* Config Items appearing twice in same Kconfig file?
From: Christoph Egger @ 2010-05-12 14:34 UTC (permalink / raw)
To: Richard Henderson, Ivan Kokshaysky, Matt Turner, Andrew Morton,
FUJITA Tomonori, Mike Frysinger, Graf Yang, Mikael Starvik,
Jesper Nilsson, Tony Luck, Fenghua Yu, Benjamin Herrenschmidt,
Paul Mackerras, Chen Liqin, Lennox Wu, Arnd Bergmann,
Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86,
Arjan van de Ven, Karsten Keil, Jesse Barnes, David S. Miller,
Alexey Kuznetsov, Pekka Savola (ipv6), James Morris,
Hideaki YOSHIFUJI, Patrick McHardy, Timo Teras, linux-alpha,
linux-kernel, uclinux-dist-devel, linux-cris-kernel, linux-ia64,
linuxppc-dev, netdev
Cc: vamos
Hi all!
First of all sorry for the huge CC, I couldn't come up with a way
to create a smaller list while reaching all the relevant people.
We, the vamos project[1] at the Friedrich Alexander University are
analything the linux configuration model trying to detect
errors. During a run with our tools we noticed the fact that the
option ALPHA_LEGACY_START_ADDRESS in arch/alpha/Kconfig.debug is
listed twice with contradictory dependencies on ALPHA_GENERIC.
Taking a closer look[2] we noticed 22 occurences where the same
config option is defined more than once in the same Kconfig file, not
all of them are, of course, contradictory as the ALPHA sample.
We"d be interested in reasons to keep this configuration split up
in the same file or if this is just a error introduced someeewhere in
linux' development that should be fixed.
Regards
Chritoph Egger
[ Please keep vamos@ Cc-ed -- we're not following all the lkml traffic
;) no need to keep me though ]
[1] http://vamos1.informatik.uni-erlangen.de
[2] grep ^config **/Kconfig* | sort | uniq -d -c
^ permalink raw reply
* Re: Config Items appearing twice in same Kconfig file?
From: Christoph Egger @ 2010-05-12 14:40 UTC (permalink / raw)
To: Richard Henderson, Ivan Kokshaysky, Matt Turner, Andrew Morton,
FUJITA Tomonori, Mike Frysinger, Graf Yang, Mikael Starvik,
Jesper Nilsson, Tony Luck, Fenghua Yu, Benjamin Herrenschmidt,
Paul Mackerras, Chen Liqin, Lennox Wu, Arnd Bergmann,
Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86,
Arjan van de Ven, Karsten Keil, Jesse Barnes, David S. Miller,
Alexey Kuznetsov, Pekka Savola (ipv6), James Morris,
Hideaki YOSHIFUJI, Patrick McHardy, Timo Teras, linux-alpha,
linux-kernel, uclinux-dist-devel, linux-cris-kernel, linux-ia64,
linuxppc-dev, netdev
Cc: vamos
In-Reply-To: <20100512143445.GA32559@faui48a.informatik.uni-erlangen.de>
Just noticed it might have been worth adding the output so everyone
out there doesn't ned to grep through his tree for reference:
% grep ^config **/Kconfig* | sort | uniq -dc
2 arch/alpha/Kconfig.debug:config ALPHA_LEGACY_START_ADDRESS
2 arch/alpha/Kconfig:config ALPHA_EV4
3 arch/alpha/Kconfig:config ALPHA_EV56
2 arch/alpha/Kconfig:config ALPHA_GAMMA
2 arch/blackfin/Kconfig:config MEM_MT48LC32M8A2_75
2 arch/cris/arch-v32/drivers/Kconfig:config ETRAX_RS485
2 arch/cris/arch-v32/drivers/Kconfig:config ETRAX_RS485_DISABLE_RECEIVER
2 arch/cris/arch-v32/drivers/Kconfig:config ETRAX_SER3_CD_BIT
2 arch/h8300/Kconfig.ide:config H8300_IDE_ALT
2 arch/h8300/Kconfig.ide:config H8300_IDE_BASE
2 arch/h8300/Kconfig.ide:config H8300_IDE_IRQ
2 arch/ia64/Kconfig:config IOMMU_HELPER
2 arch/powerpc/Kconfig:config KERNEL_START
2 arch/powerpc/Kconfig:config PAGE_OFFSET
2 arch/powerpc/Kconfig:config PHYSICAL_START
2 arch/powerpc/Kconfig:config RELOCATABLE
2 arch/score/Kconfig:config RWSEM_GENERIC_SPINLOCK
4 arch/x86/Kconfig.debug:config DEFAULT_IO_DELAY_TYPE
2 arch/x86/Kconfig:config PCI_MMCONFIG
2 arch/x86/Kconfig:config X86_EXTENDED_PLATFORM
2 drivers/isdn/hisax/Kconfig:config HISAX_AVM_A1_PCMCIA
2 net/ipv4/Kconfig:config TCP_CONG_CUBIC
^ permalink raw reply
* Re: [PATCH 1/2] Fix kexec on powerpc32
From: Simon Horman @ 2010-05-12 13:19 UTC (permalink / raw)
To: Maxim Uvarov; +Cc: Christian Kujau, linuxppc-dev, kexec
In-Reply-To: <AANLkTimSATEKI4TdO6zNWSRzl2on4OL5D2GGBlVsndMs@mail.gmail.com>
On Wed, May 12, 2010 at 10:31:26AM +0400, Maxim Uvarov wrote:
> 2010/5/12 Christian Kujau <lists@nerdbynature.de>
>
> > Maxim Uvarov wrote on 2010-05-11 10:47 :
> > > This patch is required in case if you are using new toolchains.
> >
> > I'm on Debian/stable (binutils 2.18, gcc 4.3.2), so I applied your patches
> > to the latest -git checkout:
> >
> > First, compilation fails with:
> >
> > cc1: warnings being treated as errors
> > kexec/arch/ppc/kexec-ppc.c: In function ‘get_base_ranges’:
> > kexec/arch/ppc/kexec-ppc.c:258: error: value computed is not used
> > make: *** [kexec/arch/ppc/kexec-ppc.o] Error 1
> >
> > Somehow a "-" hyphen made it into one of the functions:
> >
> > > --- a/kexec/arch/ppc/kexec-ppc.c
> > > +++ b/kexec/arch/ppc/kexec-ppc.c
> > [...]
> > > if (local_memory_ranges >= max_memory_ranges) {
> > > - fclose(file);
> > > - break;
> > > + if (realloc_memory_ranges() < 0){
> > > + - fclose(file);
> >
> > Oh, I sent this patch twice with stg -a by mistake. And second version has
> "-". I will fix and resend patches.
>
> >
> > Removing this, gets me further, until:
> >
> >
> > purgatory/arch/ppc/crt.o: In function `__lshrdi3':
> > (.text+0xf0): multiple definition of `__lshrdi3'
> > purgatory/arch/ppc/misc.o:(.text+0x50): first defined here
> > collect2: ld returned 1 exit status
> > make: *** [purgatory/purgatory.ro] Error 1
> >
> > And has been defined in purgatory/arch/ppc/crt.S and
> > ./purgatory/arch/ppc/misc.S. Removing it from e.g. crt.S
> > makes it compile, I have not tested it yet, though :-)
> >
>
> Yes, that is expected result for old tool chains. New tool chains (from code
> sourcery ) need this crt.S.
> I think we can have --with-oldtoolchain option to configure to point to old
> tool chains. (or even rename it to --with-crts,
> I don't know what is better.)
I think --with-crts would be better as --with-oldtoolchain could
mean many things.
^ permalink raw reply
* Re: Please pull 'next' branch of 4xx (again)
From: Josh Boyer @ 2010-05-12 12:22 UTC (permalink / raw)
To: benh; +Cc: linuxppc-dev
In-Reply-To: <20100507195453.GN29593@zod.rchland.ibm.com>
On Fri, May 07, 2010 at 03:54:53PM -0400, Josh Boyer wrote:
>Hi Ben,
>
>Two more quick patches from me for -next. Stefrano's is the Katmai usage of
>the reset type change that just went in. The 460SX PCI-e patch has been sitting
>for a while and seems to have been skipped over.
I've updated my next branch with two more patches. Yours for the 440SPe UART
clock fix, and the new platform addition of Icon 440SPe from Stefan. Updated
diffstat below.
Thanks,
josh
The following changes since commit 1ed31d6db90d51010545921e59d369d2f92b7ac2:
Benjamin Herrenschmidt (1):
Merge commit 'origin/master' into next
are available in the git repository at:
ssh://master.kernel.org/pub/scm/linux/kernel/git/jwboyer/powerpc-4xx.git next
Benjamin Herrenschmidt (1):
powerpc/44x: Fix UART clocks on 440SPe
Stefan Roese (2):
powerpc/44x: Add reset-type to katmai.dts
powerpc/44x: Add basic ICON PPC440SPe board support
Tirumala Marri (1):
powerpc/44x: Adding PCI-E support for PowerPC 460SX based SOC.
arch/powerpc/boot/4xx.c | 12 +-
arch/powerpc/boot/dts/icon.dts | 447 ++++++++++
arch/powerpc/boot/dts/katmai.dts | 1 +
arch/powerpc/boot/dts/redwood.dts | 122 +++
arch/powerpc/configs/44x/icon_defconfig | 1316 ++++++++++++++++++++++++++++
arch/powerpc/platforms/44x/Kconfig | 11 +
arch/powerpc/platforms/44x/ppc44x_simple.c | 3 +-
arch/powerpc/sysdev/ppc4xx_pci.c | 119 +++
arch/powerpc/sysdev/ppc4xx_pci.h | 58 ++
9 files changed, 2082 insertions(+), 7 deletions(-)
create mode 100644 arch/powerpc/boot/dts/icon.dts
create mode 100644 arch/powerpc/configs/44x/icon_defconfig
^ permalink raw reply
* Re: [PATCH 1/2] Fix kexec on powerpc32
From: Maxim Uvarov @ 2010-05-12 12:12 UTC (permalink / raw)
To: Christian Kujau; +Cc: kexec, uvarov, linuxppc-dev, horms
In-Reply-To: <4BEA5726.50806@nerdbynature.de>
[-- Attachment #1: Type: text/plain, Size: 1167 bytes --]
2010/5/12 Christian Kujau <lists@nerdbynature.de>
> On Tue, 11 May 2010 at 21:47, Maxim Uvarov wrote:
> > This patch is required in case if you are using new toolchains.
>
> After kexec-tools compiled, I actually tried it:
>
> # kexec -l /boot/2.6/zImage --command-line="root=/dev/hda6 ro"
> # echo $?
> 0
> # kexec -e
> Starting new kernel
> Bye!
>
> ....but then the system just hung there, no more messages, I had to
> powercycle it.
>
> The --debug options is listed for "dol-ppc" systems (I have a PowerBook
> G4, not sure if this falls into this category), when I use it the help
> screen of kexec is displayed again and at the bottom:
>
> Cannot load /boot/2.6/zImage
>
>
> This is with 2.6.34-rc7 (today's git):
>
> # file /boot/2.6/zImage
> /boot/2.6/zImage: ELF 32-bit MSB executable, PowerPC or cisco 4500,
> version 1 (SYSV), statically linked, with unknown capability 0x41000000
> = 0x11676e75, with unknown capability 0x10000 = 0x90402, not stripped
>
>
> .config and dmesg: http://nerdbynature.de/bits/2.6.34-rc7/kexec/
>
> Christian.
> --
> make bzImage, not war
>
Can you try vmlinux? I tested only with vmlinux.
--
Best regards,
Maxim Uvarov
[-- Attachment #2: Type: text/html, Size: 1765 bytes --]
^ permalink raw reply
* Re: [git pull] Please pull powerpc.git merge branch
From: Benjamin Herrenschmidt @ 2010-05-12 10:36 UTC (permalink / raw)
To: Kumar Gala
Cc: linuxppc-dev list, Andrew Morton, Linus Torvalds,
Linux Kernel list
In-Reply-To: <B0104965-09F9-4372-AA9F-F6D0FDEA67F2@kernel.crashing.org>
On Wed, 2010-05-12 at 05:24 -0500, Kumar Gala wrote:
> Did where you going to also send linus:
>
> http://patchwork.ozlabs.org/patch/51496/
Forgot about that one... I'll add it to the pile tomorrow.
Cheers,
Ben.
^ permalink raw reply
* Re: [PATCH] Fix UART clocks on 440SPe
From: Josh Boyer @ 2010-05-12 10:30 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: Stefan Roese, linuxppc-dev
In-Reply-To: <1273630970.21352.85.camel@pasglop>
On Wed, May 12, 2010 at 12:22:50PM +1000, Benjamin Herrenschmidt wrote:
>The code to fixup the serial ports on 440SPe uses the incorrect
>addresses for these. This fixes it.
>
>Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
>---
>
>I'm surprised nobody noticed, serial is pretty busted for me without
>that patch :-)
Odd. I don't have a 440SPe, so no surprise there but I would have thought
someone would notice. Though this is only needed if you are using an older
U-Boot that needs cuImage, and I'm guessing Stefan runs u-boot.git on his
boards ;)
>
>diff --git a/arch/powerpc/boot/4xx.c b/arch/powerpc/boot/4xx.c
>index 27db893..627de25 100644
>--- a/arch/powerpc/boot/4xx.c
>+++ b/arch/powerpc/boot/4xx.c
>@@ -544,9 +544,9 @@ void ibm440spe_fixup_clocks(unsigned int sys_clk,
> unsigned int plb_clk = __ibm440eplike_fixup_clocks(sys_clk, tmr_clk, 1);
>
> /* serial clocks beed fixup based on int/ext */
You couldn't be bothered to fixup the typo? s/beed/need ;) I'll fix it when
I bring this in.
>- eplike_fixup_uart_clk(0, "/plb/opb/serial@10000200", ser_clk, plb_clk);
>- eplike_fixup_uart_clk(1, "/plb/opb/serial@10000300", ser_clk, plb_clk);
>- eplike_fixup_uart_clk(2, "/plb/opb/serial@10000600", ser_clk, plb_clk);
>+ eplike_fixup_uart_clk(0, "/plb/opb/serial@f0000200", ser_clk, plb_clk);
>+ eplike_fixup_uart_clk(1, "/plb/opb/serial@f0000300", ser_clk, plb_clk);
>+ eplike_fixup_uart_clk(2, "/plb/opb/serial@f0000600", ser_clk, plb_clk);
> }
^ permalink raw reply
* Re: [git pull] Please pull powerpc.git merge branch
From: Kumar Gala @ 2010-05-12 10:24 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: linuxppc-dev list, Andrew Morton, Linus Torvalds,
Linux Kernel list
In-Reply-To: <1273650396.21352.121.camel@pasglop>
On May 12, 2010, at 2:46 AM, Benjamin Herrenschmidt wrote:
> Hi Linus !
>=20
> Please pull this perf event fix from Paulus. Without it, using perf =
event
> can potentially oops the machine at any time.
>=20
> Thanks !
>=20
> Cheers,
> Ben.
