summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* s390/oprofile: fix compile errorSebastian Ott2015-07-012-0/+9
| | | | | | | | | | | | | | | Fix these errors when compiling with CONFIG_OPROFILE=y and CONFIG_PERF_EVENTS=n: arch/s390/oprofile/init.c: In function ‘oprofile_hwsampler_start’: arch/s390/oprofile/init.c:93:2: error: implicit declaration of function 'perf_reserve_sampling' [-Werror=implicit-function-declaration] retval = perf_reserve_sampling(); ^ arch/s390/oprofile/init.c:99:3: error: implicit declaration of function 'perf_release_sampling' [-Werror=implicit-function-declaration] perf_release_sampling(); ^ Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/sclp: fix compile errorSebastian Ott2015-07-011-0/+1
| | | | | | | | | | | | | Fix this error when compiling with CONFIG_SMP=n and CONFIG_DYNAMIC_DEBUG=y: drivers/s390/char/sclp_early.c: In function 'sclp_read_info_early': drivers/s390/char/sclp_early.c:87:19: error: 'EBUSY' undeclared (first use in this function) } while (rc == -EBUSY); ^ Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/zcrypt: enable s390 hwrng to seed kernel entropyIngo Tuchscherer2015-07-011-0/+7
| | | | | | | | | | | Set the 'quality' property in the zcrypt rng device structure to enable the zcrypt hwrng device to take part in the kernel entropy seeding process. A module parameter named hwrng_seed will be introduced to disable the participation. By default this parameter is set to 1 (enabled). Signed-off-by: Ingo Tuchscherer <ingo.tuchscherer@de.ibm.com> Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tileLinus Torvalds2015-07-0127-254/+402
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull arch/tile updates from Chris Metcalf: "These are a grab bag of changes to improve debugging and respond to a variety of issues raised on LKML over the last couple of months" * git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile: tile: avoid a "label not used" warning in do_page_fault() tile: vdso: use raw_read_seqcount_begin() in vdso tile: force CONFIG_TILEGX if ARCH != tilepro tile: improve stack backtrace tile: fix "odd fault" warning for stack backtraces tile: set up initial stack top to honor STACK_TOP_DELTA tile: support delivering NMIs for multicore backtrace drivers/tty/hvc/hvc_tile.c: properly return -EAGAIN tile: add <asm/word-at-a-time.h> and enable support functions tile: use READ_ONCE() in arch_spin_is_locked() tile: modify arch_spin_unlock_wait() semantics
| * tile: avoid a "label not used" warning in do_page_fault()Chris Metcalf2015-06-041-7/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | There are two different ifdef cases where the label is used, but if neither is true, the label is unused and the compiler generates a warning. Refactor the code the way x86 does so that there is a do_page_fault() that just does exception handling for context tracking, and make __do_page_fault() a static inline so that various cases can just return instead of doing a jump to "done". Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: vdso: use raw_read_seqcount_begin() in vdsoChris Metcalf2015-06-041-5/+5
| | | | | | | | | | | | | | | | | | Previously we were using read_seqcount_begin(), which works fine until lockdep is enabled in the kernel, at which point lockdep locking shows up in the vdso and userspace will take a GPV accessing a kernel-only SPR when calling gettimeofday() etc. Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: force CONFIG_TILEGX if ARCH != tileproChris Metcalf2015-06-031-1/+3
| | | | | | | | | | | | | | | | This allows configuring with allnoconfig (for tilegx) or allyesconfig (for tilepro) without creating an unbuildable configuration. Suggested-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: improve stack backtraceChris Metcalf2015-05-119-82/+107
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit fixes a number of issues with the tile backtrace code. - Don't try to identify userspace shared object or executable paths if we are doing a backtrace from an interrupt; it's not legal, and also unlikely to be interesting. Likewise, don't try to do it for other address spaces, since d_path() assumes it is being called in "current" context. - Move "in_backtrace" from thread_struct to thread_info. This way we can access it even if our stack thread_info has been clobbered, which makes backtracing more robust. - Avoid using "current" directly when testing for is_sigreturn(). Since "current" may be corrupt, we're better off using kbt->task explicitly to look up the vdso_base for the current task. Conveniently, this simplifies the internal APIs (we only need one is_sigreturn() function now). - Avoid bogus "Odd fault" warning when pc/sp/ex1 are all zero, as is true for kernel threads above the last frame. - Hook into Tejun Heo's dump_stack() framework in lib/dump_stack.c. - Write last entry in save_stack_trace() as ULONG_MAX, not zero, since ftrace (at least) relies on finding that marker. - Implement save_stack_trace_regs() and save_strack_trace_user(), and set CONFIG_USER_STACKTRACE_SUPPORT. Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: fix "odd fault" warning for stack backtracesChris Metcalf2015-05-111-1/+0
| | | | | | | | | | | | | | | | | | We were setting ex1 in new kernel threads to KERNEL_PL. But since we just do a simple context-switch, not an iret, any value set here is ignored anyway, and its presence causes stack backtraces to end with a warning about an "odd fault". Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: set up initial stack top to honor STACK_TOP_DELTAChris Metcalf2015-05-111-1/+1
| | | | | | | | | | | | | | | | | | | | For some reason this was never changed to match the rest of the code where we always initialize the kernel sp 64 bytes below the top of the page. This is generally harmless, but it does mean that if you do a dump_stack() early on in kernel boot you see a bogus warning about stack overrun. Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: support delivering NMIs for multicore backtraceChris Metcalf2015-05-118-2/+197
| | | | | | | | | | | | | | | | | | | | A new hypervisor service was added some time ago (MDE 4.2.1 or later, or MDE 4.3 or later) that allows cores to request NMIs to be delivered to other cores. Use this facility to deliver a request that causes a backtrace to be generated on each core, and hook it into the magic SysRq functionality. Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * drivers/tty/hvc/hvc_tile.c: properly return -EAGAINChris Metcalf2015-05-111-1/+2
| | | | | | | | | | | | | | | | | | | | We were returning zero if no bytes could be written to the Tilera hypervisor console device, but this causes the output to be truncated. By returning -EAGAIN the tty hvc driver will come back and try again, which gives the semantics we want, and avoids dropping console output. Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: add <asm/word-at-a-time.h> and enable support functionsChris Metcalf2015-04-306-150/+48
| | | | | | | | | | | | | | | | | | | | This change enables the generic strncpy_from_user() and strnlen_user() using word-at-a-time.h. The tile implementation is trivial since both tilepro and tilegx have SIMD operations that do byte-wise comparisons against immediate zero for each byte, and return an 0x01 byte in each position where there is a 0x00 byte. Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: use READ_ONCE() in arch_spin_is_locked()Chris Metcalf2015-04-292-2/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This avoid potential issues if callers were to loop on these routines without some kind of memory barrier. Currently there are no such users in-tree, but it seems better safe than sorry. Also, in the tilepro case we read "current" before "next", which gives us a slightly better guarantee that the lock was actually unlocked at least momentarily if we return claiming that it is not locked. None of the callers actually rely on this behavior, as far as I know, however. Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
| * tile: modify arch_spin_unlock_wait() semanticsChris Metcalf2015-04-292-2/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | Rather than trying to wait until all possible lockers have unlocked the lock, we now only wait until the current locker (if any) has released the lock. The old code was correct, but the new code works more like the x86 code and thus hopefully is more appropriate under contention. See commit 78bff1c8684f ("x86/ticketlock: Fix spin_unlock_wait() livelock") for x86. Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
* | Merge branch 'for-linus' of ↵Linus Torvalds2015-07-0119-357/+401
