staging: rtl8723bs: Use % 4096u instead of & 0xfff
PATCH v5 1/1 staging: rtl8723bs: Use % 4096u instead of & 0xfff
The sequence number is constrained to a range of [0, 4095], which is a total of 4096 values. The bitmask operation using & 0xfff is used to perform this wrap-around. While this is functionally correct, it obscures the intended semantic of a 4096-based wrap.
Using a modulo operation % 4096u makes the wrap-around logic explicit and easier to understand. It clearly signals that the sequence number cycles through a range of 4096 values.
It also makes the code robust against potential changes of the 4096 upper limit, especially when it becomes a non power of 2 value while the AND(&) works solely for power of 2 values.
The use of % 4096u also guarantees that the modulo operation is performed with unsigned arithmetic, preventing potential issues with the signed types.
---
Changes in v4:
- Corrected patch to use '%' instead of '&'
- To ensure this change does not affect the functional behaviour, I compared the generated object files before and after the change using the cmp which compares the two object files byte by byte as shown below:
$ make drivers/staging/rtl8723bs/core/rtw_xmit.o $ cmp rtw_xmit_before.o rtw_xmit_after.o
No differences were found in the output, confirming that the change does not alter the compiled output.
Changes in v3:
- Added more description to the commit message.
- Removed blank line in the tag block.
- Added more patch recipients.
Changes in v2:
- Changed the commit message t a more descriptive message which makes it clear why the patch does the change.
- changed the subject title to include 4096u to show that an unsigned module is used.
Changes in v1:
- Added more patch recipients.
drivers/staging/rtl8723bs/core/rtw_xmit.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/staging/rtl8723bs/core/rtw_xmit.c +3 -3
@@ -943,7 +943,7 @@ s32 rtw_make_wlanhdr(struct adapter *padapter, u8 *hdr, struct pkt_attrib *pattrif (psta) { psta->sta_xmitpriv.txseq_tid[pattrib->priority]++; - psta->sta_xmitpriv.txseq_tid[pattrib->priority] &=0xFFF ; + psta->sta_xmitpriv.txseq_tid[pattrib->priority] %=4096u ; pattrib->seqnum = psta->sta_xmitpriv.txseq_tid[pattrib->priority];SetSeqNum (hdr, pattrib->seqnum);
@@ -963,11 +963,11 @@ s32 rtw_make_wlanhdr(struct adapter *padapter, u8 *hdr, struct pkt_attrib *pattrif (SN_LESS (pattrib->seqnum, tx_seq)) { pattrib->ampdu_en = false;/* AGG BK */ }else if (SN_EQUAL (pattrib->seqnum, tx_seq)) { - psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (tx_seq+1 )&0xfff ; + psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (tx_seq+1 )%4096u ; pattrib->ampdu_en = true;/* AGG EN */ }else { - psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (pattrib->seqnum+1 )&0xfff ; + psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (pattrib->seqnum+1 )%4096u ; pattrib->ampdu_en = true;/* AGG EN */ } }
Signature · 1 line
2.34.1
Quoted 16 lines · Abraham Samuel Adekunle
The sequence number is constrained to a range of [0, 4095], which is a total of 4096 values. The bitmask operation using
& 0xfffis used to perform this wrap-around. While this is functionally correct, it obscures the intended semantic of a 4096-based wrap.Using a modulo operation
% 4096umakes the wrap-around logic explicit and easier to understand. It clearly signals that the sequence number cycles through a range of 4096 values.
It also makes the code robust against potential changes of the 4096 upper limit, especially when it becomes a non power of 2 value while theAND(&)works solely for power of 2 values.The use of
% 4096ualso guarantees that the modulo operation is performed with unsigned arithmetic, preventing potential issues with the signed types.
...
drivers/staging/rtl8723bs/core/rtw_xmit.c s32 rtw_make_wlanhdr(struct adapter *padapter, u8 *hdr, struct pkt_attrib *pattr · line 970
Quoted 5 lines
drivers/staging/rtl8723bs/core/rtw_xmit.c +2
-2- psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (tx_seq+1 )&0xfff ; + psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (tx_seq+1 )%4096u ; - psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (pattrib->seqnum+1 )&0xfff ; + psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (pattrib->seqnum+1 )%4096u ;
To reduce churn, it would be nice to have another patch _before_ this one that makes above clearer (by adding spaces) as we already new in practice that it's hard to read:
So, this will become two lines in each of the cases, and in this patch only one of those line will be touched.
...
Besides that I haven't found in the changelog if you have looked at the entire driver code and checked that all usages of this field is done in the same / similar way and no need to convert anything more that these.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Quoted 25 lines · Andy Shevchenko
Quoted 16 lines · Abraham Samuel Adekunle
The sequence number is constrained to a range of [0, 4095], which is a total of 4096 values. The bitmask operation using
& 0xfffis used to perform this wrap-around. While this is functionally correct, it obscures the intended semantic of a 4096-based wrap.Using a modulo operation
% 4096umakes the wrap-around logic explicit and easier to understand. It clearly signals that the sequence number cycles through a range of 4096 values.
It also makes the code robust against potential changes of the 4096 upper limit, especially when it becomes a non power of 2 value while theAND(&)works solely for power of 2 values.The use of
% 4096ualso guarantees that the modulo operation is performed with unsigned arithmetic, preventing potential issues with the signed types....
Quoted 2 lines
drivers/staging/rtl8723bs/core/rtw_xmit.c +1
-1- psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (tx_seq+1 )&0xfff ; + psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (tx_seq+1 )%4096u ;
Quoted 8 lines
drivers/staging/rtl8723bs/core/rtw_xmit.c s32 rtw_make_wlanhdr(struct adapter *padapter, u8 *hdr, struct pkt_attrib *pattr · line 970
Quoted 2 lines
drivers/staging/rtl8723bs/core/rtw_xmit.c +1
-1- psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (pattrib->seqnum+1 )&0xfff ; + psta->BA_starting_seqctrl[pattrib->priority &0x0f ] = (pattrib->seqnum+1 )%4096u ;To reduce churn, it would be nice to have another patch _before_ this one that makes above clearer (by adding spaces) as we already new in practice that it's hard to read:
So, this will become two lines in each of the cases, and in this patch only one of those line will be touched.
Okay, I will send a patchset for the change. The first one will add the spaces, then the
second one will add the logic, and they can be applied in sequence.
Quoted 7 lines
...
Besides that I haven't found in the changelog if you have looked at the entire driver code and checked that all usages of this field is done in the same / similar way and no need to convert anything more that these.
No, I have not checked other drivers. I will write a semantic patch to match and transform such cases
Thanks
Adekunle.
Quoted 16 lines · Andy Shevchenko
Quoted 6 lines · Abraham Samuel Adekunle
The sequence number is constrained to a range of [0, 4095], which is a total of 4096 values. The bitmask operation using
& 0xfffis used to perform this wrap-around. While this is functionally correct, it obscures the intended semantic of a 4096-based wrap.Quoted 7 lines
Using a modulo operation
% 4096umakes the wrap-around logic explicit and easier to understand. It clearly signals that the sequence number cycles through a range of 4096 values.
It also makes the code robust against potential changes of the 4096 upper limit, especially when it becomes a non power of 2 value while theAND(&)works solely for power of 2 values.
Quoted 6 lines
...
Besides that I haven't found in the changelog if you have looked at the entire driver code and checked that all usages of this field is done in the same / similar way and no need to convert anything more that these.
I have found more cases using a small Coccinelle semantic patch. There were also cases in the media driver, but I did not touch those since it is not allowed for an outreachy applicant.
I will send a patchset soon.
Thanks
Adekunle