>=20
> The following changes since commit =
cea0d767c29669bf89f86e4aee46ef462d2ebae8:
> Linus Torvalds (1):
> Merge branch 'hwmon-for-linus' of =
git://git.kernel.org/.../jdelvare/staging
>=20
> are available in the git repository at:
>=20
> git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc.git merge
>=20
> Paul Mackerras (1):
> powerpc/perf_event: Fix oops due to perf_event_do_pending call
>=20
> arch/powerpc/include/asm/hw_irq.h | 38 -----------------------
> arch/powerpc/kernel/asm-offsets.c | 1 -
> arch/powerpc/kernel/entry_64.S | 9 -----
> arch/powerpc/kernel/irq.c | 6 ----
> arch/powerpc/kernel/time.c | 60 =
+++++++++++++++++++++++++++++-------
> 5 files changed, 48 insertions(+), 66 deletions(-)
>=20
Ben,
Did where you going to also send linus:
http://patchwork.ozlabs.org/patch/51496/
- k=
^ permalink raw reply
* Re: [RFC][PATCH 7/12 not tested yet] PPC: introduce __set_bit() like function for bitmaps in user space
From: Takuya Yoshikawa @ 2010-05-12 9:25 UTC (permalink / raw)
To: Alexander Graf
Cc: linux-arch, x86, arnd, kvm, kvm-ia64, fernando, mtosatti,
linux-kernel, kvm-ppc, linuxppc-dev, mingo, paulus, avi, hpa,
tglx, Takuya Yoshikawa
In-Reply-To: <4BE97F2A.5080900@suse.de>
>>
>> +static inline int set_bit_user_non_atomic(int nr, void __user *addr)
>> +{
>> + u8 __user *p;
>> + u8 val;
>> +
>> + p = (u8 __user *)((unsigned long)addr + nr / BITS_PER_BYTE);
>>
>
> Does C do the + or the / first? Either way, I'd like to see brackets here :)
OK, I'll change like that! I like brackets style too :)
>
>
> Alex
>
^ permalink raw reply
* Re: [RFC][PATCH 0/12] KVM, x86, ppc, asm-generic: moving dirty bitmaps to user space
From: Takuya Yoshikawa @ 2010-05-12 9:19 UTC (permalink / raw)
To: Alexander Graf
Cc: linux-arch, x86, arnd, kvm, kvm-ia64, fernando, mtosatti,
linux-kernel, kvm-ppc, linuxppc-dev, mingo, paulus, avi, hpa,
tglx, Takuya Yoshikawa
In-Reply-To: <4BE97DFA.1060408@suse.de>
>> [To ppc people]
>>
>> Hi, Benjamin, Paul, Alex,
>>
>> Please see the patches 6,7/12. I first say sorry for that I've not tested these
>> yet. In that sense, these may not be in the quality for precise reviews. But I
>> will be happy if you would give me any comments.
>>
>> Alex, could you help me? Though I have a plan to get PPC box in the future,
>> currently I cannot test these.
>>
>
> Could you please point me to a git tree where everything's readily
> applied? That would make testing a lot easier.
OK, I'll prepare one. Probably on sourceforge or somewhere like Kemari.
Thanks,
Takuya
>
> Alex
>
^ permalink raw reply
* Re: [PATCH] Fix UART clocks on 440SPe
From: Stefan Roese @ 2010-05-12 9:02 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: linuxppc-dev
In-Reply-To: <1273630970.21352.85.camel@pasglop>
On Wednesday 12 May 2010 04:22:50 Benjamin Herrenschmidt wrote:
> The code to fixup the serial ports on 440SPe uses the incorrect
> addresses for these. This fixes it.
>
> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Stefan Roese <sr@denx.de>
Cheers,
Stefan
^ permalink raw reply
* [git pull] Please pull powerpc.git merge branch
From: Benjamin Herrenschmidt @ 2010-05-12 7:46 UTC (permalink / raw)
To: Linus Torvalds; +Cc: linuxppc-dev list, Andrew Morton, Linux Kernel list
Hi Linus !
Please pull this perf event fix from Paulus. Without it, using perf event
can potentially oops the machine at any time.
Thanks !
Cheers,
Ben.
The following changes since commit cea0d767c29669bf89f86e4aee46ef462d2ebae8:
Linus Torvalds (1):
Merge branch 'hwmon-for-linus' of git://git.kernel.org/.../jdelvare/staging
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc.git merge
Paul Mackerras (1):
powerpc/perf_event: Fix oops due to perf_event_do_pending call
arch/powerpc/include/asm/hw_irq.h | 38 -----------------------
arch/powerpc/kernel/asm-offsets.c | 1 -
arch/powerpc/kernel/entry_64.S | 9 -----
arch/powerpc/kernel/irq.c | 6 ----
arch/powerpc/kernel/time.c | 60 +++++++++++++++++++++++++++++-------
5 files changed, 48 insertions(+), 66 deletions(-)
^ permalink raw reply
* Re: [PATCH] hwmon: (tmp421) Add nfactor support.
From: Jean Delvare @ 2010-05-12 7:27 UTC (permalink / raw)
To: Jeff Angielski; +Cc: lm-sensors, Linuxppc-dev, Andre Prendel
In-Reply-To: <4BE9B145.7000702@theptrgroup.com>
Hi Jeff,
On Tue, 11 May 2010 15:34:29 -0400, Jeff Angielski wrote:
> On 05/11/2010 03:03 PM, Andre Prendel wrote:
> > I'de prefer implementing the sysfs access methods in a consistent way (see other functions). That means adding the nfactor register to the tmp421_data structure and using tmp421_update_device() to update the structure.
>
> I did this on purpose since the nfactor typically only changes once at
> runtime when you program it for your sensor. It seemed like a waste of
> processing power and i2c bandwidth to read a "pseudo static" register
> over and over again.
>
> It can easily be changed if that's what will help the community the best.
I get your point and it makes sense. However, we also have to ensure
that no regular user can saturate the I2C link by repeatedly polling
for the same register value. So, any sysfs attribute which triggers an
immediate I2C transaction shouldn't be accessible by regular users
(i.e. change the mode from 0644 to 0640).
An alternative is to slit the cache into short-lived (~ 1 or 2 seconds)
and long-lived (1 to 5 minutes.) A number of drivers do this.
--
Jean Delvare
^ permalink raw reply
* Re: [PATCH 1/2] Fix kexec on powerpc32
From: Christian Kujau @ 2010-05-12 7:22 UTC (permalink / raw)
To: Maxim Uvarov; +Cc: kexec, uvarov, linuxppc-dev, horms
In-Reply-To: <20100511174749.19309.44615.stgit@muvarov>
On Tue, 11 May 2010 at 21:47, Maxim Uvarov wrote:
> This patch is required in case if you are using new toolchains.
After kexec-tools compiled, I actually tried it:
# kexec -l /boot/2.6/zImage --command-line="root=/dev/hda6 ro"
# echo $?
0
# kexec -e
Starting new kernel
Bye!
....but then the system just hung there, no more messages, I had to
powercycle it.
The --debug options is listed for "dol-ppc" systems (I have a PowerBook
G4, not sure if this falls into this category), when I use it the help
screen of kexec is displayed again and at the bottom:
Cannot load /boot/2.6/zImage
This is with 2.6.34-rc7 (today's git):
# file /boot/2.6/zImage
/boot/2.6/zImage: ELF 32-bit MSB executable, PowerPC or cisco 4500,
version 1 (SYSV), statically linked, with unknown capability 0x41000000
= 0x11676e75, with unknown capability 0x10000 = 0x90402, not stripped
.config and dmesg: http://nerdbynature.de/bits/2.6.34-rc7/kexec/
Christian.
--
make bzImage, not war
^ permalink raw reply
* [PATCH v2 2/2] powerpc new toolchains fix (crt.S)
From: Maxim Uvarov @ 2010-05-12 7:08 UTC (permalink / raw)
To: linuxppc-dev, kexec; +Cc: lists, horms
In-Reply-To: <20100512070841.9572.78661.stgit@muvarov>
Hello everybody,
Changes from previous version:
- removed bogus hyphen from the patch;
- move ifdefs to crt.S instead of Makefile
Please find here patch for user land kexec-tools application. Following
patch makes kexec-tools work for both kexec and kdump. I tested it with
git kernel (linus-tree) and Freescale/Logic MPC8360ERDK board with
mpc83xx_defconfig kernel config.
kexec:
kexec -l vmlinux --command-line="console= ... etc"
kexec -e
kdump:
kexec -p vmlinux_dump --command-line="console=... etc"
echo c > /proc/sysrq-trigger
I also think that is is reasonable:
- put GAME_CUBE specific code to separate files;
- combine ppc and ppc64 to powerpc directory (I'm planning to do it.
And that why in some places my patch have ifdefs for PPC64);
Best regards,
Maxim Uvarov.
From: Maxim Uvarov <muvarov@gmail.com>
Linker does not provide some vital functions when building freestanding
applications with a new toolchain, so we have to provide our own CRT.
p.s.
Without the CRT we won't see any build errors (since the purgatory is
linked with --no-undefined), but the purgatory code won't work,
'kexec -e' will just hang the board.
I added option to configure to keep code buildable for old toolchais.
But there should be way to do this automatically.
Author: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Maxim Uvarov <muvarov@gmail.com>
Signed-off-by: Maxim Uvarov <muvarov@gmail.com>
---
configure.ac | 9 +
purgatory/arch/ppc/Makefile | 1
purgatory/arch/ppc/crt.S | 263 +++++++++++++++++++++++++++++++++++++++++++
3 files changed, 273 insertions(+), 0 deletions(-)
create mode 100644 purgatory/arch/ppc/crt.S
diff --git a/configure.ac b/configure.ac
index fcf50e4..63606bc 100644
--- a/configure.ac
+++ b/configure.ac
@@ -159,6 +159,15 @@ if test "$with_xen" = yes ; then
AC_MSG_NOTICE([Xen support disabled])))
fi
+dnl new toolchains
+if test "$ARCH" = ppc; then
+AC_ARG_WITH([oldtoolchain],
+ AC_HELP_STRING([--with-oldtoolchain],[compile without crt.S
+ required for new toolchains]),
+ AC_DEFINE(PPC_OLDTOOLCHAIN, 1,
+ [Define to compile with old toolchains]))
+fi
+
dnl ---Sanity checks
if test "$CC" = "no"; then AC_MSG_ERROR([cc not found]); fi
if test "$CPP" = "no"; then AC_MSG_ERROR([cpp not found]); fi
diff --git a/purgatory/arch/ppc/Makefile b/purgatory/arch/ppc/Makefile
index 72289a0..4020778 100644
--- a/purgatory/arch/ppc/Makefile
+++ b/purgatory/arch/ppc/Makefile
@@ -6,6 +6,7 @@ ppc_PURGATORY_SRCS += purgatory/arch/ppc/v2wrap_32.S
ppc_PURGATORY_SRCS += purgatory/arch/ppc/misc.S
ppc_PURGATORY_SRCS += purgatory/arch/ppc/purgatory-ppc.c
ppc_PURGATORY_SRCS += purgatory/arch/ppc/console-ppc.c
+ppc_PURGATORY_SRCS += purgatory/arch/ppc/crt.S
dist += purgatory/arch/ppc/Makefile $(ppc_PURGATORY_SRCS) \
purgatory/arch/ppc/purgatory-ppc.h purgatory/arch/ppc/ppc_asm.h
diff --git a/purgatory/arch/ppc/crt.S b/purgatory/arch/ppc/crt.S
new file mode 100644
index 0000000..70f4d61
--- /dev/null
+++ b/purgatory/arch/ppc/crt.S
@@ -0,0 +1,263 @@
+/* This is from linux-2.6/arch/powerpc/lib/crtsavres.S:
+ *
+ * Special support for eabi and SVR4
+ *
+ * Copyright (C) 1995, 1996, 1998, 2000, 2001 Free Software Foundation, Inc.
+ * Copyright 2008 Freescale Semiconductor, Inc.
+ * Written By Michael Meissner
+ *
+ * Based on gcc/config/rs6000/crtsavres.asm from gcc
+ *
+ * This file is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * In addition to the permissions in the GNU General Public License, the
+ * Free Software Foundation gives you unlimited permission to link the
+ * compiled version of this file with other programs, and to distribute
+ * those programs without any restriction coming from the use of this
+ * file. (The General Public License restrictions do apply in other
+ * respects; for example, they cover modification of the file, and
+ * distribution when not linked into another program.)
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; see the file COPYING. If not, write to
+ * the Free Software Foundation, 51 Franklin Street, Fifth Floor,
+ * Boston, MA 02110-1301, USA.
+ *
+ * As a special exception, if you link this library with files
+ * compiled with GCC to produce an executable, this does not cause
+ * the resulting executable to be covered by the GNU General Public License.
+ * This exception does not however invalidate any other reasons why
+ * the executable file might be covered by the GNU General Public License.
