Skip to contentSkip to conversation

b43: Fix suspend/resume

PATCH In review Michael Buesch 2 participants 4 messages Jan 25, 2008 – Feb 7, 2008

lore
Share & tools
lore

Michael Buesch

author
#0 lore raw

PATCH 1/3 b43: Fix suspend/resume

This patch makes suspend/resume work with the b43 driver.
We must not overwrite the MAC addresses in the init function, as this would also overwrite the MAC on resume. With an all-zero MAC the device firmware is not able to ACK any received packets anymore.
Fix this by moving the initializion stuff that must be done on init but not on resume to the start function.
Also zero out filter_flags to make sure we don't have some flags from a previous instance for a tiny timeframe until mac80211 reconfigures them.

Signed-off-by Michael Buesch <[email protected]>

---

This patch is upstream inside of the netdev tree.

drivers/net/wireless/b43/main.c +7 -2
Index: linux-2.6.24/drivers/net/wireless/b43/main.c
===================================================================
@@ -3395,8 +3395,6 @@ static int b43_wireless_core_init(struct
 	b43_bluetooth_coext_enable(dev);
 
 	ssb_bus_powerup(bus, 1);	/* Enable dynamic PCTL */
-	memset(wl->bssid, 0, ETH_ALEN);
-	memset(wl->mac_addr, 0, ETH_ALEN);
 	b43_upload_card_macaddress(dev);
 	b43_security_init(dev);
 	b43_rng_init(wl);

@@ -3493,6 +3491,13 @@ static int b43_start(struct ieee80211_hw
 	int did_init = 0;
 	int err = 0;
 
+	/* Kill all old instance specific information to make sure
+	 * the card won't use it in the short timeframe between start
+	 * and mac80211 reconfiguring it. */
+	memset(wl->bssid, 0, ETH_ALEN);
+	memset(wl->mac_addr, 0, ETH_ALEN);
+	wl->filter_flags = 0;
+
 	/* First register RFkill.
 	 * LEDs that are registered later depend on it. */
 	b43_rfkill_init(dev);
#1 lore raw
Quoted 10 lines · Michael Buesch

This patch makes suspend/resume work with the b43 driver.
We must not overwrite the MAC addresses in the init function, as this would also overwrite the MAC on resume. With an all-zero MAC the device firmware is not able to ACK any received packets anymore.
Fix this by moving the initializion stuff that must be done on init but not on resume to the start function.
Also zero out filter_flags to make sure we don't have some flags from a previous instance for a tiny timeframe until mac80211 reconfigures them.

Quoted 5 lines
Signed-off-by Michael Buesch <[email protected]>

---

This patch is upstream inside of the netdev tree.

Was this series, and the other one, for .24-stable, or .23-stable?

thanks,

greg k-h

author
#2 lore raw
Quoted 11 lines · On Thursday 07 February 2008 00:14:31 Greg KH wrote:
Quoted 10 lines · Michael Buesch

This patch makes suspend/resume work with the b43 driver.
We must not overwrite the MAC addresses in the init function, as this would also overwrite the MAC on resume. With an all-zero MAC the device firmware is not able to ACK any received packets anymore.
Fix this by moving the initializion stuff that must be done on init but not on resume to the start function.
Also zero out filter_flags to make sure we don't have some flags from a previous instance for a tiny timeframe until mac80211 reconfigures them.

Quoted 7 lines
Quoted 5 lines
Signed-off-by Michael Buesch <[email protected]>

---

This patch is upstream inside of the netdev tree.

Was this series, and the other one, for .24-stable, or .23-stable?

This was for .24 stable.
.23 stable didn't have the driver, yet.

Signature · 1 line
Greetings Michael.
#3 lore raw
Quoted 14 lines · Michael Buesch
Quoted 13 lines · On Thursday 07 February 2008 00:14:31 Greg KH wrote:
Quoted 12 lines · Michael Buesch

This patch makes suspend/resume work with the b43 driver.
We must not overwrite the MAC addresses in the init function, as this would also overwrite the MAC on resume. With an all-zero MAC the device firmware is not able to ACK any received packets anymore.
Fix this by moving the initializion stuff that must be done on init but not on resume to the start function.
Also zero out filter_flags to make sure we don't have some flags from a previous instance for a tiny timeframe until mac80211 reconfigures them.

Signed-off-by Michael Buesch <[email protected]>
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

---

This patch is upstream inside of the netdev tree.

Was this series, and the other one, for .24-stable, or .23-stable?

This was for .24 stable.
.23 stable didn't have the driver, yet.

Great, thanks for letting me know.

greg k-h