Skip to contentSkip to conversation

staging: rtl8192u: ieee80211_rx: Fix incorrect type in assignments

PATCH In review Gaston Gonzalez 2 participants 6 messages Jun 21, 2015 – Sep 24, 2015 Staging

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 staging: rtl8192u: ieee80211_rx: Fix incorrect type in assignments

Fix the following incorrect type in assignments detected by sparse:

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:571:37: warning: incorrect type in assignment (different base types)
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:571:37: expected unsigned short [unsigned] [usertype] len
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:571:37: got restricted __be16 [usertype] <noident>

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1330:45: warning: incorrect type in assignment (different base types)
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1330:45: expected unsigned short [unsigned] [usertype] len
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1330:45: got restricted __be16 [usertype] <noident>

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1495:40: warning: incorrect type in assignment (different base types)
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1495:40: expected restricted __le16 <noident>
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1495:40: got int

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1497:40: warning: incorrect type in assignment (different base types)
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1497:40: expected restricted __le16 <noident>
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1497:40: got int

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1501:45: warning: incorrect type in assignment (different base types)
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1501:45: expected restricted __le16 <noident>
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c:1501:45: got unsigned short [unsigned] [usertype] <noident>

---
 drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +7 -5
index b374088..15bcf7e 100644
@@ -566,7 +566,7 @@ void ieee80211_indicate_packets(struct ieee80211_device *ieee, struct ieee80211_
 				memcpy(skb_push(sub_skb, ETH_ALEN), prxb->src, ETH_ALEN);
 				memcpy(skb_push(sub_skb, ETH_ALEN), prxb->dst, ETH_ALEN);
 			} else {
-				u16 len;
+				__be16 len;
 			/* Leave Ethernet header part of hdr and full payload */
 				len = htons(sub_skb->len);
 				memcpy(skb_push(sub_skb, 2), &len, 2);
@@ -1325,7 +1325,7 @@ int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb,
 					memcpy(skb_push(sub_skb, ETH_ALEN), src, ETH_ALEN);
 					memcpy(skb_push(sub_skb, ETH_ALEN), dst, ETH_ALEN);
 				} else {
-					u16 len;
+					__be16 len;
 					/* Leave Ethernet header part of hdr and full payload */
 					len = htons(sub_skb->len);
 					memcpy(skb_push(sub_skb, 2), &len, 2);
@@ -1492,13 +1492,15 @@ static int ieee80211_qos_convert_ac_to_parameters(struct
 		/* WMM spec P.11: The minimum value for AIFSN shall be 2 */
 		qos_param->aifs[aci] = (qos_param->aifs[aci] < 2) ? 2:qos_param->aifs[aci];
 
-		qos_param->cw_min[aci] = ac_params->ecw_min_max & 0x0F;
+		qos_param->cw_min[aci] =
+			cpu_to_le16(ac_params->ecw_min_max & 0x0F);
 
-		qos_param->cw_max[aci] = (ac_params->ecw_min_max & 0xF0) >> 4;
+		qos_param->cw_max[aci] =
+			cpu_to_le16((ac_params->ecw_min_max & 0xF0) >> 4);
 
 		qos_param->flag[aci] =
 		    (ac_params->aci_aifsn & 0x10) ? 0x01 : 0x00;
-		qos_param->tx_op_limit[aci] = le16_to_cpu(ac_params->tx_op_limit);
+		qos_param->tx_op_limit[aci] = ac_params->tx_op_limit;
 	}
 	return 0;
 }
Signature · 5 lines
2.4.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/
#1 lore raw
Quoted 7 lines · On Sunday 21 June 2015 19:12:09 Gaston Gonzalez wrote:
                /* WMM spec P.11: The minimum value for AIFSN shall be 2 */
                qos_param->aifs[aci] = (qos_param->aifs[aci] < 2) ? 2:qos_param->aifs[aci];
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +2 -1
-               qos_param->cw_min[aci] = ac_params->ecw_min_max & 0x0F;
+               qos_param->cw_min[aci] =
+                       cpu_to_le16(ac_params->ecw_min_max & 0x0F);
 

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c static int ieee80211_qos_convert_ac_to_parameters(struct · line 1505

Quoted 10 lines
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +3 -2
-               qos_param->cw_max[aci] = (ac_params->ecw_min_max & 0xF0) >> 4;
+               qos_param->cw_max[aci] =
+                       cpu_to_le16((ac_params->ecw_min_max & 0xF0) >> 4);
 
                qos_param->flag[aci] =
                    (ac_params->aci_aifsn & 0x10) ? 0x01 : 0x00;
-               qos_param->tx_op_limit[aci] = le16_to_cpu(ac_params->tx_op_limit);
+               qos_param->tx_op_limit[aci] = ac_params->tx_op_limit;
        }
        return 0;

This certainly needs a more thorough description of how you determined that the byte swaps that you add are in fact required. Did you test it on a big-endian machine?