+ */
+#include "config.h"
+
+/* On PowerPC64 Linux, these functions are provided by the linker. */
+#ifndef PPC_OLDTOOLCHAIN
+#ifndef __powerpc64__
+#define _GLOBAL(name) \
+ .type name,@function; \
+ .globl name; \
+name:
+
+/* Routines for saving integer registers, called by the compiler. */
+/* Called with r11 pointing to the stack header word of the caller of the */
+/* function, just beyond the end of the integer save area. */
+
+_GLOBAL(_savegpr_14)
+_GLOBAL(_save32gpr_14)
+ stw 14,-72(11) /* save gp registers */
+_GLOBAL(_savegpr_15)
+_GLOBAL(_save32gpr_15)
+ stw 15,-68(11)
+_GLOBAL(_savegpr_16)
+_GLOBAL(_save32gpr_16)
+ stw 16,-64(11)
+_GLOBAL(_savegpr_17)
+_GLOBAL(_save32gpr_17)
+ stw 17,-60(11)
+_GLOBAL(_savegpr_18)
+_GLOBAL(_save32gpr_18)
+ stw 18,-56(11)
+_GLOBAL(_savegpr_19)
+_GLOBAL(_save32gpr_19)
+ stw 19,-52(11)
+_GLOBAL(_savegpr_20)
+_GLOBAL(_save32gpr_20)
+ stw 20,-48(11)
+_GLOBAL(_savegpr_21)
+_GLOBAL(_save32gpr_21)
+ stw 21,-44(11)
+_GLOBAL(_savegpr_22)
+_GLOBAL(_save32gpr_22)
+ stw 22,-40(11)
+_GLOBAL(_savegpr_23)
+_GLOBAL(_save32gpr_23)
+ stw 23,-36(11)
+_GLOBAL(_savegpr_24)
+_GLOBAL(_save32gpr_24)
+ stw 24,-32(11)
+_GLOBAL(_savegpr_25)
+_GLOBAL(_save32gpr_25)
+ stw 25,-28(11)
+_GLOBAL(_savegpr_26)
+_GLOBAL(_save32gpr_26)
+ stw 26,-24(11)
+_GLOBAL(_savegpr_27)
+_GLOBAL(_save32gpr_27)
+ stw 27,-20(11)
+_GLOBAL(_savegpr_28)
+_GLOBAL(_save32gpr_28)
+ stw 28,-16(11)
+_GLOBAL(_savegpr_29)
+_GLOBAL(_save32gpr_29)
+ stw 29,-12(11)
+_GLOBAL(_savegpr_30)
+_GLOBAL(_save32gpr_30)
+ stw 30,-8(11)
+_GLOBAL(_savegpr_31)
+_GLOBAL(_save32gpr_31)
+ stw 31,-4(11)
+ blr
+
+/* Routines for restoring integer registers, called by the compiler. */
+/* Called with r11 pointing to the stack header word of the caller of the */
+/* function, just beyond the end of the integer restore area. */
+
+_GLOBAL(_restgpr_14)
+_GLOBAL(_rest32gpr_14)
+ lwz 14,-72(11) /* restore gp registers */
+_GLOBAL(_restgpr_15)
+_GLOBAL(_rest32gpr_15)
+ lwz 15,-68(11)
+_GLOBAL(_restgpr_16)
+_GLOBAL(_rest32gpr_16)
+ lwz 16,-64(11)
+_GLOBAL(_restgpr_17)
+_GLOBAL(_rest32gpr_17)
+ lwz 17,-60(11)
+_GLOBAL(_restgpr_18)
+_GLOBAL(_rest32gpr_18)
+ lwz 18,-56(11)
+_GLOBAL(_restgpr_19)
+_GLOBAL(_rest32gpr_19)
+ lwz 19,-52(11)
+_GLOBAL(_restgpr_20)
+_GLOBAL(_rest32gpr_20)
+ lwz 20,-48(11)
+_GLOBAL(_restgpr_21)
+_GLOBAL(_rest32gpr_21)
+ lwz 21,-44(11)
+_GLOBAL(_restgpr_22)
+_GLOBAL(_rest32gpr_22)
+ lwz 22,-40(11)
+_GLOBAL(_restgpr_23)
+_GLOBAL(_rest32gpr_23)
+ lwz 23,-36(11)
+_GLOBAL(_restgpr_24)
+_GLOBAL(_rest32gpr_24)
+ lwz 24,-32(11)
+_GLOBAL(_restgpr_25)
+_GLOBAL(_rest32gpr_25)
+ lwz 25,-28(11)
+_GLOBAL(_restgpr_26)
+_GLOBAL(_rest32gpr_26)
+ lwz 26,-24(11)
+_GLOBAL(_restgpr_27)
+_GLOBAL(_rest32gpr_27)
+ lwz 27,-20(11)
+_GLOBAL(_restgpr_28)
+_GLOBAL(_rest32gpr_28)
+ lwz 28,-16(11)
+_GLOBAL(_restgpr_29)
+_GLOBAL(_rest32gpr_29)
+ lwz 29,-12(11)
+_GLOBAL(_restgpr_30)
+_GLOBAL(_rest32gpr_30)
+ lwz 30,-8(11)
+_GLOBAL(_restgpr_31)
+_GLOBAL(_rest32gpr_31)
+ lwz 31,-4(11)
+ blr
+
+/* Routines for restoring integer registers, called by the compiler. */
+/* Called with r11 pointing to the stack header word of the caller of the */
+/* function, just beyond the end of the integer restore area. */
+
+_GLOBAL(_restgpr_14_x)
+_GLOBAL(_rest32gpr_14_x)
+ lwz 14,-72(11) /* restore gp registers */
+_GLOBAL(_restgpr_15_x)
+_GLOBAL(_rest32gpr_15_x)
+ lwz 15,-68(11)
+_GLOBAL(_restgpr_16_x)
+_GLOBAL(_rest32gpr_16_x)
+ lwz 16,-64(11)
+_GLOBAL(_restgpr_17_x)
+_GLOBAL(_rest32gpr_17_x)
+ lwz 17,-60(11)
+_GLOBAL(_restgpr_18_x)
+_GLOBAL(_rest32gpr_18_x)
+ lwz 18,-56(11)
+_GLOBAL(_restgpr_19_x)
+_GLOBAL(_rest32gpr_19_x)
+ lwz 19,-52(11)
+_GLOBAL(_restgpr_20_x)
+_GLOBAL(_rest32gpr_20_x)
+ lwz 20,-48(11)
+_GLOBAL(_restgpr_21_x)
+_GLOBAL(_rest32gpr_21_x)
+ lwz 21,-44(11)
+_GLOBAL(_restgpr_22_x)
+_GLOBAL(_rest32gpr_22_x)
+ lwz 22,-40(11)
+_GLOBAL(_restgpr_23_x)
+_GLOBAL(_rest32gpr_23_x)
+ lwz 23,-36(11)
+_GLOBAL(_restgpr_24_x)
+_GLOBAL(_rest32gpr_24_x)
+ lwz 24,-32(11)
+_GLOBAL(_restgpr_25_x)
+_GLOBAL(_rest32gpr_25_x)
+ lwz 25,-28(11)
+_GLOBAL(_restgpr_26_x)
+_GLOBAL(_rest32gpr_26_x)
+ lwz 26,-24(11)
+_GLOBAL(_restgpr_27_x)
+_GLOBAL(_rest32gpr_27_x)
+ lwz 27,-20(11)
+_GLOBAL(_restgpr_28_x)
+_GLOBAL(_rest32gpr_28_x)
+ lwz 28,-16(11)
+_GLOBAL(_restgpr_29_x)
+_GLOBAL(_rest32gpr_29_x)
+ lwz 29,-12(11)
+_GLOBAL(_restgpr_30_x)
+_GLOBAL(_rest32gpr_30_x)
+ lwz 30,-8(11)
+_GLOBAL(_restgpr_31_x)
+_GLOBAL(_rest32gpr_31_x)
+ lwz 0,4(11)
+ lwz 31,-4(11)
+ mtlr 0
+ mr 1,11
+ blr
+
+/* This is from linux-2.6/arch/powerpc/kernel/misc_32.S
+ *
+ * This file contains miscellaneous low-level functions.
+ * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ * Largely rewritten by Cort Dougan (cort@cs.nmt.edu)
+ * and Paul Mackerras.
+ *
+ * kexec bits:
+ * Copyright (C) 2002-2003 Eric Biederman <ebiederm@xmission.com>
+ * GameCube/ppc32 port Copyright (C) 2004 Albert Herranz
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ */
+
+_GLOBAL(__lshrdi3)
+ subfic 6,5,32
+ srw 4,4,5 # LSW = count > 31 ? 0 : LSW >> count
+ addi 7,5,32 # could be xori, or addi with -32
+ slw 6,3,6 # t1 = count > 31 ? 0 : MSW << (32-count)
+ srw 7,3,7 # t2 = count < 32 ? 0 : MSW >> (count-32)
+ or 4,4,6 # LSW |= t1
+ srw 3,3,5 # MSW = MSW >> count
+ or 4,4,7 # LSW |= t2
+ blr
+#endif
+#endif
^ permalink raw reply related
* [PATCH v2 1/2] Fix kexec on powerpc32
From: Maxim Uvarov @ 2010-05-12 7:08 UTC (permalink / raw)
To: linuxppc-dev, kexec; +Cc: lists, horms
Hello everybody,
Changes from previous version:
- removed bogus hyphen from the patch;
- move ifdefs to crt.S instead of Makefile
Please find here patch for user land kexec-tools application. Following
patch makes kexec-tools work for both kexec and kdump. I tested it with
git kernel (linus-tree) and Freescale/Logic MPC8360ERDK board with
mpc83xx_defconfig kernel config.
kexec:
kexec -l vmlinux --command-line="console= ... etc"
kexec -e
kdump:
kexec -p vmlinux_dump --command-line="console=... etc"
echo c > /proc/sysrq-trigger
I also think that is is reasonable:
- put GAME_CUBE specific code to separate files;
- combine ppc and ppc64 to powerpc directory (I'm planning to do it.
And that why in some places my patch have ifdefs for PPC64);
Best regards,
Maxim Uvarov.
From: Maxim Uvarov <muvarov@gmail.com>
Signed-off-by: Maxim Uvarov <muvarov@gmail.com>
Signed-off-by: Maxim Uvarov <muvarov@gmail.com>
---
kexec/arch/ppc/Makefile | 2
kexec/arch/ppc/crashdump-powerpc.c | 439 ++++++++++++++++++++++++++++++++++
kexec/arch/ppc/crashdump-powerpc.h | 38 +++
kexec/arch/ppc/fs2dt.c | 460 ++++++++++++++++++++++++++++++++++++
kexec/arch/ppc/kexec-elf-ppc.c | 186 +++++++++++++--
kexec/arch/ppc/kexec-ppc.c | 275 ++++++++++++++++++++--
kexec/arch/ppc/kexec-ppc.h | 32 +++
purgatory/arch/ppc/Makefile | 2
purgatory/arch/ppc/purgatory-ppc.c | 38 ++-
purgatory/arch/ppc/purgatory-ppc.h | 4
purgatory/arch/ppc/v2wrap.S | 66 -----
purgatory/arch/ppc/v2wrap_32.S | 91 +++++++
12 files changed, 1524 insertions(+), 109 deletions(-)
create mode 100644 kexec/arch/ppc/crashdump-powerpc.c
create mode 100644 kexec/arch/ppc/crashdump-powerpc.h
create mode 100644 kexec/arch/ppc/fs2dt.c
delete mode 100644 purgatory/arch/ppc/v2wrap.S
create mode 100644 purgatory/arch/ppc/v2wrap_32.S
diff --git a/kexec/arch/ppc/Makefile b/kexec/arch/ppc/Makefile
index 1c7441c..5988213 100644
--- a/kexec/arch/ppc/Makefile
+++ b/kexec/arch/ppc/Makefile
@@ -11,6 +11,8 @@ ppc_KEXEC_SRCS += kexec/arch/ppc/kexec-uImage-ppc.c
ppc_KEXEC_SRCS += kexec/arch/ppc/ppc-setup-simple.S
ppc_KEXEC_SRCS += kexec/arch/ppc/ppc-setup-dol.S
ppc_KEXEC_SRCS += kexec/arch/ppc/fixup_dtb.c
+ppc_KEXEC_SRCS += kexec/arch/ppc/fs2dt.c
+ppc_KEXEC_SRCS += kexec/arch/ppc/crashdump-powerpc.c
ppc_KEXEC_SRCS += kexec/kexec-uImage.c
libfdt_SRCS = kexec/arch/ppc/libfdt-wrapper.c
diff --git a/kexec/arch/ppc/crashdump-powerpc.c b/kexec/arch/ppc/crashdump-powerpc.c
new file mode 100644
index 0000000..7bfad20
--- /dev/null
+++ b/kexec/arch/ppc/crashdump-powerpc.c
@@ -0,0 +1,439 @@
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <limits.h>
+#include <elf.h>
+#include <dirent.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include "../../kexec.h"
+#include "../../kexec-elf.h"
+#include "../../kexec-syscall.h"
+#include "../../crashdump.h"
+#include "kexec-ppc.h"
+#include "crashdump-powerpc.h"
+
+#ifdef CONFIG_PPC64
+static struct crash_elf_info elf_info64 = {
+class: ELFCLASS64,
+data: ELFDATA2MSB,
+machine: EM_PPC64,
+backup_src_start: BACKUP_SRC_START,
+backup_src_end: BACKUP_SRC_END,
+page_offset: PAGE_OFFSET,
+lowmem_limit: MAXMEM,
+};
+#endif
+static struct crash_elf_info elf_info32 = {
+class: ELFCLASS32,
+data: ELFDATA2MSB,
+#ifdef CONFIG_PPC64
+machine: EM_PPC64,
+#else
+machine: EM_PPC,
+#endif
+backup_src_start: BACKUP_SRC_START,
+backup_src_end: BACKUP_SRC_END,
+page_offset: PAGE_OFFSET,
+lowmem_limit: MAXMEM,
+};
+
+/* Stores a sorted list of RAM memory ranges for which to create elf headers.
+ * A separate program header is created for backup region
+ */
+static struct memory_range *crash_memory_range;
+
+/* Define a variable to replace the CRASH_MAX_MEMORY_RANGES macro */
+static int crash_max_memory_ranges;
+
+/*
+ * Used to save various memory ranges/regions needed for the captured
+ * kernel to boot. (lime memmap= option in other archs)
+ */
+mem_rgns_t usablemem_rgns = {0, NULL};
+
+/*
+ * To store the memory size of the first kernel and this value will be
+ * passed to the second kernel as command line (savemaxmem=xM).
+ * The second kernel will be calculated saved_max_pfn based on this
+ * variable.
+ * Since we are creating/using usable-memory property, there is no way
+ * we can determine the RAM size unless parsing the device-tree/memoy@/reg
+ * property in the kernel.
+ */
+unsigned long long saved_max_mem;
+
+/* Reads the appropriate file and retrieves the SYSTEM RAM regions for whom to
+ * create Elf headers. Keeping it separate from get_memory_ranges() as
+ * requirements are different in the case of normal kexec and crashdumps.
+ *
+ * Normal kexec needs to look at all of available physical memory irrespective
+ * of the fact how much of it is being used by currently running kernel.
+ * Crashdumps need to have access to memory regions actually being used by
+ * running kernel. Expecting a different file/data structure than /proc/iomem
+ * to look into down the line. May be something like /proc/kernelmem or may
+ * be zone data structures exported from kernel.
+ */
+static int get_crash_memory_ranges(struct memory_range **range, int *ranges)
+{
+
+ int memory_ranges = 0;
+ char device_tree[256] = "/proc/device-tree/";
+ char fname[256];
+ char buf[MAXBYTES-1];
+ DIR *dir, *dmem;
+ FILE *file;
+ struct dirent *dentry, *mentry;
+ int i, n, crash_rng_len = 0;
+ unsigned long long start, end, cstart, cend;
+
+ crash_max_memory_ranges = max_memory_ranges + 6;
+ crash_rng_len = sizeof(struct memory_range) * crash_max_memory_ranges;
+
+ crash_memory_range = (struct memory_range *) malloc(crash_rng_len);
+ if (!crash_memory_range) {
+ fprintf(stderr, "Allocation for crash memory range failed\n");
+ return -1;
+ }
+ memset(crash_memory_range, 0, crash_rng_len);
+
+ /* create a separate program header for the backup region */
+ crash_memory_range[0].start = BACKUP_SRC_START;
+ crash_memory_range[0].end = BACKUP_SRC_END + 1;
+ crash_memory_range[0].type = RANGE_RAM;
+ memory_ranges++;
+
+ dir = opendir(device_tree);
+ if (!dir) {
+ perror(device_tree);
+ goto err;
+ }
+ while ((dentry = readdir(dir)) != NULL) {
+ if (strncmp(dentry->d_name, "memory@", 7)
+ && strcmp(dentry->d_name, "memory"))
+ continue;
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ dmem = opendir(fname);
+ if (!dmem) {
+ perror(fname);
+ closedir(dir);
+ goto err;
+ }
+ while ((mentry = readdir(dmem)) != NULL) {
+ if (strcmp(mentry->d_name, "reg"))
+ continue;
+ strcat(fname, "/reg");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ closedir(dmem);
+ closedir(dir);
+ goto err;
+ }
+ n = fread(buf, 1, MAXBYTES, file);
+ if (n < 0) {
+ perror(fname);
+ fclose(file);
+ closedir(dmem);
+ closedir(dir);
+ goto err;
+ }
+ if (memory_ranges >= (max_memory_ranges + 1)) {
+ /* No space to insert another element. */
+ fprintf(stderr,
+ "Error: Number of crash memory ranges"
+ " excedeed the max limit\n");
+ goto err;
+ }
+
+ /*
+ * FIXME: This code fails on platforms that
+ * have more than one memory range specified
+ * in the device-tree's /memory/reg property.
