(Changing subject, adding Linus and linux-kernel) Dan Carpenter <dan.carpenter@xxxxxxxxxx> writes: > On Thu, Feb 24, 2022 at 09:34:31AM +0200, Kalle Valo wrote: >> Francesco Magliocca <franciman12@xxxxxxxxx> writes: >> >> > Hi, I picked (void*) to be conformant with the other examples in htt_rx.c >> > For example at line 1431: >> >> rxd = HTT_RX_BUF_TO_RX_DESC(hw, >> >> (void *)msdu->data - hw->rx_desc_ops->rx_desc_size); >> > >> > But for me it is ok. Maybe we should fix all the occurrences of this kind. >> >> Yeah, it would be good to fix the void pointer arithmetic in a separate >> patch. I have planning to enable -Wpointer-arith in my ath10k-check and >> ath11k-check scripts, so patches are very welcome. > > Void * casts simplify a lot of code. Less noise. More readable. > They're more accurate in a sense because it's not a u8 at all. The > kernel can't compile with other compilers besides GCC and Clang so why > care about that the C standard hasn't caught up? > > What does -Wpointer-arith buy us? A good question. I have always just thought we should avoid void pointer arithmetic due to the C standard, but now that you mention it void pointers can indeed simplify the code. So I'm not so sure anymore. Any opinions? Is there a kernel wide recommendation for this? -- https://patchwork.kernel.org/project/linux-wireless/list/ https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
