summaryrefslogtreecommitdiffstats
path: root/arch
Commit message (Collapse)AuthorAgeFilesLines
* sh: use set_current_blocked() and block_sigmask()Matt Fleming2012-03-292-49/+20
| | | | | | | | | | | | | | | | | | | | As described in e6fa16ab ("signal: sigprocmask() should do retarget_shared_pending()") the modification of current->blocked is incorrect as we need to check whether the signal we're about to block is pending in the shared queue. Also, use the new helper function introduced in commit 5e6292c0f28f ("signal: add block_sigmask() for adding sigmask to current->blocked") which centralises the code for updating current->blocked after successfully delivering a signal and reduces the amount of duplicate code across architectures. In the past some architectures got this code wrong, so using this helper function should stop that from happening again. Acked-by: Oleg Nesterov <oleg@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Matt Fleming <matt.fleming@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: no need to reset handler if SA_ONESHOTMatt Fleming2012-03-292-6/+0
| | | | | | | | | | | | | | get_signal_to_deliver() already resets the signal handler if SA_ONESHOT is set in ka->sa.sa_flags, there's no need to do it again in handle_signal(). Furthermore, because we were modifying ka->sa.sa_handler (which is a copy of sighand->action[]) instead of sighand->action[] the original code had no effect on signal delivery. Acked-by: Oleg Nesterov <oleg@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Matt Fleming <matt.fleming@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: select ARCH_DISCARD_MEMBLOCK.Paul Mundt2012-03-281-0/+1
| | | | | | We don't require it outside of the __init context, so discard it. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: Consolidate duplicate _32/_64 unistd definitions.Paul Mundt2012-03-283-80/+40
| | | | | | | This does some more tidying and converging of the header split, following the x86 example. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: ecovec: switch SDHI controllers to card pollingGuennadi Liakhovetski2012-03-281-41/+63
| | | | | | | | | Polling for card hotplug events is a better option than using SDHI native hotplug interrupt, because the latter option forces the controller to stay permanently powered up and enabled. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: Avoid exporting unimplemented syscalls.Paul Mundt2012-03-284-70/+71
| | | | | | | | | | | | | | | | Now that userspace is making use of kernel-provided sanitized headers for working out supported interfaces, we need to be a bit more diligent with matching the syscall definitions with their actual wiring/support state. In theory it shouldn't hurt anything since sys_ni_syscall will ultimately do the right thing, but there's also not much need to lie about legacy x86 syscalls that we've never supported. This tightens things up a bit for uClibc at least. Suggested-by: Carmelo Amoroso <carmelo.amoroso@st.com> Acked-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: add platform_device for RSPI in setup-sh7757Shimoda, Yoshihiro2012-03-281-0/+20
| | | | | Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* SH: pci-sh7780: enable big-endian operation.Thomas Schwinge2012-03-281-3/+12
| | | | | | | | | | | If in big-endian mode, switch the PCI bus, too. Tested on both litte-endian and big-endian sh7785lcr. Signed-off-by: Thomas Schwinge <thomas@codesourcery.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: linux-sh@vger.kernel.org Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: dma: Collect up CHCR of SH7763, SH7764, SH7780 and SH7785Nobuhiro Iwamatsu2012-03-281-17/+5
| | | | | | | | CHCR of SH7763, SH7764, SH7780 and SH7785 is constitution same as SH7757. Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com> Reviewed-by: Simon Horman <horms@verge.net.au> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* sh: dma: Collect up CHCR of SH7723 and SH7730Nobuhiro Iwamatsu2012-03-281-7/+3
