net-memcg: avoid stalls when under memory pressure

jira LE-1603
commit 720ca52

    As Shakeel explains the commit under Fixes had the unintended
    side-effect of no longer pre-loading the cached memory allowance.
    Even tho we previously dropped the first packet received when
    over memory limit - the consecutive ones would get thru by using
    the cache. The charging was happening in batches of 128kB, so
    we'd let in 128kB (truesize) worth of packets per one drop.

    After the change we no longer force charge, there will be no
    cache filling side effects. This causes significant drops and
    connection stalls for workloads which use a lot of page cache,
    since we can't reclaim page cache under GFP_NOWAIT.

    Some of the latency can be recovered by improving SACK reneg
    handling but nowhere near enough to get back to the pre-5.15
    performance (the application I'm experimenting with still
    sees 5-10x worst latency).

    Apply the suggested workaround of using GFP_ATOMIC. We will now
    be more permissive than previously as we'll drop _no_ packets
    in softirq when under pressure. But I can't think of any good
    and simple way to address that within networking.

    Link: https://lore.kernel.org/all/[email protected]/
    Suggested-by: Shakeel Butt <[email protected]>
    Fixes: 4b1327b ("net-memcg: pass in gfp_t mask to mem_cgroup_charge_skmem()")
    Acked-by: Shakeel Butt <[email protected]>
    Acked-by: Roman Gushchin <[email protected]>
    Link: https://lore.kernel.org/r/[email protected]
    Signed-off-by: Jakub Kicinski <[email protected]>
    (cherry picked from commit 720ca52)

Signed-off-by: Greg Rose <[email protected]>

Author: gvrose8192

include/net/sock.h

@@ -2562,7 +2562,7 @@ static inline gfp_t gfp_any(void)
 
 static inline gfp_t gfp_memcg_charge(void)
 {
-	return in_softirq() ? GFP_NOWAIT : GFP_KERNEL;
+	return in_softirq() ? GFP_ATOMIC : GFP_KERNEL;
 }
 
 static inline long sock_rcvtimeo(const struct sock *sk, bool noblock)