| Message ID | CABPxzY+_Mjercvvh5XHmtHdNCKXDupVbL4PKXgzZPiUesRsRMA@mail.gmail.com (mailing list archive) |
|---|---|
| State | New |
| Delegated to: | Miri Korenblit |
| Headers | show |
| Series | iwlwifi: Monitor mode broken with 6.1.0 | expand |
Hi! On Mon, 2023-01-23 at 14:07 +0530, Krishna Chaitanya wrote: > Hi, > > I am using the ath.git (ea726a8d305a : was testing external Atheros > cards) + linux-firmware (bb2d42d) and the monitor mode stopped working > with iwlwifi. There is a warning from UBSAN: "UBSAN: > shift-out-of-bounds in > drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c:657:22". > > The root cause was this c6ce1c74ef292 which defaults the assignment to > 0xFFFF and then that was being used for the shift. > > I have tried to fix the warning with the below patch (just for an > experiment, not a proper one), the warning is gone, but no frames are > seen in Wireshark. > > Below is the version information, any help is appreciated, monitor > mode is important for me. > > lspci: Network controller: Intel Corporation Wi-Fi 6 AX201 (rev 20) > > [ +0.003790] iwlwifi 0000:00:14.3: api flags index 2 larger than > supported by driver > [ +0.000046] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: > 89.3.35.37 > [ +0.001519] iwlwifi 0000:00:14.3: loaded firmware version > 72.daa05125.0 QuZ-a0-hr-b0-72.ucode op_mode iwlmvm > [ +0.034887] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6 AX201 > 160MHz, REV=0x351 > [ +0.000083] thermal thermal_zone7: failed to read out thermal zone (-61) > [ +0.122144] iwlwifi 0000:00:14.3: Detected RF HR B3, rfid=0x10a100 > [ +0.065701] iwlwifi 0000:00:14.3: base HW address: 4c:79:6e:90:94:71 > [ +0.019826] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0 > [ +14.210987] device mon0 entered promiscuous mode > [Jan24 18:20] device mon0 left promiscuous mode > > --- > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > index 83abfe996138..591f9fdd0ec4 100644 > --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > @@ -654,9 +654,13 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct > iwl_mvm *mvm, > u32 action) > { > struct iwl_mac_ctx_cmd cmd = {}; > - u32 tfd_queue_msk = BIT(mvm->snif_queue); > + u32 tfd_queue_msk = 0; > int ret; > > + if (mvm->snif_queue != IWL_MVM_INVALID_QUEUE) { > + tfd_queue_msk = BIT(mvm->snif_queue); > + } > + > WARN_ON(vif->type != NL80211_IFTYPE_MONITOR); > > iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action); > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > index 515dd3e0730d..784a7f72b819 100644 > --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > @@ -2084,6 +2084,7 @@ static int iwl_mvm_add_int_sta_with_queue(struct > iwl_mvm *mvm, int macidx, > } > > *queue = txq; > + sta->tfd_queue_msk = BIT(*queue); > } > > return 0; > @@ -2092,11 +2093,15 @@ static int > iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx, > int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id) > { > int ret; > + u32 tfd_queue_msk = 0; > > lockdep_assert_held(&mvm->mutex); > + if (mvm->aux_queue != IWL_MVM_INVALID_QUEUE) { > + tfd_queue_msk = BIT(mvm->aux_queue); > + } > > /* Allocate aux station and assign to it the aux queue */ > - ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue), > + ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, tfd_queue_msk, > NL80211_IFTYPE_UNSPECIFIED, > IWL_STA_AUX_ACTIVITY); > if (ret) This bug was already fixed in our internal repository. You can use the driver from [1] until this fix will make it's way to the upstream (please use master branch). [1] https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/backport-iwlwifi.git
