Re: [PATCH] net: make SK_MEMORY_PCPU_RESERV tunable

From: Lameter, Christopher
Date: Wed Mar 06 2024 - 12:01:43 EST

Next message: Casey Schaufler: "Re: [PATCH v39 04/42] IMA: avoid label collisions with stacked LSMs"
Previous message: Puranjay Mohan: "[RFC PATCH] riscv: Implement HAVE_DYNAMIC_FTRACE_WITH_CALL_OPS"
Next in thread: Eric Dumazet: "Re: [PATCH] net: make SK_MEMORY_PCPU_RESERV tunable"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Wed, 28 Feb 2024, Eric Dumazet wrote:

__sk_mem_raise_allocated() drops to 0.4%.

I suspect some kind of flow/cpu steering issues then.
Also maybe SO_RESERVE_MEM would be better for this workload.

This is via loopback. So there is a flow steering issue in the IP stack?

Next message: Casey Schaufler: "Re: [PATCH v39 04/42] IMA: avoid label collisions with stacked LSMs"
Previous message: Puranjay Mohan: "[RFC PATCH] riscv: Implement HAVE_DYNAMIC_FTRACE_WITH_CALL_OPS"
Next in thread: Eric Dumazet: "Re: [PATCH] net: make SK_MEMORY_PCPU_RESERV tunable"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]