- 24 Apr, 2008 40 commits
-
-
Ishizaki Kou authored
This moves miscellaneous files for Beat into platforms/cell/. All files in this patch are used by celleb-beat only. Signed-off-by:
Kou Ishizaki <kou.ishizaki@toshiba.co.jp> Acked-by:
Arnd Bergmann <arnd@arndb.de> Acked-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
Paul Mackerras <paulus@samba.org>
-
Ishizaki Kou authored
This moves SPU support code on Beat into platforms/cell/. Signed-off-by:
-
Ishizaki Kou authored
This moves files for mmu and iommu on Beat into platforms/cell/. All files in this patch are used by celleb-beat only. Signed-off-by:
-
Ishizaki Kou authored
This moves files for Beat hvcall interfaces into platforms/cell/. All files in this patch are used by celleb-beat only. Signed-off-by:
-
Ishizaki Kou authored
This moves the SCC (Super Companion Chip) related code for celleb into platforms/cell/. All files in this patch are used by celleb-beat and celleb-native commonly. Signed-off-by:
-
Ishizaki Kou authored
This moves the base code for celleb support into platforms/cell/. All files in this patch are used by celleb-beat and celleb-native commonly. Signed-off-by:
-
Ishizaki Kou authored
Now, we can use generic io-workarounds mechanism and the workaround code for spider-pci. This changes Celleb PCI code to use spider-pci code. Signed-off-by:
-
Ishizaki Kou authored
This splits cell io-workaround code into spider-pci dependent code and a generic part, and also moves io-workarounds initialization into cell_setup_phb. Signed-off-by:
-
Michael Ellerman authored
Add a DEBUG config setting which turns on all (most) of the debugging under platforms/pseries. To have this take effect we need to remove all the #undef DEBUG's, in various files. We leave the #undef DEBUG in platforms/pseries/lpar.c, as this enables debugging printks from the low-level hash table routines, and tends to make your system unusable. If you want those enabled you still have to turn them on by hand. Also some of the RAS code has a DEBUG block which causes a functional change, so I've keyed this off a different (non-existant) debug #define. This is only enabled if you have PPC_EARLY_DEBUG enabled also. Signed-off-by:
Michael Ellerman <michael@ellerman.id.au> Signed-off-by:
-
Michael Ellerman authored
In pseries/lpar.c, fix some printf specifier mismatches, and add a newline to one printk. In pseries/rtasd.c add "rtasd" to some messages to make it clear where they're coming from. In pseries/scanlog.c remove the hand-rolled runtime debugging support in there. This file has been largely unchanged for eons, if we need to debug it in future we can recompile. Signed-off-by:
-
Michael Ellerman authored
On pseries LPAR we can call the udbg routines, and the udbg console very early. So mark the udbg console as safe to call early in boot, and register the udbg console as soon as the udbg routines are hooked up. This allows platforms/pseries code to use printk() and pr_debug() rather than needing to call udbg_printf() directly for early debugging. This is nice because a) it's standard, b) it goes via the printk buffer, and c) you can get printk time stamps. Signed-off-by:
-
Michael Ellerman authored
The udbg console should be safe to call basically at any time after boot. It does not need any per-cpu resources or for the cpu to be online, as long as there is a udbg_putc routine hooked up it should work. So mark it as CON_ANYTIME. Signed-off-by:
-
Michael Ellerman authored
Because the udbg_console has CON_ENABLED set, it's possible that when we register it with the console code the index won't be set. This leads to slightly confusing boot messages like: [ 0.000000] console [udbg-1] enabled We could remove CON_ENABLED, but we don't want to do that, we always want the udbg console to be activated, even if the user specified some other console on the command line. The simplest fix seems to be just to set the index to 0 by hand. There is no issue with duplicate udbg consoles, as we guard against registering multiple times in register_early_udbg_console(). Signed-off-by:
-
Tony Breeds authored
This adds the required functionality to fill in all pacas at runtime. With NR_CPUS=1024 text data bss dec hex filename 137 1704032 0 1704169 1a00e9 arch/powerpc/kernel/paca.o :Before 121 1179744 524288 1704153 1a00d9 arch/powerpc/kernel/paca.o :After Also remove unneeded #includes from arch/powerpc/kernel/paca.c Signed-off-by:
Tony Breeds <tony@bakeyournoodle.com> Signed-off-by:
-
Tony Breeds authored
