- 08 Jun, 2016 30 commits
-
-
Ricky Liang authored
commit affa80bd upstream. When running a 32-bit userspace on a 64-bit kernel, the UI_SET_PHYS ioctl needs to be treated with special care, as it has the pointer size encoded in the command. Signed-off-by:
Ricky Liang <jcliang@chromium.org> Signed-off-by:
Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Mark Brown authored
commit d3030d11 upstream. The ak4642 driver is using a regmap cache sync to restore the configuration of the chip on resume but (as Peter observed) does not actually define a register cache which means that the resume is never going to work and we trigger asserts in regmap. Fix this by enabling caching. Reported-by:
Geert Uytterhoeven <geert@linux-m68k.org> Reported-by:
Peter Ujfalusi <peter.ujfalusi@ti.com> Tested-by:
Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by:
Mark Brown <broonie@kernel.org> Signed-off-by:
-
Mikulas Patocka authored
commit 01d6e087 upstream. Commit c8f33d0b ("affs: kstrdup() memory handling") checks if the kstrdup function returns NULL due to out-of-memory condition. However, if we are remounting a filesystem with no change to filesystem-specific options, the parameter data is NULL. In this case, kstrdup returns NULL (because it was passed NULL parameter), although no out of memory condition exists. The mount syscall then fails with ENOMEM. This patch fixes the bug. We fail with ENOMEM only if data is non-NULL. The patch also changes the call to replace_mount_options - if we didn't pass any filesystem-specific options, we don't call replace_mount_options (thus we don't erase existing reported options). Fixes: c8f33d0b ("affs: kstrdup() memory handling") Signed-off-by:
Mikulas Patocka <mpatocka@redhat.com> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Maciej W. Rozycki authored
commit 94cc36b8 upstream. Avoid an aliasing issue causing a build error in VDSO: In file included from include/linux/srcu.h:34:0, from include/linux/notifier.h:15, from ./arch/mips/include/asm/uprobes.h:9, from include/linux/uprobes.h:61, from include/linux/mm_types.h:13, from ./arch/mips/include/asm/vdso.h:14, from arch/mips/vdso/vdso.h:27, from arch/mips/vdso/gettimeofday.c:11: include/linux/workqueue.h: In function 'work_static': include/linux/workqueue.h:186:2: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing] return *work_data_bits(work) & WORK_STRUCT_STATIC; ^ cc1: all warnings being treated as errors make[2]: *** [arch/mips/vdso/gettimeofday.o] Error 1 with a CONFIG_DEBUG_OBJECTS_WORK configuration and GCC 5.2.0. Include `-fno-strict-aliasing' along with compiler options used, as required for kernel code, fixing a problem present since the introduction of VDSO with commit ebb5e78c ("MIPS: Initial implementation of a VDSO"). Thanks to Tejun for diagnosing this properly! Signed-off-by:
Maciej W. Rozycki <macro@imgtec.com> Reviewed-by:
James Hogan <james.hogan@imgtec.com> Fixes: ebb5e78c ("MIPS: Initial implementation of a VDSO") Cc: Tejun Heo <tj@kernel.org> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/13357/Signed-off-by:
Ralf Baechle <ralf@linux-mips.org> Signed-off-by:
-
Harvey Hunt authored
commit aedcfbe0 upstream. On certain MIPS32 devices, the ftrace tracer "function_graph" uses __lshrdi3() during the capturing of trace data. ftrace then attempts to trace __lshrdi3() which leads to infinite recursion and a stack overflow. Fix this by marking __lshrdi3() as notrace. Mark the other compiler intrinsics as notrace in case the compiler decides to use them in the ftrace path. Signed-off-by:
Harvey Hunt <harvey.hunt@imgtec.com> Cc: <linux-mips@linux-mips.org> Cc: <linux-kernel@vger.kernel.org> Patchwork: https://patchwork.linux-mips.org/patch/13354/Signed-off-by:
-
James Hogan authored
commit bb93078e upstream. MicroMIPS kernels may be expected to run on microMIPS only cores which don't support the normal MIPS instruction set, so be sure to pass the -mmicromips flag through to the VDSO cflags. Fixes: ebb5e78c ("MIPS: Initial implementation of a VDSO") Signed-off-by:
-
James Hogan authored
commit 13eb192d upstream. In microMIPS kernels, handle_signal() sets the isa16 mode bit in the vdso address so that the sigreturn trampolines (which are offset from the VDSO) get executed as microMIPS. However commit ebb5e78c ("MIPS: Initial implementation of a VDSO") changed the offsets to come from the VDSO image, which already have the isa16 mode bit set correctly since they're extracted from the VDSO shared library symbol table. Drop the isa16 mode bit handling from handle_signal() to fix sigreturn for cores which support both microMIPS and normal MIPS. This doesn't fix microMIPS only cores, since the VDSO is still built for normal MIPS, but thats a separate problem. Fixes: ebb5e78c ("MIPS: Initial implementation of a VDSO") Signed-off-by:
-
Maciej W. Rozycki authored
