From: Johannes Berg <[email protected]>
After the fix from Jiri that disabled local IRQs instead of
just BHs (necessary to fix an issue with submitting a command
with IRQs already disabled), there was still a situation in
which we could deep in there enable BHs, if the device config
sets the apmg_wake_up_wa configuration, which is true on all
7000 series devices.
To fix that, but not require reverting commit 1ed08f6fb5ae
("iwlwifi: remove flags argument for nic_access"), split up
nic access into a version with BH manipulation to use most
of the time, and without it for this specific case where the
local IRQs are already disabled.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
---
.../wireless/intel/iwlwifi/pcie/internal.h | 5 ++++
.../net/wireless/intel/iwlwifi/pcie/trans.c | 24 ++++++++++++++++---
drivers/net/wireless/intel/iwlwifi/pcie/tx.c | 4 ++--
3 files changed, 28 insertions(+), 5 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index d9688c7bed07..76a512cd2e5c 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -447,6 +447,11 @@ struct iwl_trans
const struct iwl_cfg_trans_params *cfg_trans);
void iwl_trans_pcie_free(struct iwl_trans *trans);
+bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans);
+#define _iwl_trans_pcie_grab_nic_access(trans) \
+ __cond_lock(nic_access_nobh, \
+ likely(__iwl_trans_pcie_grab_nic_access(trans)))
+
/*****************************************************
* RX
******************************************************/
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 861dbc03d183..239bc177a3e5 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1978,12 +1978,16 @@ static void iwl_trans_pcie_removal_wk(struct work_struct *wk)
module_put(THIS_MODULE);
}
-static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
+/*
+ * This version doesn't disable BHs but rather assumes they're
+ * already disabled.
+ */
+bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
{
int ret;
struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
- spin_lock_bh(&trans_pcie->reg_lock);
+ spin_lock(&trans_pcie->reg_lock);
if (trans_pcie->cmd_hold_nic_awake)
goto out;
@@ -2068,7 +2072,7 @@ static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
}
err:
- spin_unlock_bh(&trans_pcie->reg_lock);
+ spin_unlock(&trans_pcie->reg_lock);
return false;
}
@@ -2081,6 +2085,20 @@ static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
return true;
}
+static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
+{
+ bool ret;
+
+ local_bh_disable();
+ ret = __iwl_trans_pcie_grab_nic_access(trans);
+ if (ret) {
+ /* keep BHs disabled until iwl_trans_pcie_release_nic_access */
+ return ret;
+ }
+ local_bh_enable();
+ return false;
+}
+
static void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
{
struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index 0346505351f5..4f6c187eed69 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -643,7 +643,7 @@ static int iwl_pcie_set_cmd_in_flight(struct iwl_trans *trans,
* returned. This needs to be done only on NICs that have
* apmg_wake_up_wa set (see above.)
*/
- if (!iwl_trans_grab_nic_access(trans))
+ if (!_iwl_trans_pcie_grab_nic_access(trans))
return -EIO;
/*
@@ -652,7 +652,7 @@ static int iwl_pcie_set_cmd_in_flight(struct iwl_trans *trans,
* already true, so it's OK to unconditionally set it to true.
*/
trans_pcie->cmd_hold_nic_awake = true;
- spin_unlock_bh(&trans_pcie->reg_lock);
+ spin_unlock(&trans_pcie->reg_lock);
return 0;
}
--
2.31.0
On Thu, 2021-04-15 at 16:48 +0300, Luca Coelho wrote:
> From: Johannes Berg <[email protected]>
>
> After the fix from Jiri that disabled local IRQs instead of
> just BHs (necessary to fix an issue with submitting a command
> with IRQs already disabled), there was still a situation in
> which we could deep in there enable BHs, if the device config
> sets the apmg_wake_up_wa configuration, which is true on all
> 7000 series devices.
>
> To fix that, but not require reverting commit 1ed08f6fb5ae
> ("iwlwifi: remove flags argument for nic_access"), split up
> nic access into a version with BH manipulation to use most
> of the time, and without it for this specific case where the
> local IRQs are already disabled.
>
> Signed-off-by: Johannes Berg <[email protected]>
> Signed-off-by: Luca Coelho <[email protected]>
> ---
Kalle, I'm going to assign this one to you too, but this time for -next
and not for -fixes.
This is related to the fix Jiri made, but to avoid some conflicts and a
broken v5.13-rc, we want to get it into v5.13 if still possible.
So can you please take this on top of my latest pull request?
Additionally, can you tag it for stable?
Thanks!
--
Cheers,
Luca.
Luca Coelho <[email protected]> writes:
> On Thu, 2021-04-15 at 16:48 +0300, Luca Coelho wrote:
>> From: Johannes Berg <[email protected]>
>>
>> After the fix from Jiri that disabled local IRQs instead of
>> just BHs (necessary to fix an issue with submitting a command
>> with IRQs already disabled), there was still a situation in
>> which we could deep in there enable BHs, if the device config
>> sets the apmg_wake_up_wa configuration, which is true on all
>> 7000 series devices.
>>
>> To fix that, but not require reverting commit 1ed08f6fb5ae
>> ("iwlwifi: remove flags argument for nic_access"), split up
>> nic access into a version with BH manipulation to use most
>> of the time, and without it for this specific case where the
>> local IRQs are already disabled.
>>
>> Signed-off-by: Johannes Berg <[email protected]>
>> Signed-off-by: Luca Coelho <[email protected]>
>> ---
>
> Kalle, I'm going to assign this one to you too, but this time for -next
> and not for -fixes.
>
> This is related to the fix Jiri made, but to avoid some conflicts and a
> broken v5.13-rc, we want to get it into v5.13 if still possible.
>
> So can you please take this on top of my latest pull request?
> Additionally, can you tag it for stable?
Will do.
--
https://patchwork.kernel.org/project/linux-wireless/list/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
Luca Coelho <[email protected]> wrote:
> From: Johannes Berg <[email protected]>
>
> After the fix from Jiri that disabled local IRQs instead of
> just BHs (necessary to fix an issue with submitting a command
> with IRQs already disabled), there was still a situation in
> which we could deep in there enable BHs, if the device config
> sets the apmg_wake_up_wa configuration, which is true on all
> 7000 series devices.
>
> To fix that, but not require reverting commit 1ed08f6fb5ae
> ("iwlwifi: remove flags argument for nic_access"), split up
> nic access into a version with BH manipulation to use most
> of the time, and without it for this specific case where the
> local IRQs are already disabled.
>
> Signed-off-by: Johannes Berg <[email protected]>
> Signed-off-by: Luca Coelho <[email protected]>
Patch applied to wireless-drivers-next.git, thanks.
c544d89b0d67 iwlwifi: pcie: don't enable BHs with IRQs disabled
--
https://patchwork.kernel.org/project/linux-wireless/patch/iwlwifi.20210415164821.d0f2edda1651.I75f762e0bed38914d1300ea198b86dd449b4b206@changeid/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches