kernel.git/arch/powerpc, branch linux-2.6.32.y

powerpc: Align TOC to 256 bytes

2015-09-18T11:52:01Z

commit 5e95235ccd5442d4a4fe11ec4eb99ba1b7959368 upstream. Recent toolchains force the TOC to be 256 byte aligned. We need to enforce this alignment in our linker script, otherwise pointers to our TOC variables (__toc_start, __prom_init_toc_start) could be incorrect. If they are bad, we die a few hundred instructions into boot. Signed-off-by: Anton Blanchard Signed-off-by: Michael Ellerman Signed-off-by: Ben Hutchings (cherry picked from commit 303241421684cdd2f9e931bc42b9de811320e7fd) Signed-off-by: Willy Tarreau

powerpc: Fix missing L2 cache size in /sys/devices/system/cpu

2015-09-18T11:51:58Z

commit f7e9e358362557c3aa2c1ec47490f29fe880a09e upstream. This problem appears to have been introduced in 2.6.29 by commit 93197a36a9c1 "Rewrite sysfs processor cache info code". This caused lscpu to error out on at least e500v2 devices, eg: error: cannot open /sys/devices/system/cpu/cpu0/cache/index2/size: No such file or directory Some embedded powerpc systems use cache-size in DTS for the unified L2 cache size, not d-cache-size, so we need to allow for both DTS names. Added a new CACHE_TYPE_UNIFIED_D cache_type_info structure to handle this. Fixes: 93197a36a9c1 ("powerpc: Rewrite sysfs processor cache info code") Signed-off-by: Dave Olson Signed-off-by: Michael Ellerman [bwh: Backported to 3.2: - Adjust context - Preserve __cpuinit attribute on cache_do_one_devnode_unified()] Signed-off-by: Ben Hutchings (cherry picked from commit d6de5ca93fd6425dcdb21cc341365991c0c444cc) Signed-off-by: Willy Tarreau

signal: Define __ARCH_HAS_SA_RESTORER so we know whether to clear sa_restorer

2013-06-10T09:42:21Z

Vaguely based on upstream commit 574c4866e33d 'consolidate kernel-side struct sigaction declarations'. flush_signal_handlers() needs to know whether sigaction::sa_restorer is defined, not whether SA_RESTORER is defined. Define the __ARCH_HAS_SA_RESTORER macro to indicate this. Signed-off-by: Ben Hutchings Cc: Al Viro Signed-off-by: Greg Kroah-Hartman Signed-off-by: Willy Tarreau

powerpc: Add "memory" attribute for mfmsr()

2012-10-07T21:38:00Z

commit b416c9a10baae6a177b4f9ee858b8d309542fbef upstream. Add "memory" attribute in inline assembly language as a compiler barrier to make sure 4.6.x GCC don't reorder mfmsr(). Signed-off-by: Tiejun Chen Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Greg Kroah-Hartman Signed-off-by: Willy Tarreau

powerpc/ftrace: Fix assembly trampoline register usage

2012-10-07T21:38:00Z

commit fd5a42980e1cf327b7240adf5e7b51ea41c23437 upstream. Just like the module loader, ftrace needs to be updated to use r12 instead of r11 with newer gcc's. Signed-off-by: Roger Blofeld Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Gortmaker Signed-off-by: Greg Kroah-Hartman Signed-off-by: Willy Tarreau

powerpc: Fix kernel panic during kernel module load

2012-10-07T21:37:53Z

commit 3c75296562f43e6fbc6cddd3de948a7b3e4e9bcf upstream. This fixes a problem which can causes kernel oopses while loading a kernel module. According to the PowerPC EABI specification, GPR r11 is assigned the dedicated function to point to the previous stack frame. In the powerpc-specific kernel module loader, do_plt_call() (in arch/powerpc/kernel/module_32.c), GPR r11 is also used to generate trampoline code. This combination crashes the kernel, in the case where the compiler chooses to use a helper function for saving GPRs on entry, and the module loader has placed the .init.text section far away from the .text section, meaning that it has to generate a trampoline for functions in the .init.text section to call the GPR save helper. Because the trampoline trashes r11, references to the stack frame using r11 can cause an oops. The fix just uses GPR r12 instead of GPR r11 for generating the trampoline code. According to the statements from Freescale, this is safe from an EABI perspective. I've tested the fix for kernel 2.6.33 on MPC8541. Signed-off-by: Steffen Rumler [paulus@samba.org: reworded the description] Signed-off-by: Paul Mackerras Signed-off-by: Greg Kroah-Hartman Signed-off-by: Willy Tarreau

powerpc/pmac: Fix SMP kernels on pre-core99 UP machines

2012-10-07T21:37:01Z

commit 78c5c68a4cf4329d17abfa469345ddf323d4fd62 upstream. The code for "powersurge" SMP would kick in and cause a crash at boot due to the lack of a NULL test. Adam Conrad reports that the 3.2 kernel, with CONFIG_SMP=y, will not boot on an OldWorld G3; we're unconditionally writing to psurge_start, but this is only set on powersurge machines. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Jeremy Kerr Reported-by: Adam Conrad Tested-by: Adam Conrad Signed-off-by: Willy Tarreau

Revert "powerpc/mpic: Fix problem that affinity is not updated"

2011-11-09T00:02:23Z

This reverts commit 1badd98ea79b7b20fb4ddfea110d1bb99c33a55f. It breaks the build on powerpc systems: arch/powerpc/sysdev/mpic.c: In function 'irq_choose_cpu': arch/powerpc/sysdev/mpic.c:574: error: passing argument 1 of '__cpus_equal' from incompatible pointer type Reported-by: Jiri Slaby Cc: Jiajun Wu Cc: Li Yang Cc: Benjamin Herrenschmidt Signed-off-by: Greg Kroah-Hartman

powerpc/pci: Check devices status property when scanning OF tree

2011-11-07T20:31:53Z

commit 5b339bdf164d8aee394609768f7e2e4415b0252a upstream. We ran into an issue where it looks like we're not properly ignoring a pci device with a non-good status property when we walk the device tree and instanciate the Linux side PCI devices. However, the EEH init code does look for the property and disables EEH on these devices. This leaves us in an inconsistent where we are poking at a supposedly bad piece of hardware and RTAS will block our config cycles because EEH isn't enabled anyway. Signed-of-by: Sonny Rao Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Greg Kroah-Hartman

powerpc/mpic: Fix problem that affinity is not updated

2011-11-07T20:31:52Z

commit 38e1313fc753482b93aa6c6f11cfbd43a5bcd963 upstream. Since commit 57b150cce8e004ddd36330490a68bfb59b7271e9, desc->affinity of an irq is changed after calling desc->chip->set_affinity. Therefore we need to fix the irq_choose_cpu() not to depend on the desc->affinity for new mask. Signed-off-by: Jiajun Wu Signed-off-by: Li Yang Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Greg Kroah-Hartman