- Jul 31, 2012
-
-
Kautuk Consul authored
Commits d065bd81 ("mm: retry page fault when blocking on disk transfer") and 37b23e05 ("x86,mm: make pagefault killable") introduced changes into the x86 pagefault handler for making the page fault handler retryable as well as killable. These changes reduce the mmap_sem hold time, which is crucial during OOM killer invocation. Port these changes to AVR32. [akpm@linux-foundation.org: fix comment layout] Signed-off-by:
Mohd. Faris <mohdfarisq2010@gmail.com> Signed-off-by:
Kautuk Consul <consul.kautuk@gmail.com> Acked-by:
Havard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Paul Bolle authored
There's a small group of odd looking includes in smc37c669.c. These includes appear to be if zero-ed out ever since they were added to the tree (in v2.1.89). Their purpose is unclear to me. Perhaps they were used in someones build system. Whatever their purpose was, nothing else uses something comparable. This entire if zero-ed out block might as well be removed. Signed-off-by:
Paul Bolle <pebolle@tiscali.nl> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Signed-off-by:
-
Kautuk Consul authored
Commits d065bd81 ("mm: retry page fault when blocking on disk transfer") and 37b23e05 ("x86,mm: make pagefault killable") introduced changes into the x86 pagefault handler for making the page fault handler retryable as well as killable. These changes reduce the mmap_sem hold time, which is crucial during OOM killer invocation. Port these changes to xtensa. Signed-off-by:
Chris Zankel <chris@zankel.net> Signed-off-by:
-
- Jul 27, 2012
-
-
Steven Rostedt authored
As Colin Cross ported my x86 change to ARM, he also pointed out that powerpc is also behind in this fix. The commit 722b3c74 "ftrace/graph: Trace function entry before updating index" fixes an issue with function graph tracing for x86, where if the called entry function decides not to trace interrupts, it can fail the check if an interrupt comes in just after the curr_ret_stack is updated. The solution is to call the entry function first, then update the curr_ret_stack if the entry function wants to be traced. Cc: Colin Cross <ccross@android.com> Signed-off-by:
Steven Rostedt <rostedt@goodmis.org> Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org>
-
Anton Blanchard authored
Reduce the severity of the warning given when firmware flash is not supported. Not all platforms have it. Signed-off-by:
Anton Blanchard <anton@samba.org> Signed-off-by:
-
Anton Blanchard authored
Enable the hardware RNG and crypto modules. I verified they both autoload via the VIO subsystem, so there is no need to build them in. Signed-off-by:
-
Benjamin Herrenschmidt authored
This updates the g5 defconfig to include nouveau instead of nvidiafb (which works much better nowadays, in fact the latter crashes on modern distros), and to set CONFIG_VT_HW_CONSOLE_BINDING without which takeover from the firmware offb by nouveau doesn't work properly (and leads to unexplained black screens for some users). The rest is churn of going through defconfig / savedefconfig Signed-off-by: -
Alexander Graf authored
After merging the register type check patches from Ben's tree, the hv enabled booke implementation ceased to compile. This patch fixes things up so everyone's happy again. Signed-off-by:
Alexander Graf <agraf@suse.de> Signed-off-by:
-
Stuart Yoder authored
Commit 9778b696 incorrectly changes the code setting the stack limit on entry to the kernel to mark the thread_info at the bottom of the stack out of bounds anymore. This fixes it. Signed-off-by:
Stuart Yoder <stuart.yoder@freescale.com> Signed-off-by:
-
- Jul 26, 2012
-
-
Tkhai Kirill authored
FPU state is saved into task_struct of created task. Current task continues use of the state, so it needs TIF_USEDFPU flag is not cleared. Created task receives fresh FPU and cleared TIF_USEFPU flag is required for it. Signed-off-by:
Kirill Tkhai <tkhai@yandex.ru> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Sam Ravnborg authored
Signed-off-by:
Sam Ravnborg <sam@ravnborg.org> Signed-off-by:
-
Sam Ravnborg authored
Try to keep highmem support in a more central place. Signed-off-by:
-
Sam Ravnborg authored
Only one user so move it to the file using it. It had nothing to do in fault_32. Signed-off-by:
-
Sam Ravnborg authored
Signed-off-by:
-
Sam Ravnborg authored
Signed-off-by:
-
Sam Ravnborg authored
The quicklist stuff is not used anymore - drop it. Signed-off-by:
-
Sam Ravnborg authored
We already check the model in head_32.S so no need to repeat the check here Signed-off-by:
-
Sam Ravnborg authored
The base is always the same so no need to use a variable for this. Signed-off-by:
-
Sam Ravnborg authored
This function was only used to set of_pdt_build_more to leon_node_init(). But the leon_node_init() was a nop as prom_amba_init was never assigned. Cc: Daniel Hellstrom <daniel@gaisler.com> Cc: Konrad Eisele <konrad@gaisler.com> Signed-off-by:
-
Sam Ravnborg authored
sparc32 does not support fixmaps - so do not pretend so by having the fixmap.h file. Move relevant parts to vaddrs.h. I looked at simplifying this even more but failed to understand the reasoning behind the extra guard page involved and due to missing testing possibilities only the trivial conversion was done. Signed-off-by:
-
Sam Ravnborg authored
Remove all unused stuff from fixmap.h It is only used for highmem - sparc32 has no fixmap support. Signed-off-by:
-
Sam Ravnborg authored
No users left of this function - drop it. Signed-off-by:
-
Sam Ravnborg authored
We save one page of RAM dropping swapper_pg_dir. It was only used for an assignment in init-mm.c and we redid this later in srmmu.c anyway. This is likely a left-over from the sun4c removal. To avoid a dummy variable we use a simple #define swapper_pg_dir NULL Signed-off-by:
-
Sam Ravnborg authored
Signed-off-by:
-
Sam Ravnborg authored
This allowed to us to kill a lot of casts, with no loss of readability in any places Signed-off-by:
-
Sam Ravnborg authored
Fix the most annoying issues that distracts me: - whitespace - missing space after "if" and "while" - spaces around operators and similar simple things. Signed-off-by:
-
Sam Ravnborg authored
Signed-off-by:
-
Sam Ravnborg authored
They are only used during early init so lets get rid of them after init to save some RAM. Signed-off-by:
-
Josh Boyer authored
Recently, glibc made a change to suppress sign-conversion warnings in FD_SET (glibc commit ceb9e56b3d1). This uncovered an issue with the kernel's definition of __NFDBITS if applications #include <linux/types.h> after including <sys/select.h>. A build failure would be seen when passing the -Werror=sign-compare and -D_FORTIFY_SOURCE=2 flags to gcc. It was suggested that the kernel should either match the glibc definition of __NFDBITS or remove that entirely. The current in-kernel uses of __NFDBITS can be replaced with BITS_PER_LONG, and there are no uses of the related __FDELT and __FDMASK defines. Given that, we'll continue the cleanup that was started with commit 8b3d1cda ("posix_types: Remove fd_set macros") and drop the remaining unused macros. Additionally, linux/time.h has similar macros defined that expand to nothing so we'll remove those at the same time. Reported-by:
Jeff Law <law@redhat.com> Suggested-by:
Josh Boyer <jwboyer@redhat.com> [ .. and fix up whitespace as per akpm ] Signed-off-by:
-
Shawn Guo authored
The commit 475d0094 (of: Improve prom_update_property() function) changed prom_update_property interface, and causes the compile error below. CC arch/arm/mach-mxs/mach-mxs.o arch/arm/mach-mxs/mach-mxs.c: In function ‘update_fec_mac_prop’: arch/arm/mach-mxs/mach-mxs.c:216:4: error: too many arguments to function ‘prom_update_property’ include/linux/of.h:263:12: note: declared here Update the caller update_fec_mac_prop to fix the error. Signed-off-by:
Shawn Guo <shawn.guo@linaro.org> Signed-off-by:
Olof Johansson <olof@lixom.net>
-
- Jul 25, 2012
-
-
Chris Metcalf authored