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux Pull more s390 updates from Martin Schwidefsky: "There is one larger patch for the AP bus code to make it work with the longer reset periods of the latest crypto cards. A new default configuration, a naming cleanup for SMP and a few fixes" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: s390/kdump: fix compile for !SMP s390/kdump: fix nosmt kernel parameter s390: new default configuration s390/smp: cleanup core vs. cpu in the SCLP interface s390/smp: fix sigp cpu detection loop s390/zcrypt: Fixed reset and interrupt handling of AP queues s390/kdump: fix REGSET_VX_LOW vector register ELF notes s390/bpf: Fix backward jumps
| * | s390/kdump: fix compile for !SMPHeiko Carstens2015-06-292-23/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix this compile error: arch/s390/kernel/setup.c:875:2: error: implicit declaration of function 'smp_save_dump_cpus' Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | s390/kdump: fix nosmt kernel parameterMichael Holzheu2015-06-257-61/+64
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It turned out that SIGP set-multi-threading can only be done once. Therefore switching to a different MT level after switching to sclp.mtid_prev in the dump case fails. As a symptom specifying the "nosmt" parameter currently fails for the kdump kernel and the kernel starts with multi-threading enabled. So fix this and issue diag 308 subcode 1 call after collecting the CPU states for the dump. Also enhance the diag308_reset() function to be usable also with enabled lowcore protection and prefix register != 0. After the reset it is possible to switch the MT level again. We have to do the reset very early in order not to kill the already initialized console. Therefore instead of kmalloc() the corresponding memblock functions have to be used. To avoid copying the sclp cpu code into sclp_early, we now use the simple sigp loop method for CPU detection. Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | s390: new default configurationHeiko Carstens2015-06-254-26/+30
| | | | | | | | | | | | | | | Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | s390/smp: cleanup core vs. cpu in the SCLP interfaceMartin Schwidefsky2015-06-255-52/+50
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The SCLP interface to query, configure and deconfigure CPUs actually operates on cores. For a machine without the multi-threading faciltiy a CPU and a core are equivalent but starting with System z13 a core can have multiple hardware threads, also referred to as logical CPUs. To avoid confusion replace the word 'cpu' with 'core' in the SCLP interface. Also replace MAX_CPU_ADDRESS with SCLP_MAX_CORES. The core-id is an 8-bit field, the maximum thread id is in the range 0-31. The theoretical limit for the CPU address is therefore 8191. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | s390/smp: fix sigp cpu detection loopMartin Schwidefsky2015-06-251-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | On a (theoretical) system where the read-cpu-info SCLP command does not work but SMT is enabled, the sigp detection loop may not find all configured cores. The maximum CPU address needs to be shifted with smp_cpu_mt_shift. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | s390/zcrypt: Fixed reset and interrupt handling of AP queuesIngo Tuchscherer2015-06-253-200/+239
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In case of request timeouts an AP queue reset will be triggered to recover and reinitialize the AP queue. The previous behavior was an immediate reset execution regardless of current/pending requests. Due to newly changed firmware behavior the reset may be delayed, based on the priority of pending request. The device driver's waiting time frame was limited, hence it did not received the reset response. As a consequence interrupts would not be enabled afterwards. The RAPQ (queue reset) and AQIC (interrupt control) commands will be treated fully asynchronous now. The device driver will check the reset and interrupt states periodically, thus it can handle the reinitialization properly. Signed-off-by: Ingo Tuchscherer <ingo.tuchscherer@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | s390/kdump: fix REGSET_VX_LOW vector register ELF notesMichael Holzheu2015-06-251-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The REGSET_VX_LOW ELF notes should contain the lower 64 bit halfes of the first sixteen 128 bit vector registers. Unfortunately currently we copy the upper halfes. Fix this and correctly copy the lower halfes. Fixes: a62bc0739253 ("s390/kdump: add support for vector extension") Cc: stable@vger.kernel.org # 3.18+ Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
| * | s390/bpf: Fix backward jumpsMichael Holzheu2015-06-251-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently all backward jumps crash for JITed s390x eBPF programs with an illegal instruction program check and kernel panic. Because for negative values the opcode of the jump instruction is overriden by the negative branch offset an illegal instruction is generated by the JIT: 000003ff802da378: c01100000002 lgfi %r1,2 000003ff802da37e: fffffff52065 unknown <-- illegal instruction 000003ff802da384: b904002e lgr %r2,%r14 So fix this and mask the offset in order not to damage the opcode. Cc: stable@vger.kernel.org # 4.0+ Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* | | Merge branch 'for-next' of git://git.samba.org/sfrench/cifs-2.6Linus Torvalds2015-07-0110-18/+392
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull CIFS/SMB3 updates from Steve French: "Includes two bug fixes, as well as (minimal) support for the new protocol dialect (SMB3.1.1), and support for two ioctls including reflink (duplicate extents) file copy and set integrity" * 'for-next' of git://git.samba.org/sfrench/cifs-2.6: cifs: Unset CIFS_MOUNT_POSIX_PATHS flag when following dfs mounts Update negotiate protocol for SMB3.11 dialect Add ioctl to set integrity Add Get/Set Integrity Information structure definitions Add reflink copy over SMB3.11 with new FSCTL_DUPLICATE_EXTENTS Add SMB3.11 mount option synonym for new dialect add struct FILE_STANDARD_INFO Make dialect negotiation warning message easier to read Add defines and structs for smb3.1 dialect Allow parsing vers=3.11 on cifs mount client MUST ignore EncryptionKeyLength if CAP_EXTENDED_SECURITY is set
| * | | cifs: Unset CIFS_MOUNT_POSIX_PATHS flag when following dfs mountsSachin Prabhu2015-06-291-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In a dfs setup where the client transitions from a server which supports posix paths to a server which doesn't support posix paths, the flag CIFS_MOUNT_POSIX_PATHS is not reset. This leads to the wrong directory separator being used causing smb commands to fail. Consider the following case where a dfs share on a samba server points to a share on windows smb server. # mount -t cifs -o .. //vm140-31/dfsroot/testwin/ # ls -l /mnt; touch /mnt/a total 0 touch: cannot touch ‘/mnt/a’: No such file or directory Signed-off-by: Sachin Prabhu <sprabhu@redhat.com> Acked-by: Jeff Layton <jlayton@samba.org> Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | Update negotiate protocol for SMB3.11 dialectSteve French2015-06-292-5/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Send negotiate contexts when SMB3.11 dialect is negotiated (ie the preauth and the encryption contexts) and Initialize SMB3.11 preauth negotiate context salt to random bytes Followon patch will update session setup and tree connect Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | Add ioctl to set integritySteve French2015-06-294-0/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | set integrity increases reliability of files stored on SMB3 servers. Add ioctl to allow setting this on files on SMB3 and later mounts. Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | Add Get/Set Integrity Information structure definitionsSteve French2015-06-292-0/+24
| | | | | | | | | | | | | | | | Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | Add reflink copy over SMB3.11 with new FSCTL_DUPLICATE_EXTENTSSteve French2015-06-296-3/+75
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Getting fantastic copy performance with cp --reflink over SMB3.11 using the new FSCTL_DUPLICATE_EXTENTS. This FSCTL was added in the SMB3.11 dialect (testing was against REFS file system) so have put it as a 3.11 protocol specific operation ("vers=3.1.1" on the mount). Tested at the SMB3 plugfest in Redmond. It depends on the new FS Attribute (BLOCK_REFCOUNTING) which is used to advertise support for the ability to do this ioctl (if you can support multiple files pointing to the same block than this refcounting ability or equivalent is needed to support the new reflink-like duplicate extent SMB3 ioctl. Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | Add SMB3.11 mount option synonym for new dialectSteve French2015-06-283-2/+90
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Most people think of SMB 3.1.1 as SMB version 3.11 so add synonym for "vers=3.1.1" of "vers=3.11" on mount. Also make sure that unlike SMB3.0 and 3.02 we don't send validate negotiate on mount (it is handled by negotiate contexts) - add list of SMB3.11 specific functions (distinct from 3.0 dialect). Signed-off-by: Steve French <steve.french@primarydata.com>w
| * | | add struct FILE_STANDARD_INFOSteve French2015-06-281-0/+10
| | | | | | | | | | | | | | | | | | | | Signed-off-by: Gregor Beck <gbeck@sernet.de> Reviewed-by: Jeff Layton <jlayton@primarydata.com>
| * | | Make dialect negotiation warning message easier to readSteve French2015-06-281-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Dialect version and minor version are easier to read in hex Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | Add defines and structs for smb3.1 dialectSteve French2015-06-282-6/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add new structures and defines for SMB3.11 negotiate, session setup and tcon See MS-SMB2-diff.pdf section 2.2.3 for additional protocol documentation. Reviewed-by: Jeff Layton <jlayton@primarydata.com> Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | Allow parsing vers=3.11 on cifs mountSteve French2015-06-286-3/+59
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Parses and recognizes "vers=3.1.1" on cifs mount and allows sending 0x0311 as a new CIFS/SMB3 dialect. Subsequent patches will add the new negotiate contexts and updated session setup Reviewed-by: Jeff Layton <jlayton@primarydata.com> Signed-off-by: Steve French <steve.french@primarydata.com>
| * | | client MUST ignore EncryptionKeyLength if CAP_EXTENDED_SECURITY is setNoel Power2015-06-281-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [MS-SMB] 2.2.4.5.2.1 states: "ChallengeLength (1 byte): When the CAP_EXTENDED_SECURITY bit is set, the server MUST set this value to zero and clients MUST ignore this value." Signed-off-by: Noel Power <noel.power@suse.com> Signed-off-by: Steve French <smfrench@gmail.com>
* | | | Merge tag 'xfs-for-linus-4.2-rc1' of ↵Linus Torvalds2015-07-0160-868/+1611
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs Pul xfs updates from Dave Chinner: "There's a couple of small API changes to the core DAX code which required small changes to the ext2 and ext4 code bases, but otherwise everything is within the XFS codebase. This update contains: - A new sparse on-disk inode record format to allow small extents to be used for inode allocation when free space is fragmented. - DAX support. This includes minor changes to the DAX core code to fix problems with lock ordering and bufferhead mapping abuse. - transaction commit interface cleanup - removal of various unnecessary XFS specific type definitions - cleanup and optimisation of freelist preparation before allocation - various minor cleanups - bug fixes for - transaction reservation leaks - incorrect inode logging in unwritten extent conversion - mmap lock vs freeze ordering - remote symlink mishandling - attribute fork removal issues" * tag 'xfs-for-linus-4.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs: (49 commits) xfs: don't truncate attribute extents if no extents exist xfs: clean up XFS_MIN_FREELIST macros xfs: sanitise error handling in xfs_alloc_fix_freelist xfs: factor out free space extent length check xfs: xfs_alloc_fix_freelist() can use incore perag structures xfs: remove xfs_caddr_t xfs: use void pointers in log validation helpers xfs: return a void pointer from xfs_buf_offset xfs: remove inst_t xfs: remove __psint_t and __psunsigned_t xfs: fix remote symlinks on V5/CRC filesystems xfs: fix xfs_log_done interface xfs: saner xfs_trans_commit interface xfs: remove the flags argument to xfs_trans_cancel xfs: pass a boolean flag to xfs_trans_free_items xfs: switch remaining xfs_trans_dup users to xfs_trans_roll xfs: check min blks for random debug mode sparse allocations xfs: fix sparse inodes 32-bit compile failure xfs: add initial DAX support xfs: add DAX IO path support ...
| * \ \ \ Merge branch 'xfs-misc-fixes-for-4.2-3' into for-nextDave Chinner2015-06-2315-91/+77
| |\ \ \ \
| | * | | | xfs: don't truncate attribute extents if no extents existBrian Foster2015-06-231-2/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The xfs_attr3_root_inactive() call from xfs_attr_inactive() assumes that attribute blocks exist to invalidate. It is possible to have an attribute fork without extents, however. Consider the case where the attribute fork is created towards the beginning of xfs_attr_set() but some part of the subsequent attribute set fails. If an inode in such a state hits xfs_attr_inactive(), it eventually calls xfs_dabuf_map() and possibly xfs_bmapi_read(). The former emits a filesystem corruption warning, returns an error that bubbles back up to xfs_attr_inactive(), and leads to destruction of the in-core attribute fork without an on-disk reset. If the inode happens to make it back through xfs_inactive() in this state (e.g., via a concurrent bulkstat that cycles the inode from the reclaim state and releases it), i_afp might not exist when xfs_bmapi_read() is called and causes a NULL dereference panic. A '-p 2' fsstress run to ENOSPC on a relatively small fs (1GB) reproduces these problems. The behavior is a regression caused by: 6dfe5a0 xfs: xfs_attr_inactive leaves inconsistent attr fork state behind ... which removed logic that avoided the attribute extent truncate when no extents exist. Restore this logic to ensure the attribute fork is destroyed and reset correctly if it exists without any allocated extents. cc: stable@vger.kernel.org # 3.12 to 4.0.x Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | xfs: remove xfs_caddr_tChristoph Hellwig2015-06-223-30/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Just use char pointers directly instead of the confusing typedef to a pointer type. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | xfs: use void pointers in log validation helpersChristoph Hellwig2015-06-222-17/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Compared to char pointers this saves us a lot of casting effort. Also add another local variable to make the code easier to read. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | xfs: return a void pointer from xfs_buf_offsetChristoph Hellwig2015-06-226-17/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This avoids all kinds of unessecary casts in an envrionment like Linux where we can assume that pointer arithmetics are support on void pointers. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | xfs: remove inst_tChristoph Hellwig2015-06-223-6/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We can simply use a void pointer to pass a long return addresses in the debugging helpers. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | xfs: remove __psint_t and __psunsigned_tChristoph Hellwig2015-06-224-20/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Replace uses of __psint_t with the proper uintptr_t and ptrdiff_t types, and remove the defintions of __psint_t and __psunsigned_t. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | xfs: fix remote symlinks on V5/CRC filesystemsEric Sandeen2015-06-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we create a CRC filesystem, mount it, and create a symlink with a path long enough that it can't live in the inode, we get a very strange result upon remount: # ls -l mnt total 4 lrwxrwxrwx. 1 root root 929 Jun 15 16:58 link -> XSLM XSLM is the V5 symlink block header magic (which happens to be followed by a NUL, so the string looks terminated). xfs_readlink_bmap() advanced cur_chunk by the size of the header for CRC filesystems, but never actually used that pointer; it kept reading from bp->b_addr, which is the start of the block, rather than the start of the symlink data after the header. Looks like this problem goes back to v3.10. Fixing this gets us reading the proper link target, again. Cc: stable@vger.kernel.org Signed-off-by: Eric Sandeen <sandeen@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| * | | | | Merge branch 'xfs-freelist-cleanup' into for-nextDave Chinner2015-06-237-130/+142
| |\ \ \ \ \
| | * | | | | xfs: clean up XFS_MIN_FREELIST macrosDave Chinner2015-06-227-21/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We no longer calculate the minimum freelist size from the on-disk AGF, so we don't need the macros used for this. That means the nested macros can be cleaned up, and turn this into an actual function so the logic is clear and concise. This will make it much easier to add support for the rmap btree when the time comes. This also gets rid of the XFS_AG_MAXLEVELS macro used by these freelist macros as it is simply a wrapper around a single variable. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | | xfs: sanitise error handling in xfs_alloc_fix_freelistDave Chinner2015-06-221-56/+55
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The error handling is currently an inconsistent mess as every error condition handles return values and releasing buffers individually. Clean this up by using gotos and a sane error label stack. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | | xfs: factor out free space extent length checkDave Chinner2015-06-221-27/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The longest extent length checks in xfs_alloc_fix_freelist() are now essentially identical. Factor them out into a helper function, so we know they are checking exactly the same thing before and after we lock the AGF. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
| | * | | | | xfs: xfs_alloc_fix_freelist() can use incore perag structuresDave Chinner2015-06-224-46/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | At the moment, xfs_alloc_fix_freelist() uses a mix of per-ag based access and agf buffer based access to freelist and space usage information. However, once the AGF buffer is locked inside this function, it is guaranteed that both the in-memory and on-disk values are identical. xfs_alloc_fix_freelist() doesn't modify the values in the structures directly, so it is a read-only user of the infomration, and hence can use the per-ag structure exclusively for determining what it should do. This opens up an avenue for cleaning up a lot of duplicated logic whose only difference is the structure it gets the data from, and in doing so removes a lot of needless byte swapping overhead when fixing up the free list. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-26 15:54:38 +0100