Arnd

#2 lore raw
Quoted 19 lines · Arnd Bergmann
Quoted 7 lines · On Sunday 21 June 2015 19:12:09 Gaston Gonzalez wrote:
                /* WMM spec P.11: The minimum value for AIFSN shall be 2 */
                qos_param->aifs[aci] = (qos_param->aifs[aci] < 2) ? 2:qos_param->aifs[aci];
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +2 -1
-               qos_param->cw_min[aci] = ac_params->ecw_min_max & 0x0F;
+               qos_param->cw_min[aci] =
+                       cpu_to_le16(ac_params->ecw_min_max & 0x0F);
 

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c static int ieee80211_qos_convert_ac_to_parameters(struct · line 1505

Quoted 10 lines
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +3 -2
-               qos_param->cw_max[aci] = (ac_params->ecw_min_max & 0xF0) >> 4;
+               qos_param->cw_max[aci] =
+                       cpu_to_le16((ac_params->ecw_min_max & 0xF0) >> 4);
 
                qos_param->flag[aci] =
                    (ac_params->aci_aifsn & 0x10) ? 0x01 : 0x00;
-               qos_param->tx_op_limit[aci] = le16_to_cpu(ac_params->tx_op_limit);
+               qos_param->tx_op_limit[aci] = ac_params->tx_op_limit;
        }
        return 0;
Quoted 5 lines

This certainly needs a more thorough description of how you determined that the byte swaps that you add are in fact required. Did you test it on a big-endian machine?

Arnd

Hello Arnd,

Thank you for reviewing this.
After your email and reviwing this again I'm getting a bit suspicious myself, but this is what I saw:

-- First warning:

qos_param->cw_min[aci] is defined as __le16() in ieee80211.h (ieee80211_qos_parameters structure)

ac_params-> ecw_min_max is defined as u8 in ieee80211.h (ieee80211_qos_ac_parameter structure)

So the assignment is: __le16 = u8 & 0x0F;

-- Second warning:

qos_param->cw_max[aci] is __le16()
ac_params-> ecw_min_max is u8

The assignment is: __le16 = (u8 & 0xF0) >> 4;

Thus, for the warning 1 and 2, I understand that the result won't be the same if the machine is big-endian or little-endian, and that's why we need a cpu_to_le16. Am I missing something?

-- Third warning:

In this case both sides of the assignment are already defined as __le16:

qos_param->tx_op_limit[aci] (ieee80211_qos_parameters structure defined in ieee80211.h))

ac_params->tx_op_limit (ieee80211_qos_ac_parameter structure defined in ieee80211.h)

So the assignment is: __le16() = le16_to_cpu(__le16)

Im getting suspicious now, but it sounded wrong to me.
In the case the right part is correct, I guess the left part should be u16 type?

Regarding the test: I tested it on my machine, but is of course little- endian :( I could built a qemu virtual machine to test it on a big-endian emulated platform. Should that work?

Regards,

Gaston

#3 lore raw
Quoted 44 lines · On Wednesday 24 June 2015 13:34:58 Gaston Gonzalez wrote:
Quoted 19 lines · Arnd Bergmann
Quoted 7 lines · On Sunday 21 June 2015 19:12:09 Gaston Gonzalez wrote:
                /* WMM spec P.11: The minimum value for AIFSN shall be 2 */
                qos_param->aifs[aci] = (qos_param->aifs[aci] < 2) ? 2:qos_param->aifs[aci];
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +2 -1
-               qos_param->cw_min[aci] = ac_params->ecw_min_max & 0x0F;
+               qos_param->cw_min[aci] =
+                       cpu_to_le16(ac_params->ecw_min_max & 0x0F);
 

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c static int ieee80211_qos_convert_ac_to_parameters(struct · line 1505

Quoted 10 lines
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +3 -2
-               qos_param->cw_max[aci] = (ac_params->ecw_min_max & 0xF0) >> 4;
+               qos_param->cw_max[aci] =
+                       cpu_to_le16((ac_params->ecw_min_max & 0xF0) >> 4);
 
                qos_param->flag[aci] =
                    (ac_params->aci_aifsn & 0x10) ? 0x01 : 0x00;
-               qos_param->tx_op_limit[aci] = le16_to_cpu(ac_params->tx_op_limit);
+               qos_param->tx_op_limit[aci] = ac_params->tx_op_limit;
        }
        return 0;
Quoted 5 lines

This certainly needs a more thorough description of how you determined that the byte swaps that you add are in fact required. Did you test it on a big-endian machine?

Arnd

Hello Arnd,

Thank you for reviewing this.
After your email and reviwing this again I'm getting a bit suspicious myself, but this is what I saw:

-- First warning:

qos_param->cw_min[aci] is defined as __le16() in ieee80211.h (ieee80211_qos_parameters structure)

ac_params-> ecw_min_max is defined as u8 in ieee80211.h (ieee80211_qos_ac_parameter structure)

So the assignment is: __le16 = u8 & 0x0F;

-- Second warning:

Quoted 8 lines

qos_param->cw_max[aci] is __le16()
ac_params-> ecw_min_max is u8

The assignment is: __le16 = (u8 & 0xF0) >> 4;

Thus, for the warning 1 and 2, I understand that the result won't be the same if the machine is big-endian or little-endian, and that's why we need a cpu_to_le16. Am I missing something?

I think your analysis is right, as long as the __le16 annotation is actually correct. It usually helps to look at the git history to see what the intent of the patch was that introduced the assignment and the patch that introduced the __le16 type. Presumably one of them was incorrect, and it would be good to figure out where it went wrong, and to add a 'Fixes:' tag in your patch description that pinpoints the exact mistake.

Quoted 7 lines

-- Third warning:

In this case both sides of the assignment are already defined as __le16:

qos_param->tx_op_limit[aci] (ieee80211_qos_parameters structure defined in ieee80211.h))

Quoted 8 lines

ac_params->tx_op_limit (ieee80211_qos_ac_parameter structure defined in ieee80211.h)

So the assignment is: __le16() = le16_to_cpu(__le16)

Im getting suspicious now, but it sounded wrong to me.
In the case the right part is correct, I guess the left part should be u16 type?

Again, your logic sounds good: there is clearly something wrong here, but it's not obvious to conclude whether it is an incorrect annotation or an extraneous byte swap. Besides looking at the git history, it also helps to look at all other uses of the two sides of the assignment:

See if qos_param->tx_op_limit is in fact used as a little-endian value (e.g. by copying to memory or a register), and if the value that got written to ac_params->tx_op_limit was byte-swapped already at the time of assignment.

Quoted 3 lines

Regarding the test: I tested it on my machine, but is of course little- endian :( I could built a qemu virtual machine to test it on a big-endian emulated platform. Should that work?

Yes, that would work: you can assign the USB device to the qemu machine and run a kernel in there. The easiest emulation to try is probably a PowerPC PAPR machine with a file system from
https://people.debian.org/~aurel32/qemu/powerpc/.
MIPS should work as well.

Arnd

#4 lore raw
Quoted 54 lines · Arnd Bergmann
Quoted 44 lines · On Wednesday 24 June 2015 13:34:58 Gaston Gonzalez wrote:
Quoted 19 lines · Arnd Bergmann
Quoted 7 lines · On Sunday 21 June 2015 19:12:09 Gaston Gonzalez wrote:
                /* WMM spec P.11: The minimum value for AIFSN shall be 2 */
                qos_param->aifs[aci] = (qos_param->aifs[aci] < 2) ? 2:qos_param->aifs[aci];
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +2 -1
-               qos_param->cw_min[aci] = ac_params->ecw_min_max & 0x0F;
+               qos_param->cw_min[aci] =
+                       cpu_to_le16(ac_params->ecw_min_max & 0x0F);
 

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c static int ieee80211_qos_convert_ac_to_parameters(struct · line 1505

Quoted 10 lines
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +3 -2
-               qos_param->cw_max[aci] = (ac_params->ecw_min_max & 0xF0) >> 4;
+               qos_param->cw_max[aci] =
+                       cpu_to_le16((ac_params->ecw_min_max & 0xF0) >> 4);
 
                qos_param->flag[aci] =
                    (ac_params->aci_aifsn & 0x10) ? 0x01 : 0x00;
-               qos_param->tx_op_limit[aci] = le16_to_cpu(ac_params->tx_op_limit);
+               qos_param->tx_op_limit[aci] = ac_params->tx_op_limit;
        }
        return 0;
Quoted 5 lines

This certainly needs a more thorough description of how you determined that the byte swaps that you add are in fact required. Did you test it on a big-endian machine?

Arnd

Hello Arnd,

Thank you for reviewing this.
After your email and reviwing this again I'm getting a bit suspicious myself, but this is what I saw:

-- First warning:

qos_param->cw_min[aci] is defined as __le16() in ieee80211.h (ieee80211_qos_parameters structure)

ac_params-> ecw_min_max is defined as u8 in ieee80211.h (ieee80211_qos_ac_parameter structure)

So the assignment is: __le16 = u8 & 0x0F;

-- Second warning:

Quoted 8 lines

qos_param->cw_max[aci] is __le16()
ac_params-> ecw_min_max is u8

The assignment is: __le16 = (u8 & 0xF0) >> 4;

Thus, for the warning 1 and 2, I understand that the result won't be the same if the machine is big-endian or little-endian, and that's why we need a cpu_to_le16. Am I missing something?

Quoted 8 lines

I think your analysis is right, as long as the __le16 annotation is actually correct. It usually helps to look at the git history to see what the intent of the patch was that introduced the assignment and the patch that introduced the __le16 type. Presumably one of them was incorrect, and it would be good to figure out where it went wrong, and to add a 'Fixes:' tag in your patch description that pinpoints the exact mistake.

Ok, will do.

Quoted 21 lines
Quoted 7 lines

-- Third warning:

In this case both sides of the assignment are already defined as __le16:

qos_param->tx_op_limit[aci] (ieee80211_qos_parameters structure defined in ieee80211.h))

Quoted 8 lines

ac_params->tx_op_limit (ieee80211_qos_ac_parameter structure defined in ieee80211.h)

So the assignment is: __le16() = le16_to_cpu(__le16)

Im getting suspicious now, but it sounded wrong to me.
In the case the right part is correct, I guess the left part should be u16 type?

Again, your logic sounds good: there is clearly something wrong here, but it's not obvious to conclude whether it is an incorrect annotation or an extraneous byte swap. Besides looking at the git history, it also helps to look at all other uses of the two sides of the assignment:

Quoted 5 lines

See if qos_param->tx_op_limit is in fact used as a little-endian value (e.g. by copying to memory or a register), and if the value that got written to ac_params->tx_op_limit was byte-swapped already at the time of assignment.

Ok, I'll do it too.

Quoted 10 lines
Quoted 3 lines

Regarding the test: I tested it on my machine, but is of course little- endian :( I could built a qemu virtual machine to test it on a big-endian emulated platform. Should that work?

Yes, that would work: you can assign the USB device to the qemu machine and run a kernel in there. The easiest emulation to try is probably a PowerPC PAPR machine with a file system from
https://people.debian.org/~aurel32/qemu/powerpc/.
MIPS should work as well.

Ok, thanks a lot for all the pointers.

Gaston

Quoted 1 line

Arnd

#5 lore raw
Quoted 105 lines · Gaston Gonzalez
Quoted 54 lines · Arnd Bergmann
Quoted 44 lines · On Wednesday 24 June 2015 13:34:58 Gaston Gonzalez wrote:
Quoted 19 lines · Arnd Bergmann
Quoted 7 lines · On Sunday 21 June 2015 19:12:09 Gaston Gonzalez wrote:
                /* WMM spec P.11: The minimum value for AIFSN shall be 2 */
                qos_param->aifs[aci] = (qos_param->aifs[aci] < 2) ? 2:qos_param->aifs[aci];
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +2 -1
-               qos_param->cw_min[aci] = ac_params->ecw_min_max & 0x0F;
+               qos_param->cw_min[aci] =
+                       cpu_to_le16(ac_params->ecw_min_max & 0x0F);
 

drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c static int ieee80211_qos_convert_ac_to_parameters(struct · line 1505

Quoted 10 lines
drivers/staging/rtl8192u/ieee80211/ieee80211_rx.c +3 -2
-               qos_param->cw_max[aci] = (ac_params->ecw_min_max & 0xF0) >> 4;
+               qos_param->cw_max[aci] =
+                       cpu_to_le16((ac_params->ecw_min_max & 0xF0) >> 4);
 
                qos_param->flag[aci] =
                    (ac_params->aci_aifsn & 0x10) ? 0x01 : 0x00;
-               qos_param->tx_op_limit[aci] = le16_to_cpu(ac_params->tx_op_limit);
+               qos_param->tx_op_limit[aci] = ac_params->tx_op_limit;
        }
        return 0;
Quoted 5 lines

This certainly needs a more thorough description of how you determined that the byte swaps that you add are in fact required. Did you test it on a big-endian machine?

Arnd

Hello Arnd,

Thank you for reviewing this.
After your email and reviwing this again I'm getting a bit suspicious myself, but this is what I saw:

-- First warning:

qos_param->cw_min[aci] is defined as __le16() in ieee80211.h (ieee80211_qos_parameters structure)

ac_params-> ecw_min_max is defined as u8 in ieee80211.h (ieee80211_qos_ac_parameter structure)

So the assignment is: __le16 = u8 & 0x0F;

-- Second warning:

Quoted 8 lines

qos_param->cw_max[aci] is __le16()
ac_params-> ecw_min_max is u8

The assignment is: __le16 = (u8 & 0xF0) >> 4;

Thus, for the warning 1 and 2, I understand that the result won't be the same if the machine is big-endian or little-endian, and that's why we need a cpu_to_le16. Am I missing something?

Quoted 8 lines

I think your analysis is right, as long as the __le16 annotation is actually correct. It usually helps to look at the git history to see what the intent of the patch was that introduced the assignment and the patch that introduced the __le16 type. Presumably one of them was incorrect, and it would be good to figure out where it went wrong, and to add a 'Fixes:' tag in your patch description that pinpoints the exact mistake.

Ok, will do.

Quoted 21 lines
Quoted 7 lines

-- Third warning:

In this case both sides of the assignment are already defined as __le16:

qos_param->tx_op_limit[aci] (ieee80211_qos_parameters structure defined in ieee80211.h))

Quoted 8 lines

ac_params->tx_op_limit (ieee80211_qos_ac_parameter structure defined in ieee80211.h)

So the assignment is: __le16() = le16_to_cpu(__le16)

Im getting suspicious now, but it sounded wrong to me.
In the case the right part is correct, I guess the left part should be u16 type?

Again, your logic sounds good: there is clearly something wrong here, but it's not obvious to conclude whether it is an incorrect annotation or an extraneous byte swap. Besides looking at the git history, it also helps to look at all other uses of the two sides of the assignment:

Quoted 5 lines

See if qos_param->tx_op_limit is in fact used as a little-endian value (e.g. by copying to memory or a register), and if the value that got written to ac_params->tx_op_limit was byte-swapped already at the time of assignment.

Ok, I'll do it too.

Quoted 10 lines
Quoted 3 lines

Regarding the test: I tested it on my machine, but is of course little- endian :( I could built a qemu virtual machine to test it on a big-endian emulated platform. Should that work?

Yes, that would work: you can assign the USB device to the qemu machine and run a kernel in there. The easiest emulation to try is probably a PowerPC PAPR machine with a file system from
https://people.debian.org/~aurel32/qemu/powerpc/.
MIPS should work as well.

Quoted 4 lines

Ok, thanks a lot for all the pointers.

Gaston

Quoted 1 line

Arnd

So more than two months have passed without any reply from my side.

The thing is I'm struggling to get a new hardware with this chipset. Today I received my fourth rtl8192cu device to my collection of wrong devices, which also comprises one ralink and some other chip. All of them specified as rtl8192u on the sellers site...

I have one or two options left to try to get a replacement rtl8192u device, but for the moment I don't have the hardware.

Though I know this 'change'is almost insignificant I wanted to give some life signal and not leave this thread as abandoned.

regards,

Gaston