summaryrefslogtreecommitdiffstats
path: root/drivers/net/wireless/iwlwifi/iwl-devtrace.h
Commit message (Collapse)AuthorAgeFilesLines
* iwlwifi: Update Copyright to 2014Emmanuel Grumbach2013-12-311-1/+1
| | | | | | Happy new year! Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
* mac80211: add control port protocol TX control flagJohannes Berg2013-08-121-3/+4
| | | | | | | | | | | | | A lot of drivers check the frame protocol for ETH_P_PAE, for various reasons (like making those more reliable). Add a new flags bitmap to the TX control info and a new flag indicating the control port protocol is in use to let all drivers also apply such logic to other control port protocols, should they be configured. Also use the new flag in the iwlwifi drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: use __get_str in tracingJohannes Berg2013-03-061-1/+1
| | | | | | | | | | Instead of using (char *)__get_dynamic_array use __get_str. The latter is actually a macro that expands to the former in the code, but trace-cmd in userspace can parse __get_str only. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: rename IWL_MAX_CMD_TFDS to IWL_MAX_CMD_TBS_PER_TFDJohannes Berg2013-02-281-1/+1
| | | | | | | | | | The IWL_MAX_CMD_TFDS name for this constant is wrong, the constant really indicates how many TBs we can use in the driver for a single command TFD, rename the constant and also add a comment explaining it. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: always copy first 16 bytes of commandsJohannes Berg2013-02-261-6/+4
| | | | | | | | | | | | | | | | | | | The FH hardware will always write back to the scratch field in commands, even host commands not just TX commands, which can overwrite parts of the command. This is problematic if the command is re-used (with IWL_HCMD_DFL_NOCOPY) and can cause calibration issues. Address this problem by always putting at least the first 16 bytes into the buffer we also use for the command header and therefore make the DMA engine write back into this. For commands that are smaller than 16 bytes also always map enough memory for the DMA engine to write back to. Cc: stable@vger.kernel.org Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: update copyrightJohannes Berg2013-01-241-1/+1
| | | | | | | Update Copyright notices to 2013. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: move prph handling into the transportEmmanuel Grumbach2012-12-061-0/+34
| | | | | | | | | | | | | New transports may handle it internally for better performance. Also move the tracing inside PRPH access which will make the output more readable: iwlwifi_dev_ioread_prph32: Read 0x0 from SCD_AGGR_SEL (32-bit) instead of the corresponding accesses to HBUS_TARG_PRPH_*. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: fix typo in RX data tracingJohannes Berg2012-11-071-1/+1
| | | | | | | The printk message should say RX, not TX. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: make data frame tracing optionalJohannes Berg2012-10-161-8/+87
| | | | | | | | | | | | | | | | | | | When tracing in iwlwifi, we get all data. Most of the time, we don't need it, and it just takes up a lot of extra space in the trace. Make this optional by recording the data into two separate trace events if it is needed. Without it, record only the content of non-data and EAPOL TX frames. As a result, tracing without the data tracepoints will record meta information including the 802.11 headers for all frames but will not record the contents of data frames to reduce trace overhead. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: reduce overhead if tracing disabledJohannes Berg2012-09-051-13/+21
| | | | | | | | | | | | | | | | | | Tracing commands builds an array of trace data items even when the tracepoint is disabled. Instead, loop in the tracepoint assignment. This reduces overhead if tracing is compiled into the driver but not enabled and slightly reduces overall driver size as well: text data bss dec hex filename 114514 6509 48 121071 1d8ef before/iwlwifi.ko 114189 6509 48 120746 1d7aa after/iwlwifi.ko Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Max Stepanov <Max.Stepanov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: use __get_str in tracingJohannes Berg2012-06-251-1/+1
| | | | | | | | | __get_str() is identical to (char *)__get_dynamic_array() that is in the code now, substitute __get_str to make the code more readable. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: bump trace message limitJohannes Berg2012-06-251-1/+1
| | | | | | | | | There's one message that goes just over the limit of 100 characters, so bump the limit to 110 to get rid of the warning from that. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: move DVM code into subdirectoryJohannes Berg2012-06-061-0/+1
| | | | | | | | | | Since we're working on another mode/driver inside iwlwifi, move the current one into a subdirectory to more cleanly separate the code. While at it, rename all the files. Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* iwlwifi: make tracing use device as identifierJohannes Berg2012-03-071-69/+72
| | | | | | | | | | | | | | Tracing used the priv pointer as an identifier, which has the problem that we don't have it in all code, and also some people say no pointers should be "leaked" to userspace. Use the device name instead, it is more useful anyway. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: trace debug messagesJohannes Berg2012-02-171-0/+65
| | | | | | | | | | | | | Make iwlwifi record all debug messages into tracing, even if debug_level is not enabled. Due to the lack of APIs, the debug messages are now recorded up to a max length of 100, the only one above that is the RXON which is not needed if you trace the commands as well as it only dumps the command contents. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Wey-Yi W Guy <wey-yi.w.guy@intel.com>
* iwlwifi: update CopyrightWey-Yi Guy2012-01-241-1/+1
| | | | | | | Update Copyright to 2012 Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlagn: add IRQ tracingJohannes Berg2011-12-191-0/+29
| | | | | | | | | | | | The legacy IRQs could be read from a trace by their IO accesses, but reading the ICT doesn't leave any trace (pun intended ;-) ) so in order to see what input they get we need to add specific tracepoints. While at it, fix whitespace in two related places. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
* iwlagn: remove unnecessary type for tracing operationsDon Fry2011-11-081-12/+11
| | | | | | | | | | | The device tracing routines only use the priv pointer as an opaque value. Change from a typed iwl_priv pointer to a null pointer and eliminate the need to include iwl_priv.h. CMD_ASYNC is defined in iwl_shared.h which is the only reason it is included. Signed-off-by: Don Fry <donald.h.fry@intel.com> Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlagn: all function iwl-io.c receive iwl_busEmmanuel Grumbach2011-08-291-0/+2
| | | | | | | | Which means that iwl-io.c doesn't need to include iwl-dev.h any more. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlagn: more ucode error log infoWey-Yi Guy2011-05-131-9/+32
| | | | | | | No functional changes, just logging more information when uCode crash, also change change the format. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
* iwlagn: support multiple TBs per commandJohannes Berg2011-05-131-5/+12
| | | | | | | | | | | | | | | | | | | | | | | | | The current "huge" command handling is a bit confusing, and very limited since only one command may be huge at a time. Additionally, we often copy data around quite pointlessly since we could instead map the existing scan buffer for example and use it directly. This patch makes that possible. The first change is that multiple buffers may be given to each command (this change was prepared earlier so callsites don't need to change). Each of those can be mapped attached to a TB in the TFD, and the command header can use a TB (the first one) in the TFD as well. Doing this allows getting rid of huge commands in favour of mapping existing buffers. The beacon transmission is also optimised to not copy the SKB at all but use multiple TBs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
* iwlagn: change Copyright to 2011Wey-Yi Guy2011-04-071-1/+1
| | | | | Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: fix TX tracerJohannes Berg2010-08-061-1/+1
| | | | | | | | | | The TX tracing code copies with the wrong length, which will typically copy too little data. Fix this by using the correct length variable. Cc: stable@kernel.org [2.6.32+] Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: don't include iwl-dev.h from iwl-devtrace.hTejun Heo2010-03-301-1/+0
| | | | | | | | | | | | | | | iwl-devtrace.h is used to declare and define trace points and including iwl-dev.h from the file, which in turn includes other generic headers, can lead to problems like generating duplicate copies of generic trace points depending on the order of includes. Don't include iwl-dev.h from iwl-devtrace.h but include it from its users - iwl-io.h and iwl-devtrace.c. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Reinette Chatre <reinette.chatre@intel.com> Cc: Zhu Yi <yi.zhu@intel.com> Cc: Intel Linux Wireless <ilw@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com>
* Merge branch 'master' of ↵David S. Miller2010-01-231-0/+26
|\ | | | | | | master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
| * iwlwifi: add license to tracing filesReinette Chatre2010-01-181-0/+26
| | | | | | | | | | Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* | iwlwifi: add continuous uCode event log capabilityWey-Yi Guy2009-12-211-0/+44
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In order to help uCode debugging, adding the capability to provide continuous uCode event logging function. uCode events is located in round-robin event queue and filled by uCode, by enable continuous event logging, driver check the write pointer and log the newly added events in iwl_bg_ucode_trace() timer function. There is still possibility of missing events if event queue being wrapped before next event dump; but with this capability, we can have much better understanding of the uCode behavior during runtime; it can help to debug the uCode related issues. Methods to enable/disable the continuous event log: step 1: enable ucode trace timer "echo 1 > /sys/kernel/debug/ieee80211/phyX/iwlagn/debug/ucode_tracing" step 2: start ftrace sudo ./trace-cmd record -e iwlwifi_ucode:* sleep 1d step 3: stop ftrace sudo ./trace-cmd report trace.dat step 4: disable ucode trace timer "echo 0 > /sys/kernel/debug/ieee80211/phyX/iwlagn/debug/ucode_tracing" use "ucode_tracing" debugfs file to display number of event queue wrapped when driver attempt the continuous event logging. If event queue being wrapped more than once when driver has opportunity to log the event; it indicated there are events missing in the event log trace. This continuous event log function only available for 4965 and newer NICs. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: separate IO tracingJohannes Berg2009-11-231-1/+4
| | | | | | | | | | | | | | Since IO tracing is usually not needed and generates a lot of data, separate it into its own trace system so that we can always enable iwlwifi:* and not have to worry about getting too much data. If IO tracing is then really needed we can enable iwlwifi_io:* in addition and get that data. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: Add iwl_write8()Ben Cahill2009-11-231-0/+16
| | | | | | | | | To support byte writes to CSR_INT_COALESCING and CSR_INT_PERIODIC registers, add iwl_write8(), including debug/trace support. Signed-off-by: Ben Cahill <ben.m.cahill@intel.com> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
* iwlwifi: device tracingJohannes Berg2009-10-071-0/+178
In order to have an easier way to debug issues, create trace events (using the ftrace framework) that will allow us to follow exactly what the driver is doing with the device. The text format isn't all that useful, but the binary format can also be obtained easily via debugfs and then analysed on the fly or offline with debugging tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
OpenPOWER on IntegriCloud