[PATCH] [REBASED for 4.14] once: add DO_ONCE_SLOW() for sleepable contexts

[Greg KH]

On Tue, Dec 13, 2022 at 01:22:40PM +0100, Christophe Leroy wrote:
> From: Eric Dumazet <edumazet at google.com>
> 
> [ Upstream commit 62c07983bef9d3e78e71189441e1a470f0d1e653 ]
> 
> Christophe Leroy reported a ~80ms latency spike
> happening at first TCP connect() time.
> 
> This is because __inet_hash_connect() uses get_random_once()
> to populate a perturbation table which became quite big
> after commit 4c2c8f03a5ab ("tcp: increase source port perturb table to 2^16")
> 
> get_random_once() uses DO_ONCE(), which block hard irqs for the duration
> of the operation.
> 
> This patch adds DO_ONCE_SLOW() which uses a mutex instead of a spinlock
> for operations where we prefer to stay in process context.
> 
> Then __inet_hash_connect() can use get_random_slow_once()
> to populate its perturbation table.
> 
> Fixes: 4c2c8f03a5ab ("tcp: increase source port perturb table to 2^16")
> Fixes: 190cc82489f4 ("tcp: change source port randomizarion at connect() time")
> Reported-by: Christophe Leroy <christophe.leroy at csgroup.eu>
> Link: https://lore.kernel.org/netdev/CANn89iLAEYBaoYajy0Y9UmGFff5GPxDUoG-ErVB2jDdRNQ5Tug@mail.gmail.com/T/#t
> Signed-off-by: Eric Dumazet <edumazet at google.com>
> Cc: Willy Tarreau <w at 1wt.eu>
> Tested-by: Christophe Leroy <christophe.leroy at csgroup.eu>
> Signed-off-by: David S. Miller <davem at davemloft.net>
> Signed-off-by: Sasha Levin <sashal at kernel.org>
> Signed-off-by: Christophe Leroy <christophe.leroy at csgroup.eu>
> ---
>  include/linux/once.h       | 28 ++++++++++++++++++++++++++++
>  lib/once.c                 | 30 ++++++++++++++++++++++++++++++
>  net/ipv4/inet_hashtables.c |  4 ++--
>  3 files changed, 60 insertions(+), 2 deletions(-)

Now queued up, thanks.

greg k-h