+ * or where the #address-cells and #size-cells
+ * are not identical.
+ *
+ * We should interpret the /memory/reg property
+ * based on the values of the #address-cells and
+ * #size-cells properites.
+ */
+ if (n == (sizeof(unsigned long) * 2)) {
+ start = ((unsigned long *)buf)[0];
+ end = start + ((unsigned long *)buf)[1];
+ } else {
+ start = ((unsigned long long *)buf)[0];
+ end = start + ((unsigned long long *)buf)[1];
+ }
+ if (start == 0 && end >= (BACKUP_SRC_END + 1))
+ start = BACKUP_SRC_END + 1;
+
+ cstart = crash_base;
+ cend = crash_base + crash_size;
+ /*
+ * Exclude the region that lies within crashkernel
+ */
+ if (cstart < end && cend > start) {
+ if (start < cstart && end > cend) {
+ crash_memory_range[memory_ranges].start
+ = start;
+ crash_memory_range[memory_ranges].end
+ = cstart;
+ crash_memory_range[memory_ranges].type
+ = RANGE_RAM;
+ memory_ranges++;
+ crash_memory_range[memory_ranges].start
+ = cend;
+ crash_memory_range[memory_ranges].end
+ = end;
+ crash_memory_range[memory_ranges].type
+ = RANGE_RAM;
+ memory_ranges++;
+ } else if (start < cstart) {
+ crash_memory_range[memory_ranges].start
+ = start;
+ crash_memory_range[memory_ranges].end
+ = cstart;
+ crash_memory_range[memory_ranges].type
+ = RANGE_RAM;
+ memory_ranges++;
+ } else if (end > cend) {
+ crash_memory_range[memory_ranges].start
+ = cend;
+ crash_memory_range[memory_ranges].end
+ = end;
+ crash_memory_range[memory_ranges].type
+ = RANGE_RAM;
+ memory_ranges++;
+ }
+ } else {
+ crash_memory_range[memory_ranges].start = start;
+ crash_memory_range[memory_ranges].end = end;
+ crash_memory_range[memory_ranges].type
+ = RANGE_RAM;
+ memory_ranges++;
+ }
+ fclose(file);
+ }
+ closedir(dmem);
+ }
+ closedir(dir);
+
+ /*
+ * If RTAS region is overlapped with crashkernel, need to create ELF
+ * Program header for the overlapped memory.
+ */
+ if (crash_base < rtas_base + rtas_size &&
+ rtas_base < crash_base + crash_size) {
+ cstart = rtas_base;
+ cend = rtas_base + rtas_size;
+ if (cstart < crash_base)
+ cstart = crash_base;
+ if (cend > crash_base + crash_size)
+ cend = crash_base + crash_size;
+ crash_memory_range[memory_ranges].start = cstart;
+ crash_memory_range[memory_ranges++].end = cend;
+ }
+ /*
+ * Can not trust the memory regions order that we read from
+ * device-tree. Hence, get the MAX end value.
+ */
+ for (i = 0; i < memory_ranges; i++)
+ if (saved_max_mem < crash_memory_range[i].end)
+ saved_max_mem = crash_memory_range[i].end;
+
+ *range = crash_memory_range;
+ *ranges = memory_ranges;
+#if DEBUG
+ int j;
+ printf("CRASH MEMORY RANGES\n");
+ for (j = 0; j < *ranges; j++) {
+ start = crash_memory_range[j].start;
+ end = crash_memory_range[j].end;
+ fprintf(stderr, "%016Lx-%016Lx\n", start, end);
+ }
+#endif
+ return 0;
+
+err:
+ if (crash_memory_range)
+ free(crash_memory_range);
+ return -1;
+}
+
+/* Converts unsigned long to ascii string. */
+static void ulltoa(unsigned long long i, char *str)
+{
+ int j = 0, k;
+ char tmp;
+
+ do {
+ str[j++] = i % 10 + '0';
+ } while ((i /= 10) > 0);
+ str[j] = '\0';
+
+ /* Reverse the string. */
+ for (j = 0, k = strlen(str) - 1; j < k; j++, k--) {
+ tmp = str[k];
+ str[k] = str[j];
+ str[j] = tmp;
+ }
+}
+
+static int add_cmdline_param(char *cmdline, unsigned long long addr,
+ char *cmdstr, char *byte)
+{
+ int cmdlen, len, align = 1024;
+ char str[COMMAND_LINE_SIZE], *ptr;
+
+ /* Passing in =xxxK / =xxxM format. Saves space required in cmdline.*/
+ switch (byte[0]) {
+ case 'K':
+ if (addr%align)
+ return -1;
+ addr = addr/align;
+ break;
+ case 'M':
+ addr = addr/(align *align);
+ break;
+ }
+ ptr = str;
+ strcpy(str, cmdstr);
+ ptr += strlen(str);
+ ulltoa(addr, ptr);
+ strcat(str, byte);
+ len = strlen(str);
+ cmdlen = strlen(cmdline) + len;
+ if (cmdlen > (COMMAND_LINE_SIZE - 1))
+ die("Command line overflow\n");
+ strcat(cmdline, str);
+#if DEBUG
+ fprintf(stderr, "Command line after adding elfcorehdr: %s\n", cmdline);
+#endif
+ return 0;
+}
+
+/* Loads additional segments in case of a panic kernel is being loaded.
+ * One segment for backup region, another segment for storing elf headers
+ * for crash memory image.
+ */
+int load_crashdump_segments(struct kexec_info *info, char *mod_cmdline,
+ unsigned long max_addr, unsigned long min_base)
+{
+ void *tmp;
+ unsigned long sz, elfcorehdr;
+ int nr_ranges, align = 1024, i;
+ unsigned long long end;
+ struct memory_range *mem_range;
+
+ if (get_crash_memory_ranges(&mem_range, &nr_ranges) < 0)
+ return -1;
+
+ /* Create a backup region segment to store backup data*/
+ sz = (BACKUP_SRC_SIZE + align - 1) & ~(align - 1);
+ tmp = xmalloc(sz);
+ memset(tmp, 0, sz);
+ info->backup_start = add_buffer(info, tmp, sz, sz, align,
+ 0, max_addr, 1);
+ reserve(info->backup_start, sz);
+
+ /* On powerpc memory ranges in device-tree is denoted as start
+ * and size rather than start and end, as is the case with
+ * other architectures like i386 . Because of this when loading
+ * the memory ranges in crashdump-elf.c the filesz calculation
+ * [ end - start + 1 ] goes for a toss.
+ *
+ * To be in sync with other archs adjust the end value for
+ * every crash memory range before calling the generic function
+ */
+
+ for (i = 0; i < nr_ranges; i++) {
+ end = crash_memory_range[i].end - 1;
+ crash_memory_range[i].end = end;
+ }
+
+
+#ifdef CONFIG_PPC64
+ /* Create elf header segment and store crash image data. */
+ if (arch_options.core_header_type == CORE_TYPE_ELF64) {
+ if (crash_create_elf64_headers(info, &elf_info64,
+ crash_memory_range, nr_ranges, &tmp,
+ &sz, ELF_CORE_HEADER_ALIGN) < 0)
+ return -1;
+ } else if (crash_create_elf32_headers(info, &elf_info32,
+ crash_memory_range, nr_ranges, &tmp, &sz,
+ ELF_CORE_HEADER_ALIGN) < 0)
+ return -1;
+#else
+ if (crash_create_elf32_headers(info, &elf_info32, crash_memory_range,
+ nr_ranges, &tmp, &sz, ELF_CORE_HEADER_ALIGN)
+ < 0)
+ return -1;
+#endif
+
+ elfcorehdr = add_buffer(info, tmp, sz, sz, align,
+ min_base, max_addr, 1);
+ reserve(elfcorehdr, sz);
+ /* modify and store the cmdline in a global array. This is later
+ * read by flatten_device_tree and modified if required
+ */
+ add_cmdline_param(mod_cmdline, elfcorehdr, " elfcorehdr=", "K");
+ add_cmdline_param(mod_cmdline, saved_max_mem, " savemaxmem=", "M");
+ return 0;
+}
+
+/*
+ * Used to save various memory regions needed for the captured kernel.
+ */
+
+void add_usable_mem_rgns(unsigned long long base, unsigned long long size)
+{
+ int i;
+ unsigned long long end = base + size;
+ unsigned long long ustart, uend;
+
+ base = _ALIGN_DOWN(base, getpagesize());
+ end = _ALIGN_UP(end, getpagesize());
+
+ for (i = 0; i < usablemem_rgns.size; i++) {
+ ustart = usablemem_rgns.ranges[i].start;
+ uend = usablemem_rgns.ranges[i].end;
+ if (base < uend && end > ustart) {
+ if ((base >= ustart) && (end <= uend))
+ return;
+ if (base < ustart && end > uend) {
+ usablemem_rgns.ranges[i].start = base;
+ usablemem_rgns.ranges[i].end = end;
+ return;
+ } else if (base < ustart) {
+ usablemem_rgns.ranges[i].start = base;
+ return;
+ } else if (end > uend) {
+ usablemem_rgns.ranges[i].end = end;
+ return;
+ }
+ }
+ }
+ usablemem_rgns.ranges[usablemem_rgns.size].start = base;
+ usablemem_rgns.ranges[usablemem_rgns.size++].end = end;
+
+#ifdef DEBUG
+ fprintf(stderr, "usable memory rgns size:%u base:%llx size:%llx\n",
+ usablemem_rgns.size, base, size);
+#endif
+}
+
+int is_crashkernel_mem_reserved(void)
+{
+ int fd;
+
+ fd = open("/proc/device-tree/chosen/linux,crashkernel-base", O_RDONLY);
+ if (fd < 0)
+ return 0;
+ close(fd);
+ return 1;
+}
+
diff --git a/kexec/arch/ppc/crashdump-powerpc.h b/kexec/arch/ppc/crashdump-powerpc.h
new file mode 100644
index 0000000..dc2772d
--- /dev/null
+++ b/kexec/arch/ppc/crashdump-powerpc.h
@@ -0,0 +1,38 @@
+#ifndef CRASHDUMP_POWERPC_H
+#define CRASHDUMP_POWERPC_H
+
+struct kexec_info;
+int load_crashdump_segments(struct kexec_info *info, char *mod_cmdline,
+ unsigned long max_addr, unsigned long min_base);
+void add_usable_mem_rgns(unsigned long long base, unsigned long long size);
+
+extern struct arch_options_t arch_options;
+
+#ifdef CONFIG_PPC64
+#define PAGE_OFFSET 0xC000000000000000UL
+#define VMALLOCBASE 0xD000000000000000UL
+#define MAXMEM (-KERNELBASE-VMALLOCBASE)
+#else
+#define PAGE_OFFSET 0xC0000000
+#define MAXMEM 0x30000000 /* Use CONFIG_LOWMEM_SIZE from kernel */
+#endif
+
+#define KERNELBASE PAGE_OFFSET
+#define __pa(x) ((unsigned long)(x)-PAGE_OFFSET)
+
+#define COMMAND_LINE_SIZE 512 /* from kernel */
+/* Backup Region, First 64K of System RAM. */
+#define BACKUP_SRC_START 0x0000
+#define BACKUP_SRC_END 0xffff
+#define BACKUP_SRC_SIZE (BACKUP_SRC_END - BACKUP_SRC_START + 1)
+
+#define KDUMP_BACKUP_LIMIT BACKUP_SRC_SIZE
+#define _ALIGN_UP(addr, size) (((addr)+((size)-1))&(~((size)-1)))
+#define _ALIGN_DOWN(addr, size) ((addr)&(~((size)-1)))
+
+extern unsigned long long crash_base;
+extern unsigned long long crash_size;
+extern unsigned int rtas_base;
+extern unsigned int rtas_size;
+
+#endif /* CRASHDUMP_POWERPC_H */
diff --git a/kexec/arch/ppc/fs2dt.c b/kexec/arch/ppc/fs2dt.c
new file mode 100644
index 0000000..238a3f2
--- /dev/null
+++ b/kexec/arch/ppc/fs2dt.c
@@ -0,0 +1,460 @@
+/*
+ * fs2dt: creates a flattened device-tree
+ *
+ * Copyright (C) 2004,2005 Milton D Miller II, IBM Corporation
+ * Copyright (C) 2005 R Sharada (sharada@in.ibm.com), IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation (version 2 of the License).
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <sys/types.h>
+#include <sys/stat.h>
+
+#include <fcntl.h>
+#include <dirent.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <stdio.h>
+#include "../../kexec.h"
+#include "kexec-ppc.h"
+
+#define MAXPATH 1024 /* max path name length */
+#define NAMESPACE 16384 /* max bytes for property names */
+#define TREEWORDS 65536 /* max 32 bit words for properties */
+#define MEMRESERVE 256 /* max number of reserved memory blks */
+#define MAX_MEMORY_RANGES 1024
+#define COMMAND_LINE_SIZE 512 /* from kernel */
+
+static char pathname[MAXPATH];
+static char propnames[NAMESPACE] = { 0 };
+static unsigned dtstruct[TREEWORDS], *dt;
+static unsigned long long mem_rsrv[2*MEMRESERVE] = { 0, 0 };
+
+static int crash_param;
+static char local_cmdline[COMMAND_LINE_SIZE] = { "" };
+static unsigned *dt_len; /* changed len of modified cmdline
+ in flat device-tree */
+static struct bootblock bb[1];
+
+void reserve(unsigned long long where, unsigned long long length)
+{
+ size_t offset;
+
+ for (offset = 0; mem_rsrv[offset + 1]; offset += 2)
+ ;
+
+ if (offset + 4 >= 2 * MEMRESERVE)
+ die("unrecoverable error: exhasuted reservation meta data\n");
+
+ mem_rsrv[offset] = where;
+ mem_rsrv[offset + 1] = length;
+ mem_rsrv[offset + 3] = 0; /* N.B: don't care about offset + 2 */
+}
+
+/* look for properties we need to reserve memory space for */
+static void checkprop(char *name, unsigned *data, int len)
+{
+ static unsigned long long base, size, end;
+
+ if ((data == NULL) && (base || size || end))
+ die("unrecoverable error: no property data");
+ else if (!strcmp(name, "linux,rtas-base"))
+ base = *data;
+ else if (!strcmp(name, "linux,tce-base"))
+ base = *(unsigned long long *) data;
+ else if (!strcmp(name, "rtas-size") ||
+ !strcmp(name, "linux,tce-size"))
+ size = *data;
+ else if (reuse_initrd && !strcmp(name, "linux,initrd-start"))
+ if (len == 8)
+ base = *(unsigned long long *) data;
+ else
+ base = *data;
+ else if (reuse_initrd && !strcmp(name, "linux,initrd-end"))
+ end = *(unsigned long long *) data;
+
+ if (size && end)
+ die("unrecoverable error: size and end set at same time\n");
+ if (base && size) {
+ reserve(base, size);
+ base = 0;
+ size = 0;
+ }
+ if (base && end) {
+ reserve(base, end-base);
+ base = 0;
+ end = 0;
+ }
+}
+
+/*
+ * return the property index for a property name, creating a new one
+ * if needed.
+ */
+static unsigned propnum(const char *name)
+{
+ unsigned offset = 0;
+
+ while (propnames[offset])
+ if (strcmp(name, propnames+offset))
+ offset += strlen(propnames+offset)+1;
+ else
+ return offset;
+
+ if (NAMESPACE - offset < strlen(name) + 1)
+ die("unrecoverable error: propnames overrun\n");
+
+ strcpy(propnames+offset, name);
+
+ return offset;
+}
+
+static void add_usable_mem_property(int fd, int len)
+{
+ char fname[MAXPATH], *bname;
+ unsigned long buf[2];
+ unsigned long ranges[2*MAX_MEMORY_RANGES];
+ unsigned long long base, end, loc_base, loc_end;
+ int range, rlen = 0;
+
+ strcpy(fname, pathname);
+ bname = strrchr(fname, '/');
+ bname[0] = '\0';
+ bname = strrchr(fname, '/');
+ if (strncmp(bname, "/memory@", 8) && strcmp(bname, "/memory"))
+ return;
+
+ if (len < 2 * sizeof(unsigned long))
+ die("unrecoverable error: not enough data for mem property\n");
+ len = 2 * sizeof(unsigned long);
+
+ if (lseek(fd, 0, SEEK_SET) < 0)
+ die("unrecoverable error: error seeking in \"%s\": %s\n",
+ pathname, strerror(errno));
+ if (read(fd, buf, len) != len)
+ die("unrecoverable error: error reading \"%s\": %s\n",
+ pathname, strerror(errno));
+
+ if (~0ULL - buf[0] < buf[1])
+ die("unrecoverable error: mem property overflow\n");
+ base = buf[0];
+ end = base + buf[1];
+
+ for (range = 0; range < usablemem_rgns.size; range++) {
+ loc_base = usablemem_rgns.ranges[range].start;
+ loc_end = usablemem_rgns.ranges[range].end;
+ if (loc_base >= base && loc_end <= end) {
+ ranges[rlen++] = loc_base;
+ ranges[rlen++] = loc_end - loc_base;
+ } else if (base < loc_end && end > loc_base) {
+ if (loc_base < base)
+ loc_base = base;
+ if (loc_end > end)
+ loc_end = end;
+ ranges[rlen++] = loc_base;
+ ranges[rlen++] = loc_end - loc_base;
+ }
+ }
+
+ if (!rlen) {
+ /*
+ * User did not pass any ranges for thsi region. Hence, write
+ * (0,0) duple in linux,usable-memory property such that
+ * this region will be ignored.
+ */
+ ranges[rlen++] = 0;
+ ranges[rlen++] = 0;
+ }
+
+ rlen = rlen * sizeof(unsigned long);
+ /*
+ * No add linux,usable-memory property.
+ */
+ *dt++ = 3;
+ *dt++ = rlen;
+ *dt++ = propnum("linux,usable-memory");
+ memcpy(dt, &ranges, rlen);
+ dt += (rlen + 3)/4;
+}
+
+/* put all properties (files) in the property structure */
+static void putprops(char *fn, struct dirent **nlist, int numlist)
+{
+ struct dirent *dp;
+ int i = 0, fd, len;
+ struct stat statbuf;
+
+ for (i = 0; i < numlist; i++) {
+ dp = nlist[i];
+ strcpy(fn, dp->d_name);
+
+ if (!strcmp(dp->d_name, ".") || !strcmp(dp->d_name, ".."))
+ continue;
+
+ if (lstat(pathname, &statbuf))
+ die("unrecoverable error: could not stat \"%s\": %s\n",
+ pathname, strerror(errno));
+
+ if (!crash_param && !strcmp(fn, "linux,crashkernel-base"))
+ continue;
+
+ if (!crash_param && !strcmp(fn, "linux,crashkernel-size"))
+ continue;
+
+ /*
+ * This property will be created for each node during kexec
+ * boot. So, ignore it.
+ */
+ if (!strcmp(dp->d_name, "linux,pci-domain") ||
+ !strcmp(dp->d_name, "linux,htab-base") ||
+ !strcmp(dp->d_name, "linux,htab-size") ||
+ !strcmp(dp->d_name, "linux,kernel-end") ||
+ !strcmp(dp->d_name, "linux,usable-memory"))
+ continue;
+
+ /* This property will be created/modified later in putnode()
+ * So ignore it, unless we are reusing the initrd.
+ */
+ if ((!strcmp(dp->d_name, "linux,initrd-start") ||
+ !strcmp(dp->d_name, "linux,initrd-end")) &&
+ !reuse_initrd)
+ continue;
+
+ if (!S_ISREG(statbuf.st_mode))
+ continue;
+
+ len = statbuf.st_size;
+
+ *dt++ = 3;
+ dt_len = dt;
+ *dt++ = len;
+ *dt++ = propnum(fn);
+
+ fd = open(pathname, O_RDONLY);
+ if (fd == -1)
+ die("unrecoverable error: could not open \"%s\": %s\n",
+ pathname, strerror(errno));
+
+ if (read(fd, dt, len) != len)
+ die("unrecoverable error: could not read \"%s\": %s\n",
+ pathname, strerror(errno));
+
+ checkprop(fn, dt, len);
+
+ /* Get the cmdline from the device-tree and modify it */
+ if (!strcmp(dp->d_name, "bootargs")) {
+ int cmd_len;
+ char temp_cmdline[COMMAND_LINE_SIZE] = { "" };
+ char *param = NULL;
+ cmd_len = strlen(local_cmdline);
+ if (cmd_len != 0) {
+ param = strstr(local_cmdline, "crashkernel=");
+ if (param)
+ crash_param = 1;
+ param = strstr(local_cmdline, "root=");
+ }
+ if (!param) {
+ char *old_param;
+ memcpy(temp_cmdline, dt, len);
+ param = strstr(temp_cmdline, "root=");
+ if (param) {
+ old_param = strtok(param, " ");
+ if (cmd_len != 0)
+ strcat(local_cmdline, " ");
+ strcat(local_cmdline, old_param);
+ }
+ }
+ strcat(local_cmdline, " ");
+ cmd_len = strlen(local_cmdline);
+ cmd_len = cmd_len + 1;
+ memcpy(dt, local_cmdline, cmd_len);
+ len = cmd_len;
+ *dt_len = cmd_len;
+#if DEBUG
+ fprintf(stderr, "Modified cmdline:%s\n", local_cmdline);
+#endif
+ }
+
+ dt += (len + 3)/4;
+ if (!strcmp(dp->d_name, "reg") && usablemem_rgns.size)
+ add_usable_mem_property(fd, len);
+ close(fd);
+ }
+
+ fn[0] = '\0';
+ checkprop(pathname, NULL, 0);
+}
+
+/*
+ * Compare function used to sort the device-tree directories
+ * This function will be passed to scandir.
+ */
+static int comparefunc(const void *dentry1, const void *dentry2)
+{
+ char *str1 = (*(struct dirent **)dentry1)->d_name;
+ char *str2 = (*(struct dirent **)dentry2)->d_name;
+
+ /*
+ * strcmp scans from left to right and fails to idetify for some
+ * strings such as memory@10000000 and memory@f000000.
+ * Therefore, we get the wrong sorted order like memory@10000000 and
+ * memory@f000000.
+ */
+ if (strchr(str1, '@') && strchr(str2, '@') &&
+ (strlen(str1) > strlen(str2)))
+ return 1;
+
+ return strcmp(str1, str2);
+}
+
+/*
+ * put a node (directory) in the property structure. first properties
+ * then children.
+ */
+static void putnode(void)
+{
+ char *dn;
+ struct dirent *dp;
+ char *basename;
+ struct dirent **namelist;
+ int numlist, i;
+ struct stat statbuf;
+
+ numlist = scandir(pathname, &namelist, 0, comparefunc);
+ if (numlist < 0)
+ die("unrecoverable error: could not scan \"%s\": %s\n",
+ pathname, strerror(errno));
+ if (numlist == 0)
+ die("unrecoverable error: no directory entries in \"%s\"",
+ pathname);
+
+ basename = strrchr(pathname, '/') + 1;
+
+ *dt++ = 1;
+ strcpy((void *)dt, *basename ? basename : "");
+ dt += strlen((void *)dt) / sizeof(unsigned) + 1;
+
+ strcat(pathname, "/");
+ dn = pathname + strlen(pathname);
+
+ putprops(dn, namelist, numlist);
+
+ /* Add initrd entries to the second kernel */
+ if (initrd_base && !strcmp(basename, "chosen/")) {
+ int len = 8;
+ unsigned long long initrd_end;
+ *dt++ = 3;
+ *dt++ = len;
+ *dt++ = propnum("linux,initrd-start");
+
+ memcpy(dt, &initrd_base, len);
+ dt += (len + 3)/4;
+
+ len = 8;
+ *dt++ = 3;
+ *dt++ = len;
+ *dt++ = propnum("linux,initrd-end");
+
+ initrd_end = initrd_base + initrd_size;
+
+ memcpy(dt, &initrd_end, len);
+ dt += (len + 3)/4;
+
+ reserve(initrd_base, initrd_size);
+ }
+
+ for (i = 0; i < numlist; i++) {
+ dp = namelist[i];
+ strcpy(dn, dp->d_name);
+ free(namelist[i]);
+
+ if (!strcmp(dn, ".") || !strcmp(dn, ".."))
+ continue;
+
+ if (lstat(pathname, &statbuf))
+ die("unrecoverable error: could not stat \"%s\": %s\n",
+ pathname, strerror(errno));
+
+ if (S_ISDIR(statbuf.st_mode))
+ putnode();
+ }
+
+ *dt++ = 2;
+ dn[-1] = '\0';
+ free(namelist);
+}
+
+int create_flatten_tree(struct kexec_info *info, unsigned char **bufp,
+ unsigned long *sizep, char *cmdline)
+{
+ unsigned long len;
+ unsigned long tlen;
+ unsigned char *buf;
+ unsigned long me;
+
+ me = 0;
+
+ strcpy(pathname, "/proc/device-tree/");
+
+ dt = dtstruct;
+
+ if (cmdline)
+ strcpy(local_cmdline, cmdline);
+
+ putnode();
+ *dt++ = 9;
+
+ len = sizeof(bb[0]);
+ len += 7; len &= ~7;
+
+ bb->off_mem_rsvmap = len;
+
+ for (len = 1; mem_rsrv[len]; len += 2)
+ ;
+ len += 3;
+ len *= sizeof(mem_rsrv[0]);
+
+ bb->off_dt_struct = bb->off_mem_rsvmap + len;
+
+ len = dt - dtstruct;
+ len *= sizeof(unsigned);
+ bb->dt_struct_size = len;
+ bb->off_dt_strings = bb->off_dt_struct + len;
+
+ len = propnum("");
+ bb->dt_strings_size = len;
+ len += 3; len &= ~3;
+ bb->totalsize = bb->off_dt_strings + len;
+
+ bb->magic = 0xd00dfeed;
+ bb->version = 17;
+ bb->last_comp_version = 16;
+
+ reserve(me, bb->totalsize); /* patched later in kexec_load */
+
+ buf = (unsigned char *) malloc(bb->totalsize);
+ *bufp = buf;
+ memcpy(buf, bb, bb->off_mem_rsvmap);
+ tlen = bb->off_mem_rsvmap;
+ memcpy(buf+tlen, mem_rsrv, bb->off_dt_struct - bb->off_mem_rsvmap);
+ tlen = tlen + (bb->off_dt_struct - bb->off_mem_rsvmap);
+ memcpy(buf+tlen, dtstruct, bb->off_dt_strings - bb->off_dt_struct);
+ tlen = tlen + (bb->off_dt_strings - bb->off_dt_struct);
+ memcpy(buf+tlen, propnames, bb->totalsize - bb->off_dt_strings);
+ tlen = tlen + bb->totalsize - bb->off_dt_strings;
+ *sizep = tlen;
+ return 0;
+}
diff --git a/kexec/arch/ppc/kexec-elf-ppc.c b/kexec/arch/ppc/kexec-elf-ppc.c
index a54a5d5..4bcac26 100644
--- a/kexec/arch/ppc/kexec-elf-ppc.c
+++ b/kexec/arch/ppc/kexec-elf-ppc.c
@@ -11,6 +11,7 @@
#include <string.h>
#include <stdlib.h>
#include <errno.h>
+#include <limits.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
@@ -23,15 +24,22 @@
#include "../../kexec-elf.h"
#include "kexec-ppc.h"
#include <arch/options.h>
+#include "../../kexec-syscall.h"
+#include "crashdump-powerpc.h"
#include "config.h"
#include "fixup_dtb.h"
static const int probe_debug = 0;
-#define MAX_COMMAND_LINE 256
+unsigned long long initrd_base, initrd_size;
+unsigned char reuse_initrd;
+const char *ramdisk;
+int create_flatten_tree(struct kexec_info *, unsigned char **, unsigned long *,
+ char *);
#define UPSZ(X) ((sizeof(X) + 3) & ~3)
+#ifdef WITH_GAMECUBE
static struct boot_notes {
Elf_Bhdr hdr;
Elf_Nhdr bl_hdr;
@@ -65,7 +73,7 @@ static struct boot_notes {
.n_type = EBN_COMMAND_LINE,
},
};
-
+#endif
int elf_ppc_probe(const char *buf, off_t len)
{
@@ -92,6 +100,7 @@ int elf_ppc_probe(const char *buf, off_t len)
return result;
}
+#ifdef WITH_GAMECUBE
static void gamecube_hack_addresses(struct mem_ehdr *ehdr)
{
struct mem_phdr *phdr, *phdr_end;
@@ -112,6 +121,7 @@ static void gamecube_hack_addresses(struct mem_ehdr *ehdr)
phdr->p_paddr &= ~0xf0000000; /* clear bits 0-3, ibm syntax */
}
}
+#endif
#define OPT_APPEND (OPT_ARCH_MAX+0)
#define OPT_GAMECUBE (OPT_ARCH_MAX+1)
@@ -145,10 +155,24 @@ int elf_ppc_load(int argc, char **argv, const char *buf, off_t len,
struct kexec_info *info)
{
struct mem_ehdr ehdr;
- char *command_line;
+ char *command_line, *crash_cmdline, *cmdline_buf;
int command_line_len;
char *dtb;
int result;
+ unsigned long max_addr, hole_addr;
+ struct mem_phdr *phdr;
+ size_t size;
+ unsigned long long *rsvmap_ptr;
+ struct bootblock *bb_ptr;
+ unsigned int nr_segments;
+ unsigned long my_kernel, my_dt_offset;
+ unsigned long my_stack, my_backup_start;
+#ifdef CONFIG_PPC64
+ unsigned long toc_addr;
+#endif
+ unsigned int slave_code[256 / sizeof(unsigned int)], master_entry;
+ unsigned char *seg_buf = NULL;
+ off_t seg_size = 0;
#ifdef WITH_GAMECUBE
int target_is_gamecube = 1;
char *arg_buf;
@@ -170,6 +194,9 @@ int elf_ppc_load(int argc, char **argv, const char *buf, off_t len,
command_line = NULL;
dtb = NULL;
+ max_addr = LONG_MAX;
+ hole_addr = 0;
+
while ((opt = getopt_long(argc, argv, short_options, options, 0)) != -1) {
switch (opt) {
default:
@@ -209,15 +236,45 @@ int elf_ppc_load(int argc, char **argv, const char *buf, off_t len,
fixup_nodes[cur_fixup] = NULL;
+ /* Need to append some command line parameters internally in case of
+ * taking crash dumps.
+ */
+ if (info->kexec_flags & KEXEC_ON_CRASH) {
+ crash_cmdline = xmalloc(COMMAND_LINE_SIZE);
+ memset((void *)crash_cmdline, 0, COMMAND_LINE_SIZE);
+ } else
+ crash_cmdline = NULL;
+
/* Parse the Elf file */
result = build_elf_exec_info(buf, len, &ehdr, 0);
if (result < 0) {
free_elf_info(&ehdr);
return result;
}
+
+#ifdef WITH_GAMECUBE
if (target_is_gamecube) {
gamecube_hack_addresses(&ehdr);
}
+#endif
+
+ /* Load the Elf data. Physical load addresses in elf64 header do not
+ * show up correctly. Use user supplied address for now to patch the
+ * elf header
+ */
+
+ phdr = &ehdr.e_phdr[0];
+ size = phdr->p_filesz;
+ if (size > phdr->p_memsz)
+ size = phdr->p_memsz;
+
+ hole_addr = locate_hole(info, size, 0, 0, max_addr, 1);
+#ifdef CONFIG_PPC64
+ ehdr.e_phdr[0].p_paddr = (Elf64_Addr)hole_addr;
+#else
+ ehdr.e_phdr[0].p_paddr = hole_addr;
+#endif
+
/* Load the Elf data */
result = elf_exec_load(&ehdr, info);
if (result < 0) {
@@ -225,6 +282,27 @@ int elf_ppc_load(int argc, char **argv, const char *buf, off_t len,
return result;
}
+ /* If panic kernel is being loaded, additional segments need
+ * to be created.
+ */
+ if (info->kexec_flags & KEXEC_ON_CRASH) {
+ result = load_crashdump_segments(info, crash_cmdline,
+ max_addr, 0);
+ if (result < 0) {
+ free(crash_cmdline);
+ return -1;
+ }
+ }
+
+ cmdline_buf = xmalloc(COMMAND_LINE_SIZE);
+ memset((void *)cmdline_buf, 0, COMMAND_LINE_SIZE);
+ if (command_line)
+ strncat(cmdline_buf, command_line, command_line_len);
+ if (crash_cmdline)
+ strncat(cmdline_buf, crash_cmdline,
+ sizeof(crash_cmdline) -
+ strlen(crash_cmdline) - 1);
+
/*
* In case of a toy we take the hardcoded things and an easy setup via
* one of the assembly startups. Every thing else should be grown up
@@ -269,31 +347,105 @@ int elf_ppc_load(int argc, char **argv, const char *buf, off_t len,
blob_buf = slurp_file(dtb, &blob_size);
if (!blob_buf || !blob_size)
die("Device tree seems to be an empty file.\n");
- blob_buf = fixup_dtb_nodes(blob_buf, &blob_size, fixup_nodes, command_line);
+ blob_buf = fixup_dtb_nodes(blob_buf, &blob_size, fixup_nodes,
+ cmdline_buf);
dtb_addr = add_buffer(info, blob_buf, blob_size, blob_size, 0, 0,
KERNEL_ACCESS_TOP, -1);
} else {
- dtb_addr = 0;
+ /* create from fs2dt */
+ seg_buf = NULL;
+ seg_size = 0;
+ create_flatten_tree(info, (unsigned char **)&seg_buf,
+ (unsigned long *)&seg_size, cmdline_buf);
+ add_buffer(info, seg_buf, seg_size, seg_size,
+#ifdef CONFIG_PPC64
+ 0, 0, max_addr, -1);
+#else
+ /* load dev tree at 16 Mb offset from kernel load address */
+ 0, 0, ehdr.e_phdr[0].p_paddr + SIZE_16M, -1);
+#endif
}
- /* set various variables for the purgatory */
- addr = ehdr.e_entry;
- elf_rel_set_symbol(&info->rhdr, "kernel", &addr, sizeof(addr));
- addr = dtb_addr;
- elf_rel_set_symbol(&info->rhdr, "dt_offset", &addr, sizeof(addr));
+ if (dtb) {
+ /* set various variables for the purgatory */
+ addr = ehdr.e_entry;
+ elf_rel_set_symbol(&info->rhdr, "kernel", &addr, sizeof(addr));
+
+ addr = dtb_addr;
+ elf_rel_set_symbol(&info->rhdr, "dt_offset",
+ &addr, sizeof(addr));
- addr = rmo_top;
- elf_rel_set_symbol(&info->rhdr, "mem_size", &addr, sizeof(addr));
+ addr = rmo_top;
+
+ elf_rel_set_symbol(&info->rhdr, "mem_size",
+ &addr, sizeof(addr));
#define PUL_STACK_SIZE (16 * 1024)
- addr = locate_hole(info, PUL_STACK_SIZE, 0, 0, elf_max_addr(&ehdr), 1);
- addr += PUL_STACK_SIZE;
- elf_rel_set_symbol(&info->rhdr, "pul_stack", &addr, sizeof(addr));
+ addr = locate_hole(info, PUL_STACK_SIZE, 0, 0,
+ elf_max_addr(&ehdr), 1);
+ addr += PUL_STACK_SIZE;
+ elf_rel_set_symbol(&info->rhdr, "stack", &addr, sizeof(addr));
#undef PUL_STACK_SIZE
- addr = elf_rel_get_addr(&info->rhdr, "purgatory_start");
- info->entry = (void *)addr;
+ addr = elf_rel_get_addr(&info->rhdr, "purgatory_start");
+ info->entry = (void *)addr;
+
+ } else { /*from fs2dt*/
+
+ /* patch reserve map address for flattened device-tree
+ * find last entry (both 0) in the reserve mem list. Assume DT
+ * entry is before this one
+ */
+ bb_ptr = (struct bootblock *)(
+ (unsigned char *)info->segment[(info->nr_segments) -
+ 1].buf);
+ rsvmap_ptr = (unsigned long long *)(
+ (unsigned char *)info->segment[(info->nr_segments) -
+ 1].buf + bb_ptr->off_mem_rsvmap);
+ while (*rsvmap_ptr || *(rsvmap_ptr + 1))
+ rsvmap_ptr += 2;
+ rsvmap_ptr -= 2;
+ *rsvmap_ptr = (unsigned long)(
+ info->segment[(info->nr_segments)-1].mem);
+ rsvmap_ptr++;
+ *rsvmap_ptr = (unsigned long long)bb_ptr->totalsize;
+
+ nr_segments = info->nr_segments;
+
+ /* Set kernel */
+ my_kernel = (unsigned long)info->segment[0].mem;
+ elf_rel_set_symbol(&info->rhdr, "kernel", &my_kernel,
+ sizeof(my_kernel));
+
+ /* Set dt_offset */
+ my_dt_offset = (unsigned long)info->segment[nr_segments -
+ 1].mem;
+ elf_rel_set_symbol(&info->rhdr, "dt_offset", &my_dt_offset,
+ sizeof(my_dt_offset));
+
+ /* get slave code from new kernel, put in purgatory */
+ elf_rel_get_symbol(&info->rhdr, "purgatory_start", slave_code,
+ sizeof(slave_code));
+ master_entry = slave_code[0];
+ memcpy(slave_code, info->segment[0].buf, sizeof(slave_code));
+ slave_code[0] = master_entry;
+ elf_rel_set_symbol(&info->rhdr, "purgatory_start", slave_code,
+ sizeof(slave_code));
+
+ /* Set stack address */
+ my_stack = locate_hole(info, 16*1024, 0, 0, max_addr, 1);
+ my_stack += 16*1024;
+ elf_rel_set_symbol(&info->rhdr, "stack", &my_stack,
+ sizeof(my_stack));
+ }
+
+ if (info->kexec_flags & KEXEC_ON_CRASH) {
+ /* Set backup address */
+ my_backup_start = info->backup_start;
+ elf_rel_set_symbol(&info->rhdr, "backup_start",
+ &my_backup_start, sizeof(my_backup_start));
+ }
#endif
return 0;
}
diff --git a/kexec/arch/ppc/kexec-ppc.c b/kexec/arch/ppc/kexec-ppc.c
index f552d79..255d541 100644
--- a/kexec/arch/ppc/kexec-ppc.c
+++ b/kexec/arch/ppc/kexec-ppc.c
@@ -21,6 +21,7 @@
#include "../../kexec.h"
#include "../../kexec-syscall.h"
#include "kexec-ppc.h"
+#include "crashdump-powerpc.h"
#include <arch/options.h>
#include "config.h"
@@ -45,13 +46,14 @@ static int get_memory_ranges_gc(struct memory_range **range, int *ranges,
}
#else
static int use_new_dtb;
-static int max_memory_ranges;
+int max_memory_ranges;
static int nr_memory_ranges, nr_exclude_ranges;
static struct memory_range *exclude_range;
static struct memory_range *memory_range;
static struct memory_range *base_memory_range;
static uint64_t memory_max;
uint64_t rmo_top;
+unsigned long long crash_base, crash_size;
unsigned int rtas_base, rtas_size;
/*
@@ -174,6 +176,40 @@ static int sort_base_ranges(void)
#define MAXBYTES 128
+static int realloc_memory_ranges(void)
+{
+ size_t memory_range_len;
+
+ max_memory_ranges++;
+ memory_range_len = sizeof(struct memory_range) * max_memory_ranges;
+
+ memory_range = (struct memory_range *) malloc(memory_range_len);
+ if (!memory_range)
+ goto err;
+
+ base_memory_range = (struct memory_range *) realloc(memory_range,
+ memory_range_len);
+ if (!base_memory_range)
+ goto err;
+
+ exclude_range = (struct memory_range *) realloc(exclude_range,
+ memory_range_len);
+ if (!exclude_range)
+ goto err;
+
+ usablemem_rgns.ranges = (struct memory_range *)
+ realloc(usablemem_rgns.ranges,
+ memory_range_len);
+ if (!(usablemem_rgns.ranges))
+ goto err;
+
+ return 0;
+
+err:
+ fprintf(stderr, "memory range structure re-allocation failure\n");
+ return -1;
+}
+
/* Get base memory ranges */
static int get_base_ranges(void)
{
@@ -219,8 +255,10 @@ static int get_base_ranges(void)
return -1;
}
if (local_memory_ranges >= max_memory_ranges) {
- fclose(file);
- break;
+ if (realloc_memory_ranges() < 0){
+ fclose(file);
+ break;
+ }
}
base_memory_range[local_memory_ranges].start =
((uint32_t *)buf)[0];
@@ -250,16 +288,23 @@ static int get_base_ranges(void)
/* Get devtree details and create exclude_range array
* Also create usablemem_ranges for KEXEC_ON_CRASH
*/
-static int get_devtree_details(unsigned long UNUSED(kexec_flags))
+static int get_devtree_details(unsigned long kexec_flags)
{
uint64_t rmo_base;
- char buf[MAXBYTES];
+ unsigned long long tce_base;
+ unsigned int tce_size;
+ unsigned long long htab_base, htab_size;
+ unsigned long long kernel_end;
+ unsigned long long initrd_start, initrd_end;
+ char buf[MAXBYTES-1];
char device_tree[256] = "/proc/device-tree/";
char fname[256];
DIR *dir, *cdir;
FILE *file;
struct dirent *dentry;
+ struct stat fstat;
int n, i = 0;
+ unsigned long tmp_long;
if ((dir = opendir(device_tree)) == NULL) {
perror(device_tree);
@@ -269,10 +314,10 @@ static int get_devtree_details(unsigned long UNUSED(kexec_flags))
while ((dentry = readdir(dir)) != NULL) {
if (strncmp(dentry->d_name, "chosen", 6) &&
strncmp(dentry->d_name, "memory@", 7) &&
- strcmp(dentry->d_name, "memory") &&
+ strncmp(dentry->d_name, "memory", 6) &&
+ strncmp(dentry->d_name, "pci@", 4) &&
strncmp(dentry->d_name, "rtas", 4))
continue;
-
strcpy(fname, device_tree);
strcat(fname, dentry->d_name);
if ((cdir = opendir(fname)) == NULL) {
@@ -280,13 +325,172 @@ static int get_devtree_details(unsigned long UNUSED(kexec_flags))
goto error_opendir;
}
+ if (strncmp(dentry->d_name, "chosen", 6) == 0) {
+ strcat(fname, "/linux,kernel-end");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ goto error_opencdir;
+ }
+ if (fread(&tmp_long, sizeof(unsigned long), 1, file)
+ != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ kernel_end = tmp_long;
+ fclose(file);
+
+ /* Add kernel memory to exclude_range */
+ exclude_range[i].start = 0x0UL;
+ exclude_range[i].end = kernel_end;
+ i++;
+ if (i >= max_memory_ranges)
+ realloc_memory_ranges();
+ if (kexec_flags & KEXEC_ON_CRASH) {
+ memset(fname, 0, sizeof(fname));
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ strcat(fname, "/linux,crashkernel-base");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ goto error_opencdir;
+ }
+ if (fread(&tmp_long, sizeof(unsigned long), 1,
+ file) != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ crash_base = tmp_long;
+ fclose(file);
+
+ memset(fname, 0, sizeof(fname));
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ strcat(fname, "/linux,crashkernel-size");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ goto error_opencdir;
+ }
+ if (fread(&tmp_long, sizeof(unsigned long), 1,
+ file) != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ crash_size = tmp_long;
+
+ if (crash_base > mem_min)
+ mem_min = crash_base;
+ if (crash_base + crash_size < mem_max)
+ mem_max = crash_base + crash_size;
+
+ add_usable_mem_rgns(0, crash_base + crash_size);
+ reserve(KDUMP_BACKUP_LIMIT,
+ crash_base-KDUMP_BACKUP_LIMIT);
+ }
+ memset(fname, 0, sizeof(fname));
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ strcat(fname, "/linux,htab-base");
+ file = fopen(fname, "r");
+ if (!file) {
+ closedir(cdir);
+ if (errno == ENOENT) {
+ /* Non LPAR */
+ errno = 0;
+ continue;
+ }
+ perror(fname);
+ goto error_opendir;
+ }
+ if (fread(&htab_base, sizeof(unsigned long), 1, file)
+ != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ memset(fname, 0, sizeof(fname));
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ strcat(fname, "/linux,htab-size");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ goto error_opencdir;
+ }
+ if (fread(&htab_size, sizeof(unsigned long), 1, file)
+ != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ /* Add htab address to exclude_range - NON-LPAR only */
+ exclude_range[i].start = htab_base;
+ exclude_range[i].end = htab_base + htab_size;
+ i++;
+ if (i >= max_memory_ranges)
+ realloc_memory_ranges();
+
+ /* reserve the initrd_start and end locations. */
+ if (reuse_initrd) {
+ memset(fname, 0, sizeof(fname));
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ strcat(fname, "/linux,initrd-start");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ goto error_opencdir;
+ }
+ /* check for 4 and 8 byte initrd offset sizes */
+ if (stat(fname, &fstat) != 0) {
+ perror(fname);
+ goto error_openfile;
+ }
+ if (fread(&initrd_start, fstat.st_size, 1, file)
+ != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ fclose(file);
+
+ memset(fname, 0, sizeof(fname));
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ strcat(fname, "/linux,initrd-end");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ goto error_opencdir;
+ }
+ /* check for 4 and 8 byte initrd offset sizes */
+ if (stat(fname, &fstat) != 0) {
+ perror(fname);
+ goto error_openfile;
+ }
+ if (fread(&initrd_end, fstat.st_size, 1, file)
+ != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ fclose(file);
+
+ /* Add initrd address to exclude_range */
+ exclude_range[i].start = initrd_start;
+ exclude_range[i].end = initrd_end;
+ i++;
+ if (i >= max_memory_ranges)
+ realloc_memory_ranges();
+ }
+ } /* chosen */
+
if (strncmp(dentry->d_name, "rtas", 4) == 0) {
strcat(fname, "/linux,rtas-base");
if ((file = fopen(fname, "r")) == NULL) {
perror(fname);
goto error_opencdir;
}
- if (fread(&rtas_base, sizeof(unsigned int), 1, file) != 1) {
+ if (fread(&rtas_base, sizeof(unsigned int), 1, file)
+ != 1) {
perror(fname);
goto error_openfile;
}
@@ -298,7 +502,8 @@ static int get_devtree_details(unsigned long UNUSED(kexec_flags))
perror(fname);
goto error_opencdir;
}
- if (fread(&rtas_size, sizeof(unsigned int), 1, file) != 1) {
+ if (fread(&rtas_size, sizeof(unsigned int), 1, file)
+ != 1) {
perror(fname);
goto error_openfile;
}
@@ -307,6 +512,8 @@ static int get_devtree_details(unsigned long UNUSED(kexec_flags))
exclude_range[i].start = rtas_base;
exclude_range[i].end = rtas_base + rtas_size;
i++;
+ if (kexec_flags & KEXEC_ON_CRASH)
+ add_usable_mem_rgns(rtas_base, rtas_size);
} /* rtas */
if (!strncmp(dentry->d_name, "memory@", 7) ||
@@ -336,6 +543,48 @@ static int get_devtree_details(unsigned long UNUSED(kexec_flags))
fclose(file);
closedir(cdir);
} /* memory */
+
+ if (strncmp(dentry->d_name, "pci@", 4) == 0) {
+ strcat(fname, "/linux,tce-base");
+ file = fopen(fname, "r");
+ if (!file) {
+ closedir(cdir);
+ if (errno == ENOENT) {
+ /* Non LPAR */
+ errno = 0;
+ continue;
+ }
+ perror(fname);
+ goto error_opendir;
+ }
+ if (fread(&tce_base, sizeof(unsigned long), 1, file)
+ != 1) {
+ perror(fname);
+ goto error_openfile;
+ return -1;
+ }
+ memset(fname, 0, sizeof(fname));
+ strcpy(fname, device_tree);
+ strcat(fname, dentry->d_name);
+ strcat(fname, "/linux,tce-size");
+ file = fopen(fname, "r");
+ if (!file) {
+ perror(fname);
+ goto error_opencdir;
+ }
+ if (fread(&tce_size, sizeof(unsigned int), 1, file)
+ != 1) {
+ perror(fname);
+ goto error_openfile;
+ }
+ /* Add tce to exclude_range - NON-LPAR only */
+ exclude_range[i].start = tce_base;
+ exclude_range[i].end = tce_base + tce_size;
+ i++;
+ if (kexec_flags & KEXEC_ON_CRASH)
+ add_usable_mem_rgns(tce_base, tce_size);
+ closedir(cdir);
+ } /* pci */
}
closedir(dir);
@@ -347,8 +596,8 @@ static int get_devtree_details(unsigned long UNUSED(kexec_flags))
int k;
for (k = 0; k < i; k++)
fprintf(stderr, "exclude_range sorted exclude_range[%d] "
- "start:%llx, end:%llx\n", k, exclude_range[k].start,
- exclude_range[k].end);
+ "start:%llx, end:%llx\n", k, exclude_range[k].start,
+ exclude_range[k].end);
#endif
return 0;
@@ -532,7 +781,3 @@ void arch_update_purgatory(struct kexec_info *UNUSED(info))
{
}
-int is_crashkernel_mem_reserved(void)
-{
- return 0; /* kdump is not supported on this platform (yet) */
-}
diff --git a/kexec/arch/ppc/kexec-ppc.h b/kexec/arch/ppc/kexec-ppc.h
index 6cec467..fc0471f 100644
--- a/kexec/arch/ppc/kexec-ppc.h
+++ b/kexec/arch/ppc/kexec-ppc.h
@@ -1,6 +1,11 @@
#ifndef KEXEC_PPC_H
#define KEXEC_PPC_H
+#define MAXBYTES 128
+#define MAX_LINE 160
+#define CORE_TYPE_ELF32 1
+#define CORE_TYPE_ELF64 2
+
extern unsigned char setup_simple_start[];
extern uint32_t setup_simple_size;
@@ -16,6 +21,8 @@ extern struct {
uint32_t spr8;
} setup_dol_regs;
+#define SIZE_16M (16*1024*1024UL)
+
int elf_ppc_probe(const char *buf, off_t len);
int elf_ppc_load(int argc, char **argv, const char *buf, off_t len,
struct kexec_info *info);
@@ -37,4 +44,29 @@ void dol_ppc_usage(void);
*/
#define KERNEL_ACCESS_TOP (24 * 1024 * 1024)
+/* boot block version 17 as defined by the linux kernel */
+struct bootblock {
+ unsigned magic,
+ totalsize,
+ off_dt_struct,
+ off_dt_strings,
+ off_mem_rsvmap,
+ version,
+ last_comp_version,
+ boot_physid,
+ dt_strings_size,
+ dt_struct_size;
+};
+
+typedef struct mem_rgns {
+ unsigned int size;
+ struct memory_range *ranges;
+} mem_rgns_t;
+extern mem_rgns_t usablemem_rgns;
+extern int max_memory_ranges;
+extern unsigned long long initrd_base, initrd_size;
+extern unsigned char reuse_initrd;
+#define COMMAND_LINE_SIZE 512 /* from kernel */
+/*fs2dt*/
+void reserve(unsigned long long where, unsigned long long length);
#endif /* KEXEC_PPC_H */
diff --git a/purgatory/arch/ppc/Makefile b/purgatory/arch/ppc/Makefile
index 0dd18b6..72289a0 100644
--- a/purgatory/arch/ppc/Makefile
+++ b/purgatory/arch/ppc/Makefile
@@ -2,7 +2,7 @@
# Purgatory ppc
#
-ppc_PURGATORY_SRCS += purgatory/arch/ppc/v2wrap.S
+ppc_PURGATORY_SRCS += purgatory/arch/ppc/v2wrap_32.S
ppc_PURGATORY_SRCS += purgatory/arch/ppc/misc.S
ppc_PURGATORY_SRCS += purgatory/arch/ppc/purgatory-ppc.c
ppc_PURGATORY_SRCS += purgatory/arch/ppc/console-ppc.c
diff --git a/purgatory/arch/ppc/purgatory-ppc.c b/purgatory/arch/ppc/purgatory-ppc.c
index 01d0f38..3d7d484 100644
--- a/purgatory/arch/ppc/purgatory-ppc.c
+++ b/purgatory/arch/ppc/purgatory-ppc.c
@@ -1,19 +1,41 @@
+/*
+ * kexec: Linux boots Linux
+ *
+ * Created by: Mohan Kumar M (mohan@in.ibm.com)
+ *
+ * Copyright (C) IBM Corporation, 2005. All rights reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation (version 2 of the License).
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
#include <purgatory.h>
#include "purgatory-ppc.h"
-unsigned int pul_stack = 0;
-unsigned int dt_offset = 0;
-unsigned int kernel = 0;
-unsigned int epapr_magic = 0;
-unsigned int mem_size = 0;
+unsigned int panic_kernel = 0;
+unsigned long backup_start = 0;
+unsigned long stack = 0;
+unsigned long dt_offset = 0;
+unsigned long my_toc = 0;
+unsigned long kernel = 0;
void setup_arch(void)
{
- /* Nothing for now */
+ return;
}
-/* This function can be used to execute after the SHA256 verification. */
void post_verification_setup_arch(void)
{
- /* Nothing for now */
+ if (panic_kernel)
+ crashdump_backup_memory();
}
diff --git a/purgatory/arch/ppc/purgatory-ppc.h b/purgatory/arch/ppc/purgatory-ppc.h
index e931cae..7eff8aa 100644
--- a/purgatory/arch/ppc/purgatory-ppc.h
+++ b/purgatory/arch/ppc/purgatory-ppc.h
@@ -1,6 +1,6 @@
#ifndef PURGATORY_PPC_H
#define PURGATORY_PPC_H
-/* nothing yet */
-
+void crashdump_backup_memory(void);
+void post_verification_setup_arch(void);
#endif /* PURGATORY_PPC_H */
diff --git a/purgatory/arch/ppc/v2wrap.S b/purgatory/arch/ppc/v2wrap.S
deleted file mode 100644
index 79d188f..0000000
--- a/purgatory/arch/ppc/v2wrap.S
+++ /dev/null
@@ -1,66 +0,0 @@
-#
-# kexec: Linux boots Linux
-#
-# Copyright (C) 2004 - 2005, Milton D Miller II, IBM Corporation
-# Copyright (C) 2006, Mohan Kumar M (mohan@in.ibm.com), IBM Corporation
-# Copyright (C) 2008, Sebastian Andrzej Siewior (bigeasy@linutronix.de), linutronix
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation (version 2 of the License).
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-#
-
-#include "ppc_asm.h"
-
-# v2wrap.S
-# a wrapper to call purgatory code
-# Invokes ppc kernel with the arguments according to ePAPR v1.0
-# It assumes that the MSR is allready correct.
-
-# calling convention:
-# no register are considred
-#
-
-#define LOADADDR(rn,name) \
- lis rn,name##@h; \
- ori rn,rn,name##@l; \
-
- .globl purgatory_start
-purgatory_start:
-
- LOADADDR(r6,pul_stack)
- lwz r1,0(r6) #setup stack
-
- subi r1, r1, 112
- bl purgatory
- nop
-
- LOADADDR(r6,kernel)
- lwz r4,0(r6) # load the kernel address
- mtlr r4 # prepare branch too
-
- LOADADDR(r6, dt_offset)
- lwz r3, 0(r6) # load device-tree address
-
- li r4, 0
- li r5, 0
-
- LOADADDR(r6, epapr_magic) # ePAPR magic value
- lwz r6, 0(r6)
-
- LOADADDR(r7, mem_size) # the Initial Mapped Area
- lwz r7, 0(r6)
-
- li r8, 0
- li r9, 0
-
- blr # start kernel
diff --git a/purgatory/arch/ppc/v2wrap_32.S b/purgatory/arch/ppc/v2wrap_32.S
new file mode 100644
index 0000000..8442d16
--- /dev/null
+++ b/purgatory/arch/ppc/v2wrap_32.S
@@ -0,0 +1,91 @@
+#
+# kexec: Linux boots Linux
+#
+# Copyright (C) 2004 - 2005, Milton D Miller II, IBM Corporation
+# Copyright (C) 2006, Mohan Kumar M (mohan@in.ibm.com), IBM Corporation
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation (version 2 of the License).
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+#
+
+# v2wrap.S
+# a wrapper to call purgatory code to backup first
+# 32kB of first kernel into the backup region
+# reserved by kexec-tools.
+# Invokes powerpc kernel with the expected arguments
+# of kernel(device-tree, phys-offset, 0)
+
+#
+# calling convention:
+# r3 = physical number of this cpu (all cpus)
+# r4 = address of this chunk (master only)
+# master enters at purgatory_start (aka first byte of this chunk)
+# slaves (additional cpus), if any, enter a copy of the
+# first 0x100 bytes of this code relocated to 0x0
+#
+# in other words,
+# a copy of the first 0x100 bytes of this code is copied to 0
+# and the slaves are sent to address 0x60
+# with r3 = their physical cpu number.
+
+ .globl purgatory_start
+purgatory_start: b master
+ .org purgatory_start + 0x60 # ABI: slaves start at 60 with r3=phys
+slave: b $
+ .org purgatory_start + 0x100 # ABI: end of copied region
+ .size purgatory_start, . - purgatory_start
+
+#
+# The above 0x100 bytes at purgatory_start are replaced with the
+# code from the kernel (or next stage) by kexec/arch/powerpc/kexec-powerpc.c
+#
+
+master:
+ or 1,1,1 # low priority to let other threads catchup
+ isync
+ mr 17,3 # save cpu id to r17
+ mr 15,4 # save physical address in reg15
+
+ lis 6,stack@h
+ ori 6,6,stack@l
+ lwz 1,0(6) #setup stack
+
+ subi 1,1,112
+ bl purgatory
+ nop
+
+ or 3,3,3 # ok now to high priority, lets boot
+ lis 6,0x1
+ mtctr 6 # delay a bit for slaves to catch up
+83: bdnz 83b # before we overwrite 0-100 again
+
+ lis 6,dt_offset@h
+ ori 6,6,dt_offset@l
+ lwz 3,0(6) # load device-tree address
+ lwz 6,20(3) # fetch version number
+ cmpwi 0,6,2 # v2 ?
+ blt 80f
+ stw 17,28(3) # save my cpu number as boot_cpu_phys
+80:
+ lis 6,kernel@h
+ ori 6,6,kernel@l
+ lwz 4,0(6) # load the kernel address
+ li 5,0 # r5 will be 0 for kernel
+ li 6,0 # clear r6 for good measure
+ mtctr 4 # prepare branch too
+
+ lwz 8,0(4) # get the first instruction that we stole
+ stw 8,0(0) # and put it in the slave loop at 0
+ # skip cache flush, do we care?
+
+ bctr # start kernel
^ permalink raw reply related
* Re: [PATCH 1/2] Fix kexec on powerpc32
From: Christian Kujau @ 2010-05-12 6:56 UTC (permalink / raw)
To: Maxim Uvarov; +Cc: kexec, linuxppc-dev
In-Reply-To: <AANLkTimSATEKI4TdO6zNWSRzl2on4OL5D2GGBlVsndMs@mail.gmail.com>
On Wed, 12 May 2010 at 10:31, Maxim Uvarov wrote:
> Yes, that is expected result for old tool chains. New tool chains (from code
> sourcery ) need this crt.S.
> I think we can have --with-oldtoolchain option to configure to point to old
> tool chains. (or even rename it to --with-crts,
> I don't know what is better.)
Yes, I've seen this --with-oldtoolchain flag in ./configure, but: when
exactly is my toolchain old enough so that this flag has to be used? Is my
binutils 2.18 too old?
So, I'm all in favour of --with-crts (which I would use when compilation breaks
with some descriptive error message) or, even better: can't we set this
flag during ./configure automatically, based on the actual toolchain installed
on the system? ./configure should be able to determine this, no?
> This -Werror came from mips. kexec/arch/mips/Makefile. It is already removed
> in the git. We already saw this problem.
Powerpc including a mops Makefile, whooha :-)
Thanks,
Christian.
