- 27 Feb, 2009 1 commit
-
-
Steven Rostedt authored
This patch creates an options directory in the debugfs, that contains the available tracing options. These files contain 1 or 0, where 1 is the option is enabled and 0 it is disabled. Simply echoing in 1 will enable the option and 0 will disable it. This patch only contains the core options, not the tracer options. Signed-off-by:Steven Rostedt <srostedt@redhat.com>
-
- 26 Feb, 2009 11 commits
-
-
Ingo Molnar authored
Conflicts: kernel/sched_clock.c
-
Ingo Molnar authored
If the TSC is constant and non-stop, also set it reliable. (We will turn this off in DMI quirks for multi-chassis systems) The performance number on a 16-way Nehalem system running 32 tasks that context-switch between each other is significant: sched_clock_stable=0 sched_clock_stable=1 .................... .................... 22.456925 million/sec 24.306972 million/sec [+8.2%] lmbench's "lat_ctx -s 0 2" goes from 0.63 microseconds to 0.59 microseconds - a 6.7% increase in context-switching performance. Perfstat of 1 million pipe context switches between two tasks: Performance counter stats for './pipe-test-1m': [before] [after] ............ ............ 37621.421089 36436.848378 task clock ticks (msecs) 0 0 CPU migrations (events) 2000274 2000189 context switches (events) 194 193 pagefaults (events) 8433799643 8171016416 CPU cycles (events) -3.21% 8370133368 8180999694 instructions (events) -2.31% 4158565 3895941 cache references (events) -6.74% 44312 46264 cache misses (events) 2349.287976 2279.362465 wall-time (msecs) -3.06% The speedup comes straight from the reduction in the instruction count. sched_clock_cpu() got simpler and the whole workload thus executes faster. Signed-off-by:Ingo Molnar <mingo@elte.hu>
-
Ingo Molnar authored
Allow CONFIG_HAVE_UNSTABLE_SCHED_CLOCK architectures to still specify that their sched_clock() implementation is reliable. This will be used by x86 to switch on a faster sched_clock_cpu() implementation on certain CPU types. Signed-off-by: -
Ingo Molnar authored
Impact: implement new tracing timestamp APIs Add three trace clock variants, with differing scalability/precision tradeoffs: - local: CPU-local trace clock - medium: scalable global clock with some jitter - global: globally monotonic, serialized clock Make the ring-buffer use the local trace clock internally. Acked-by:
Peter Zijlstra <peterz@infradead.org> Acked-by:
Steven Rostedt <rostedt@goodmis.org> Signed-off-by:
-
Ingo Molnar authored
make sure we dont execute more complex sched_clock() code in NMI context. Acked-by:
-
Jason Baron authored
Impact: add new tracepoints Add them to the generic IRQ code, that way every architecture gets these new tracepoints, not just x86. Using Steve's new 'TRACE_FORMAT', I can get function graph trace as follows using the original two IRQ tracepoints: 3) | handle_IRQ_event() { 3) | /* (irq_handler_entry) irq=28 handler=eth0 */ 3) | e1000_intr_msi() { 3) 2.460 us | __napi_schedule(); 3) 9.416 us | } 3) | /* (irq_handler_exit) irq=28 handler=eth0 return=handled */ 3) + 22.935 us | } Signed-off-by:Jason Baron <jbaron@redhat.com> Signed-off-by:
Masami Hiramatsu <mhiramat@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Mathieu Desnoyers <compudj@krystal.dyndns.org> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by:
-
Frederic Weisbecker authored
Impact: restructure the VFS layout of per CPU trace buffers The per cpu trace files are all in a single directory: /debug/tracing/per_cpu. In case of a large number of cpu, the content of this directory becomes messy so we create now one directory per cpu inside /debug/tracing/per_cpu which contain each their own trace_pipe and trace files. Ie: /debug/tracing$ ls -R per_cpu per_cpu: cpu0 cpu1 per_cpu/cpu0: trace trace_pipe per_cpu/cpu1: trace trace_pipe Signed-off-by:
Frederic Weisbecker <fweisbec@gmail.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by:
-
Ingo Molnar authored
Merge branch 'tip/tracing/ftrace' of ssh://master.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace into tracing/ftrace
-
Ingo Molnar authored
-
Steven Rostedt authored
Peter Zijlstra warned that TPPROTO and TPARGS might become something other than a simple copy of itself. To prevent this from having side effects in the TRACE_FORMAT macro in tracepoint.h, we add a PARAMS() macro to be defined as just a wrapper. Reported-by:
-
Steven Rostedt authored