commit abf378be upstream. Correct the cases missed with commit 9b26616c ("MIPS: Respect the ISA level in FCSR handling") and prevent writes to read-only FCSR bits there. This in particular applies to FP context initialisation where any IEEE 754-2008 bits preset by `mips_set_personality_nan' are cleared before the relevant ptrace(2) call takes effect and the PTRACE_POKEUSR request addressing FPC_CSR where no masking of read-only FCSR bits is done. Remove the FCSR clearing from FP context initialisation then and unify PTRACE_POKEUSR/FPC_CSR and PTRACE_SETFPREGS handling, by factoring out code from `ptrace_setfpregs' and calling it from both places. This mostly matters to soft float configurations where the emulator can be switched this way to a mode which should not be accessible and cannot be set with the CTC1 instruction. With hard float configurations any effect is transient anyway as read-only bits will retain their values at the time the FP context is restored. Signed-off-by:
-
Maciej W. Rozycki authored
commit 42495484 upstream. Fix a floating-point context restoration regression introduced with commit 9b26616c ("MIPS: Respect the ISA level in FCSR handling") that causes a Floating Point exception and consequently a kernel oops with hard float configurations when one or more FCSR Enable and their corresponding Cause bits are set both at a time via a ptrace(2) call. To do so reinstate Cause bit masking originally introduced with commit b1442d39 ("MIPS: Prevent user from setting FCSR cause bits") to address this exact problem and then inadvertently removed from the PTRACE_SETFPREGS request with the commit referred above. Signed-off-by:
-
Paul Burton authored
commit bd239f1e upstream. Whilst a PR_SET_FP_MODE prctl is performed there are decisions made based upon whether the task is executing on the current CPU. This may change if we're preempted, so disable preemption to avoid such changes for the lifetime of the mode switch. Signed-off-by:
Paul Burton <paul.burton@imgtec.com> Fixes: 9791554b ("MIPS,prctl: add PR_[GS]ET_FP_MODE prctl options for MIPS") Reviewed-by:
Aurelien Jarno <aurelien@aurel32.net> Cc: Adam Buchbinder <adam.buchbinder@gmail.com> Cc: James Hogan <james.hogan@imgtec.com> Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/13144/Signed-off-by:
-
Paul Burton authored
commit 6533af4d upstream. If a kernel doesn't support MSA context (ie. CONFIG_CPU_HAS_MSA=n) then it will only keep 64 bits per FP register in thread context, and the calls to set_fpr64 in restore_msa_extcontext will overrun the end of the FP register context into the FCSR & MSACSR values. GCC 6.x has become smart enough to detect this & complain like so: arch/mips/kernel/signal.c: In function 'protected_restore_fp_context': ./arch/mips/include/asm/processor.h:114:17: error: array subscript is above array bounds [-Werror=array-bounds] fpr->val##width[FPR_IDX(width, idx)] = val; \ ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~ ./arch/mips/include/asm/processor.h:118:1: note: in expansion of macro 'BUILD_FPR_ACCESS' BUILD_FPR_ACCESS(64) The only way to trigger this code to run would be for a program to set up an artificial extended MSA context structure following a sigframe & execute sigreturn. Whilst this doesn't allow a program to write to any state that it couldn't already, it makes little sense to allow this "restoration" of MSA context in a system that doesn't support MSA. Fix this by killing a program with SIGSYS if it tries something as crazy as "restoring" fake MSA context in this way, also fixing the build error & allowing for most of restore_msa_extcontext to be optimised out of kernels without support for MSA. Signed-off-by:
Michal Toman <michal.toman@imgtec.com> Fixes: bf82cb30 ("MIPS: Save MSA extended context around signals") Tested-by:
Aaro Koskinen <aaro.koskinen@iki.fi> Cc: James Hogan <james.hogan@imgtec.com> Cc: Michal Toman <michal.toman@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/13164/Signed-off-by:
-
Paul Burton authored
commit 6b832257 upstream. Commit 9791554b ("MIPS,prctl: add PR_[GS]ET_FP_MODE prctl options for MIPS") added support for the PR_SET_FP_MODE prctl, which allows a userland program to modify its FP mode at runtime. This is most notably required if dynamic linking leads to the FP mode requirement changing at runtime from that indicated in the initial executable's ELF header. In order to avoid overhead in the general FP context restore code, it aimed to have threads in the process become unable to enable the FPU during a mode switch & have the thread calling the prctl syscall wait for all other threads in the process to be context switched at least once. Once that happens we can know that no thread in the process whose mode will be switched has live FP context, and it's safe to perform the mode switch. However in the (rare) case of modeswitches occurring in multithreaded programs this can lead to indeterminate delays for the thread invoking the prctl syscall, and the code monitoring for those context switches was woefully inadequate for all but the simplest cases. Fix this by broadcasting an IPI if other CPUs may have live FP context for an affected thread, with a handler causing those CPUs to relinquish their FPU ownership. Threads will then be allowed to continue running but will stall on the wait_on_atomic_t in enable_restore_fp_context if they attempt to use FP again whilst the mode switch is still in progress. The end result is less fragile poking at scheduler context switch counts & a more expedient completion of the mode switch. Signed-off-by:
-
James Hogan authored
commit ea168857 upstream. The MSA ld_*/st_* assembler macros for when the toolchain doesn't support MSA use addu to offset the base address. However it is a virtual memory pointer so fix it to use PTR_ADDU which expands to daddu for 64-bit kernels. Signed-off-by:
-
Paul Burton authored
commit 8a3c8b48 upstream. In revision 1.12 of the MSA specification, the copy_u.w instruction has been removed for MIPS32 & the copy_u.d instruction has been removed for MIPS64. Newer toolchains (eg. Codescape SDK essentials 2015.10) will complain about this like so: arch/mips/kernel/r4k_fpu.S:290: Error: opcode not supported on this processor: mips32r2 (mips32r2) `copy_u.w $1,$w26[3]' Since we always copy to the width of a GPR, simply use copy_s instead of copy_u to fix this. Signed-off-by:
-
Huacai Chen authored
commit 3484de7b upstream. Due to datasheet, reserving 0xff800000~0xffffffff (8MB below 4GB) is not enough for RS780E integrated GPU's TOM (top of memory) registers and MSI/MSI-x memory region, so we reserve 0xfe000000~0xffffffff (32MB below 4GB). Signed-off-by:
Huacai Chen <chenhc@lemote.com> Cc: Aurelien Jarno <aurelien@aurel32.net> Cc: Steven J . Hill <sjhill@realitydiluted.com> Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Zhangjin Wu <wuzhangjin@gmail.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/12889/Signed-off-by:
-
Huacai Chen authored
commit a95d0692 upstream. After commit 92923ca3 ("mm: meminit: only set page reserved in the memblock region"), the MIPS hibernation is broken. Because pages in nosave data section should be "reserved", but currently they aren't set to "reserved" at initialization. This patch makes hibernation work again. Signed-off-by:
-
Matthias Schiffer authored
commit f5b556c9 upstream. This makes the ath79 bootconsole behave the same way as the generic 8250 bootconsole. Also waiting for TEMT (transmit buffer is empty) instead of just THRE (transmit buffer is not full) ensures that all characters have been transmitted before the real serial driver starts reconfiguring the serial controller (which would sometimes result in garbage being transmitted.) This change does not cause a visible performance loss. In addition, this seems to fix a hang observed in certain configurations on many AR7xxx/AR9xxx SoCs during autoconfig of the real serial driver. A more complete follow-up patch will disable 8250 autoconfig for ath79 altogether (the serial controller is detected as a 16550A, which is not fully compatible with the ath79 serial, and the autoconfig may lead to undefined behavior on ath79.) Signed-off-by:
Matthias Schiffer <mschiffer@universe-factory.net> Signed-off-by:
-
Huacai Chen authored
commit 820880cd upstream. Commit d5ece1cb ("Fix ld-version.sh to handle large 3rd version part") modifies the ld version description. This causes a build error on Loongson-3, so fix it. Signed-off-by:
-
Paul Burton authored
commit 37d22a0d upstream. It's possible for pages to become visible prior to update_mmu_cache running if a thread within the same address space preempts the current thread or runs simultaneously on another CPU. That is, the following scenario is possible: CPU0 CPU1 write to page flush_dcache_page flush_icache_page set_pte_at map page update_mmu_cache If CPU1 maps the page in between CPU0's set_pte_at, which marks it valid & visible, and update_mmu_cache where the dcache flush occurs then CPU1s icache will fill from stale data (unless it fills from the dcache, in which case all is good, but most MIPS CPUs don't have this property). Commit 4d46a67a ("MIPS: Fix race condition in lazy cache flushing.") attempted to fix that by performing the dcache flush in flush_icache_page such that it occurs before the set_pte_at call makes the page visible. However it has the problem that not all code that writes to pages exposed to userland call flush_icache_page. There are many callers of set_pte_at under mm/ and only 2 of them do call flush_icache_page. Thus the race window between a page becoming visible & being coherent between the icache & dcache remains open in some cases. To illustrate some of the cases, a WARN was added to __update_cache with this patch applied that triggered in cases where a page about to be flushed from the dcache was not the last page provided to flush_icache_page. That is, backtraces were obtained for cases in which the race window is left open without this patch. The 2 standout examples follow. When forking a process: [ 15.271842] [<80417630>] __update_cache+0xcc/0x188 [ 15.277274] [<80530394>] copy_page_range+0x56c/0x6ac [ 15.282861] [<8042936c>] copy_process.part.54+0xd40/0x17ac [ 15.289028] [<80429f80>] do_fork+0xe4/0x420 [ 15.293747] [<80413808>] handle_sys+0x128/0x14c When exec'ing an ELF binary: [ 14.445964] [<80417630>] __update_cache+0xcc/0x188 [ 14.451369] [<80538d88>] move_page_tables+0x414/0x498 [ 14.457075] [<8055d848>] setup_arg_pages+0x220/0x318 [ 14.462685] [<805b0f38>] load_elf_binary+0x530/0x12a0 [ 14.468374] [<8055ec3c>] search_binary_handler+0xbc/0x214 [ 14.474444] [<8055f6c0>] do_execveat_common+0x43c/0x67c [ 14.480324] [<8055f938>] do_execve+0x38/0x44 [ 14.485137] [<80413808>] handle_sys+0x128/0x14c These code paths write into a page, call flush_dcache_page then call set_pte_at without flush_icache_page inbetween. The end result is that the icache can become corrupted & userland processes may execute unexpected or invalid code, typically resulting in a reserved instruction exception, a trap or a segfault. Fix this race condition fully by performing any cache maintenance required to keep the icache & dcache in sync in set_pte_at, before the page is made valid. This has the added bonus of ensuring the cache maintenance always happens in one location, rather than being duplicated in flush_icache_page & update_mmu_cache. It also matches the way other architectures solve the same problem (see arm, ia64 & powerpc). Signed-off-by:
Ionela Voinescu <ionela.voinescu@imgtec.com> Cc: Lars Persson <lars.persson@axis.com> Fixes: 4d46a67a ("MIPS: Fix race condition in lazy cache flushing.") Cc: Steven J. Hill <sjhill@realitydiluted.com> Cc: David Daney <david.daney@cavium.com> Cc: Huacai Chen <chenhc@lemote.com> Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Jerome Marchand <jmarchan@redhat.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/12722/Signed-off-by:
-
Paul Burton authored
commit f4281bba upstream. The following patch will expose __update_cache to highmem pages. Handle them by mapping them in for the duration of the cache maintenance, just like in __flush_dcache_page. The code for that isn't shared because we need the page address in __update_cache so sharing became messy. Given that the entirity is an extra 5 lines, just duplicate it. Signed-off-by:
-
Paul Burton authored
commit 234859e4 upstream. When flush_dcache_page is called on an executable page, that page is about to be provided to userland & we can presume that the icache contains no valid entries for its address range. However if the icache does not fill from the dcache then we cannot presume that the pages content has been written back as far as the memories that the dcache will fill from (ie. L2 or further out). This was being done for lowmem pages, but not for highmem which can lead to icache corruption. Fix this by mapping highmem pages & flushing their content from the dcache in __flush_dcache_page before providing the page to userland, just as is done for lowmem pages. Signed-off-by:
-
James Hogan authored
commit a7e89326 upstream. Commit f51246ef ("MIPS: Get rid of finish_arch_switch().") moved the __restore_watch() call from finish_arch_switch() (i.e. after resume() returns) to before the resume() call in switch_to(). This results in watchpoints only being restored when a task is descheduled, preventing the watchpoints from being effective most of the time, except due to chance before the watchpoints are lazily removed. Fix the call sequence from switch_to() through to mips_install_watch_registers() to pass the task_struct pointer of the next task, instead of using current. This allows the watchpoints for the next (non-current) task to be restored without reintroducing finish_arch_switch(). Fixes: f51246ef ("MIPS: Get rid of finish_arch_switch().") Signed-off-by:
-
James Hogan authored
commit 987e5b83 upstream. Since commit 8cb48fe1 ("MIPS: Provide correct siginfo_t.si_stime"), MIPS' uapi/asm/siginfo.h has included uapi/asm-generic/siginfo.h directly before defining MIPS' struct siginfo, in order to get the necessary definitions needed for the siginfo struct without the generic copy_siginfo() hitting compiler errors due to struct siginfo not yet being defined. Now that the generic copy_siginfo() is moved out to linux/signal.h we can safely include asm-generic/siginfo.h before defining the MIPS specific struct siginfo, which avoids the uapi/ include as well as breakage due to generic copy_siginfo() being defined before struct siginfo. Reported-by:
Christopher Ferris <cferris@google.com> Fixes: 8cb48fe1 ("MIPS: Provide correct siginfo_t.si_stime") Signed-off-by:
-
James Hogan authored
commit 5daebc47 upstream. Commit 85efde6f ("make exported headers use strict posix types") changed the asm-generic siginfo.h to use the __kernel_* types, and commit 3a471cbc ("remove __KERNEL_STRICT_NAMES") make the internal types accessible only to the kernel, but the MIPS implementation hasn't been updated to match. Switch to proper types now so that the exported asm/siginfo.h won't produce quite so many compiler errors when included alone by a user program. Signed-off-by:
-
James Hogan authored