| | | | | | | | CHCR of SH7723 and SH7730 is constitution same as SH7724. Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com> Reviewed-by: Simon Horman <horms@verge.net.au> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* Merge branch 'sh/cpufreq' into sh-latestPaul Mundt2012-03-281-39/+82
|\
| * sh: cpufreq: Wire up scaling_available_freqs support.Paul Mundt2012-01-271-0/+6
| | | | | | | | | | | | | | | | | | scaling_available_freqs is provided generically for drivers that are using frequency table based rounding. This will be optional for our case, but the generic code already takes that in to consideration, so we can simply wire it up outright. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
| * sh: cpufreq: notify about rate rounding fallback.Paul Mundt2012-01-271-3/+9
| | | | | | | | | | | | | | | | | | The general case for platforms that support the clock framework fully will be rate table rounding, while others will have to fall back on much coarser general rate rounding. Notify about it during boot so the limited functionality for the given subtype is appropriately noted. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
| * sh: cpufreq: Support CPU clock frequency table.Paul Mundt2012-01-271-28/+46
| | | | | | | | | | | | | | | | | | This adds support for the frequency table provided by the clock framework under the struct clk definition (if available). In cases where no table is generated or otherwise supported, we fall back on coarse grained scaling via clock framework rounding, as before. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
| * sh: cpufreq: struct device lookup from CPU topology.Paul Mundt2012-01-271-10/+17
| | | | | | | | | | | | | | | | | | The struct device pointer associated with the CPU we're on can be fetched via the topology information. Tie this in to localize the CPU clock lookup. While we're at it, tidy up some of the debug/info printing notices too. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
| * sh: cpufreq: percpu struct clk accounting.Paul Mundt2012-01-271-9/+15
| | | | | | | | | | | | | | | | At the moment there is simply a global struct clk pointer for the CPU frequency, which is fundamentally broken in the SMP case. This moves to fix it up by switching to a percpu case. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* | sh/next: Fix build fail by asm/system.h in asm/bitops.hNobuhiro Iwamatsu2012-03-281-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Other files are not done include definitely by asm/system.h made include in asm/bitops.h. This patch remove asm/system.h from asm/bitops.h. ----- LANG=C make ARCH=sh CROSS_COMPILE=sh4-linux-gnu- CHK include/linux/version.h CHK include/generated/utsrelease.h make[1]: `include/generated/machtypes.h' is up to date. CC arch/sh/kernel/asm-offsets.s In file included from include/asm-generic/getorder.h:7, from /home/iwamatsu/work/kernel/sh-2.6-devel/arch/sh/include/asm/page.h:187, from /home/iwamatsu/work/kernel/sh-2.6-devel/arch/sh/include/asm/mmu.h:38, from /home/iwamatsu/work/kernel/sh-2.6-devel/arch/sh/include/asm/system_32.h:5, from /home/iwamatsu/work/kernel/sh-2.6-devel/arch/sh/include/asm/system.h:179, from /home/iwamatsu/work/kernel/sh-2.6-devel/arch/sh/include/asm/bitops.h:10, from include/linux/bitops.h:22, from include/linux/thread_info.h:52, from include/linux/preempt.h:9, from include/linux/spinlock.h:50, from include/linux/mmzone.h:7, from include/linux/gfp.h:4, from include/linux/mm.h:8, from arch/sh/kernel/asm-offsets.c:13: include/linux/log2.h: In function '__ilog2_u32': include/linux/log2.h:34: error: implicit declaration of function 'fls' include/linux/log2.h: In function '__ilog2_u64': include/linux/log2.h:42: error: implicit declaration of function 'fls64' include/linux/log2.h: In function '__roundup_pow_of_two': include/linux/log2.h:63: error: implicit declaration of function 'fls_long' ----- Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
* | Merge branch 'master' of ↵Paul Mundt2012-03-282011-46837/+47579
|\ \ | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into sh-latest
| * \ Merge branch 'for-linus' of ↵Linus Torvalds2012-03-2710-34/+969
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux Pull s390 patches part 2 from Martin Schwidefsky: "Some minor improvements and one additional feature for the 3.4 merge window: Hendrik added perf support for the s390 CPU counters." * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: [S390] register cpu devices for SMP=n [S390] perf: add support for s390x CPU counters [S390] oprofile: Allow multiple users of the measurement alert interrupt [S390] qdio: log all adapter characteristics [S390] Remove unncessary export of arch_pick_mmap_layout
| | * | [S390] register cpu devices for SMP=nMartin Schwidefsky2012-03-231-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A kernel compiled with SMP=n will not register any cpu devices. The resulting kernel image will not boot with this error message: kernel BUG at fs/sysfs/group.c:65! Use GENERIC_CPU_DEVICES=y if SMP=n to get the missing cpu device. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| | * | [S390] perf: add support for s390x CPU countersHendrik Brueckner2012-03-235-5/+912
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add a perf PMU to access the CPU-measurement counter facility CPUM CF. CPUM CF provides multiple counter sets for measuring generic, problem-state, and crypto activaties. Also an extended counter set for the IBM System z10 and IBM z196 mainframes is available. Counters from the basic and problem-state counter set are mapped to generic perf hardware events. Other counters are accessible through raw events. For a list of available counter sets and counters, see: - The Load-Program-Parameter and the CPU-Measurement Facilities (SA23-2260) - The CPU-Measurement Facility Extended Counters Definition for z10 and z196 (SA23-2261) Reviewed-by: Jan Glauber <jang@linux.vnet.ibm.com> Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| | * | [S390] oprofile: Allow multiple users of the measurement alert interruptJan Glauber2012-03-234-30/+59
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Prepare the measurement facility which is currently only used by oprofile for multiple users. To achieve that the measurement alert interrupt control bit needs to be protected. The measurement alert definitions are moved to a header file and an interrupt mask is added so that users can discard interrupts if they are for a different measurement subsystem. Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| | * | [S390] Remove unncessary export of arch_pick_mmap_layoutBen Hutchings2012-03-231-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This function is defined for use in exec, not in modules. No other architecture exports its implementation. Signed-off-by: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | | Merge branch 'for-linus-3.4-rc1' of ↵Linus Torvalds2012-03-2745-728/+794
| |\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml Pull UML changes from Richard Weinberger: "Mostly bug fixes and cleanups" * 'for-linus-3.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml: (35 commits) um: Update defconfig um: Switch to large mcmodel on x86_64 MTD: Relax dependencies um: Wire CONFIG_GENERIC_IO up um: Serve io_remap_pfn_range() Introduce CONFIG_GENERIC_IO um: allow SUBARCH=x86 um: most of the SUBARCH uses can be killed um: deadlock in line_write_interrupt() um: don't bother trying to rebuild CHECKFLAGS for USER_OBJS um: use the right ifdef around exports in user_syms.c um: a bunch of headers can be killed by using generic-y um: ptrace-generic.h doesn't need user.h um: kill HOST_TASK_PID um: remove pointless include of asm/fixmap.h from asm/pgtable.h um: asm-offsets.h might as well come from underlying arch... um: merge processor_{32,64}.h a bit... um: switch close_chan() to struct line um: race fix: initialize delayed_work *before* registering IRQ um: line->have_irq is never checked... ...
| | * | | um: Update defconfigRichard Weinberger2012-03-251-184/+471
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Enable ext4, cgroups and devtmpfs. Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: Switch to large mcmodel on x86_64Richard Weinberger2012-03-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | x86_64 UML is unable to load modules if more than 504MiB of memory are used. This happens because on x86_64 the UML process has a quite high start address (typically around 0x6000000). If UML's memory is larger than 504MiB VMALLOC_START happens to be after 0x8000000. This is no problem unless one loads a module which was built with R_X86_64_32S relocations. Symbols with a location > 0x8000000 cannot be used with R_X86_64_32S To deal with this x86_64 UML has to be compiled with -mcmodel=large such that no R_X86_64_32S relocations are used. Signed-off-by: Richard Weinberger <richard@nod.at> Reported-by: 전하늘 <allskyee@gmail.com>
| | * | | um: Wire CONFIG_GENERIC_IO upRichard Weinberger2012-03-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | UML has no io memory but implements everything defined in generic-asm/io.h. Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: Serve io_remap_pfn_range()Richard Weinberger2012-03-251-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | At some places io_remap_pfn_range() is needed. UML has to serve it like all other archs do. Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: allow SUBARCH=x86Al Viro2012-03-251-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | nicked from patch by dwmw2 back in July Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: most of the SUBARCH uses can be killedAl Viro2012-03-253-7/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> [richard@nod.at: Re-export SUBARCH in arch/um/Makefile] Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: deadlock in line_write_interrupt()Al Viro2012-03-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: don't bother trying to rebuild CHECKFLAGS for USER_OBJSAl Viro2012-03-251-4/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ... just strip NOSTDINC_FLAGS out of it for those Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: use the right ifdef around exports in user_syms.cAl Viro2012-03-252-3/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ... the same one that controls whether elf_aux.o is included into the build, bringing the vsyscall_e... into it. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: a bunch of headers can be killed by using generic-yAl Viro2012-03-257-124/+1
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: ptrace-generic.h doesn't need user.hAl Viro2012-03-251-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: kill HOST_TASK_PIDAl Viro2012-03-254-8/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | just provide get_current_pid() to the userland side of things instead of get_current() + manual poking in its results Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: remove pointless include of asm/fixmap.h from asm/pgtable.hAl Viro2012-03-251-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: asm-offsets.h might as well come from underlying arch...Al Viro2012-03-251-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: merge processor_{32,64}.h a bit...Al Viro2012-03-253-20/+10
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: switch close_chan() to struct lineAl Viro2012-03-253-9/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ... and switch chan_interrupt() to directly calling close_one_chan(), so we can lose delay_free_irq argument of close_chan() as well. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: race fix: initialize delayed_work *before* registering IRQAl Viro2012-03-253-18/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ... since chan_interrupt() might schedule it if there's too much incoming data. Kill task argument of chan_interrupt(), while we are at it - it's always &line->task. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: line->have_irq is never checked...Al Viro2012-03-252-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | looks like a half-arsed duplicate of line->enabled Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: chan_init_pri is dead nowAl Viro2012-03-251-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: switch users of ->chan_list to ->chan_{in,out} (easy cases)Al Viro2012-03-255-128/+88
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: sorting out the chan mess, part 1Al Viro2012-03-252-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | put references to in and out chans associated with line into explicit struct chan * fields in it. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: finally kill ->init_str leaksAl Viro2012-03-251-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | now we can do that... Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: get rid of lines_init()Al Viro2012-03-254-46/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | move config-independent parts of initialization into register_lines(), call setup_one_line() after it instead of abusing ->init_str. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: switch line.c tty drivers to dynamic device creationAl Viro2012-03-254-42/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Current code doesn't update the symlinks in /sys/dev/char when we add/remove tty lines. Fixing that allows to stop messing with ->valid before the driver registration, which is a Good Thing(tm) - we shouldn't have it set before we really have the things set up and ready for line_open(). We need tty_driver available to call tty_{un,}register_device(), so we just stash a reference to it into struct line_driver. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: fix races between line_open() and line_config()Al Viro2012-03-252-28/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull parse_chan_pair() call into setup_one_line(), under the mutex. We really don't want open() to succeed before parse_chan_pair() had been done (or after it has failed, BTW). We also want "remove con<n>" to free irqs, etc., same as "config con<n>=none". Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
| | * | | um: convert count_lock to mutex, fix a race in line_open()Al Viro2012-03-254-20/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If two processes are opening the same line, the second to get into line_open() will decide that it doesn't need to do anything (correctly) or wait for anything. The latter, unfortunately, is incorrect - the first opener might not be through yet. We need to have exclusion covering the entire line_init(), including the blocking parts. Moreover, the next patch will need to widen the exclusion on mconsole side of things, also including the blocking bits, so let's just convert that sucker to mutex... Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Richard Weinberger <richard@nod.at>
generated by cgit v1.2.3 (git 2.25.1) at 2024-09-29 15:06:03 +0000