> I have also removed all the code that thereby serves no purpose. [ snip ] > @@ -743,7 +743,7 @@ static u8 phy_path_a_iqk(struct adapter *adapt, bool config_pathb) > reg_eac = phy_query_bb_reg(adapt, rRx_Power_After_IQK_A_2, bMaskDWord); > reg_e94 = phy_query_bb_reg(adapt, rTx_Power_Before_IQK_A, bMaskDWord); > reg_e9c = phy_query_bb_reg(adapt, rTx_Power_After_IQK_A, bMaskDWord); > - reg_ea4 = phy_query_bb_reg(adapt, rRx_Power_Before_IQK_A_2, bMaskDWord); > + phy_query_bb_reg(adapt, rRx_Power_Before_IQK_A_2, bMaskDWord); > > if (!(reg_eac & BIT28) && > (((reg_e94 & 0x03FF0000)>>16) != 0x142) && You clearly didn't remove *all* the "code that thereby serves no purpose". Could we call a cease fire on these and stop sending them until Greg or someone else weighs in? It's hard for me to comment on 50 patches that I don't like. To re-iterate: 1) Removing part of the line is 90% likely to be wrong. We can't know without knowing the code better, knowing the hardware, or testing. 2) Silencing static checker warnings in the wrong way, negates the usefulness of static checkers and makes bad code hard to find. Some of these "unused" variable warnings could easily be bugs so we are hiding bugs. 3) From an aesthetic point of view the code was easier to understand before. The line you have deleted was clearly copy and pasted from the line before. But now it looks deliberate and the information we need to understand the original intent is hidden in the git log. regards, dan carpenter _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