There's been a bit confusion to whether DEFINE/DECLARE_TRACE_FMT should be a DEFINE or a DECLARE. Ingo Molnar suggested simply calling it TRACE_FORMAT. Reported-by:
-
- 25 Feb, 2009 9 commits
-
-
Frederic Weisbecker authored
Now that several per-cpu files can be read or spliced at the same, we want the read/splice callbacks for tracing files to be reentrants. Until now, a single global mutex (trace_types_lock) serialized the access to tracing_read_pipe(), tracing_splice_read_pipe(), and the seq helpers. Ie: it means that if a user tries to read trace_pipe0 and trace_pipe1 at the same time, the access to the function tracing_read_pipe() is contended and one reader must wait for the other to finish its read call. The trace_type_lock mutex is mostly here to serialize the access to the global current tracer (current_trace), which can be changed concurrently. Although the iter struct keeps a private pointer to this tracer, its callbacks can be changed by another function. The method used here is to not keep anymore private reference to the tracer inside the iterator but to make a copy of it inside the iterator. Then it checks on subsequents read calls if the tracer has changed. This is not costly because the current tracer is not expected to be changed often, so we use a branch prediction for that. Moreover, we add a private mutex to the iterator (there is one iterator per file descriptor) to serialize the accesses in case of multiple consumers per file descriptor (which would be a silly idea from the user). Note that this is not to protect the ring buffer, since the ring buffer already serializes the readers accesses. This is to prevent from traces weirdness in case of concurrent consumers. But these mutexes can be dropped anyway, that would not result in any crash. Just tell me what you think about it. Signed-off-by:
-
Frederic Weisbecker authored
Impact: split up tracing output per cpu Currently, on the tracing debugfs directory, three files are available to the user to let him extracting the trace output: - trace is an iterator through the ring-buffer. It's a reader but not a consumer It doesn't block when no more traces are available. - trace pretty similar to the former, except that it adds more informations such as prempt count, irq flag, ... - trace_pipe is a reader and a consumer, it will also block waiting for traces if necessary (heh, yes it's a pipe). The traces coming from different cpus are curretly mixed up inside these files. Sometimes it messes up the informations, sometimes it's useful, depending on what does the tracer capture. The tracing_cpumask file is useful to filter the output and select only the traces captured a custom defined set of cpus. But still it is not enough powerful to extract at the same time one trace buffer per cpu. So this patch creates a new directory: /debug/tracing/per_cpu/. Inside this directory, you will now find one trace_pipe file and one trace file per cpu. Which means if you have two cpus, you will have: trace0 trace1 trace_pipe0 trace_pipe1 And of course, reading these files will have the same effect than with the usual tracing files, except that you will only see the traces from the given cpu. The original all-in-one cpu trace file are still available on their original place. Until now, only one consumer was allowed on trace_pipe to avoid racy consuming on the ring-buffer. Now the approach changed a bit, you can have only one consumer per cpu. Which means you are allowed to read concurrently trace_pipe0 and trace_pipe1 But you can't have two readers on trace_pipe0 or trace_pipe1. Following the same logic, if there is one reader on the common trace_pipe, you can not have at the same time another reader on trace_pipe0 or in trace_pipe1. Because in trace_pipe is already a consumer in all cpu buffers in essence. Signed-off-by:
-
Ingo Molnar authored
Merge branch 'tip/tracing/ftrace' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace into tracing/ftrace
-
Ingo Molnar authored
Impact: remove old debug/tracing API /debug/tracing/latency_trace is an old legacy format we kept from the old latency tracer. Remove the file for now. If there's any useful bit missing then we'll propagate any useful output bits into the /debug/tracing/trace output. Reported-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Ingo Molnar authored
Impact: fix CPU hotplug lockup bts_hotcpu_handler() is called with irqs disabled, so using mutex_lock() is a no-no. All the BTS codepaths here are atomic (they do not schedule), so using a spinlock is the right solution. Cc: Markus Metzger <markus.t.metzger@intel.com> Signed-off-by: -
