpc64-unknown-linux-gnu-gcc (GCC) 4.1.1
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# make
kexec/arch/ppc64/kexec-elf-rel-ppc64.c: In function 'machine_apply_elf_rel':
kexec/arch/ppc64/kexec-elf-rel-ppc64.c:66: warning: cast from pointer to integer of different size
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
The consumer of the purgatory buffer, elf_rel_build_load()
expects a signed char buffer.
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
|
Signed-off-by: Simon Horman <horms@verge.net.au>
|
