Re: [PATCH] net: rds: fix per-cpu helper usage

2014-01-17 Thread David Miller
From: Gerald Schaefer Date: Thu, 16 Jan 2014 16:54:48 +0100 > commit ae4b46e9d "net: rds: use this_cpu_* per-cpu helper" broke per-cpu > handling for rds. chpfirst is the result of __this_cpu_read(), so it is > an absolute pointer and not __percpu. Therefore, __this_cpu_write() > should not

Re: [PATCH] net: rds: fix per-cpu helper usage

2014-01-17 Thread David Miller
From: Gerald Schaefer gerald.schae...@de.ibm.com Date: Thu, 16 Jan 2014 16:54:48 +0100 commit ae4b46e9d net: rds: use this_cpu_* per-cpu helper broke per-cpu handling for rds. chpfirst is the result of __this_cpu_read(), so it is an absolute pointer and not __percpu. Therefore,

[PATCH] net: rds: fix per-cpu helper usage

2014-01-16 Thread Gerald Schaefer
commit ae4b46e9d "net: rds: use this_cpu_* per-cpu helper" broke per-cpu handling for rds. chpfirst is the result of __this_cpu_read(), so it is an absolute pointer and not __percpu. Therefore, __this_cpu_write() should not operate on chpfirst, but rather on cache->percpu->first, just like

[PATCH] net: rds: fix per-cpu helper usage

2014-01-16 Thread Gerald Schaefer
commit ae4b46e9d net: rds: use this_cpu_* per-cpu helper broke per-cpu handling for rds. chpfirst is the result of __this_cpu_read(), so it is an absolute pointer and not __percpu. Therefore, __this_cpu_write() should not operate on chpfirst, but rather on cache-percpu-first, just like