The bombing to convert pci_bus.subordinate to busn_res.end accidentally modified a "struct pci_dev" site, causing this file not to compile. This commit reverts that code to use dev->subordinate again. Signed-off-by:Chris Metcalf <cmetcalf@tilera.com>
-
Chris Metcalf authored
Yinghai Lu removed pci_bus.subordinate in pci-next, which meant that the tile-next changes to add tilegx PCI support don't build. This was expected (seen in linux-next) and this one-line fix is along the same lines as commit b918c62e for all other architectures. Acked-by:
Bjorn Helgaas <bhelgaas@google.com> Signed-off-by:
-
Stephen Warren authored
This was accidentally disabled by commit 2a5fdc9a "ARM: dt: tegra: invert status=disable vs status=okay". Cc: <stable@vger.kernel.org> # v3.5 (file is named tegra-trimslice.dts there) Signed-off-by:
Stephen Warren <swarren@nvidia.com> Signed-off-by:
Arnd Bergmann <arnd@arndb.de>
-
Stephen Warren authored
On TrimSlice, Tegra's USB1 port may be routed to either an external micro USB port, or an internal USB->SATA bridge for SSD or HDD. This muxing is controlled by a GPIO. Whilst not strictly a VBUS GPIO, the TrimSlice board files caused this GPIO to be set appropriately to enable the SATA bridge by passing it as the VBUS GPIO to the USB driver. Echo this same configuration in device tree to enable the SATA bridge. An alternative might be to implement a full USB bus mux driver. However, that seems over-complex right now. Cc: <stable@vger.kernel.org> # v3.5 (file is named tegra-trimslice.dts there) Signed-off-by:
-
Andrew Lunn authored
It has been decided to use marvell, not mrvl, in the compatibility property. Search & replace. Signed-off-by:Andrew Lunn <andrew@lunn.ch>
-
Hannes Reinecke authored
The last argument of orion_ge00_init() is actually the error interrupt, so we should be using the correct value here. Signed-off-by:
Hannes Reinecke <hare@suse.de> Signed-off-by:
-
Simon Baatz authored
Commit 98d9986c (ARM: Kirkwood: Replace clock gating) and the fix 5fb2ce (ARM: Kirkwood: clk_register_gate_fn: add fn assignment) introduced a custom variant of clock gating which allows to define a function to be called before gating the clock off. This is used to disable the SATA and PCIe PHYs if the respective clocks are unused after initialization. However, of these two drivers, the SATA driver may be compiled as a module. The driver re-enables the clocks at module init but the PHYs stay disabled. Since the custom clock gating disabled the PHYs when gating the clock off, it should also re-enable them when enabling the clock gate. This is done by adding a second function that may be used to enable the PHYs. Signed-off-by:
Simon Baatz <gmbnomis@gmail.com> Signed-off-by:
-
Andrew Lunn authored
Marvell engineers tell us: It seems that many units use the RUNIT clock. SPI, UART, NAND, TWSI, ... So it's not possible to clock gate it. Currently the SPI, NAND and TWSI driver will clk_prepaure_enable() this clk, but since we have no idea what ... is, and turning this clk off results in a hard lock, unconditionally enable runit. Signed-off-by:
-
Pawel Moll authored
Due to automatic merge of a613163d "ARM: integrator: convert to common clock" and 56a34b03 "ARM: versatile: Make plat-versatile clock optional" two platforms: Integrator and Versatile Express now select both COMMON_CLK and PLAT_VERSATILE_CLOCK which breaks building them. Fixed now. Signed-off-by:
Pawel Moll <pawel.moll@arm.com> Signed-off-by:
-
- Jul 24, 2012
-
-
Laxman Dewangan authored
Name the regulator as per board schematics and adds its supply name info in regulator data. Add the always on fixed regulator to refer the battery supply. Use this fixed regulator for input supply of some of PMIC regulator This patch was originally part of a 2-patch series. Patch 2 got applied to the regulator tree as 7c7fac30 "regulator: tps6586x: add support for input supply" without this patch. This broke regulator support on Harmony. Signed-off-by:
Laxman Dewangan <ldewangan@nvidia.com> [swarren: added dependency info to commit description] Signed-off-by:
-