On Mon, Jan 23, 2023 at 6:29 PM Greenman, Gregory <gregory.greenman@intel.com> wrote: > > Hi! > > On Mon, 2023-01-23 at 14:07 +0530, Krishna Chaitanya wrote: > > Hi, > > > > I am using the ath.git (ea726a8d305a : was testing external Atheros > > cards) + linux-firmware (bb2d42d) and the monitor mode stopped working > > with iwlwifi. There is a warning from UBSAN: "UBSAN: > > shift-out-of-bounds in > > drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c:657:22". > > > > The root cause was this c6ce1c74ef292 which defaults the assignment to > > 0xFFFF and then that was being used for the shift. > > > > I have tried to fix the warning with the below patch (just for an > > experiment, not a proper one), the warning is gone, but no frames are > > seen in Wireshark. > > > > Below is the version information, any help is appreciated, monitor > > mode is important for me. > > > > lspci: Network controller: Intel Corporation Wi-Fi 6 AX201 (rev 20) > > > > [ +0.003790] iwlwifi 0000:00:14.3: api flags index 2 larger than > > supported by driver > > [ +0.000046] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: > > 89.3.35.37 > > [ +0.001519] iwlwifi 0000:00:14.3: loaded firmware version > > 72.daa05125.0 QuZ-a0-hr-b0-72.ucode op_mode iwlmvm > > [ +0.034887] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6 AX201 > > 160MHz, REV=0x351 > > [ +0.000083] thermal thermal_zone7: failed to read out thermal zone (-61) > > [ +0.122144] iwlwifi 0000:00:14.3: Detected RF HR B3, rfid=0x10a100 > > [ +0.065701] iwlwifi 0000:00:14.3: base HW address: 4c:79:6e:90:94:71 > > [ +0.019826] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0 > > [ +14.210987] device mon0 entered promiscuous mode > > [Jan24 18:20] device mon0 left promiscuous mode > > > > --- > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > index 83abfe996138..591f9fdd0ec4 100644 > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > @@ -654,9 +654,13 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct > > iwl_mvm *mvm, > > u32 action) > > { > > struct iwl_mac_ctx_cmd cmd = {}; > > - u32 tfd_queue_msk = BIT(mvm->snif_queue); > > + u32 tfd_queue_msk = 0; > > int ret; > > > > + if (mvm->snif_queue != IWL_MVM_INVALID_QUEUE) { > > + tfd_queue_msk = BIT(mvm->snif_queue); > > + } > > + > > WARN_ON(vif->type != NL80211_IFTYPE_MONITOR); > > > > iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action); > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > index 515dd3e0730d..784a7f72b819 100644 > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > @@ -2084,6 +2084,7 @@ static int iwl_mvm_add_int_sta_with_queue(struct > > iwl_mvm *mvm, int macidx, > > } > > > > *queue = txq; > > + sta->tfd_queue_msk = BIT(*queue); > > } > > > > return 0; > > @@ -2092,11 +2093,15 @@ static int > > iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx, > > int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id) > > { > > int ret; > > + u32 tfd_queue_msk = 0; > > > > lockdep_assert_held(&mvm->mutex); > > + if (mvm->aux_queue != IWL_MVM_INVALID_QUEUE) { > > + tfd_queue_msk = BIT(mvm->aux_queue); > > + } > > > > /* Allocate aux station and assign to it the aux queue */ > > - ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue), > > + ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, tfd_queue_msk, > > NL80211_IFTYPE_UNSPECIFIED, > > IWL_STA_AUX_ACTIVITY); > > if (ret) > > This bug was already fixed in our internal repository. You can use the driver from [1] > until this fix will make it's way to the upstream (please use master branch). > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/backport-iwlwifi.git Excellent, thanks for the reply. The driver in [1] was not building due to 20b0b53aca436 but it was simple to fix, after installation (and reboot) the warning is gone, but still unable to sniff. FW is now updated to: loaded firmware version 77.2dda880d.0 QuZ-a0-hr-b0-77.ucode op_mode iwlmvm
On Mon, Jan 23, 2023 at 9:41 PM Krishna Chaitanya <chaitanya.mgit@gmail.com> wrote: > > On Mon, Jan 23, 2023 at 6:29 PM Greenman, Gregory > <gregory.greenman@intel.com> wrote: > > > > Hi! > > > > On Mon, 2023-01-23 at 14:07 +0530, Krishna Chaitanya wrote: > > > Hi, > > > > > > I am using the ath.git (ea726a8d305a : was testing external Atheros > > > cards) + linux-firmware (bb2d42d) and the monitor mode stopped working > > > with iwlwifi. There is a warning from UBSAN: "UBSAN: > > > shift-out-of-bounds in > > > drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c:657:22". > > > > > > The root cause was this c6ce1c74ef292 which defaults the assignment to > > > 0xFFFF and then that was being used for the shift. > > > > > > I have tried to fix the warning with the below patch (just for an > > > experiment, not a proper one), the warning is gone, but no frames are > > > seen in Wireshark. > > > > > > Below is the version information, any help is appreciated, monitor > > > mode is important for me. > > > > > > lspci: Network controller: Intel Corporation Wi-Fi 6 AX201 (rev 20) > > > > > > [ +0.003790] iwlwifi 0000:00:14.3: api flags index 2 larger than > > > supported by driver > > > [ +0.000046] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: > > > 89.3.35.37 > > > [ +0.001519] iwlwifi 0000:00:14.3: loaded firmware version > > > 72.daa05125.0 QuZ-a0-hr-b0-72.ucode op_mode iwlmvm > > > [ +0.034887] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6 AX201 > > > 160MHz, REV=0x351 > > > [ +0.000083] thermal thermal_zone7: failed to read out thermal zone (-61) > > > [ +0.122144] iwlwifi 0000:00:14.3: Detected RF HR B3, rfid=0x10a100 > > > [ +0.065701] iwlwifi 0000:00:14.3: base HW address: 4c:79:6e:90:94:71 > > > [ +0.019826] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0 > > > [ +14.210987] device mon0 entered promiscuous mode > > > [Jan24 18:20] device mon0 left promiscuous mode > > > > > > --- > > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > index 83abfe996138..591f9fdd0ec4 100644 > > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > @@ -654,9 +654,13 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct > > > iwl_mvm *mvm, > > > u32 action) > > > { > > > struct iwl_mac_ctx_cmd cmd = {}; > > > - u32 tfd_queue_msk = BIT(mvm->snif_queue); > > > + u32 tfd_queue_msk = 0; > > > int ret; > > > > > > + if (mvm->snif_queue != IWL_MVM_INVALID_QUEUE) { > > > + tfd_queue_msk = BIT(mvm->snif_queue); > > > + } > > > + > > > WARN_ON(vif->type != NL80211_IFTYPE_MONITOR); > > > > > > iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action); > > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > index 515dd3e0730d..784a7f72b819 100644 > > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > @@ -2084,6 +2084,7 @@ static int iwl_mvm_add_int_sta_with_queue(struct > > > iwl_mvm *mvm, int macidx, > > > } > > > > > > *queue = txq; > > > + sta->tfd_queue_msk = BIT(*queue); > > > } > > > > > > return 0; > > > @@ -2092,11 +2093,15 @@ static int > > > iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx, > > > int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id) > > > { > > > int ret; > > > + u32 tfd_queue_msk = 0; > > > > > > lockdep_assert_held(&mvm->mutex); > > > + if (mvm->aux_queue != IWL_MVM_INVALID_QUEUE) { > > > + tfd_queue_msk = BIT(mvm->aux_queue); > > > + } > > > > > > /* Allocate aux station and assign to it the aux queue */ > > > - ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue), > > > + ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, tfd_queue_msk, > > > NL80211_IFTYPE_UNSPECIFIED, > > > IWL_STA_AUX_ACTIVITY); > > > if (ret) > > > > This bug was already fixed in our internal repository. You can use the driver from [1] > > until this fix will make it's way to the upstream (please use master branch). > > > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/backport-iwlwifi.git > Excellent, thanks for the reply. > > The driver in [1] was not building due to 20b0b53aca436 but it was simple > to fix, after installation (and reboot) the warning is gone, but still > unable to sniff. > > FW is now updated to: > loaded firmware version 77.2dda880d.0 QuZ-a0-hr-b0-77.ucode op_mode iwlmvm Any ideas why it still doesn't work? Should I try something else, please suggest.
On Thu, 2023-02-09 at 20:21 +0530, Krishna Chaitanya wrote: > On Mon, Jan 23, 2023 at 9:41 PM Krishna Chaitanya > <chaitanya.mgit@gmail.com> wrote: > > > > On Mon, Jan 23, 2023 at 6:29 PM Greenman, Gregory > > <gregory.greenman@intel.com> wrote: > > > > > > Hi! > > > > > > On Mon, 2023-01-23 at 14:07 +0530, Krishna Chaitanya wrote: > > > > Hi, > > > > > > > > I am using the ath.git (ea726a8d305a : was testing external Atheros > > > > cards) + linux-firmware (bb2d42d) and the monitor mode stopped working > > > > with iwlwifi. There is a warning from UBSAN: "UBSAN: > > > > shift-out-of-bounds in > > > > drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c:657:22". > > > > > > > > The root cause was this c6ce1c74ef292 which defaults the assignment to > > > > 0xFFFF and then that was being used for the shift. > > > > > > > > I have tried to fix the warning with the below patch (just for an > > > > experiment, not a proper one), the warning is gone, but no frames are > > > > seen in Wireshark. > > > > > > > > Below is the version information, any help is appreciated, monitor > > > > mode is important for me. > > > > > > > > lspci: Network controller: Intel Corporation Wi-Fi 6 AX201 (rev 20) > > > > > > > > [ +0.003790] iwlwifi 0000:00:14.3: api flags index 2 larger than > > > > supported by driver > > > > [ +0.000046] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: > > > > 89.3.35.37 > > > > [ +0.001519] iwlwifi 0000:00:14.3: loaded firmware version > > > > 72.daa05125.0 QuZ-a0-hr-b0-72.ucode op_mode iwlmvm > > > > [ +0.034887] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6 AX201 > > > > 160MHz, REV=0x351 > > > > [ +0.000083] thermal thermal_zone7: failed to read out thermal zone (-61) > > > > [ +0.122144] iwlwifi 0000:00:14.3: Detected RF HR B3, rfid=0x10a100 > > > > [ +0.065701] iwlwifi 0000:00:14.3: base HW address: 4c:79:6e:90:94:71 > > > > [ +0.019826] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0 > > > > [ +14.210987] device mon0 entered promiscuous mode > > > > [Jan24 18:20] device mon0 left promiscuous mode > > > > > > > > --- > > > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > index 83abfe996138..591f9fdd0ec4 100644 > > > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > @@ -654,9 +654,13 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct > > > > iwl_mvm *mvm, > > > > u32 action) > > > > { > > > > struct iwl_mac_ctx_cmd cmd = {}; > > > > - u32 tfd_queue_msk = BIT(mvm->snif_queue); > > > > + u32 tfd_queue_msk = 0; > > > > int ret; > > > > > > > > + if (mvm->snif_queue != IWL_MVM_INVALID_QUEUE) { > > > > + tfd_queue_msk = BIT(mvm->snif_queue); > > > > + } > > > > + > > > > WARN_ON(vif->type != NL80211_IFTYPE_MONITOR); > > > > > > > > iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action); > > > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > index 515dd3e0730d..784a7f72b819 100644 > > > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > @@ -2084,6 +2084,7 @@ static int iwl_mvm_add_int_sta_with_queue(struct > > > > iwl_mvm *mvm, int macidx, > > > > } > > > > > > > > *queue = txq; > > > > + sta->tfd_queue_msk = BIT(*queue); > > > > } > > > > > > > > return 0; > > > > @@ -2092,11 +2093,15 @@ static int > > > > iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx, > > > > int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id) > > > > { > > > > int ret; > > > > + u32 tfd_queue_msk = 0; > > > > > > > > lockdep_assert_held(&mvm->mutex); > > > > + if (mvm->aux_queue != IWL_MVM_INVALID_QUEUE) { > > > > + tfd_queue_msk = BIT(mvm->aux_queue); > > > > + } > > > > > > > > /* Allocate aux station and assign to it the aux queue */ > > > > - ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue), > > > > + ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, tfd_queue_msk, > > > > NL80211_IFTYPE_UNSPECIFIED, > > > > IWL_STA_AUX_ACTIVITY); > > > > if (ret) > > > > > > This bug was already fixed in our internal repository. You can use the driver from [1] > > > until this fix will make it's way to the upstream (please use master branch). > > > > > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/backport-iwlwifi.git > > Excellent, thanks for the reply. > > > > The driver in [1] was not building due to 20b0b53aca436 but it was simple > > to fix, after installation (and reboot) the warning is gone, but still > > unable to sniff. > > > > FW is now updated to: > > loaded firmware version 77.2dda880d.0 QuZ-a0-hr-b0-77.ucode op_mode iwlmvm > Any ideas why it still doesn't work? Should I try something else, > please suggest. We'll try to reproduce it internally, if it'll work for us, I'll ask for some debug data. Can you please open a new ticket in bugzilla, so it'll be easier for us to track? Here is the wiki page regarding bugzilla for iwlwifi: https://wireless.wiki.kernel.org/en/users/drivers/iwlwifi/debugging
On Sun, Feb 12, 2023 at 3:35 PM Greenman, Gregory <gregory.greenman@intel.com> wrote: > > On Thu, 2023-02-09 at 20:21 +0530, Krishna Chaitanya wrote: > > On Mon, Jan 23, 2023 at 9:41 PM Krishna Chaitanya > > <chaitanya.mgit@gmail.com> wrote: > > > > > > On Mon, Jan 23, 2023 at 6:29 PM Greenman, Gregory > > > <gregory.greenman@intel.com> wrote: > > > > > > > > Hi! > > > > > > > > On Mon, 2023-01-23 at 14:07 +0530, Krishna Chaitanya wrote: > > > > > Hi, > > > > > > > > > > I am using the ath.git (ea726a8d305a : was testing external Atheros > > > > > cards) + linux-firmware (bb2d42d) and the monitor mode stopped working > > > > > with iwlwifi. There is a warning from UBSAN: "UBSAN: > > > > > shift-out-of-bounds in > > > > > drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c:657:22". > > > > > > > > > > The root cause was this c6ce1c74ef292 which defaults the assignment to > > > > > 0xFFFF and then that was being used for the shift. > > > > > > > > > > I have tried to fix the warning with the below patch (just for an > > > > > experiment, not a proper one), the warning is gone, but no frames are > > > > > seen in Wireshark. > > > > > > > > > > Below is the version information, any help is appreciated, monitor > > > > > mode is important for me. > > > > > > > > > > lspci: Network controller: Intel Corporation Wi-Fi 6 AX201 (rev 20) > > > > > > > > > > [ +0.003790] iwlwifi 0000:00:14.3: api flags index 2 larger than > > > > > supported by driver > > > > > [ +0.000046] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: > > > > > 89.3.35.37 > > > > > [ +0.001519] iwlwifi 0000:00:14.3: loaded firmware version > > > > > 72.daa05125.0 QuZ-a0-hr-b0-72.ucode op_mode iwlmvm > > > > > [ +0.034887] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6 AX201 > > > > > 160MHz, REV=0x351 > > > > > [ +0.000083] thermal thermal_zone7: failed to read out thermal zone (-61) > > > > > [ +0.122144] iwlwifi 0000:00:14.3: Detected RF HR B3, rfid=0x10a100 > > > > > [ +0.065701] iwlwifi 0000:00:14.3: base HW address: 4c:79:6e:90:94:71 > > > > > [ +0.019826] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0 > > > > > [ +14.210987] device mon0 entered promiscuous mode > > > > > [Jan24 18:20] device mon0 left promiscuous mode > > > > > > > > > > --- > > > > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > > b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > > index 83abfe996138..591f9fdd0ec4 100644 > > > > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c > > > > > @@ -654,9 +654,13 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct > > > > > iwl_mvm *mvm, > > > > > u32 action) > > > > > { > > > > > struct iwl_mac_ctx_cmd cmd = {}; > > > > > - u32 tfd_queue_msk = BIT(mvm->snif_queue); > > > > > + u32 tfd_queue_msk = 0; > > > > > int ret; > > > > > > > > > > + if (mvm->snif_queue != IWL_MVM_INVALID_QUEUE) { > > > > > + tfd_queue_msk = BIT(mvm->snif_queue); > > > > > + } > > > > > + > > > > > WARN_ON(vif->type != NL80211_IFTYPE_MONITOR); > > > > > > > > > > iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action); > > > > > diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > > b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > > index 515dd3e0730d..784a7f72b819 100644 > > > > > --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > > +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c > > > > > @@ -2084,6 +2084,7 @@ static int iwl_mvm_add_int_sta_with_queue(struct > > > > > iwl_mvm *mvm, int macidx, > > > > > } > > > > > > > > > > *queue = txq; > > > > > + sta->tfd_queue_msk = BIT(*queue); > > > > > } > > > > > > > > > > return 0; > > > > > @@ -2092,11 +2093,15 @@ static int > > > > > iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx, > > > > > int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id) > > > > > { > > > > > int ret; > > > > > + u32 tfd_queue_msk = 0; > > > > > > > > > > lockdep_assert_held(&mvm->mutex); > > > > > + if (mvm->aux_queue != IWL_MVM_INVALID_QUEUE) { > > > > > + tfd_queue_msk = BIT(mvm->aux_queue); > > > > > + } > > > > > > > > > > /* Allocate aux station and assign to it the aux queue */ > > > > > - ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue), > > > > > + ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, tfd_queue_msk, > > > > > NL80211_IFTYPE_UNSPECIFIED, > > > > > IWL_STA_AUX_ACTIVITY); > > > > > if (ret) > > > > > > > > This bug was already fixed in our internal repository. You can use the driver from [1] > > > > until this fix will make it's way to the upstream (please use master branch). > > > > > > > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/backport-iwlwifi.git > > > Excellent, thanks for the reply. > > > > > > The driver in [1] was not building due to 20b0b53aca436 but it was simple > > > to fix, after installation (and reboot) the warning is gone, but still > > > unable to sniff. > > > > > > FW is now updated to: > > > loaded firmware version 77.2dda880d.0 QuZ-a0-hr-b0-77.ucode op_mode iwlmvm > > Any ideas why it still doesn't work? Should I try something else, > > please suggest. > > We'll try to reproduce it internally, if it'll work for us, I'll ask for some debug data. > Can you please open a new ticket in bugzilla, so it'll be easier for us to track? > Here is the wiki page regarding bugzilla for iwlwifi: > https://wireless.wiki.kernel.org/en/users/drivers/iwlwifi/debugging FYI, https://bugzilla.kernel.org/show_bug.cgi?id=217051
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c index 83abfe996138..591f9fdd0ec4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c @@ -654,9 +654,13 @@ static int iwl_mvm_mac_ctxt_cmd_listener(struct iwl_mvm *mvm, u32 action) { struct iwl_mac_ctx_cmd cmd = {}; - u32 tfd_queue_msk = BIT(mvm->snif_queue); + u32 tfd_queue_msk = 0; int ret; + if (mvm->snif_queue != IWL_MVM_INVALID_QUEUE) { + tfd_queue_msk = BIT(mvm->snif_queue); + } + WARN_ON(vif->type != NL80211_IFTYPE_MONITOR); iwl_mvm_mac_ctxt_cmd_common(mvm, vif, &cmd, NULL, action); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 515dd3e0730d..784a7f72b819 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -2084,6 +2084,7 @@ static int iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx, } *queue = txq; + sta->tfd_queue_msk = BIT(*queue); }
Hi, I am using the ath.git (ea726a8d305a : was testing external Atheros cards) + linux-firmware (bb2d42d) and the monitor mode stopped working with iwlwifi. There is a warning from UBSAN: "UBSAN: shift-out-of-bounds in drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c:657:22". The root cause was this c6ce1c74ef292 which defaults the assignment to 0xFFFF and then that was being used for the shift. I have tried to fix the warning with the below patch (just for an experiment, not a proper one), the warning is gone, but no frames are seen in Wireshark. Below is the version information, any help is appreciated, monitor mode is important for me. lspci: Network controller: Intel Corporation Wi-Fi 6 AX201 (rev 20) [ +0.003790] iwlwifi 0000:00:14.3: api flags index 2 larger than supported by driver [ +0.000046] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: 89.3.35.37 [ +0.001519] iwlwifi 0000:00:14.3: loaded firmware version 72.daa05125.0 QuZ-a0-hr-b0-72.ucode op_mode iwlmvm [ +0.034887] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6 AX201 160MHz, REV=0x351 [ +0.000083] thermal thermal_zone7: failed to read out thermal zone (-61) [ +0.122144] iwlwifi 0000:00:14.3: Detected RF HR B3, rfid=0x10a100 [ +0.065701] iwlwifi 0000:00:14.3: base HW address: 4c:79:6e:90:94:71 [ +0.019826] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0 [ +14.210987] device mon0 entered promiscuous mode [Jan24 18:20] device mon0 left promiscuous mode --- return 0; @@ -2092,11 +2093,15 @@ static int iwl_mvm_add_int_sta_with_queue(struct iwl_mvm *mvm, int macidx, int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm, u32 lmac_id) { int ret; + u32 tfd_queue_msk = 0; lockdep_assert_held(&mvm->mutex); + if (mvm->aux_queue != IWL_MVM_INVALID_QUEUE) { + tfd_queue_msk = BIT(mvm->aux_queue); + } /* Allocate aux station and assign to it the aux queue */ - ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue), + ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, tfd_queue_msk, NL80211_IFTYPE_UNSPECIFIED, IWL_STA_AUX_ACTIVITY); if (ret)