From: Julian Wiedmann <jwi@linux.ibm.com>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
linux-s390@vger.kernel.org
Cc: Karsten Graul <kgraul@linux.ibm.com>,
Heiko Carstens <hca@linux.ibm.com>,
Vasily Gorbik <gor@linux.ibm.com>,
Christian Borntraeger <borntraeger@de.ibm.com>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH 2/6] s390/ctcm: Put struct qllc on stack.
Date: Thu, 19 Nov 2020 09:56:20 +0200 [thread overview]
Message-ID: <8dd75a21-0e50-d445-9532-9a0950b884ea@linux.ibm.com> (raw)
In-Reply-To: <20201118105317.605671-3-bigeasy@linutronix.de>
On 18.11.20 12:53, Sebastian Andrzej Siewior wrote:
> The size of struct qllc is 2 byte. The memory for is allocated, initialized,
> used and deallocated a few lines later.
>
> It is more efficient to avoid the allocation/free dance and keeping the
> variable on stack. Especially since the compiler is smart enough to not
> allocate the memory on stack but assign the values directly.
>
> Rename `qllcptr' to `qllc' and use it on stack.
>
Can we please shrink this down to
qllcptr = (struct qllc *) skb_put(...);
qllcptr->foo = bar;
...
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> ---
> drivers/s390/net/ctcm_mpc.c | 24 ++++++------------------
> 1 file changed, 6 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/s390/net/ctcm_mpc.c b/drivers/s390/net/ctcm_mpc.c
> index 04a51cc89e74c..4ff51af44d338 100644
> --- a/drivers/s390/net/ctcm_mpc.c
> +++ b/drivers/s390/net/ctcm_mpc.c
> @@ -2049,11 +2049,10 @@ static void mpc_action_rcvd_xid7(fsm_instance *fsm, int event, void *arg)
> */
> static int mpc_send_qllc_discontact(struct net_device *dev)
> {
> - __u32 new_len = 0;
> struct sk_buff *skb;
> - struct qllc *qllcptr;
> struct ctcm_priv *priv = dev->ml_priv;
> struct mpc_group *grp = priv->mpcg;
> + struct qllc qllc;
>
> CTCM_PR_DEBUG("%s: GROUP STATE: %s\n",
> __func__, mpcg_state_names[grp->saved_state]);
> @@ -2080,31 +2079,20 @@ static int mpc_send_qllc_discontact(struct net_device *dev)
> case MPCG_STATE_FLOWC:
> case MPCG_STATE_READY:
> grp->send_qllc_disc = 2;
> - new_len = sizeof(struct qllc);
> - qllcptr = kzalloc(new_len, gfp_type() | GFP_DMA);
> - if (qllcptr == NULL) {
> - CTCM_DBF_TEXT_(MPC_ERROR, CTC_DBF_ERROR,
> - "%s(%s): qllcptr allocation error",
> - CTCM_FUNTAIL, dev->name);
> - return -ENOMEM;
> - }
> -
> - qllcptr->qllc_address = 0xcc;
> - qllcptr->qllc_commands = 0x03;
> -
> - skb = __dev_alloc_skb(new_len, GFP_ATOMIC);
>
> + skb = __dev_alloc_skb(sizeof(struct qllc), GFP_ATOMIC);
> if (skb == NULL) {
> CTCM_DBF_TEXT_(MPC_ERROR, CTC_DBF_ERROR,
> "%s(%s): skb allocation error",
> CTCM_FUNTAIL, dev->name);
> priv->stats.rx_dropped++;
> - kfree(qllcptr);
> return -ENOMEM;
> }
>
> - skb_put_data(skb, qllcptr, new_len);
> - kfree(qllcptr);
> + qllc.qllc_address = 0xcc;
> + qllc.qllc_commands = 0x03;
> +
> + skb_put_data(skb, &qllc, sizeof(struct qllc));
>
> if (skb_headroom(skb) < 4) {
> CTCM_DBF_TEXT_(MPC_ERROR, CTC_DBF_ERROR,
>
next prev parent reply other threads:[~2020-11-19 7:56 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-18 10:53 [PATCH 0/6] s390/ctcm: Remove gfp_type() usage Sebastian Andrzej Siewior
2020-11-18 10:53 ` [PATCH 1/6] s390/ctcm: Put struct th_header and th_sweep on stack Sebastian Andrzej Siewior
2020-11-19 7:45 ` Julian Wiedmann
2020-11-19 8:12 ` Sebastian Andrzej Siewior
2020-11-19 8:16 ` Julian Wiedmann
2020-11-19 9:37 ` Sebastian Andrzej Siewior
2020-11-18 10:53 ` [PATCH 2/6] s390/ctcm: Put struct qllc " Sebastian Andrzej Siewior
2020-11-19 7:56 ` Julian Wiedmann [this message]
2020-11-18 10:53 ` [PATCH 3/6] s390/ctcm: Put struct pdu " Sebastian Andrzej Siewior
2020-11-19 7:52 ` Julian Wiedmann
2020-11-18 10:53 ` [PATCH 4/6] s390/ctcm: Use explicit allocation mask in ctcmpc_unpack_skb() Sebastian Andrzej Siewior
2020-11-19 7:28 ` Julian Wiedmann
2020-11-18 10:53 ` [PATCH 5/6] s390/ctcm: Use GFP_KERNEL in add_channel() Sebastian Andrzej Siewior
2020-11-19 7:30 ` Julian Wiedmann
2020-11-18 10:53 ` [PATCH 6/6] s390/ctcm: Use GFP_ATOMIC in ctcmpc_tx() Sebastian Andrzej Siewior
2020-11-19 7:34 ` Julian Wiedmann
2020-11-19 8:01 ` [PATCH 0/6] s390/ctcm: Remove gfp_type() usage Julian Wiedmann
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=8dd75a21-0e50-d445-9532-9a0950b884ea@linux.ibm.com \
--to=jwi@linux.ibm.com \
--cc=bigeasy@linutronix.de \
--cc=borntraeger@de.ibm.com \
--cc=gor@linux.ibm.com \
--cc=hca@linux.ibm.com \
--cc=kgraul@linux.ibm.com \
--cc=linux-s390@vger.kernel.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox