Skip to contentSkip to conversation

Silence r8712u info and add verbose option v2

PATCH In review Alexandre Demers 3 participants 5 messages Nov 12, 2013 Staging

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Silence r8712u info and add verbose option v2

r8712u pollutes dmesg and logs. Silence it and add a verbose option to KConfig if we ever really want to hear about it.

v2: keep netdev_info instead of replacing it by a KERN_NOTICE

---
 drivers/staging/rtl8712/Kconfig        | 7 +++++++
 drivers/staging/rtl8712/rtl871x_mlme.c | 7 ++++---
 2 files changed, 11 insertions(+), 3 deletions(-)
drivers/staging/rtl8712/Kconfig +7 -0
index 6a43312..7cae8f8 100644
@@ -16,4 +16,11 @@ config R8712_TX_AGGR
 	---help---
 	This option provides transmit aggregation for the Realtek RTL8712 USB device.
 
+config R8712U_DEBUG
+	bool "Realtek RTL8712U Wireless verbose debug"
+	depends on R8712U
+	help
+	  Say Y here in order to have the rtl8712u code generate
+	  verbose debugging messages.
+
 
drivers/staging/rtl8712/rtl871x_mlme.c +1 -0
index 6596154..1226009 100644
@@ -1043,9 +1043,12 @@ void r8712_got_addbareq_event_callback(struct _adapter *adapter, u8 *pbuf)
 	struct	sta_priv *pstapriv = &adapter->stapriv;
 	struct	recv_reorder_ctrl *precvreorder_ctrl = NULL;
 
+#ifdef CONFIG_R8712U_DEBUG
	netdev_info(adapter->pnetdev, "%s: mac = %pM, seq = %d, tid = %d\n",
		    __func__, pAddbareq_pram->MacAddress,
	    pAddbareq_pram->StartSeqNum, pAddbareq_pram->tid);

+#endif
+
psta = r8712_get_stainfo(pstapriv, pAddbareq_pram->MacAddress); if (psta) {
precvreorder_ctrl =

Signature · 1 line
1.8.4.2
#1 lore raw
Quoted 9 lines · Alexandre Demers

r8712u pollutes dmesg and logs. Silence it and add a verbose option to KConfig if we ever really want to hear about it.

v2: keep netdev_info instead of replacing it by a KERN_NOTICE

---
  drivers/staging/rtl8712/Kconfig        | 7 +++++++
  drivers/staging/rtl8712/rtl871x_mlme.c | 7 ++++---
  2 files changed, 11 insertions(+), 3 deletions(-)

Do you really want the "v2" to be part of the patch title in the git repos? I expect not, thus you should put it inside the []. All that stuff is stripped off when the patch is merged.

Similarly, do you want the v2 line in the patch description to be part of the permanent record? If not, move it below the --- line.

You are also missing a Signed-off-by: line. Have you read Documentation/SubmittingPatches? At least you should run every patch through scripts/checkpatch.pl before submitting it.

Finally, why do you want to add another Kconfig variable? I suggest either deleting the logging statement, or commenting it out. I'm not sure that this info would ever need logging.

I NACK this patch.

Larry

#2 lore raw
Quoted 18 lines · Larry Finger
Quoted 9 lines · Alexandre Demers

r8712u pollutes dmesg and logs. Silence it and add a verbose option to KConfig if we ever really want to hear about it.

v2: keep netdev_info instead of replacing it by a KERN_NOTICE

---
  drivers/staging/rtl8712/Kconfig        | 7 +++++++
  drivers/staging/rtl8712/rtl871x_mlme.c | 7 ++++---
  2 files changed, 11 insertions(+), 3 deletions(-)

Do you really want the "v2" to be part of the patch title in the git repos? I expect not, thus you should put it inside the []. All that stuff is stripped off when the patch is merged.

Similarly, do you want the v2 line in the patch description to be part of the permanent record? If not, move it below the --- line.

Quoted 7 lines

You are also missing a Signed-off-by: line. Have you read Documentation/SubmittingPatches? At least you should run every patch through scripts/checkpatch.pl before submitting it.

Finally, why do you want to add another Kconfig variable? I suggest either deleting the logging statement, or commenting it out. I'm not sure that this info would ever need logging.

To be a bit more clear here, we are activly _removing_ driver and subsystem specific DEBUG configuration flags, and moving everything to using the dynamic debug infrastructure in the kernel, as no user ever rebuilds their kernels/drivers, and we want to be able to get debugging information from them at times.

thanks,

greg k-h

#3 lore raw

Hi Larry and Greg,
You are right about the v2 thing and the Signed-off-by. It's the first time I submit a patch directly to the kernel list, I didn't go to the bottom of the SubmittingPatches, I should have read the whole document and not stop in the middle of it. I'll finish reading it and do what is expected. Greg, I understand your comment: I would have had the answer if I had read the whole document (it's in there, I've just seen it).

However, I'll take the opportunity you are offering me, Larry, with your last question: I was asking myself the same question about simply deleting the statement which seems pretty useless to me. On the other hand, should we continue writing patches against r8712u? I'm not aware of the conclusion about replacing r8712u by r92su. Was there any decision taken on the subject?

Thank you,
Alexandre D

Quoted 36 lines · Greg Kroah-Hartman
Quoted 18 lines · Larry Finger
Quoted 9 lines · Alexandre Demers

r8712u pollutes dmesg and logs. Silence it and add a verbose option to KConfig if we ever really want to hear about it.

v2: keep netdev_info instead of replacing it by a KERN_NOTICE

---
  drivers/staging/rtl8712/Kconfig        | 7 +++++++
  drivers/staging/rtl8712/rtl871x_mlme.c | 7 ++++---
  2 files changed, 11 insertions(+), 3 deletions(-)

Do you really want the "v2" to be part of the patch title in the git repos? I expect not, thus you should put it inside the []. All that stuff is stripped off when the patch is merged.

Similarly, do you want the v2 line in the patch description to be part of the permanent record? If not, move it below the --- line.

Quoted 7 lines

You are also missing a Signed-off-by: line. Have you read Documentation/SubmittingPatches? At least you should run every patch through scripts/checkpatch.pl before submitting it.

Finally, why do you want to add another Kconfig variable? I suggest either deleting the logging statement, or commenting it out. I'm not sure that this info would ever need logging.

To be a bit more clear here, we are activly _removing_ driver and subsystem specific DEBUG configuration flags, and moving everything to using the dynamic debug infrastructure in the kernel, as no user ever rebuilds their kernels/drivers, and we want to be able to get debugging information from them at times.

thanks,

greg k-h

#4 lore raw
Quoted 9 lines · Alexandre Demers

Hi Larry and Greg,
You are right about the v2 thing and the Signed-off-by. It's the first time I submit a patch directly to the kernel list, I didn't go to the bottom of the SubmittingPatches, I should have read the whole document and not stop in the middle of it. I'll finish reading it and do what is expected. Greg, I understand your comment: I would have had the answer if I had read the whole document (it's in there, I've just seen it).

Quoted 6 lines

However, I'll take the opportunity you are offering me, Larry, with your last question: I was asking myself the same question about simply deleting the statement which seems pretty useless to me. On the other hand, should we continue writing patches against r8712u? I'm not aware of the conclusion about replacing r8712u by r92su. Was there any decision taken on the subject?

I am not involved in the r92su project. My impression is that the latest proponent of that change was not happy with the information that r92su would still be in staging, but without mac80211 there is no other choice.

If there are parts of r8712u that are problems for you, then patch them. I will review them, but I am not likely to contribute other than for things that are bugs.

Larry