Currently all iSeries secondary CPUs spin directly on the cpu_start field in their paca. Make them spin on the global __secondary_hold_spinloop until after the pacas have been initialised. As Stephen Rothwell points out, this works at the moment because __secondary_hold_spinloop is being set already, but iSeries isn't looking at it :) Signed-off-by:
Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
-
Kumar Gala authored
* Removed get_msr(), get_srr0(), and get_srr1() - not used anywhere * Use STACK_FRAME_OVERHEAD instead of magic number Signed-off-by:
Kumar Gala <galak@kernel.crashing.org> Signed-off-by:
-
Michael Ellerman authored
Replace two open-coded occurences of the of_get_next_parent() logic. Signed-off-by:
-
Michael Ellerman authored
As BenH said the other day, it is an "accident" that prom_init.o is linked with the rest of the kernel. The truth is a little more subtle, prom_init isn't truly bootloader, it does access kernel data in a few places. What we can do is discourage people from adding new code that accesses data outside of prom_init. And hence this patch; from the script: # This script checks prom_init.o to see what external symbols it # is using, if it finds symbols not in the whitelist it returns # an error. The point of this is to discourage people from # intentionally or accidentally adding new code to prom_init.c # which has side effects on other parts of the kernel. Signed-off-by:
-
Timur Tabi authored
Add Timur Tabi as the maintainer for the Freescale QE library, the Freescale QE UART device driver, the Freescale SOC sound drivers, and the Crystal Semiconductor CS4270 device driver. Signed-off-by:
Timur Tabi <timur@freescale.com> Signed-off-by:
-
Kumar Gala authored
* Removed TI_EXECDOMAIN define as its not used anywhere * Use STACK_INT_FRAME_SIZE to allow common define of INT_FRAME_SIZE * Define TI_CPU on both ppc32 & ppc64 (removes an ifdef). Signed-off-by:
-
Kumar Gala authored
Use (31-THREAD_SHIFT) to get to thread_info from stack pointer. This makes the code a bit easier to read and more robust if we ever change THREAD_SHIFT. Signed-off-by:
-
Christoph Hellwig authored
Remove the inclusion of asm-offsets.h from stacktrace.c. It isn't supposed to be included in C code and it causes problems with multiple definitions of things. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Kumar Gala authored
The fixmap code from x86 allows us to have compile time virtual addresses that we change the physical addresses of at run time. This is useful for applications like kmap_atomic, PCI config that is done via direct memory map, kexec/kdump. We got ride of CONFIG_HIGHMEM_START as we can now determine a more optimal location for PKMAP_BASE based on where the fixmap addresses start and working back from there. Additionally, the kmap code in asm-powerpc/highmem.h always had debug enabled. Moved to using CONFIG_DEBUG_HIGHMEM to determine if we should have the extra debug checking. Signed-off-by:
-
Adrian Bunk authored
This patch fixes the following build error: <-- snip --> ... CC [M] drivers/char/xilinx_hwicap/xilinx_hwicap.o ... /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/char/xilinx_hwicap/xilinx_hwicap.c:806: error: hwicap_of_match causes a section type conflict /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/char/xilinx_hwicap/xilinx_hwicap.c:806: error: hwicap_of_match causes a section type conflict make[4]: *** [drivers/char/xilinx_hwicap/xilinx_hwicap.o] Error 1 <-- snip --> Signed-off-by:
Adrian Bunk <bunk@kernel.org> Signed-off-by:
-
Adrian Bunk authored
After commit 585468e5 ([POWERPC] i2c: Fix build breakage introduced by OF helpers) drivers/of/of_i2c.c needs a MODULE_LICENSE. Signed-off-by:
-
Kumar Gala authored
Added support to allow an 85xx kernel to be run from a non-zero physical address (useful for cooperative asymmetric multiprocessing situations and kdump). The support can be configured at compile time by setting CONFIG_PAGE_OFFSET, CONFIG_KERNEL_START, and CONFIG_PHYSICAL_START as desired. Alternatively, the kernel build can set CONFIG_RELOCATABLE. Setting this config option causes the kernel to determine at runtime the physical addresses of CONFIG_PAGE_OFFSET and CONFIG_KERNEL_START. If CONFIG_RELOCATABLE is set, then CONFIG_PHYSICAL_START has no meaning. However, CONFIG_PHYSICAL_START will always be used to set the LOAD program header physical address field in the resulting ELF image. Currently we are limited to running at a physical address that is a multiple of 256M. This is due to how we map TLBs to cover lowmem. This should be fixed to allow 64M or maybe even 16M alignment in the future. It is considered an error to try and run a kernel at a non-aligned physical address. All the magic for this support is accomplished by proper initialization of the kernel memory subsystem and use of ARCH_PFN_OFFSET. The use of ARCH_PFN_OFFSET only affects normal memory and not IO mappings. ioremap uses map_page and isn't affected by ARCH_PFN_OFFSET. /dev/mem continues to allow access to any physical address in the system regardless of how CONFIG_PHYSICAL_START is set. Signed-off-by:
-
Josh Boyer authored
Commit 0119536c added an assembly version of strncmp to PowerPC. However, it changed a common header file between arch/ppc and arch/powerpc without adding strncmp to arch/ppc. This fixes that omission so that arch/ppc links again. Signed-off-by:
Josh Boyer <jwboyer@linux.vnet.ibm.com> Signed-off-by:
-
Josh Boyer authored
Commit d04ceb3f moved phys_addr_t definitions to include/asm-powerpc/types.h. However, arch/ppc 440 builds had a duplicate definition in include/asm-ppc/mmu.h that caused the build to fail. This removes the duplicate definition in arch/ppc. Signed-off-by:
-
Remi Machet authored
The MPSC driver and prpmc2800.dts have been modified to use property 'cell-index' as the serial port number, but the early serial console driver for the mv64x60 has not been modified to use this new property. This fixes it. Signed-off-by: Remi Machet (rmachet@slac.stanford.edu) Acked-by:
Dale Farnsworth <dale@farnsworth.org> Signed-off-by:
-
Remi Machet authored
If one of the devices of the mv64x60 init fails, the remaining devices are not initialized. This changes the code to display an error and continue the initialization. Signed-off-by: Remi Machet (rmachet@slac.stanford.edu) Signed-off-by: -
Remi Machet authored
I2C parameters freq_m and freq_n are assigned defaults in the code, but if properties for those parameters are not found in the open firmware description the init routine returns an error and doesn't create the platform device. This changes the code so that it doesn't return an error if the properties are not found but instead uses the default values. Signed-off-by: Remi Machet (rmachet@slac.stanford.edu) Acked-by:
-
Benjamin Herrenschmidt authored
Use the __weak macro instead of the longer __attribute__ ((weak)) form in one place in init/main.c. Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> -- init/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Signed-off-by:
-
Benjamin Herrenschmidt authored
The powerpc kernel stacks need to be naturally aligned, as they contain the thread info at the bottom, which is obtained by clearing the low bits of the stack pointer. However, when using 64K pages, the stack is smaller than a page, so we use kmalloc to allocate it, but that doesn't provide the alignment guarantee we need. It appeared to work so far... until one enables SLUB debugging which then returns unaligned pointers. Ooops... This fixes it by using a slab cache with enforced alignment. It relies on my previous patch that adds a thread_info_cache_init() callback. Signed-off-by:
-
Benjamin Herrenschmidt authored
Some architectures need to maintain a kmem cache for thread info structures. The next commit adds that to powerpc to fix an alignment problem. There is no good arch callback to use to initialize that cache that I can find, so this adds a new one in the form of a weak function whose default is empty. Signed-off-by:
-
Geoff Levand authored
The PS3 gelic network driver depends on the wake-on-lan support provided by the PS3 sys manager driver. Add that dependency to the GELIC_NET Kconfig option. Prevents these build errors: ps3_gelic_net.c:1277: undefined reference to `.ps3_sys_manager_get_wol' ps3_gelic_net.c:1337: undefined reference to `.ps3_sys_manager_set_wol' CC: Masakazu Mokuno <mokuno@sm.sony.co.jp> CC: Jeff Garzik <jgarzik@pobox.com> Signed-off-by:
Geoff Levand <geoffrey.levand@am.sony.com> Signed-off-by:
-
Michael Ellerman authored
os-area.c requires routines declared in linux/of.h, so should include it. Signed-off-by:
-
Michael Ellerman authored
numa.c requires routines declared in linux/of.h, so should include it. Signed-off-by:
-
Paul Mackerras authored
-
Kay Sievers authored
Since 43cc71ee, the platform modalias is prefixed with "platform:". Add MODULE_ALIAS() to the hotpluggable "macintosh" platform drivers, to re-enable auto loading. [dbrownell@users.sourceforge.net: registration fixes] Signed-off-by:
Kay Sievers <kay.sievers@vrfy.org> Signed-off-by:
David Brownell <dbrownell@users.sourceforge.net> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Paul Mackerras authored
This reverts commit e4cc5894, as requested by Roland McGrath, because compat_ptrace_request (added in commit e16b2781, "ptrace: compat_ptrace_request siginfo") now handles this case. Signed-off-by:
-
