- 12 Apr, 2011 1 commit
-
-
Vivek Natarajan authored
In a highly noisy environment, the tx rate of the driver drops and the application slows down since it has not yet received ACKs for the frames already queued in the hardware. Since this ACK may take more than 100ms, stopping the dev queues for entering PS at this stage breaks applications, WMM test cases in my testing. If there are frames already pending in the tx queue, postponing the PS logic helps to avoid redundant queue stops. When power save is enabled by default and in a noisy environment, this API certainly helps in improving the average throughput. Signed-off-by:
Vivek Natarajan <vnatarajan@atheros.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
- 08 Apr, 2011 16 commits
-
-
Johannes Berg authored
If we maintain API properly, then there isn't really a reason to warn about this since we'll just be adding things that are safe to ignore, so downgrade the warning to debug info level. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
Wey-Yi Guy <wey-yi.w.guy@intel.com>
-
Johannes Berg authored
There's no need to keep both normal and BT statistics versions around all the time in memory when we only use a subset of both. So keep only the subsets that we need in memory, depending on the debug config). Also, in doing so, we can remove all the calls to iwl_bt_statistics() in the driver as we'll just access the copied statistics now. Finally, also remove this call from the one place where it might still be needed and automatically detect what kind of statistics the device is sending based on their size. This way, we don't need to keep track of which devices do what any more, which is good since this is subject to change based on the ucode version (as some ucode even for non-BT devices will in fact use BT statistics). Warn upon encountering a statistics command from the ucode that isn't known, so we will find such issues earlier in the future. Signed-off-by:
Don Fry <donald.h.fry@intel.com> Signed-off-by:
-
Wey-Yi Guy authored
_agn driver should perform both sensitivity and chain noise calib. Signed-off-by: -
Wey-Yi Guy authored
Remove the config flag for tx power calib Signed-off-by: -
Wey-Yi Guy authored
Remove broken_power_save checking Signed-off-by: -
Wey-Yi Guy authored
More cleanup code, no functional changes Signed-off-by: -
Wey-Yi Guy authored
After driver split, set_l0s config is no longer needed, remove it Signed-off-by: -
Wey-Yi Guy authored
Remove the reference to both 3945 and 4965 in LED code Signed-off-by: -
Wey-Yi Guy authored
More clean up after driver split Signed-off-by: -
Wey-Yi Guy authored
3945 no longer apply Signed-off-by: -
Wey-Yi Guy authored
Remove 3945 define Signed-off-by: -
Wey-Yi Guy authored
For 2000 series NICs, disable OTP refresh in order to read correct PAPD table from high OTP block Signed-off-by: -
Johannes Berg authored
If change_interface gets invoked during a firmware restart, it may crash; prevent that from happening by checking if ctx->vif is assigned. Additionally, in my initial commit I forgot to set the vif->p2p variable correctly, so fix that too. Cc: stable@kernel.org [2.6.38+] Signed-off-by:
-
Garen Tamrazian authored
The microcode may sometimes reject TX frames when on a radar channel even after we associated as it clears information during association and needs to receive a new beacon before allowing that channel again. This manifests itself as a TX status value of TX_STATUS_FAIL_PASSIVE_NO_RX. So in this case, stop the corresponding queue and give the frame back to mac80211 for retransmission. We start the queue again when a beacon from the AP is received which will make the regulatory enforcement in the device allow transmitting again. Signed-off-by:
Garen Tamrazian <garenx.tamrazian@intel.com> Signed-off-by:
-
Wey-Yi Guy authored
After driver split, no need to use function pointer for those event and register dump function. Signed-off-by: -
root authored
define of firmware filenames use extra macro to build the files name. Signed-off-by:
Jay Sternberg <jay.e.sternberg@intel.com> Signed-off-by:
-
- 07 Apr, 2011 23 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6John W. Linville authored
Conflicts: drivers/net/wireless/rtlwifi/efuse.c drivers/net/wireless/rtlwifi/rtl8192c/fw_common.c net/bluetooth/mgmt.c
-
Luis R. Rodriguez authored
We have no other option but to inform userspace that we have queued up their regulatory hint request when we are given one given that nl80211 operates atomically on user requests. The best we can do is accept the request, and add a delayed work item for processing failure and cancel it if we succeeed. Upon failure we restore the regulatory settings and ignore the user input. This fixes this reported bug: https://bugzilla.kernel.org/show_bug.cgi?id=28112 Reported-by: gregoryx.alagnou@intel.com Signed-off-by:
Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by:
-
Luis R. Rodriguez authored
When we restore regulatory settings its possible CRDA will not reply because of a bogus user entry. In this case the bogus entry will prevent any further processing on cfg80211 for regulatory domains even if we restore regulatory settings. To prevent this we suck out all pending requests when restoring regulatory settings and add them back into the queue after we have queued up the reset work. The impact of not having this applied is that a user with privileges can issue a userspace regulatory hint while we are disasocciating and this would prevent any further processing of regulatory domains. Signed-off-by:
-
Johannes Berg authored
This generates a massive reduction in module size: with debug: text data bss dec hex filename 670300 13136 420 683856 a6f50 iwlagn.ko (before) 388347 13136 408 401891 621e3 iwlagn.ko (after) without debug: text data bss dec hex filename 528575 13072 420 542067 84573 iwlagn.ko (before) 294192 13072 408 307672 4b1d8 iwlagn.ko (after) This also removes all the IO debug functionality since it can easily be replaced by tracing, and makes the code unnecessarily complex. I haven't done any CPU utilisation measurements, but given that the hotpaths don't use much IO it is not likely to have a negative impact; in fact, the size reduction will reduce cache pressure which possibly improves performance. Finally, an unused function or two were removed. Signed-off-by:
-
Johannes Berg authored
The ISR (interrupt service routine) ops are now no longer necessary since they are the same for all devices this driver now handles. Signed-off-by:
-
Johannes Berg authored
We never use the value in alloc_rxb_page, so there's no point in keeping it either. Signed-off-by:
-
Johannes Berg authored
The rev_id variable is only printed, we don't need to store it. Signed-off-by:
-
Johannes Berg authored
The hw_rev variable is used only during init, so there's no need to keep it around. Signed-off-by:
-
Johannes Berg authored
The variable is never used. Signed-off-by:
-
Johannes Berg authored
Again, a 4965 specific code path that we no longer need in iwlagn. Signed-off-by:
-
Johannes Berg authored
We know after loading the ucode whether it will support PAN or not, so we can also initialise the cmd_queue variable much earlier. Signed-off-by:
-
Johannes Berg authored
Devices newer than 4965 don't actually send two different versions of the ALIVE command, so we always had a bug here since before this patch we copy more data than we got. Remove the iwl_init_alive_resp struct and don't use it. Since we also really don't need to track all the data received in ALIVE as we only use the error and log event tables later, we can also save space by just keeping those and not more data around in memory. Signed-off-by:
-
Johannes Berg authored
My previous patch left a message talking about bootstrap, but that's clearly bogus. Signed-off-by:
-
Johannes Berg authored
Some new devices and microcode files will a greater variety of features, so the TLV-per-feature approach we took before will quickly make things harder to manage and increase the file size. Add a new TLV that has feature flags. Currently, it will contain: 1) a PAN feature flag, which moves from a separate TLV 2) a new BT stats bit that indicates whether the microcode image uses bluetooth statistics 3) a new MFP flag for management frame protection which can be enabled once the device/microcode supports it Signed-off-by: -
Wey-Yi Guy authored
In alive notify, we should check return code instead of assume everything ok Signed-off-by:
-
Wey-Yi Guy authored
Signed-off-by:
-
Johannes Berg authored
This was used only on 4965 in conjunction with the bootstrap ucode. Signed-off-by:
-
Johannes Berg authored
The ucode restart has to take into account a number of things, like clearing the HCMD_ACTIVE and other status bits, and waking up the wait_command_queue. Currently, however, there are a number of places that neither do that, nor actually set the FW error bit that leads to proper restart handling, which means that in those cases things will probably just hang completely. To clean this up, make all ucode restart go through a single function, except for the cases where it's called during firmware loading. Also fix a bug in wimax coexist restart avoidance, it needs to first clear the status bits (and it has to clear the HCMD_ACTIVE one as well) and then wake up anything waiting on wait_command_queue. Signed-off-by:
-
Johannes Berg authored
Only 4965 had a bootstrap microcode image, so the agn driver can completely ignore that and we can remove some code from it. Signed-off-by:
-
Johannes Berg authored
When we loaded a ucode, there's no point in checking any one that is present, we know which one is supposed to be present so also verify that it is exactly the right one. That also simplifies the code and makes it faster since it doesn't have to check all. Signed-off-by:
-
Johannes Berg authored
The code in iwlcore_verify_inst_sparse really doesn't need to keep track of the number of errors it encountered since a single one is fatal. Also, the code in iwl_verify_inst_full is just used to print out some things, so rename it to iwl_print_inst and don't give it a return code and just make it print out the values. Signed-off-by:
-
Wey-Yi Guy authored
In alive notification call, return the status from iwl_send_calib_results() Signed-off-by:
-
Johannes Berg authored
A number of places just use a variable to return it right away, which is useless, so let's remove the variables there. Signed-off-by:
-
