- 07 Apr, 2009 23 commits
-
-
Markus Metzger authored
Add debug store support for Core i7. Core i7 adds a reset value for each performance counter and a new PEBS record format. Signed-off-by:
Markus Metzger <markus.t.metzger@intel.com> Cc: roland@redhat.com Cc: eranian@googlemail.com Cc: oleg@redhat.com Cc: juan.villacis@intel.com Cc: ak@linux.jf.intel.com LKML-Reference: <20090403144607.088997000@intel.com> Signed-off-by:
Ingo Molnar <mingo@elte.hu>
-
Markus Metzger authored
Check the buffer size more precisely to allow buffers for exactly one element provided the base address is already properly aligned. Add a debug store selftest. Reported-by:
Stephane Eranian <eranian@googlemail.com> Signed-off-by:
-
Markus Metzger authored
Ds_reset_pebs() passed the wrong qualifier to a shared function resulting in a reset of bts, rather than pebs. Reported-by:
-
Markus Metzger authored
Debug store already uses TIF_DS_AREA_MSR to trigger debug store context switch handling. No need to use TIF_DEBUGCTLMSR, as well. Signed-off-by:
-
Markus Metzger authored
Add the ptrace bts context field to task_struct unconditionally. Initialize the field directly in copy_process(). Remove all the unneeded functionality used to initialize that field. Signed-off-by:
-
Markus Metzger authored
Use a single configuration for all cpus. Reported-by:
-
Markus Metzger authored
Add a warning in case a debug store context is not removed before the task it is attached to is freed. Remove the old warning at thread exit. It is too early. Declare the debug store context field in thread_struct unconditionally. Remove ds_copy_thread() and ds_exit_thread() and do the work directly in process*.c. Signed-off-by:
-
Markus Metzger authored
Add selftests to cover per-task branch tracing. Signed-off-by:
-
Markus Metzger authored
Perform debug store selftests on each cpu. Cover both the normal and the _noirq variant of the debug store interface. Signed-off-by:
-
Markus Metzger authored
Fix a bad bounds check in the debug store selftest. Signed-off-by:
-
Markus Metzger authored
Size_t is defined differently on i386 and x86_64. Change type to avoid compiler warning. Signed-off-by:
-
Markus Metzger authored
Allocate the trace_iterator for the hw-branch-tracer selftest on the heap. Signed-off-by:
-
Markus Metzger authored
The hw-branch-tracer uses debug store functions from an on_each_cpu() context, which is simply wrong since the functions may sleep. Add _noirq variants for most functions, which may be called with interrupts disabled. Separate per-cpu and per-task tracing and allow per-cpu tracing to be controlled from any cpu. Make the hw-branch-tracer use the new debug store interface, synchronize with hotplug cpu event using get/put_online_cpus(), and remove the unnecessary spinlock. Make the ptrace bts and the ds selftest code use the new interface. Defer the ds selftest. Signed-off-by:
-
Markus Metzger authored
Add functions to get and set the debugctlmsr on different cpus. Signed-off-by:
-
Markus Metzger authored
Rename the bts_struct timestamp field to event. Use trace_clock_global() for time measurement. Reported-by:
-
Markus Metzger authored
Per-task branch tracing installs a debug store context with the traced task. This immediately results in the branch trace control bits to be cleared for the next context switch of that task, if not set before. Either per-cpu or per-task tracing are allowed at the same time. An active per-cpu tracing would be disabled even if the per-task tracing request is rejected and the task debug store context removed. Check the tracing type (per-cpu or per-task) before installing a task debug store context. Signed-off-by:
-
Markus Metzger authored
In order to stop branch tracing for a running task, we need to first clear the branch tracing control bits before we may free the tracing buffer. If the traced task is running, the cpu might still trace that task after the branch trace control bits have cleared. Wait until the traced task has been scheduled out before proceeding. A similar problem affects the task debug store context. We first remove the context, then we need to wait until the task has been scheduled out before we can free the context memory. Reviewed-by:
Oleg Nesterov <oleg@redhat.com> Signed-off-by:
-
Markus Metzger authored
When a ptraced task is unlinked, we need to stop branch tracing for that task. Since the unlink is called with interrupts disabled, and we need interrupts enabled to stop branch tracing, we defer the work. Collect all branch tracing related stuff in a branch tracing context. Reviewed-by:
-
Markus Metzger authored
Add a function to wait until some other task has been switched out at least once. This differs from wait_task_inactive() subtly, in that the latter will wait until the task has left the CPU. Signed-off-by:
-
Markus Metzger authored
When the bts tracer is removed while the traced task is running, the write to clear the bts tracer pointer races with context switch code. Read the tracer once during a context switch. When a new tracer is installed, the bts tracer is set in the ds context before the tracer is initialized in order to claim the context for that tracer. This may result in write accesses using an uninitialized trace configuration when scheduling timestamps have been requested. Store active tracing flags separately and only set active flags after the tracing configuration has been initialized. Reviewed-by:
-
Ingo Molnar authored
Merge reason: update to latest tracing and ptrace APIs Signed-off-by: -
