Re: [PATCH 1/8] /proc/net/ entries for bluetooth protocols
From: Andrei Emeltchenko <hidden>
Date: 2012-06-18 08:27:15
Also in:
lkml
Hi Masatake, On Sun, Jun 17, 2012 at 09:26:37AM +0900, Masatake YAMATO wrote:
Hi, Thank you for reviewing. I have a question.quoted
quoted
+ sk = sk_entry(v); + bt = bt_sk(sk); + seq_printf(seq, "%pK %-6d %-6u %-6u %-6u %-6lu", + sk, + atomic_read(&sk->sk_refcnt), + sk_rmem_alloc_get(sk), + sk_wmem_alloc_get(sk), + sock_i_uid(sk), + sock_i_ino(sk) + ); + seq_puts(seq, batostr(&bt->src));batostr looks OK now but this will be outdated soon by %pMR.As far as reading Documentation/printk-formats.txt and pointer() in lib/vsprintf.c, %pMR is not defined. I guess you mean %pM, don't you?
Sorry for confusion, I meant my patch adding new printf specifier which is not yet in the tree. I think it will take longer time to be accepted then your patch.
I've tried following code, and it works as expected. baswap(&src_baswapped, &bt->src); baswap(&dst_baswapped, &bt->dst); seq_printf(seq, "%pK %-6d %-6u %-6u %-6u %-6lu %pM %pM %-6lu", sk, atomic_read(&sk->sk_refcnt), sk_rmem_alloc_get(sk), sk_wmem_alloc_get(sk), sock_i_uid(sk), sock_i_ino(sk), &src_baswapped, &dst_baswapped, bt->parent? sock_i_ino(bt->parent): 0LU);
Please use whatever method best suits for you although I prefer second solution since it is easier to change to %pMR later. Best regards Andrei Emeltchenko