commit 81a76d71 upstream. When showing backtraces in response to traps, for example crashes and address errors (usually unaligned accesses) when they are set in debugfs to be reported, unwind_stack will be used if the PC was in the kernel text address range. However since EVA it is possible for user and kernel address ranges to overlap, and even without EVA userland can still trigger an address error by jumping to a KSeg0 address. Adjust the check to also ensure that it was running in kernel mode. I don't believe any harm can come of this problem, since unwind_stack() is sufficiently defensive, however it is only meant for unwinding kernel code, so to be correct it should use the raw backtracing instead. Signed-off-by:
Leonid Yegoshin <Leonid.Yegoshin@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/11701/Signed-off-by:
-
James Hogan authored
commit a816b306 upstream. When unwinding through IRQs and exceptions, the unwinding only continues if the PC is a kernel text address, however since EVA it is possible for user and kernel address ranges to overlap, potentially allowing unwinding to continue to user mode if the user PC happens to be in the kernel text address range. Adjust the check to also ensure that the register state from before the exception is actually running in kernel mode, i.e. !user_mode(regs). I don't believe any harm can come of this problem, since the PC is only output, the stack pointer is checked to ensure it resides within the task's stack page before it is dereferenced in search of the return address, and the return address register is similarly only output (if the PC is in a leaf function or the beginning of a non-leaf function). However unwind_stack() is only meant for unwinding kernel code, so to be correct the unwind should stop there. Signed-off-by:
-
Maciej W. Rozycki authored
commit e49d3848 upstream. Fix a build regression from commit c9017757 ("MIPS: init upper 64b of vector registers when MSA is first used"): arch/mips/built-in.o: In function `enable_restore_fp_context': traps.c:(.text+0xbb90): undefined reference to `_init_msa_upper' traps.c:(.text+0xbb90): relocation truncated to fit: R_MIPS_26 against `_init_msa_upper' traps.c:(.text+0xbef0): undefined reference to `_init_msa_upper' traps.c:(.text+0xbef0): relocation truncated to fit: R_MIPS_26 against `_init_msa_upper' to !CONFIG_CPU_HAS_MSA configurations with older GCC versions, which are unable to figure out that calls to `_init_msa_upper' are indeed dead. Of the many ways to tackle this failure choose the approach we have already taken in `thread_msa_context_live'. [ralf@linux-mips.org: Drop patch segment to junk file.] Signed-off-by:
-
Paul Burton authored
commit ab4a92e6 upstream. When emulating a jalr instruction with rd == $0, the code in isBranchInstr was incorrectly writing to GPR $0 which should actually always remain zeroed. This would lead to any further instructions emulated which use $0 operating on a bogus value until the task is next context switched, at which point the value of $0 in the task context would be restored to the correct zero by a store in SAVE_SOME. Fix this by not writing to rd if it is $0. Fixes: 102cedc3 ("MIPS: microMIPS: Floating point support.") Signed-off-by:
-
Leonid Yegoshin authored
commit 41fa29e4 upstream. Error recovery pointers for fixups was improperly set as ".word" which is unsuitable for MIPS64. Replaced by STR(PTR) [ralf@linux-mips.org: Apply changes as requested in the review process.] Signed-off-by:
Markos Chandras <markos.chandras@imgtec.com> Fixes: b0a668fb ("MIPS: kernel: mips-r2-to-r6-emul: Add R2 emulator for MIPS R6") Cc: macro@linux-mips.org Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/9911/Signed-off-by:
-
Chao Yu authored
commit ab47036d upstream. Below backtrace info was reported by Yunlei He: Call Trace: [<ffffffff817a9395>] schedule+0x35/0x80 [<ffffffff817abb7d>] rwsem_down_read_failed+0xed/0x130 [<ffffffff813c12a8>] call_rwsem_down_read_failed+0x18/0x [<ffffffff817ab1d0>] down_read+0x20/0x30 [<ffffffffa02a1a12>] f2fs_evict_inode+0x242/0x3a0 [f2fs] [<ffffffff81217057>] evict+0xc7/0x1a0 [<ffffffff81217cd6>] iput+0x196/0x200 [<ffffffff812134f9>] __dentry_kill+0x179/0x1e0 [<ffffffff812136f9>] dput+0x199/0x1f0 [<ffffffff811fe77b>] __fput+0x18b/0x220 [<ffffffff811fe84e>] ____fput+0xe/0x10 [<ffffffff81097427>] task_work_run+0x77/0x90 [<ffffffff81074d62>] exit_to_usermode_loop+0x73/0xa2 [<ffffffff81003b7a>] do_syscall_64+0xfa/0x110 [<ffffffff817acf65>] entry_SYSCALL64_slow_path+0x25/0x25 Call Trace: [<ffffffff817a9395>] schedule+0x35/0x80 [<ffffffff81216dc3>] __wait_on_freeing_inode+0xa3/0xd0 [<ffffffff810bc300>] ? autoremove_wake_function+0x40/0x4 [<ffffffff8121771d>] find_inode_fast+0x7d/0xb0 [<ffffffff8121794a>] ilookup+0x6a/0xd0 [<ffffffffa02bc740>] sync_node_pages+0x210/0x650 [f2fs] [<ffffffff8122e690>] ? do_fsync+0x70/0x70 [<ffffffffa02b085e>] block_operations+0x9e/0xf0 [f2fs] [<ffffffff8137b795>] ? bio_endio+0x55/0x60 [<ffffffffa02b0942>] write_checkpoint+0x92/0xba0 [f2fs] [<ffffffff8117da57>] ? mempool_free_slab+0x17/0x20 [<ffffffff8117de8b>] ? mempool_free+0x2b/0x80 [<ffffffff8122e690>] ? do_fsync+0x70/0x70 [<ffffffffa02a53e3>] f2fs_sync_fs+0x63/0xd0 [f2fs] [<ffffffff8129630f>] ? ext4_sync_fs+0xbf/0x190 [<ffffffff8122e6b0>] sync_fs_one_sb+0x20/0x30 [<ffffffff812002e9>] iterate_supers+0xb9/0x110 [<ffffffff8122e7b5>] sys_sync+0x55/0x90 [<ffffffff81003ae9>] do_syscall_64+0x69/0x110 [<ffffffff817acf65>] entry_SYSCALL64_slow_path+0x25/0x25 With following excuting serials, we will set inline_node in inode page after inode was unlinked, result in a deadloop described as below: 1. open file 2. write file 3. unlink file 4. write file 5. close file Thread A Thread B - dput - iput_final - inode->i_state |= I_FREEING - evict - f2fs_evict_inode - f2fs_sync_fs - write_checkpoint - block_operations - f2fs_lock_all (down_write(cp_rwsem)) - f2fs_lock_op (down_read(cp_rwsem)) - sync_node_pages - ilookup - find_inode_fast - __wait_on_freeing_inode (wait on I_FREEING clear) Here, we change to set inline_node flag only for linked inode for fixing. Reported-by:
Yunlei He <heyunlei@huawei.com> Signed-off-by:
Chao Yu <yuchao0@huawei.com> Tested-by:
Jaegeuk Kim <jaegeuk@kernel.org> Signed-off-by:
-
- 01 Jun, 2016 10 commits
-
-
Greg Kroah-Hartman authored
-
Arnd Bergmann authored
commit c9c6837d upstream. gcc-6 started warning by default about variables that are not used anywhere and that are marked 'const', generating many false positives in an allmodconfig build, e.g.: arch/arm/mach-davinci/board-da830-evm.c:282:20: warning: 'da830_evm_emif25_pins' defined but not used [-Wunused-const-variable=] arch/arm/plat-omap/dmtimer.c:958:34: warning: 'omap_timer_match' defined but not used [-Wunused-const-variable=] drivers/bluetooth/hci_bcm.c:625:39: warning: 'acpi_bcm_default_gpios' defined but not used [-Wunused-const-variable=] drivers/char/hw_random/omap-rng.c:92:18: warning: 'reg_map_omap4' defined but not used [-Wunused-const-variable=] drivers/devfreq/exynos/exynos5_bus.c:381:32: warning: 'exynos5_busfreq_int_pm' defined but not used [-Wunused-const-variable=] drivers/dma/mv_xor.c:1139:34: warning: 'mv_xor_dt_ids' defined but not used [-Wunused-const-variable=] This is similar to the existing -Wunused-but-set-variable warning that was added in an earlier release and that we disable by default now and only enable when W=1 is set, so it makes sense to do the same here. Once we have eliminated the majority of the warnings for both, we can put them back into the default list. We probably want this in backport kernels as well, to allow building them with gcc-6 without introducing extra warnings. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Acked-by:
Olof Johansson <olof@lixom.net> Acked-by:
Lee Jones <lee.jones@linaro.org> Signed-off-by:
Michal Marek <mmarek@suse.com> Signed-off-by:
-
Johannes Thumshirn authored
commit 305c2e71 upstream. Now that we've done a more comprehensive fix with the intermediate target state we can remove the previous hack introduced with commit 90a88d6e ("scsi: fix soft lockup in scsi_remove_target() on module removal"). Signed-off-by:
Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by:
Ewan D. Milne <emilne@redhat.com> Reviewed-by:
Hannes Reinecke <hare@suse.com> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by:
-
Johannes Thumshirn authored
commit f05795d3 upstream. Add intermediate STARGET_REMOVE state to scsi_target_state to avoid running into the BUG_ON() in scsi_target_reap(). The STARGET_REMOVE state is only valid in the path from scsi_remove_target() to scsi_target_destroy() indicating this target is going to be removed. This re-fixes the problem introduced in commits bc3f02a7 ("[SCSI] scsi_remove_target: fix softlockup regression on hot remove") and 40998193 ("scsi: restart list search after unlock in scsi_remove_target") in a more comprehensive way. [mkp: Included James' fix for scsi_target_destroy()] Signed-off-by:
Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Tested-by:
James Bottomley <jejb@linux.vnet.ibm.com> Signed-off-by:
-
Mikulas Patocka authored
commit 037369b8 upstream. The HPFS filesystem used generic_show_options to produce string that is displayed in /proc/mounts. However, there is a problem that the options may disappear after remount. If we mount the filesystem with option1 and then remount it with option2, /proc/mounts should show both option1 and option2, however it only shows option2 because the whole option string is replaced with replace_mount_options in hpfs_remount_fs. To fix this bug, implement the hpfs_show_options function that prints options that are currently selected. Signed-off-by:
-
Mikulas Patocka authored
commit 44d51706 upstream. Commit ce657611 ("hpfs: kstrdup() out of memory handling") checks if the kstrdup function returns NULL due to out-of-memory condition. However, if we are remounting a filesystem with no change to filesystem-specific options, the parameter data is NULL. In this case, kstrdup returns NULL (because it was passed NULL parameter), although no out of memory condition exists. The mount syscall then fails with ENOMEM. This patch fixes the bug. We fail with ENOMEM only if data is non-NULL. The patch also changes the call to replace_mount_options - if we didn't pass any filesystem-specific options, we don't call replace_mount_options (thus we don't erase existing reported options). Fixes: ce657611 ("hpfs: kstrdup() out of memory handling") Signed-off-by:
-
Richard Weinberger authored
commit 1900149c upstream. Ezequiel reported that he's facing UBI going into read-only mode after power cut. It turned out that this behavior happens only when updating a static volume is interrupted and Fastmap is used. A possible trace can look like: ubi0 warning: ubi_io_read_vid_hdr [ubi]: no VID header found at PEB 2323, only 0xFF bytes ubi0 warning: ubi_eba_read_leb [ubi]: switch to read-only mode CPU: 0 PID: 833 Comm: ubiupdatevol Not tainted 4.6.0-rc2-ARCH #4 Hardware name: SAMSUNG ELECTRONICS CO., LTD. 300E4C/300E5C/300E7C/NP300E5C-AD8AR, BIOS P04RAP 10/15/2012 0000000000000286 00000000eba949bd ffff8800c45a7b38 ffffffff8140d841 ffff8801964be000 ffff88018eaa4800 ffff8800c45a7bb8 ffffffffa003abf6 ffffffff850e2ac0 8000000000000163 ffff8801850e2ac0 ffff8801850e2ac0 Call Trace: [<ffffffff8140d841>] dump_stack+0x63/0x82 [<ffffffffa003abf6>] ubi_eba_read_leb+0x486/0x4a0 [ubi] [<ffffffffa00453b3>] ubi_check_volume+0x83/0xf0 [ubi] [<ffffffffa0039d97>] ubi_open_volume+0x177/0x350 [ubi] [<ffffffffa00375d8>] vol_cdev_open+0x58/0xb0 [ubi] [<ffffffff8124b08e>] chrdev_open+0xae/0x1d0 [<ffffffff81243bcf>] do_dentry_open+0x1ff/0x300 [<ffffffff8124afe0>] ? cdev_put+0x30/0x30 [<ffffffff81244d36>] vfs_open+0x56/0x60 [<ffffffff812545f4>] path_openat+0x4f4/0x1190 [<ffffffff81256621>] do_filp_open+0x91/0x100 [<ffffffff81263547>] ? __alloc_fd+0xc7/0x190 [<ffffffff812450df>] do_sys_open+0x13f/0x210 [<ffffffff812451ce>] SyS_open+0x1e/0x20 [<ffffffff81a99e32>] entry_SYSCALL_64_fastpath+0x1a/0xa4 UBI checks static volumes for data consistency and reads the whole volume upon first open. If the volume is found erroneous users of UBI cannot read from it, but another volume update is possible to fix it. The check is performed by running ubi_eba_read_leb() on every allocated LEB of the volume. For static volumes ubi_eba_read_leb() computes the checksum of all data stored in a LEB. To verify the computed checksum it has to read the LEB's volume header which stores the original checksum. If the volume header is not found UBI treats this as fatal internal error and switches to RO mode. If the UBI device was attached via a full scan the assumption is correct, the volume header has to be present as it had to be there while scanning to get known as mapped. If the attach operation happened via Fastmap the assumption is no longer correct. When attaching via Fastmap UBI learns the mapping table from Fastmap's snapshot of the system state and not via a full scan. It can happen that a LEB got unmapped after a Fastmap was written to the flash. Then UBI can learn the LEB still as mapped and accessing it returns only 0xFF bytes. As UBI is not a FTL it is allowed to have mappings to empty PEBs, it assumes that the layer above takes care of LEB accounting and referencing. UBIFS does so using the LEB property tree (LPT). For static volumes UBI blindly assumes that all LEBs are present and therefore special actions have to be taken. The described situation can happen when updating a static volume is interrupted, either by a user or a power cut. The volume update code first unmaps all LEBs of a volume and then writes LEB by LEB. If the sequence of operations is interrupted UBI detects this either by the absence of LEBs, no volume header present at scan time, or corrupted payload, detected via checksum. In the Fastmap case the former method won't trigger as no scan happened and UBI automatically thinks all LEBs are present. Only by reading data from a LEB it detects that the volume header is missing and incorrectly treats this as fatal error. To deal with the situation ubi_eba_read_leb() from now on checks whether we attached via Fastmap and handles the absence of a volume header like a data corruption error. This way interrupted static volume updates will correctly get detected also when Fastmap is used. Reported-by:
Ezequiel Garcia <ezequiel@vanguardiasur.com.ar> Tested-by:
Richard Weinberger <richard@nod.at> Signed-off-by:
-
Chris Mason authored
commit 56244ef1 upstream. When btrfs_copy_from_user isn't able to copy all of the pages, we need to adjust our accounting to reflect the work that was actually done. Commit 2e78c927 changed around the decisions a little and we ended up skipping the accounting adjustments some of the time. This commit makes sure that when we don't copy anything at all, we still hop into the adjustments, and switches to release_bytes instead of write_bytes, since write_bytes isn't aligned. The accounting errors led to warnings during btrfs_destroy_inode: [ 70.847532] WARNING: CPU: 10 PID: 514 at fs/btrfs/inode.c:9350 btrfs_destroy_inode+0x2b3/0x2c0 [ 70.847536] Modules linked in: i2c_piix4 virtio_net i2c_core input_leds button led_class serio_raw acpi_cpufreq sch_fq_codel autofs4 virtio_blk [ 70.847538] CPU: 10 PID: 514 Comm: umount Tainted: G W 4.6.0-rc6_00062_g2997da1-dirty #23 [ 