Steven Rostedt authored
This patch adds the directory /debug/tracing/events/ that will contain all the registered trace points. # ls /debug/tracing/events/ sched_kthread_stop sched_process_fork sched_switch sched_kthread_stop_ret sched_process_free sched_wait_task sched_migrate_task sched_process_wait sched_wakeup sched_process_exit sched_signal_send sched_wakeup_new # ls /debug/tracing/events/sched_switch/ enable # cat /debug/tracing/events/sched_switch/enable 1 # cat /debug/tracing/set_event sched_switch Signed-off-by: -
Steven Rostedt authored
This patch changes the trace/sched.h to use the DECLARE_TRACE_FMT such that they are automatically registered with the event tracer. And it also adds the tracing sched headers to kernel/trace/events.c Signed-off-by: -
Steven Rostedt authored
This patch creates the event tracing infrastructure of ftrace. It will create the files: /debug/tracing/available_events /debug/tracing/set_event The available_events will list the trace points that have been registered with the event tracer. set_events will allow the user to enable or disable an event hook. example: # echo sched_wakeup > /debug/tracing/set_event Will enable the sched_wakeup event (if it is registered). # echo "!sched_wakeup" >> /debug/tracing/set_event Will disable the sched_wakeup event (and only that event). # echo > /debug/tracing/set_event Will disable all events (notice the '>') # cat /debug/tracing/available_events > /debug/tracing/set_event Will enable all registered event hooks. Signed-off-by: -
Steven Rostedt authored
This patch creates a DEFINE_TRACE_FMT to map to DECLARE_TRACE. This allows for the developers to place format strings and args in with their tracepoint declaration. A tracer may now override the DEFINE_TRACE_FMT macro and use it to record a default format. Signed-off-by:
-
- 24 Feb, 2009 19 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/adobriyan/procLinus Torvalds authored
* 'proc-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/adobriyan/proc: proc: fix PG_locked reporting in /proc/kpageflags
-
git://jdelvare.pck.nerim.net/jdelvare-2.6Linus Torvalds authored
* 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6: Add i2c_board_info for RiscPC PCF8583 i2c: Make sure i2c_algo_bit_data.timeout is HZ-independent i2c-dev: Clarify the unit of ioctl I2C_TIMEOUT i2c: Timeouts reach -1 i2c: Fix misplaced parentheses
-
Linus Torvalds authored
Merge branch 'firedtv-merge' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6 * 'firedtv-merge' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6: firedtv: dvb_frontend_info for FireDTV S2, fix "frequency limits undefined" error firedtv: massive refactoring firedtv: rename files, variables, functions from firesat to firedtv firedtv: Use DEFINE_SPINLOCK firedtv: fix registration - adapter number could only be zero firedtv: use length_field() of PMT as length firedtv: fix returned struct for ca_info firedtv: cleanups and minor fixes ieee1394: remove superfluous assertions ieee1394: inherit ud vendor_id from node vendor_id ieee1394: add hpsb_node_read() and hpsb_node_lock() ieee1394: use correct barrier types between accesses of nodeid and generation firesat: copyrights, rename to firedtv, API conversions, fix remote control input firesat: avc resend firesat: update isochronous interface, add CI support firesat: add DVB-S support for DVB-S2 devices firesat: fix DVB-S2 device recognition DVB: add firesat driver
-
git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4Linus Torvalds authored
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: ext4: Fix deadlock in ext4_write_begin() and ext4_da_write_begin() ext4: Add fallback for find_group_flex
-
Russell King authored
Add the necessary i2c_board_info structure to fix the lack of PCF8583 RTC on RiscPC. Signed-off-by:
Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by:
Jean Delvare <khali@linux-fr.org> Cc: Alessandro Zummo <a.zummo@towertech.it>
-
Jean Delvare authored
i2c_algo_bit_data.timeout is supposed to be in jiffies, so drivers should use set this value in terms of HZ. Ultimately I think this field should be discarded in favor of i2c_adapter.timeout, but that's left for a future patch. Signed-off-by:
Lennert Buytenhek <kernel@wantstofly.org> Acked-by:
Len Sorensen <lsorense@csclub.uwaterloo.ca>
-
Jean Delvare authored