Trond Myklebust authored
Due to an apparent typo, commit a67d18f8 (NFS: load the rpc/rdma transport module automatically) lead to the 'proto=' mount option doing a double free, while Opt_mountproto leaks a string. Signed-off-by:
Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Linus Torvalds authored
This makes the defautl ext3 data ordering mode (when no explicit ordering is set) configurable, so as to allow people to default to 'data=writeback' and get the resulting latency improvements. This is a non-issue if a filesystem has been explicitly set to some ordering (with 'tune2fs'). Signed-off-by:
-
- 06 Apr, 2009 17 commits
-
-
git://git.infradead.org/ubifs-2.6Linus Torvalds authored
* 'linux-next' of git://git.infradead.org/ubifs-2.6: UBIFS: fix recovery bug UBIFS: add R/O compatibility UBIFS: fix compiler warnings UBIFS: fully sort GCed nodes UBIFS: fix commentaries UBIFS: introduce a helpful variable UBIFS: use KERN_CONT UBIFS: fix lprops committing bug UBIFS: fix bogus assertion UBIFS: fix bug where page is marked uptodate when out of space UBIFS: amend key_hash return value UBIFS: improve find function interface UBIFS: list usage cleanup UBIFS: fix dbg_chk_lpt_sz()
-
git://git.infradead.org/ubi-2.6Linus Torvalds authored
* 'linux-next' of git://git.infradead.org/ubi-2.6: UBI: add fsync capability
-
Oleg Nesterov authored
The CAP_KILL check in exit_notify() looks just wrong, kill it. Whatever logic we have to reset ->exit_signal, the malicious user can bypass it if it execs the setuid application before exiting. Signed-off-by:
Serge Hallyn <serue@us.ibm.com> Acked-by:
Roland McGrath <roland@redhat.com> Signed-off-by:
-
git://git.infradead.org/mtd-2.6Linus Torvalds authored
* git://git.infradead.org/mtd-2.6: (53 commits) [MTD] struct device - replace bus_id with dev_name(), dev_set_name() [MTD] [NOR] Fixup for Numonyx M29W128 chips [MTD] mtdpart: Make ecc_stats more realistic. powerpc/85xx: TQM8548: Update DTS file for multi-chip support powerpc: NAND: FSL UPM: document new bindings [MTD] [NAND] FSL-UPM: Add wait flags to support board/chip specific delays [MTD] [NAND] FSL-UPM: add multi chip support [MTD] [NOR] Add device parent info to physmap_of [MTD] [NAND] Add support for NAND on the Socrates board [MTD] [NAND] Add support for 4KiB pages. [MTD] sysfs support should not depend on CONFIG_PROC_FS [MTD] [NAND] Add parent info for CAFÉ controller [MTD] support driver model updates [MTD] driver model updates (part 2) [MTD] driver model updates [MTD] [NAND] move gen_nand's probe function to .devinit.text [MTD] [MAPS] move sa1100 flash's probe function to .devinit.text [MTD] fix use after free in register_mtd_blktrans [MTD] [MAPS] Drop now unused sharpsl-flash map [MTD] ofpart: Check name property to determine partition nodes. ... Manually fix trivial conflict in drivers/mtd/maps/Makefile
-
Alan Cox authored
As noted by Janne Grunau it would be good if the date was also right. (Web site also resynched) Signed-off-by:
Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by:
-
Alexander Beregalov authored
Fix this build error: drivers/md/raid1.c: In function 'raid1_congested': drivers/md/raid1.c:589: error: 'BDI_write_congested' undeclared BDI_write_congested was changed in commit 1faa16d2 ("block: change the request allocation/congestion logic to be sync/async based") Signed-off-by:
Alexander Beregalov <a.beregalov@gmail.com> Cc: Neil Brown <neilb@suse.de> Cc: Jens Axboe <jens.axboe@oracle.com> Signed-off-by:
-
Alan Cox authored
The 8430 patch was short a const so caused a warning. Signed-off-by:
-
Roel Kluin authored
Signed-off-by:
Roel Kluin <roel.kluin@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Alan Cox authored
There are various bits of code here that are unfinished and instead of being harmless either confuse or spew stuff into the logs at higher than debug level. They can and should go away. Also remove the bogus use of tty->lowlatency. We fixed the need for this hack long ago (with the flip buffer rewrite) but people keep copying it into drivers. Signed-off-by:
-
Alan Cox authored
Closes bug 9065 Signed-off-by:
-
Alan Cox authored
Bugzilla #9095 and a couple of other confirmations Signed-off-by:
-
Breno Leitao authored
Signed-off-by:
Breno Leitao <leitao@linux.vnet.ibm.com> Signed-off-by:
-
Alexander Beregalov authored
Fixes: In file included from drivers/serial/mux.c:37: include/linux/serial_core.h: In function 'uart_handle_sysrq_char': include/linux/serial_core.h:467: error: 'struct uart_port' has no member named 'sysrq' include/linux/serial_core.h:468: error: 'struct uart_port' has no member named 'sysrq' Signed-off-by:
-
Breno Leitao authored
As it was, the retval was never returned, so its assignments were silly. Just consolidate everything to rc, and remove the unused retval variable. Signed-off-by:
Alan Cox <alan@etchedpixels.co.uk> Signed-off-by:
-
Breno Leitao authored
Signed-off-by:
-
Jiri Slaby authored
mxser_check_modem_status is called with tty parameter, so the reference should be increased by callers already -- for ioctl syscall it is held whole time gap since open to close, for interrupt, the reference count is increased in the irq handler. There is no tty_kref_put in that function, so this also fixes a refcounting bug. Signed-off-by:
Jiri Slaby <jirislaby@gmail.com> Signed-off-by:
-
Jiri Slaby authored
There is missing tty_kref_put on some paths in moxa_poll_port, although the reference is always taken. Fix it. Signed-off-by:
Jan 'Yenya' Kasprzak <kas@fi.muni.cz> Signed-off-by:
-