70.847539] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.0-1.fc24 04/01/2014 [ 70.847542] 0000000000000000 ffff880ff5cafab8 ffffffff8149d5e9 0000000000000202 [ 70.847543] 0000000000000000 0000000000000000 0000000000000000 ffff880ff5cafb08 [ 70.847547] ffffffff8107bdfd ffff880ff5cafaf8 000024868120013d ffff880ff5cafb28 [ 70.847547] Call Trace: [ 70.847550] [<ffffffff8149d5e9>] dump_stack+0x51/0x78 [ 70.847551] [<ffffffff8107bdfd>] __warn+0xfd/0x120 [ 70.847553] [<ffffffff8107be3d>] warn_slowpath_null+0x1d/0x20 [ 70.847555] [<ffffffff8139c9e3>] btrfs_destroy_inode+0x2b3/0x2c0 [ 70.847556] [<ffffffff812003a1>] ? __destroy_inode+0x71/0x140 [ 70.847558] [<ffffffff812004b3>] destroy_inode+0x43/0x70 [ 70.847559] [<ffffffff810b7b5f>] ? wake_up_bit+0x2f/0x40 [ 70.847560] [<ffffffff81200c68>] evict+0x148/0x1d0 [ 70.847562] [<ffffffff81398ade>] ? start_transaction+0x3de/0x460 [ 70.847564] [<ffffffff81200d49>] dispose_list+0x59/0x80 [ 70.847565] [<ffffffff81201ba0>] evict_inodes+0x180/0x190 [ 70.847566] [<ffffffff812191ff>] ? __sync_filesystem+0x3f/0x50 [ 70.847568] [<ffffffff811e95f8>] generic_shutdown_super+0x48/0x100 [ 70.847569] [<ffffffff810b75c0>] ? woken_wake_function+0x20/0x20 [ 70.847571] [<ffffffff811e9796>] kill_anon_super+0x16/0x30 [ 70.847573] [<ffffffff81365cde>] btrfs_kill_super+0x1e/0x130 [ 70.847574] [<ffffffff811e99be>] deactivate_locked_super+0x4e/0x90 [ 70.847576] [<ffffffff811e9e61>] deactivate_super+0x51/0x70 [ 70.847577] [<ffffffff8120536f>] cleanup_mnt+0x3f/0x80 [ 70.847579] [<ffffffff81205402>] __cleanup_mnt+0x12/0x20 [ 70.847581] [<ffffffff81098358>] task_work_run+0x68/0xa0 [ 70.847582] [<ffffffff810022b6>] exit_to_usermode_loop+0xd6/0xe0 [ 70.847583] [<ffffffff81002e1d>] do_syscall_64+0xbd/0x170 [ 70.847586] [<ffffffff817d4dbc>] entry_SYSCALL64_slow_path+0x25/0x25 This is the test program I used to force short returns from btrfs_copy_from_user void *dontneed(void *arg) { char *p = arg; int ret; while(1) { ret = madvise(p, BUFSIZE/4, MADV_DONTNEED); if (ret) { perror("madvise"); exit(1); } } } int main(int ac, char **av) { int ret; int fd; char *filename; unsigned long offset; char *buf; int i; pthread_t tid; if (ac != 2) { fprintf(stderr, "usage: dammitdave filename\n"); exit(1); } buf = mmap(NULL, BUFSIZE, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0); if (buf == MAP_FAILED) { perror("mmap"); exit(1); } memset(buf, 'a', BUFSIZE); filename = av[1]; ret = pthread_create(&tid, NULL, dontneed, buf); if (ret) { fprintf(stderr, "error %d from pthread_create\n", ret); exit(1); } ret = pthread_detach(tid); if (ret) { fprintf(stderr, "pthread detach failed %d\n", ret); exit(1); } while (1) { fd = open(filename, O_RDWR | O_CREAT, 0600); if (fd < 0) { perror("open"); exit(1); } for (i = 0; i < ROUNDS; i++) { int this_write = BUFSIZE; offset = rand() % MAXSIZE; ret = pwrite(fd, buf, this_write, offset); if (ret < 0) { perror("pwrite"); exit(1); } else if (ret != this_write) { fprintf(stderr, "short write to %s offset %lu ret %d\n", filename, offset, ret); exit(1); } if (i == ROUNDS - 1) { ret = sync_file_range(fd, offset, 4096, SYNC_FILE_RANGE_WRITE); if (ret < 0) { perror("sync_file_range"); exit(1); } } } ret = ftruncate(fd, 0); if (ret < 0) { perror("ftruncate"); exit(1); } ret = close(fd); if (ret) { perror("close"); exit(1); } ret = unlink(filename); if (ret) { perror("unlink"); exit(1); } } return 0; } Signed-off-by:
Chris Mason <clm@fb.com> Reported-by:
Dave Jones <dsj@fb.com> Fixes: 2e78c927Signed-off-by:
-
Luke Dashjr authored
commit 4c63c245 upstream. 32-bit ioctl uses these rather than the regular FS_IOC_* versions. They can be handled in btrfs using the same code. Without this, 32-bit {ch,ls}attr fail. Signed-off-by:
Luke Dashjr <luke-jr+git@utopios.org> Reviewed-by:
Josef Bacik <jbacik@fb.com> Reviewed-by:
David Sterba <dsterba@suse.com> Signed-off-by:
-
James Hogan authored
commit ca9eb49a upstream. The generic copy_siginfo() is currently defined in asm-generic/siginfo.h, after including uapi/asm-generic/siginfo.h which defines the generic struct siginfo. However this makes it awkward for an architecture to use it if it has to define its own struct siginfo (e.g. MIPS and potentially IA64), since it means that asm-generic/siginfo.h can only be included after defining the arch-specific siginfo, which may be problematic if the arch-specific definition needs definitions from uapi/asm-generic/siginfo.h. It is possible to work around this by first including uapi/asm-generic/siginfo.h to get the constants before defining the arch-specific siginfo, and include asm-generic/siginfo.h after. However uapi headers can't be included by other uapi headers, so that first include has to be in an ifdef __kernel__, with the non __kernel__ case including the non-UAPI header instead. Instead of that mess, move the generic copy_siginfo() definition into linux/signal.h, which allows an arch-specific uapi/asm/siginfo.h to include asm-generic/siginfo.h and define the arch-specific siginfo, and for the generic copy_siginfo() to see that arch-specific definition. Signed-off-by:
-