The unit in which user-space can set the bus timeout value is jiffies for historical reasons (back when HZ was always 100.) This is however not good because user-space doesn't know how long a jiffy lasts. The timeout value should instead be set in a fixed time unit. Given the original value of HZ, this unit should be 10 ms, for compatibility. Signed-off-by:
Wolfram Sang <w.sang@pengutronix.de>
-
Roel Kluin authored
With a postfix decrement these timeouts reach -1 rather than 0, but after the loop it is tested whether they have become 0. As pointed out by Jean Delvare, the condition we are waiting for should also be tested before the timeout. With the current order, you could exit with a timeout error while the job is actually done. Signed-off-by:
Roel Kluin <roel.kluin@gmail.com> Signed-off-by:
-
Roel Kluin authored
Fix misplaced parentheses. Signed-off-by:
-
Helge Bahmann authored
Expr always evaluates to zero. Cc: Matt Mackall <mpm@selenic.com> Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com>
-
Markus Metzger authored
Remove unnecessary CONFIG guards around type declarations and macro definitions. Reported-by:
Cyrill Gorcunov <gorcunov@gmail.com> Signed-off-by:
Markus Metzger <markus.t.metzger@intel.com> Cc: markus.t.metzger@gmail.com Cc: roland@redhat.com Cc: eranian@googlemail.com Cc: oleg@redhat.com Cc: juan.villacis@intel.com Signed-off-by:
-
Markus Metzger authored
Fix an invalid memory reference problem when cpu hotplug support is disabled and the hw-branch-tracer is set as current tracer. Initializing the tracer calls bts_trace_init() which has already been freed at this time. Reported-by:
-
Ingo Molnar authored
-
Beat Michel Liechti authored
I found that the function fdtv_frontend_init in the file firedtv-fe.c was missing a case for FIREDTV_DVB_S2 which resulted in "frequency limits undefined" errors in syslog. Signed-off-by:
Beat Michel Liechti <bml303@gmail.com> Change by Stefan R: combine it with case case FIREDTV_DVB_S as originally suggested by Beat Michel. This enables FE_CAN_FEC_AUTO also for FireDTV-S2 devices which is possible as long as only DVB-S channels are used. FE_CAN_FEC_AUTO would be wrong for DVB-S2 channels, but those cannot be used yet since the driver is not yet converted to S2API. Signed-off-by:
Stefan Richter <stefanr@s5r6.in-berlin.de>
-
Stefan Richter authored
Combination of the following changes: Mon, 23 Feb 2009 14:21:10 +0100 (CET) firedtv: reinstate debug logging option Henrik Kurelid tells me that FCP debug logging (which I removed during cleanups) is still useful when working on driver issues together with end users. So bring it back in an updated form with only 60% of the original code footprint. Logging can be enabled with # echo -1 > /sys/module/firedtv/parameters/debug 1 instead of -1 enables only FCP header logging, 2 instead of -1 enables only hexdumps of the entire FCP frames. 0 switches logging off again. Fri, 20 Feb 2009 20:54:27 +0100 (CET) firedtv: build fix for INPUT=m and DVB_FIREDTV=y Thu, 19 Feb 2009 20:40:39 +0100 firedtv: use msecs_to_jiffies Pointed out by Mauro Carvalho Chehab. Sun Feb 15 20:50:46 CET 2009 firedtv: some more housekeeping Fix an old checkpatch warning and a new compiler warning. Sun Feb 15 15:33:17 CET 2009 firedtv: rename a file once more At the moment, about a third of avc.c is specific to FireDTVs rather than generic AV/C code. Rename it to firedtv-avc.c. Sun Feb 15 15:33:17 CET 2009 firedtv: dvb demux: more compact channels backing store Replace struct firedtv_channel { bool active; int pid; } channel[16]; by unsigned long channel_active; u16 channel_pid[16];. Sun Feb 15 15:33:17 CET 2009 firedtv: dvb demux: some simplifications c->active was unnecessarily cleared twice. Also, by marking the channel inactive before the for loop, the loop becomes identical with fdtv_channel_collect(). Sun Feb 15 15:33:17 CET 2009 firedtv: dvb demux: remove a bogus loop This loop is unnecessary because - only active channel[].pid's will be sent to the device, - when a channel is activated, its pid is set to dvbdmxfeed->pid. Perhaps the original code was there because it was initially not fully covered by the fdtv->demux_mutex. Sun Feb 15 15:33:17 CET 2009 firedtv: dvb demux: fix mutex protection fdtv_start_feed() accessed the channel list unsafely. Fully serialize it with itself and fdtv_stop_feed(). Sun Feb 15 15:33:17 CET 2009 firedtv: dvb demux: fix missing braces Original code was: ... case DMX_TS_PES_OTHER: //Dirty fix to keep firesat->channel pid-list up to date for(k=0;k<16;k++){ if(firesat->channel[k].active == 0) firesat->channel[k].pid = dvbdmxfeed->pid; break; } channel = firesat_channel_allocate(firesat); break; default: ... Looks bogus in several respects. For now let's just add braces to the if because that seems to be what the author meant. Sun Feb 15 15:33:17 CET 2009 firedtv: allow build without input subsystem !CONFIG_INPUT is very unlikely on systems on which firedtv is of interest. But we can easily support it. Sun Feb 15 15:33:17 CET 2009 firedtv: replace EXTRA_CFLAGS by ccflags The former are deprecated. The latter can depend on Kconfig variables. Sun Feb 15 15:33:17 CET 2009 firedtv: concentrate ieee1394 dependencies Move the entire interface with drivers/ieee1394 to firedtv-1394.c. Move 1394-independent module initialization code to firedtv-dvb.c. This prepares interfacing with drivers/firewire. Sun Feb 15 15:33:17 CET 2009 firedtv: amend Kconfig menu prompt Sun Feb 15 15:33:17 CET 2009 firedtv: remove kernel version compatibility macro Sun Feb 15 15:33:17 CET 2009 firedtv: combine header files avc.h and firedtv-*.h are small and currently not shared with other drivers, hence concatenate them all into firedtv.h. Sun Feb 15 15:33:17 CET 2009 firedtv: misc style touch-ups Standardize on lower-case hexadecimal constants. Adjust whitespace. Omit unnecessary pointer type casts and an unnecessary list head initialization. Use dev_printk. Wed Feb 11 21:21:04 CET 2009 firedtv: avc, ci: remove unused constants Wed Feb 11 21:21:04 CET 2009 firedtv: avc: remove bitfields from read descriptor response operands Don't use bitfields in struct types of on-the-wire data. Wed Feb 11 21:21:04 CET 2009 firedtv: avc: remove bitfields from DSD command operands Don't use bitfields in struct types of on-the-wire data. Wed Feb 11 21:21:04 CET 2009 firedtv: avc: header file cleanup Remove unused constants and declarations. Move privately used constants into .c files. Wed Feb 11 21:21:04 CET 2009 firedtv: avc: remove bitfields from FCP frame types Don't use bitfields in struct types of on-the-wire data. Also move many privately used constants from avc.h to avc.c and remove some unused constants. Sun, 18 Jan 2009 16:30:00 +0100 (CET) firedtv: avc: fix offset in avc_tuner_get_ts The parentheses were wrong. It didn't matter though because this code only writes a 0 into an area which is already initialized to 0. Sun, 18 Jan 2009 16:30:00 +0100 (CET) firedtv: avc: reduce stack usage, remove two typedefs It is safe to share a memory buffer for command frame and response frame because the response data come in after the command frame was last used. Even less stack would be required if only the actual required frame size instead of the entire FCP register size was allocated. Also, rename the defined types AVCCmdFrm and AVCRspFrm to struct avc_command_frame and struct avc_response_frame. TODO: Remove the bitfields in these types. Sun, 18 Jan 2009 16:30:00 +0100 (CET) firedtv: cmp: move code to avc Sun, 18 Jan 2009 16:30:00 +0100 (CET) firedtv: iso: move code to firedtv-1394 Sun, 18 Jan 2009 16:30:00 +0100 (CET) firedtv: iso: remove unnecessary struct type definitions Sun, 18 Jan 2009 16:30:00 +0100 (CET) firedtv: iso: style changes and fixlets Add cleanup after failure in setup_iso_channel. Replace printk() by dv_err(). Decrease indentation level in rawiso_activity_cb(). Signed-off-by: -
Rambaldi authored
Combination of the following changes: Sat, 17 Jan 2009 14:47:34 +0100 firedtv: rename variables and functions from firesat to firedtv Signed-off-by:
Rambaldi <Rambaldi@xs4all.nl> Additional changes by Stefan Richter: Renamed struct firedtv *firedtv to struct firedtv *fdtv and firedtv_foo_bar() to fdtv_foo_bar() for brevity. Sat, 17 Jan 2009 13:07:44 +0100 firedtv: rename files from firesat to firedtv Signed-off-by:
-
Julia Lawall authored
SPIN_LOCK_UNLOCKED is deprecated. The following makes the change suggested in Documentation/spinlocks.txt Signed-off-by:
Julia Lawall <julia@diku.dk> Signed-off-by:
-
Henrik Kurelid authored
There was a bug causing the initialization to fail if adapter number was greater than zero. The adapter was however registered which caused the driver to oops the second time initialization was tried. Signed-off-by:
Henrik Kurelid <henrik@kurelid.se> Signed-off-by:
-
Henrik Kurelid authored
Parsed and used the length_field() of the PMT message instead of using the length field of the message struct, which does not seem to be filled correctly by e.g. MythTV. Signed-off-by:
-