--
BOFH excuse #91:
Mouse chewed through power cable
^ permalink raw reply
* Re: [Patch 2/2] PPC64-HWBKPT: Implement hw-breakpoints for PPC64
From: Paul Mackerras @ 2010-05-12 6:32 UTC (permalink / raw)
To: K.Prasad
Cc: Michael Neuling, Benjamin Herrenschmidt, shaggy,
Frederic Weisbecker, David Gibson, linuxppc-dev@ozlabs.org,
Alan Stern, Roland McGrath
In-Reply-To: <20100504203302.GA3894@in.ibm.com>
On Wed, May 05, 2010 at 02:03:03AM +0530, K.Prasad wrote:
> It is true that the breakpoint exceptions will go amiss following the
> alignment exception, and be restored when the thread single-steps due
> to other requests causing undesirable effects. (Borrowing from some of
> the discussions I had with BenH, earlier) There can be two ways of
> changing the implementation to counter it:
>
> - To sense that the impending exception (alignment, page-fault,
> single-step) is a successor to a hw-breakpoint exception (and that
> restoration of debug register values is necessary), somewhere early in
> exceptions-64s.S and jump to a common handler, say
> do_single_step_dabr() which does a majority of
> single_step_dabr_instruction().
> - To modify emulate_single_step() to also do a notify_die(DIE_SSTEP,...)
> in addition to its existing code. This would invoke
> single_step_dabr_instruction() where the breakpoints can be restored.
I thought you would change the explicit regs->msr modification in
single_step_exception() to clear_single_step(), then just make
emulate_single_step() call single_step_exception().
> I must admit that it is not clear to me when you say "doesn't return to
> the instruction" and "instruction has been changed underneath". Are you
Suppose the address at which the data breakpoint has been unmapped,
and the process has a handler for the SIGSEGV signal. When we try to
single-step the load or store, we will get a DSI (0x300) interrupt,
call into do_page_fault, and end up sending the process a SIGSEGV.
That will invoke the signal handler, which can then do anything it
likes. It can do a blocking system call, it can longjmp() back into
its main event, or it can return from the signal handler. Only in the
last case will it retry the load or store, and then only if the signal
handler hasn't modified the NIP value in the signal frame. That's
what I mean by "doesn't return to the instruction".
The instruction could be changed underneath us if the program is
multi-threaded and another thread writes another instruction to the
instruction word where the load or store is. Or it could use mmap()
to map some other page at the address of the load or store. Either
way we could end up with a different instruction there.
> referring to the fact that the thread which generated breakpoints hits
> could have moved out of the CPU due to a scheduler induced context
> switch (which is an apparent cause for current->thread.last_hit_ubp to
> turn stale) or is there a different source for such a change that I
> don't realise?
>
> Given that 'last_hit_ubp' is safely ensconced inside 'thread_struct',
> the ill-effects of a possible pre-emption during a page-fault will be
> suitably handled i.e. the pending single-step exception will be
> generated on the processor to which 'current' is migrated to, and the
> breakpoint will be set on the new processor.
If we do get a context switch, e.g. as a result of a page fault, and
then switch back to the task, it looks to me like we will end up with
MSR_SE and DABR both set. I don't suppose that will actually cause
any real problem besides double-counting the hit.
> However, the possibility that current->thread.last_hit_ubp points to a
> perf_event structure that is unregistered and freed does exist, and I
> did not foresee the risk. An arch-specific function that hooks onto
> release_bp_slot() would be required to perform the cleanup. I will
> submit modify the patch to that effect. Thanks for pointing it out.
Yes, I think we need that.
> In conjunction with what you've stated below, do you suggest that
> emulate_step() be replaced with fix_alignment() which appears to be more
> powerful at emulation (or carve out a helper function for fix_alignment()
> that does only emulation and which can be invoked here)?
Something like that eventually, but not for a first pass.
Paul.
^ permalink raw reply
* Re: [PATCH 1/2] Fix kexec on powerpc32
From: Maxim Uvarov @ 2010-05-12 6:31 UTC (permalink / raw)
To: Christian Kujau; +Cc: kexec, linuxppc-dev
In-Reply-To: <4BEA2375.7040802@nerdbynature.de>
[-- Attachment #1: Type: text/plain, Size: 2207 bytes --]
2010/5/12 Christian Kujau <lists@nerdbynature.de>
> Maxim Uvarov wrote on 2010-05-11 10:47 :
> > This patch is required in case if you are using new toolchains.
>
> I'm on Debian/stable (binutils 2.18, gcc 4.3.2), so I applied your patches
> to the latest -git checkout:
>
> First, compilation fails with:
>
> cc1: warnings being treated as errors
> kexec/arch/ppc/kexec-ppc.c: In function ‘get_base_ranges’:
> kexec/arch/ppc/kexec-ppc.c:258: error: value computed is not used
> make: *** [kexec/arch/ppc/kexec-ppc.o] Error 1
>
> Somehow a "-" hyphen made it into one of the functions:
>
> > --- a/kexec/arch/ppc/kexec-ppc.c
> > +++ b/kexec/arch/ppc/kexec-ppc.c
> [...]
> > if (local_memory_ranges >= max_memory_ranges) {
> > - fclose(file);
> > - break;
> > + if (realloc_memory_ranges() < 0){
> > + - fclose(file);
>
> Oh, I sent this patch twice with stg -a by mistake. And second version has
"-". I will fix and resend patches.
>
> Removing this, gets me further, until:
>
>
> purgatory/arch/ppc/crt.o: In function `__lshrdi3':
> (.text+0xf0): multiple definition of `__lshrdi3'
> purgatory/arch/ppc/misc.o:(.text+0x50): first defined here
> collect2: ld returned 1 exit status
> make: *** [purgatory/purgatory.ro] Error 1
>
> And has been defined in purgatory/arch/ppc/crt.S and
> ./purgatory/arch/ppc/misc.S. Removing it from e.g. crt.S
> makes it compile, I have not tested it yet, though :-)
>
Yes, that is expected result for old tool chains. New tool chains (from code
sourcery ) need this crt.S.
I think we can have --with-oldtoolchain option to configure to point to old
tool chains. (or even rename it to --with-crts,
I don't know what is better.)
>
> Thanks,
> Christian.
>
> PS: Where is -Werror defined? I could not find it anywhere in the kexec
> source (execpt in arch/mips)...
> --
>
This -Werror came from mips. kexec/arch/mips/Makefile. It is already removed
in the git. We already saw this problem.
> make bzImage, not war
>
--
Best regards,
Maxim Uvarov
[-- Attachment #2: Type: text/html, Size: 3314 bytes --]
^ permalink raw reply
* Re: [RFC][PATCH RFC 10/12] KVM: move dirty bitmaps to user space
From: Takuya Yoshikawa @ 2010-05-12 6:27 UTC (permalink / raw)
To: Marcelo Tosatti
Cc: linux-arch, arnd, kvm, kvm-ia64, fernando, x86, agraf, kvm-ppc,
linux-kernel, linuxppc-dev, mingo, paulus, avi, hpa, tglx,
Takuya Yoshikawa
In-Reply-To: <20100511032827.GA3458@amt.cnet>
>> r = 0;
>> @@ -1195,11 +1232,16 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn)
>> gfn = unalias_gfn(kvm, gfn);
>> memslot = gfn_to_memslot_unaliased(kvm, gfn);
>> if (memslot&& memslot->dirty_bitmap) {
>> - unsigned long rel_gfn = gfn - memslot->base_gfn;
>> + int nr = generic_le_bit_offset(gfn - memslot->base_gfn);
>>
>> - generic___set_le_bit(rel_gfn, memslot->dirty_bitmap);
>> + if (kvm_set_bit_user(nr, memslot->dirty_bitmap))
>> + goto out_fault;
>
> mark_page_dirty is called with the mmu_lock spinlock held in set_spte.
> Must find a way to move it outside of the spinlock section.
>
Oh, it's a serious problem. I have to consider it.
Thanks,
Takuya
^ permalink raw reply
* Re: [RFC][PATCH 11/12] KVM: introduce new API for getting/switching dirty bitmaps
From: Takuya Yoshikawa @ 2010-05-12 6:03 UTC (permalink / raw)
To: Marcelo Tosatti
Cc: linux-arch, arnd, kvm, kvm-ia64, fernando, x86, agraf, kvm-ppc,
linux-kernel, linuxppc-dev, mingo, paulus, avi, hpa, tglx,
Takuya Yoshikawa
In-Reply-To: <20100511140712.GA7063@amt.cnet>
>>> One alternative would be:
>>>
>>> KVM_SWITCH_DIRTY_LOG passing the address of a bitmap. If the active
>>> bitmap was clean, it returns 0, no switch performed. If the active
>>> bitmap was dirty, the kernel switches to the new bitmap and returns 1.
>>>
>>> And the responsability of cleaning the new bitmap could also be left
>>> for userspace.
>>>
>>
>> That is a beautiful approach but we can do that only when we give up using
>> GET api.
>>
>>
>> I follow you and Avi's advice about that kind of maintenance policy!
>> What do you think?
>
> If you introduce a switch ioctl that frees the bitmap vmalloc'ed by the
> current set_memory_region (if its not freed already), after pointing the
> memslot to the user supplied one, it should be fine?
>
You mean switching from vmalloc'ed(not do_mmap'ed) one to user supplied one?
It may be possible but makes things really complicated in my view:
until some point we use set_bit, and then use set_bit_user, etc.
IMO:
- # of slots is limited and the size of dirty_bitmap_old pointer is not problematic.
- Both user side and kernel side need not allocate buffers every time and once
paired buffers are registered, we will reuse the buffers until user side orders
to stop logging.
- We have a tiny advantage if we need not copy_from_user to get a bitmap address
for switch ioctl.
=> So I think having two __user bitmaps is not a bad thing.
^ permalink raw reply
* 8640D Erratum fixup
From: Thirumalai @ 2010-05-12 4:32 UTC (permalink / raw)
To: Kumar Gala; +Cc: linuxppc-dev
Hi kumar,
I just want to know whether the erratum of MPC8640D has fixed on
u-boot or linux ?
-Thirumalai
**************** CAUTION - Disclaimer *****************This email may contain confidential and privileged material for the
sole use of the intended recipient(s). Any review, use, retention, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message. Also, email is susceptible to data corruption, interception, tampering, unauthorized amendment and viruses. We only send and receive emails on the basis that we are not liable for any such corruption, interception, tampering, amendment or viruses or any consequence thereof.
*********** End of Disclaimer ***********DataPatterns ITS Group**********
^ permalink raw reply
* Re: [PATCH 1/2] Fix kexec on powerpc32
From: Christian Kujau @ 2010-05-12 3:44 UTC (permalink / raw)
To: Maxim Uvarov; +Cc: kexec, uvarov, linuxppc-dev
In-Reply-To: <20100511174749.19309.44615.stgit@muvarov>
Maxim Uvarov wrote on 2010-05-11 10:47 :
> This patch is required in case if you are using new toolchains.
I'm on Debian/stable (binutils 2.18, gcc 4.3.2), so I applied your patches=
=20
to the latest -git checkout:
First, compilation fails with:
cc1: warnings being treated as errors
kexec/arch/ppc/kexec-ppc.c: In function =E2=80=98get_base_ranges=E2=80=99:
kexec/arch/ppc/kexec-ppc.c:258: error: value computed is not used
make: *** [kexec/arch/ppc/kexec-ppc.o] Error 1
Somehow a "-" hyphen made it into one of the functions:
> --- a/kexec/arch/ppc/kexec-ppc.c
> +++ b/kexec/arch/ppc/kexec-ppc.c
[...]
> =09=09=09if (local_memory_ranges >=3D max_memory_ranges) {
> -=09=09=09=09fclose(file);
> -=09=09=09=09break;
> +=09=09=09=09if (realloc_memory_ranges() < 0){
> +=09-=09=09=09=09fclose(file);
Removing this, gets me further, until:
purgatory/arch/ppc/crt.o: In function `__lshrdi3':
(.text+0xf0): multiple definition of `__lshrdi3'
purgatory/arch/ppc/misc.o:(.text+0x50): first defined here
collect2: ld returned 1 exit status
make: *** [purgatory/purgatory.ro] Error 1
And has been defined in purgatory/arch/ppc/crt.S and
=2E/purgatory/arch/ppc/misc.S. Removing it from e.g. crt.S
makes it compile, I have not tested it yet, though :-)
Thanks,
Christian.
PS: Where is -Werror defined? I could not find it anywhere in the kexec=20
source (execpt in arch/mips)...
--=20
make bzImage, not war
^ permalink raw reply
* RE: 440SPe/Katmai PCIe problems
From: Pravin Bathija @ 2010-05-12 3:30 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Stefan Roese; +Cc: linuxppc-dev
In-Reply-To: <1273633366.21352.100.camel@pasglop>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^ permalink raw reply
* Re: [Patch 2/2] PPC64-HWBKPT: Implement hw-breakpoints for PPC64
From: K.Prasad @ 2010-05-12 3:34 UTC (permalink / raw)
To: Paul Mackerras
Cc: Michael Neuling, Benjamin Herrenschmidt, shaggy,
Frederic Weisbecker, David Gibson, linuxppc-dev@ozlabs.org,
Alan Stern, Roland McGrath
In-Reply-To: <20100504203302.GA3894@in.ibm.com>
On Wed, May 05, 2010 at 02:03:03AM +0530, K.Prasad wrote:
> On Mon, May 03, 2010 at 04:23:30PM +1000, Paul Mackerras wrote:
> > On Wed, Apr 14, 2010 at 09:18:27AM +0530, K.Prasad wrote:
> >
[snipped]
>
> It has been pointed out to me before (Roland's mail Ref:linuxppc-dev
> message-id: 20100119100335.3EB621DE@magilla.sf.frob.com) that there will
> be too many corner cases that will be difficult to foresee, however your
> above list appears to be exhaustive. While the alternatives to this being
> a fallback to one-shot breakpoints (thereby leading to confusing
> hw-breakpoint interface semantics), this is an attempt to generate
> continuous and 'trigger-after-execute' (for non-ptrace requests)
> breakpoint exceptions. I believe that, with the addressal of concerns
> cited above, the resultant patchset would be one that achieves the
> stated design goals with no loss to existing functionality.
>
Hi Paul,
> I intend to send out another version of this patchset with fixes as
> described in my replies above (unless I hear objections to it :-)).
>
Meanwhile, a little sickness had kept me away from working on this
patchset. I have now posted a new version of the same here () which
contains changes as described above.
A few more changes to the patch is impending post merger
of Frederic's patches (which are now in -tip) into mainline (ref: commit
73266fc1df2f94cf72b3beba3eee3b88ed0b0664 to
777d0411cd1e384115985dac5ccd42031e3eee2b);
mainly due to the new ability for a per-task breakpoint to request
kernel-space breakpoints (the notion of kernel- vs user-bp would also
become obsolete, it is better to call them per-cpu vs per-task
breakpoints).
Also, I find that possibility of a kernel-thread specific breakpoint
(which can migrate across CPUs) has not been thought and implemented well
in this patch (will be much easier after merger of Frederic's patch). I
would prefer to have atleast some version of the patch included in
mainline before bringing in support for the same.
Thanks,
K.Prasad
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox