* [PATCH net 2/2] s390: qeth: Fix potential array overrun in cmd/rc lookup
From: Julian Wiedmann @ 2018-09-26 16:07 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Jean Delvare, Julian Wiedmann
In-Reply-To: <20180926160710.64739-1-jwi@linux.ibm.com>
From: Jean Delvare <jdelvare@suse.de>
Functions qeth_get_ipa_msg and qeth_get_ipa_cmd_name are modifying
the last member of global arrays without any locking that I can see.
If two instances of either function are running at the same time,
it could cause a race ultimately leading to an array overrun (the
contents of the last entry of the array is the only guarantee that
the loop will ever stop).
Performing the lookups without modifying the arrays is admittedly
slower (two comparisons per iteration instead of one) but these
are operations which are rare (should only be needed in error
cases or when debugging, not during successful operation) and it
seems still less costly than introducing a mutex to protect the
arrays in question.
As a side bonus, it allows us to declare both arrays as const data.
Signed-off-by: Jean Delvare <jdelvare@suse.de>
Cc: Julian Wiedmann <jwi@linux.ibm.com>
Cc: Ursula Braun <ubraun@linux.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 2 +-
drivers/s390/net/qeth_core_mpc.c | 30 ++++++++++++++++--------------
drivers/s390/net/qeth_core_mpc.h | 4 ++--
3 files changed, 19 insertions(+), 17 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index de8282420f96..ffce6f39828a 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -610,7 +610,7 @@ static void qeth_put_reply(struct qeth_reply *reply)
static void qeth_issue_ipa_msg(struct qeth_ipa_cmd *cmd, int rc,
struct qeth_card *card)
{
- char *ipa_name;
+ const char *ipa_name;
int com = cmd->hdr.command;
ipa_name = qeth_get_ipa_cmd_name(com);
if (rc)
diff --git a/drivers/s390/net/qeth_core_mpc.c b/drivers/s390/net/qeth_core_mpc.c
index e8263ded0af0..e891c0b52f4c 100644
--- a/drivers/s390/net/qeth_core_mpc.c
+++ b/drivers/s390/net/qeth_core_mpc.c
@@ -148,10 +148,10 @@ EXPORT_SYMBOL_GPL(IPA_PDU_HEADER);
struct ipa_rc_msg {
enum qeth_ipa_return_codes rc;
- char *msg;
+ const char *msg;
};
-static struct ipa_rc_msg qeth_ipa_rc_msg[] = {
+static const struct ipa_rc_msg qeth_ipa_rc_msg[] = {
{IPA_RC_SUCCESS, "success"},
{IPA_RC_NOTSUPP, "Command not supported"},
{IPA_RC_IP_TABLE_FULL, "Add Addr IP Table Full - ipv6"},
@@ -219,22 +219,23 @@ static struct ipa_rc_msg qeth_ipa_rc_msg[] = {
-char *qeth_get_ipa_msg(enum qeth_ipa_return_codes rc)
+const char *qeth_get_ipa_msg(enum qeth_ipa_return_codes rc)
{
- int x = 0;
- qeth_ipa_rc_msg[ARRAY_SIZE(qeth_ipa_rc_msg) - 1].rc = rc;
- while (qeth_ipa_rc_msg[x].rc != rc)
- x++;
+ int x;
+
+ for (x = 0; x < ARRAY_SIZE(qeth_ipa_rc_msg) - 1; x++)
+ if (qeth_ipa_rc_msg[x].rc == rc)
+ return qeth_ipa_rc_msg[x].msg;
return qeth_ipa_rc_msg[x].msg;
}
struct ipa_cmd_names {
enum qeth_ipa_cmds cmd;
- char *name;
+ const char *name;
};
-static struct ipa_cmd_names qeth_ipa_cmd_names[] = {
+static const struct ipa_cmd_names qeth_ipa_cmd_names[] = {
{IPA_CMD_STARTLAN, "startlan"},
{IPA_CMD_STOPLAN, "stoplan"},
{IPA_CMD_SETVMAC, "setvmac"},
@@ -266,11 +267,12 @@ static struct ipa_cmd_names qeth_ipa_cmd_names[] = {
{IPA_CMD_UNKNOWN, "unknown"},
};
-char *qeth_get_ipa_cmd_name(enum qeth_ipa_cmds cmd)
+const char *qeth_get_ipa_cmd_name(enum qeth_ipa_cmds cmd)
{
- int x = 0;
- qeth_ipa_cmd_names[ARRAY_SIZE(qeth_ipa_cmd_names) - 1].cmd = cmd;
- while (qeth_ipa_cmd_names[x].cmd != cmd)
- x++;
+ int x;
+
+ for (x = 0; x < ARRAY_SIZE(qeth_ipa_cmd_names) - 1; x++)
+ if (qeth_ipa_cmd_names[x].cmd == cmd)
+ return qeth_ipa_cmd_names[x].name;
return qeth_ipa_cmd_names[x].name;
}
diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h
index aa8b9196b089..aa5de1fe01e1 100644
--- a/drivers/s390/net/qeth_core_mpc.h
+++ b/drivers/s390/net/qeth_core_mpc.h
@@ -797,8 +797,8 @@ enum qeth_ipa_arp_return_codes {
QETH_IPA_ARP_RC_Q_NO_DATA = 0x0008,
};
-extern char *qeth_get_ipa_msg(enum qeth_ipa_return_codes rc);
-extern char *qeth_get_ipa_cmd_name(enum qeth_ipa_cmds cmd);
+extern const char *qeth_get_ipa_msg(enum qeth_ipa_return_codes rc);
+extern const char *qeth_get_ipa_cmd_name(enum qeth_ipa_cmds cmd);
#define QETH_SETASS_BASE_LEN (sizeof(struct qeth_ipacmd_hdr) + \
sizeof(struct qeth_ipacmd_setassparms_hdr))
--
2.16.4
^ permalink raw reply related
* Re: [PATCH net v2] bnxt_en: Fix TX timeout during netpoll.
From: Song Liu @ 2018-09-26 16:11 UTC (permalink / raw)
To: Michael Chan
Cc: edumazet@google.com, davem@davemloft.net, netdev@vger.kernel.org
In-Reply-To: <1537936864-23472-1-git-send-email-michael.chan@broadcom.com>
> On Sep 25, 2018, at 9:41 PM, Michael Chan <michael.chan@broadcom.com> wrote:
>
> The current netpoll implementation in the bnxt_en driver has problems
> that may miss TX completion events. bnxt_poll_work() in effect is
> only handling at most 1 TX packet before exiting. In addition,
> there may be in flight TX completions that ->poll() may miss even
> after we fix bnxt_poll_work() to handle all visible TX completions.
> netpoll may not call ->poll() again and HW may not generate IRQ
> because the driver does not ARM the IRQ when the budget (0 for netpoll)
> is reached.
>
> We fix it by handling all TX completions and to always ARM the IRQ
> when we exit ->poll() with 0 budget.
>
> Also, the logic to ACK the completion ring in case it is almost filled
> with TX completions need to be adjusted to take care of the 0 budget
> case, as discussed with Eric Dumazet <edumazet@google.com>
>
> Reported-by: Song Liu <songliubraving@fb.com>
> Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Reviewed-and-tested-by: Song Liu <songliubraving@fb.com>
> ---
> drivers/net/ethernet/broadcom/bnxt/bnxt.c | 13 ++++++++++---
> 1 file changed, 10 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> index 61957b0..0478e56 100644
> --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> @@ -1884,8 +1884,11 @@ static int bnxt_poll_work(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
> if (TX_CMP_TYPE(txcmp) == CMP_TYPE_TX_L2_CMP) {
> tx_pkts++;
> /* return full budget so NAPI will complete. */
> - if (unlikely(tx_pkts > bp->tx_wake_thresh))
> + if (unlikely(tx_pkts > bp->tx_wake_thresh)) {
> rx_pkts = budget;
> + raw_cons = NEXT_RAW_CMP(raw_cons);
> + break;
> + }
> } else if ((TX_CMP_TYPE(txcmp) & 0x30) == 0x10) {
> if (likely(budget))
> rc = bnxt_rx_pkt(bp, bnapi, &raw_cons, &event);
> @@ -1913,7 +1916,7 @@ static int bnxt_poll_work(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
> }
> raw_cons = NEXT_RAW_CMP(raw_cons);
>
> - if (rx_pkts == budget)
> + if (rx_pkts && rx_pkts == budget)
> break;
> }
>
> @@ -2027,8 +2030,12 @@ static int bnxt_poll(struct napi_struct *napi, int budget)
> while (1) {
> work_done += bnxt_poll_work(bp, bnapi, budget - work_done);
>
> - if (work_done >= budget)
> + if (work_done >= budget) {
> + if (!budget)
> + BNXT_CP_DB_REARM(cpr->cp_doorbell,
> + cpr->cp_raw_cons);
> break;
> + }
>
> if (!bnxt_has_work(bp, cpr)) {
> if (napi_complete_done(napi, work_done))
> --
> 2.5.1
>
^ permalink raw reply
* Re: [PATCH net-next v6 07/23] zinc: ChaCha20 ARM and ARM64 implementations
From: Andy Lutomirski @ 2018-09-26 16:21 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: Jason A. Donenfeld, Herbert Xu, Thomas Gleixner,
Linux Kernel Mailing List, <netdev@vger.kernel.org>,
open list:HARDWARE RANDOM NUMBER GENERATOR CORE, David S. Miller,
Greg Kroah-Hartman, Samuel Neves, Andy Lutomirski,
Jean-Philippe Aumasson, Russell King, linux-arm-kernel
In-Reply-To: <CAKv+Gu8ih-TsASRGqK+ST_5+EQ0=Zo-zhGCadOdGyPjucMFTCg@mail.gmail.com>
> On Sep 26, 2018, at 7:02 AM, Ard Biesheuvel <ard.biesheuvel@linaro.org> wrote:
>
> (+ Herbert, Thomas)
>
>> On Wed, 26 Sep 2018 at 15:33, Jason A. Donenfeld <Jason@zx2c4.com> wrote:
>>
>> Hi Ard,
>> .
>
>> And if it becomes one,
>> this is something we can address *later*, but certainly there's no use
>> of adding additional complexity to the initial patchset to do this
>> now.
>>
>
> You are introducing a very useful SIMD abstraction, but it lets code
> run with preemption disabled for unbounded amounts of time, and so now
> is the time to ensure we get it right.
>
> Part of the [justified] criticism on the current state of the crypto
> API is on its complexity, and so I don't think it makes sense to keep
> it simple now and add the complexity later (and the same concern
> applies to async support btw).
Are, is what you’re saying that the Zinc chacha20 functions should call simd_relax() every n bytes automatically for some reasonable value of n? If so, seems sensible, except that some care might be needed to make sure they interact with preemption correctly.
What I mean is: the public Zinc entry points should either be callable in an atomic context or they should not be. I think this should be checked at runtime in an appropriate place with an __might_sleep or similar. Or simd_relax should learn to *not* schedule if the result of preempt_enable() leaves it atomic. (And the latter needs to be done in a way that works even on non-preempt kernels, and I don’t remember whether that’s possible.). And this should happen regardless of how many bytes are processed. IOW, calling into Zinc should be equally not atomic-safe for 100 bytes and for 10 MB.
As for async, ISTM a really good WireGuard accelerator would expose a different interface than crypto API supports, and it probably makes sense to wait for such hardware to show up before figuring out how to use it. And no matter what form it takes, I don’t think it should complicate the basic Zinc crypto entry points.
^ permalink raw reply
* [PATCH net-next 00/15] s390/net: updates 2018-09-26
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
Hi Dave,
please apply one more series of cleanups and small improvements for qeth
to net-next. Note that one patch needs to touch both af_iucv and qeth, in
order to untangle their receive paths.
Thanks,
Julian
Julian Wiedmann (15):
s390/qeth: convert layer attribute to enum
s390/qeth: use DEFINE_MUTEX for qeth_mod_mutex
s390/qeth: fix discipline unload after setup error
s390/qeth: on gdev release, reset drvdata
net/af_iucv: locate IUCV header via skb_network_header()
s390/qeth: replace open-coded skb_queue_walk()
s390/qeth: remove additional skb refcount
s390/qeth: re-use qeth_notify_skbs()
s390/qeth: pass card pointer in iob callback
s390/qeth: remove CARD_FROM_CDEV helper
s390/qeth: remove various redundant code
s390/qeth: consume local address events
s390/qeth: re-indent qeth_check_ipa_data()
s390/qeth: clean up drop conditions for received cmds
s390/qeth: remove duplicated carrier state tracking
drivers/s390/net/qeth_core.h | 21 ++-
drivers/s390/net/qeth_core_main.c | 373 ++++++++++++++++----------------------
drivers/s390/net/qeth_core_sys.c | 15 +-
drivers/s390/net/qeth_l2_main.c | 12 +-
drivers/s390/net/qeth_l3_main.c | 24 +--
include/net/iucv/af_iucv.h | 5 +
net/iucv/af_iucv.c | 42 ++---
7 files changed, 200 insertions(+), 292 deletions(-)
--
2.16.4
^ permalink raw reply
* [PATCH net-next 01/15] s390/qeth: convert layer attribute to enum
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
While the raw values are fixed due to their use in a sysfs attribute,
we can still use the proper QETH_DISCIPLINE_* enum within the driver.
Also move the initialization into qeth_set_initial_options(), along with
all other user-configurable fields.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core.h | 17 ++++++++++-------
drivers/s390/net/qeth_core_main.c | 22 +++++++++-------------
drivers/s390/net/qeth_core_sys.c | 8 ++++----
drivers/s390/net/qeth_l2_main.c | 2 +-
drivers/s390/net/qeth_l3_main.c | 2 +-
5 files changed, 25 insertions(+), 26 deletions(-)
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 0dbe81f958f0..630a01b3212c 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -671,6 +671,12 @@ struct qeth_card_info {
__u32 hwtrap;
};
+enum qeth_discipline_id {
+ QETH_DISCIPLINE_UNDETERMINED = -1,
+ QETH_DISCIPLINE_LAYER3 = 0,
+ QETH_DISCIPLINE_LAYER2 = 1,
+};
+
struct qeth_card_options {
struct qeth_routing_info route4;
struct qeth_ipa_info ipa4;
@@ -680,7 +686,7 @@ struct qeth_card_options {
struct qeth_sbp_info sbp; /* SETBRIDGEPORT options */
struct qeth_vnicc_info vnicc; /* VNICC options */
int fake_broadcast;
- int layer2;
+ enum qeth_discipline_id layer;
int performance_stats;
int rx_sg_cb;
enum qeth_ipa_isolation_modes isolation;
@@ -690,6 +696,9 @@ struct qeth_card_options {
char hsuid[9];
};
+#define IS_LAYER2(card) ((card)->options.layer == QETH_DISCIPLINE_LAYER2)
+#define IS_LAYER3(card) ((card)->options.layer == QETH_DISCIPLINE_LAYER3)
+
/*
* thread bits for qeth_card thread masks
*/
@@ -702,12 +711,6 @@ struct qeth_osn_info {
int (*data_cb)(struct sk_buff *skb);
};
-enum qeth_discipline_id {
- QETH_DISCIPLINE_UNDETERMINED = -1,
- QETH_DISCIPLINE_LAYER3 = 0,
- QETH_DISCIPLINE_LAYER2 = 1,
-};
-
struct qeth_discipline {
const struct device_type *devtype;
int (*process_rx_buffer)(struct qeth_card *card, int budget, int *done);
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 89e09e7b8fff..4fd9bdc2d0ae 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -1429,6 +1429,7 @@ static void qeth_set_initial_options(struct qeth_card *card)
card->options.rx_sg_cb = QETH_RX_SG_CB;
card->options.isolation = ISOLATION_MODE_NONE;
card->options.cq = QETH_CQ_DISABLED;
+ card->options.layer = QETH_DISCIPLINE_UNDETERMINED;
}
static int qeth_do_start_thread(struct qeth_card *card, unsigned long thread)
@@ -1522,7 +1523,6 @@ static struct qeth_card *qeth_alloc_card(struct ccwgroup_device *gdev)
goto out_channel;
if (qeth_setup_channel(&card->data, false))
goto out_data;
- card->options.layer2 = -1;
card->qeth_service_level.seq_print = qeth_core_sl_print;
register_service_level(&card->qeth_service_level);
return card;
@@ -2291,7 +2291,7 @@ static int qeth_update_max_mtu(struct qeth_card *card, unsigned int max_mtu)
if (dev->mtu)
new_mtu = dev->mtu;
/* default MTUs for first setup: */
- else if (card->options.layer2)
+ else if (IS_LAYER2(card))
new_mtu = ETH_DATA_LEN;
else
new_mtu = ETH_DATA_LEN - 8; /* allow for LLC + SNAP */
@@ -2358,7 +2358,7 @@ static u8 qeth_mpc_select_prot_type(struct qeth_card *card)
{
if (IS_OSN(card))
return QETH_PROT_OSN2;
- return (card->options.layer2 == 1) ? QETH_PROT_LAYER2 : QETH_PROT_TCPIP;
+ return IS_LAYER2(card) ? QETH_PROT_LAYER2 : QETH_PROT_TCPIP;
}
static int qeth_ulp_enable(struct qeth_card *card)
@@ -2896,10 +2896,7 @@ static void qeth_fill_ipacmd_header(struct qeth_card *card,
/* cmd->hdr.seqno is set by qeth_send_control_data() */
cmd->hdr.adapter_type = qeth_get_ipa_adp_type(card->info.link_type);
cmd->hdr.rel_adapter_no = (u8) card->dev->dev_port;
- if (card->options.layer2)
- cmd->hdr.prim_version_no = 2;
- else
- cmd->hdr.prim_version_no = 1;
+ cmd->hdr.prim_version_no = IS_LAYER2(card) ? 2 : 1;
cmd->hdr.param_count = 1;
cmd->hdr.prot_version = prot;
}
@@ -4278,8 +4275,7 @@ static int qeth_setadpparms_change_macaddr_cb(struct qeth_card *card,
if (qeth_setadpparms_inspect_rc(cmd))
return 0;
- if (!card->options.layer2 ||
- !(card->info.mac_bits & QETH_LAYER2_MAC_READ)) {
+ if (IS_LAYER3(card) || !(card->info.mac_bits & QETH_LAYER2_MAC_READ)) {
ether_addr_copy(card->dev->dev_addr,
cmd->data.setadapterparms.data.change_addr.addr);
card->info.mac_bits |= QETH_LAYER2_MAC_READ;
@@ -4633,9 +4629,9 @@ static int qeth_snmp_command(struct qeth_card *card, char __user *udata)
return -EOPNOTSUPP;
if ((!qeth_adp_supported(card, IPA_SETADP_SET_SNMP_CONTROL)) &&
- (!card->options.layer2)) {
+ IS_LAYER3(card))
return -EOPNOTSUPP;
- }
+
/* skip 4 bytes (data_len struct member) to get req_len */
if (copy_from_user(&req_len, udata + sizeof(int), sizeof(int)))
return -EFAULT;
@@ -5623,7 +5619,7 @@ int qeth_core_load_discipline(struct qeth_card *card,
void qeth_core_free_discipline(struct qeth_card *card)
{
- if (card->options.layer2)
+ if (IS_LAYER2(card))
symbol_put(qeth_l2_discipline);
else
symbol_put(qeth_l3_discipline);
@@ -6146,7 +6142,7 @@ void qeth_core_get_drvinfo(struct net_device *dev,
{
struct qeth_card *card = dev->ml_priv;
- strlcpy(info->driver, card->options.layer2 ? "qeth_l2" : "qeth_l3",
+ strlcpy(info->driver, IS_LAYER2(card) ? "qeth_l2" : "qeth_l3",
sizeof(info->driver));
strlcpy(info->version, "1.0", sizeof(info->version));
strlcpy(info->fw_version, card->info.mcl_level,
diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c
index 25d0be25bcb3..fdb67af811f4 100644
--- a/drivers/s390/net/qeth_core_sys.c
+++ b/drivers/s390/net/qeth_core_sys.c
@@ -228,7 +228,7 @@ static ssize_t qeth_dev_prioqing_store(struct device *dev,
card->qdio.do_prio_queueing = QETH_PRIO_Q_ING_TOS;
card->qdio.default_out_queue = QETH_DEFAULT_QUEUE;
} else if (sysfs_streq(buf, "prio_queueing_vlan")) {
- if (!card->options.layer2) {
+ if (IS_LAYER3(card)) {
rc = -ENOTSUPP;
goto out;
}
@@ -379,7 +379,7 @@ static ssize_t qeth_dev_layer2_show(struct device *dev,
if (!card)
return -EINVAL;
- return sprintf(buf, "%i\n", card->options.layer2);
+ return sprintf(buf, "%i\n", card->options.layer);
}
static ssize_t qeth_dev_layer2_store(struct device *dev,
@@ -413,7 +413,7 @@ static ssize_t qeth_dev_layer2_store(struct device *dev,
goto out;
}
- if (card->options.layer2 == newdis)
+ if (card->options.layer == newdis)
goto out;
if (card->info.layer_enforced) {
/* fixed layer, can't switch */
@@ -432,7 +432,7 @@ static ssize_t qeth_dev_layer2_store(struct device *dev,
card->discipline->remove(card->gdev);
qeth_core_free_discipline(card);
- card->options.layer2 = -1;
+ card->options.layer = QETH_DISCIPLINE_UNDETERMINED;
free_netdev(card->dev);
card->dev = ndev;
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 12858ffa28cf..afa7a005b21e 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -806,7 +806,7 @@ static int qeth_l2_probe_device(struct ccwgroup_device *gdev)
}
INIT_LIST_HEAD(&card->vid_list);
hash_init(card->mac_htable);
- card->options.layer2 = 1;
+ card->options.layer = QETH_DISCIPLINE_LAYER2;
card->info.hwtrap = 0;
qeth_l2_vnicc_set_defaults(card);
return 0;
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 8930d2a9fcad..552bfad20f85 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2489,7 +2489,7 @@ static int qeth_l3_probe_device(struct ccwgroup_device *gdev)
}
hash_init(card->ip_htable);
hash_init(card->ip_mc_htable);
- card->options.layer2 = 0;
+ card->options.layer = QETH_DISCIPLINE_LAYER3;
card->info.hwtrap = 0;
return 0;
}
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 02/15] s390/qeth: use DEFINE_MUTEX for qeth_mod_mutex
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
Consolidate declaration and initialization of a static variable.
While at it reduce its scope in qeth_core_load_discipline(), and simplify
the return logic accordingly.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 4fd9bdc2d0ae..c3068f680f67 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -62,7 +62,6 @@ static struct kmem_cache *qeth_qdio_outbuf_cache;
static struct device *qeth_core_root_dev;
static struct lock_class_key qdio_out_skb_queue_key;
-static struct mutex qeth_mod_mutex;
static void qeth_send_control_data_cb(struct qeth_channel *,
struct qeth_cmd_buffer *);
@@ -5589,11 +5588,11 @@ static int qeth_register_dbf_views(void)
return 0;
}
+static DEFINE_MUTEX(qeth_mod_mutex); /* for synchronized module loading */
+
int qeth_core_load_discipline(struct qeth_card *card,
enum qeth_discipline_id discipline)
{
- int rc = 0;
-
mutex_lock(&qeth_mod_mutex);
switch (discipline) {
case QETH_DISCIPLINE_LAYER3:
@@ -5607,14 +5606,15 @@ int qeth_core_load_discipline(struct qeth_card *card,
default:
break;
}
+ mutex_unlock(&qeth_mod_mutex);
if (!card->discipline) {
dev_err(&card->gdev->dev, "There is no kernel module to "
"support discipline %d\n", discipline);
- rc = -EINVAL;
+ return -EINVAL;
}
- mutex_unlock(&qeth_mod_mutex);
- return rc;
+
+ return 0;
}
void qeth_core_free_discipline(struct qeth_card *card)
@@ -6623,7 +6623,6 @@ static int __init qeth_core_init(void)
INIT_LIST_HEAD(&qeth_core_card_list.list);
INIT_LIST_HEAD(&qeth_dbf_list);
rwlock_init(&qeth_core_card_list.rwlock);
- mutex_init(&qeth_mod_mutex);
qeth_wq = create_singlethread_workqueue("qeth_wq");
if (!qeth_wq) {
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 03/15] s390/qeth: fix discipline unload after setup error
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
Device initialization code usually first loads a subdriver
(via qeth_core_load_discipline()), and then runs its setup() callback.
If this fails, it rolls back the load via qeth_core_free_discipline().
qeth_core_free_discipline() expects the options.layer attribute to be
initialized, but on error in setup() that's currently not the case.
Resulting in misbalanced symbol_put() calls.
Fix this by setting options.layer when loading the subdriver.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 2 ++
drivers/s390/net/qeth_core_sys.c | 2 --
drivers/s390/net/qeth_l2_main.c | 1 -
drivers/s390/net/qeth_l3_main.c | 1 -
4 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index c3068f680f67..1fed8f113f40 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -5614,6 +5614,7 @@ int qeth_core_load_discipline(struct qeth_card *card,
return -EINVAL;
}
+ card->options.layer = discipline;
return 0;
}
@@ -5623,6 +5624,7 @@ void qeth_core_free_discipline(struct qeth_card *card)
symbol_put(qeth_l2_discipline);
else
symbol_put(qeth_l3_discipline);
+ card->options.layer = QETH_DISCIPLINE_UNDETERMINED;
card->discipline = NULL;
}
diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c
index fdb67af811f4..970f6c71a66e 100644
--- a/drivers/s390/net/qeth_core_sys.c
+++ b/drivers/s390/net/qeth_core_sys.c
@@ -432,8 +432,6 @@ static ssize_t qeth_dev_layer2_store(struct device *dev,
card->discipline->remove(card->gdev);
qeth_core_free_discipline(card);
- card->options.layer = QETH_DISCIPLINE_UNDETERMINED;
-
free_netdev(card->dev);
card->dev = ndev;
}
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index afa7a005b21e..02566af7e23d 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -806,7 +806,6 @@ static int qeth_l2_probe_device(struct ccwgroup_device *gdev)
}
INIT_LIST_HEAD(&card->vid_list);
hash_init(card->mac_htable);
- card->options.layer = QETH_DISCIPLINE_LAYER2;
card->info.hwtrap = 0;
qeth_l2_vnicc_set_defaults(card);
return 0;
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 552bfad20f85..2756795f7708 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2489,7 +2489,6 @@ static int qeth_l3_probe_device(struct ccwgroup_device *gdev)
}
hash_init(card->ip_htable);
hash_init(card->ip_mc_htable);
- card->options.layer = QETH_DISCIPLINE_LAYER3;
card->info.hwtrap = 0;
return 0;
}
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 04/15] s390/qeth: on gdev release, reset drvdata
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
qeth_core_probe_device() sets the gdev's drvdata, but doesn't reset it
on a subsequent error. Move the (re-)setting around a bit, so that it
happens symmetrically on allocating/freeing the qeth_card struct.
This is no actual problem, as the ccwgroup core will discard the gdev
on a probe error. But from qeth's perspective the gdev is an external
resource, so it's best to manage it cleanly.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 1fed8f113f40..a6c632d36df8 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -1513,6 +1513,7 @@ static struct qeth_card *qeth_alloc_card(struct ccwgroup_device *gdev)
QETH_DBF_HEX(SETUP, 2, &card, sizeof(void *));
card->gdev = gdev;
+ dev_set_drvdata(&gdev->dev, card);
CARD_RDEV(card) = gdev->cdev[0];
CARD_WDEV(card) = gdev->cdev[1];
CARD_DDEV(card) = gdev->cdev[2];
@@ -1531,6 +1532,7 @@ static struct qeth_card *qeth_alloc_card(struct ccwgroup_device *gdev)
out_channel:
qeth_clean_channel(&card->read);
out_ip:
+ dev_set_drvdata(&gdev->dev, NULL);
kfree(card);
out:
return NULL;
@@ -5074,6 +5076,7 @@ static void qeth_core_free_card(struct qeth_card *card)
qeth_clean_channel(&card->data);
qeth_free_qdio_buffers(card);
unregister_service_level(&card->qeth_service_level);
+ dev_set_drvdata(&card->gdev->dev, NULL);
kfree(card);
}
@@ -5788,7 +5791,6 @@ static int qeth_core_probe_device(struct ccwgroup_device *gdev)
goto err_card;
}
- dev_set_drvdata(&gdev->dev, card);
qeth_setup_card(card);
qeth_update_from_chp_desc(card);
@@ -5851,7 +5853,6 @@ static void qeth_core_remove_device(struct ccwgroup_device *gdev)
write_unlock_irq(&qeth_core_card_list.rwlock);
free_netdev(card->dev);
qeth_core_free_card(card);
- dev_set_drvdata(&gdev->dev, NULL);
put_device(&gdev->dev);
}
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 06/15] s390/qeth: replace open-coded skb_queue_walk()
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
To match the use of __skb_queue_purge(), also make the skb's enqueue in
qeth_fill_buffer() lockless.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 19 ++++---------------
1 file changed, 4 insertions(+), 15 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index a6c632d36df8..c5c40c6d8b7d 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -1177,10 +1177,7 @@ static void qeth_notify_skbs(struct qeth_qdio_out_q *q,
{
struct sk_buff *skb;
- if (skb_queue_empty(&buf->skb_list))
- goto out;
- skb = skb_peek(&buf->skb_list);
- while (skb) {
+ skb_queue_walk(&buf->skb_list, skb) {
QETH_CARD_TEXT_(q->card, 5, "skbn%d", notification);
QETH_CARD_TEXT_(q->card, 5, "%lx", (long) skb);
if (be16_to_cpu(skb->protocol) == ETH_P_AF_IUCV) {
@@ -1189,13 +1186,7 @@ static void qeth_notify_skbs(struct qeth_qdio_out_q *q,
iucv->sk_txnotify(skb, notification);
}
}
- if (skb_queue_is_last(&buf->skb_list, skb))
- skb = NULL;
- else
- skb = skb_queue_next(&buf->skb_list, skb);
}
-out:
- return;
}
static void qeth_release_skbs(struct qeth_qdio_out_buffer *buf)
@@ -1210,8 +1201,7 @@ static void qeth_release_skbs(struct qeth_qdio_out_buffer *buf)
/* release may never happen from within CQ tasklet scope */
WARN_ON_ONCE(atomic_read(&buf->state) == QETH_QDIO_BUF_IN_CQ);
- skb = skb_dequeue(&buf->skb_list);
- while (skb) {
+ skb_queue_walk(&buf->skb_list, skb) {
QETH_CARD_TEXT(buf->q->card, 5, "skbr");
QETH_CARD_TEXT_(buf->q->card, 5, "%lx", (long) skb);
if (notify_general_error &&
@@ -1222,9 +1212,8 @@ static void qeth_release_skbs(struct qeth_qdio_out_buffer *buf)
}
}
refcount_dec(&skb->users);
- dev_kfree_skb_any(skb);
- skb = skb_dequeue(&buf->skb_list);
}
+ __skb_queue_purge(&buf->skb_list);
}
static void qeth_clear_output_buffer(struct qeth_qdio_out_q *queue,
@@ -4000,7 +3989,7 @@ static int qeth_fill_buffer(struct qeth_qdio_out_q *queue,
int flush_cnt = 0;
refcount_inc(&skb->users);
- skb_queue_tail(&buf->skb_list, skb);
+ __skb_queue_tail(&buf->skb_list, skb);
/* build dedicated header element */
if (hd_len) {
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 05/15] net/af_iucv: locate IUCV header via skb_network_header()
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
This patch attempts to untangle the TX and RX code in qeth from
af_iucv's respective HiperTransport path:
On the TX side, pointing skb_network_header() at the IUCV header
means that qeth_l3_fill_af_iucv_hdr() no longer needs a magical offset
to access the header.
On the RX side, qeth pulls the (fake) L2 header off the skb like any
normal ethernet driver would. This makes working with the IUCV header
in af_iucv easier, since we no longer have to assume a fixed skb layout.
While at it, replace the open-coded length checks in af_iucv's RX path
with pskb_may_pull().
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_l3_main.c | 12 +++++-------
include/net/iucv/af_iucv.h | 5 +++++
net/iucv/af_iucv.c | 42 +++++++++++++----------------------------
3 files changed, 23 insertions(+), 36 deletions(-)
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 2756795f7708..7148ef71ac78 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1348,6 +1348,7 @@ static void qeth_l3_rebuild_skb(struct qeth_card *card, struct sk_buff *skb,
static int qeth_l3_process_inbound_buffer(struct qeth_card *card,
int budget, int *done)
{
+ struct net_device *dev = card->dev;
int work_done = 0;
struct sk_buff *skb;
struct qeth_hdr *hdr;
@@ -1369,11 +1370,10 @@ static int qeth_l3_process_inbound_buffer(struct qeth_card *card,
magic = *(__u16 *)skb->data;
if ((card->info.type == QETH_CARD_TYPE_IQD) &&
(magic == ETH_P_AF_IUCV)) {
- skb->protocol = cpu_to_be16(ETH_P_AF_IUCV);
len = skb->len;
- card->dev->header_ops->create(skb, card->dev, 0,
- card->dev->dev_addr, "FAKELL", len);
- skb_reset_mac_header(skb);
+ dev_hard_header(skb, dev, ETH_P_AF_IUCV,
+ dev->dev_addr, "FAKELL", len);
+ skb->protocol = eth_type_trans(skb, dev);
netif_receive_skb(skb);
} else {
qeth_l3_rebuild_skb(card, skb, hdr);
@@ -2005,17 +2005,15 @@ static void qeth_l3_fill_af_iucv_hdr(struct qeth_hdr *hdr, struct sk_buff *skb,
unsigned int data_len)
{
char daddr[16];
- struct af_iucv_trans_hdr *iucv_hdr;
hdr->hdr.l3.id = QETH_HEADER_TYPE_LAYER3;
hdr->hdr.l3.length = data_len;
hdr->hdr.l3.flags = QETH_HDR_IPV6 | QETH_CAST_UNICAST;
- iucv_hdr = (struct af_iucv_trans_hdr *)(skb_mac_header(skb) + ETH_HLEN);
memset(daddr, 0, sizeof(daddr));
daddr[0] = 0xfe;
daddr[1] = 0x80;
- memcpy(&daddr[8], iucv_hdr->destUserID, 8);
+ memcpy(&daddr[8], iucv_trans_hdr(skb)->destUserID, 8);
memcpy(hdr->hdr.l3.next_hop.ipv6_addr, daddr, 16);
}
diff --git a/include/net/iucv/af_iucv.h b/include/net/iucv/af_iucv.h
index f4c21b5a1242..14a490246be9 100644
--- a/include/net/iucv/af_iucv.h
+++ b/include/net/iucv/af_iucv.h
@@ -80,6 +80,11 @@ struct af_iucv_trans_hdr {
u8 pad; /* total 104 bytes */
} __packed;
+static inline struct af_iucv_trans_hdr *iucv_trans_hdr(struct sk_buff *skb)
+{
+ return (struct af_iucv_trans_hdr *)skb_network_header(skb);
+}
+
enum iucv_tx_notify {
/* transmission of skb is completed and was successful */
TX_NOTIFY_OK = 0,
diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c
index 5b68ee908107..45115c125569 100644
--- a/net/iucv/af_iucv.c
+++ b/net/iucv/af_iucv.c
@@ -320,13 +320,9 @@ static int afiucv_hs_send(struct iucv_message *imsg, struct sock *sock,
struct sk_buff *nskb;
int err, confirm_recv = 0;
- memset(skb->head, 0, ETH_HLEN);
- phs_hdr = skb_push(skb, sizeof(struct af_iucv_trans_hdr));
- skb_reset_mac_header(skb);
+ phs_hdr = skb_push(skb, sizeof(*phs_hdr));
+ memset(phs_hdr, 0, sizeof(*phs_hdr));
skb_reset_network_header(skb);
- skb_push(skb, ETH_HLEN);
- skb_reset_mac_header(skb);
- memset(phs_hdr, 0, sizeof(struct af_iucv_trans_hdr));
phs_hdr->magic = ETH_P_AF_IUCV;
phs_hdr->version = 1;
@@ -350,6 +346,9 @@ static int afiucv_hs_send(struct iucv_message *imsg, struct sock *sock,
if (imsg)
memcpy(&phs_hdr->iucv_hdr, imsg, sizeof(struct iucv_message));
+ skb_push(skb, ETH_HLEN);
+ memset(skb->data, 0, ETH_HLEN);
+
skb->dev = iucv->hs_dev;
if (!skb->dev) {
err = -ENODEV;
@@ -1943,8 +1942,7 @@ static void iucv_callback_shutdown(struct iucv_path *path, u8 ipuser[16])
/***************** HiperSockets transport callbacks ********************/
static void afiucv_swap_src_dest(struct sk_buff *skb)
{
- struct af_iucv_trans_hdr *trans_hdr =
- (struct af_iucv_trans_hdr *)skb->data;
+ struct af_iucv_trans_hdr *trans_hdr = iucv_trans_hdr(skb);
char tmpID[8];
char tmpName[8];
@@ -1967,13 +1965,12 @@ static void afiucv_swap_src_dest(struct sk_buff *skb)
**/
static int afiucv_hs_callback_syn(struct sock *sk, struct sk_buff *skb)
{
+ struct af_iucv_trans_hdr *trans_hdr = iucv_trans_hdr(skb);
struct sock *nsk;
struct iucv_sock *iucv, *niucv;
- struct af_iucv_trans_hdr *trans_hdr;
int err;
iucv = iucv_sk(sk);
- trans_hdr = (struct af_iucv_trans_hdr *)skb->data;
if (!iucv) {
/* no sock - connection refused */
afiucv_swap_src_dest(skb);
@@ -2034,15 +2031,13 @@ static int afiucv_hs_callback_syn(struct sock *sk, struct sk_buff *skb)
static int afiucv_hs_callback_synack(struct sock *sk, struct sk_buff *skb)
{
struct iucv_sock *iucv = iucv_sk(sk);
- struct af_iucv_trans_hdr *trans_hdr =
- (struct af_iucv_trans_hdr *)skb->data;
if (!iucv)
goto out;
if (sk->sk_state != IUCV_BOUND)
goto out;
bh_lock_sock(sk);
- iucv->msglimit_peer = trans_hdr->window;
+ iucv->msglimit_peer = iucv_trans_hdr(skb)->window;
sk->sk_state = IUCV_CONNECTED;
sk->sk_state_change(sk);
bh_unlock_sock(sk);
@@ -2098,8 +2093,6 @@ static int afiucv_hs_callback_fin(struct sock *sk, struct sk_buff *skb)
static int afiucv_hs_callback_win(struct sock *sk, struct sk_buff *skb)
{
struct iucv_sock *iucv = iucv_sk(sk);
- struct af_iucv_trans_hdr *trans_hdr =
- (struct af_iucv_trans_hdr *)skb->data;
if (!iucv)
return NET_RX_SUCCESS;
@@ -2107,7 +2100,7 @@ static int afiucv_hs_callback_win(struct sock *sk, struct sk_buff *skb)
if (sk->sk_state != IUCV_CONNECTED)
return NET_RX_SUCCESS;
- atomic_sub(trans_hdr->window, &iucv->msg_sent);
+ atomic_sub(iucv_trans_hdr(skb)->window, &iucv->msg_sent);
iucv_sock_wake_msglim(sk);
return NET_RX_SUCCESS;
}
@@ -2170,22 +2163,13 @@ static int afiucv_hs_rcv(struct sk_buff *skb, struct net_device *dev,
int err = NET_RX_SUCCESS;
char nullstring[8];
- if (skb->len < (ETH_HLEN + sizeof(struct af_iucv_trans_hdr))) {
- WARN_ONCE(1, "AF_IUCV too short skb, len=%d, min=%d",
- (int)skb->len,
- (int)(ETH_HLEN + sizeof(struct af_iucv_trans_hdr)));
+ if (!pskb_may_pull(skb, sizeof(*trans_hdr))) {
+ WARN_ONCE(1, "AF_IUCV failed to receive skb, len=%u", skb->len);
kfree_skb(skb);
return NET_RX_SUCCESS;
}
- if (skb_headlen(skb) < (ETH_HLEN + sizeof(struct af_iucv_trans_hdr)))
- if (skb_linearize(skb)) {
- WARN_ONCE(1, "AF_IUCV skb_linearize failed, len=%d",
- (int)skb->len);
- kfree_skb(skb);
- return NET_RX_SUCCESS;
- }
- skb_pull(skb, ETH_HLEN);
- trans_hdr = (struct af_iucv_trans_hdr *)skb->data;
+
+ trans_hdr = iucv_trans_hdr(skb);
EBCASC(trans_hdr->destAppName, sizeof(trans_hdr->destAppName));
EBCASC(trans_hdr->destUserID, sizeof(trans_hdr->destUserID));
EBCASC(trans_hdr->srcAppName, sizeof(trans_hdr->srcAppName));
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 07/15] s390/qeth: remove additional skb refcount
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
This was presumably left over from back when qeth recursed into
dev_queue_xmit().
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index c5c40c6d8b7d..012519ffd8de 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -1211,7 +1211,6 @@ static void qeth_release_skbs(struct qeth_qdio_out_buffer *buf)
iucv->sk_txnotify(skb, TX_NOTIFY_GENERALERROR);
}
}
- refcount_dec(&skb->users);
}
__skb_queue_purge(&buf->skb_list);
}
@@ -3988,7 +3987,6 @@ static int qeth_fill_buffer(struct qeth_qdio_out_q *queue,
bool is_first_elem = true;
int flush_cnt = 0;
- refcount_inc(&skb->users);
__skb_queue_tail(&buf->skb_list, skb);
/* build dedicated header element */
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 08/15] s390/qeth: re-use qeth_notify_skbs()
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
When not using the CQ, this allows us avoid the second skb queue walk
in qeth_release_skbs().
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 29 +++++------------------------
1 file changed, 5 insertions(+), 24 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 012519ffd8de..1b853f900720 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -1180,38 +1180,19 @@ static void qeth_notify_skbs(struct qeth_qdio_out_q *q,
skb_queue_walk(&buf->skb_list, skb) {
QETH_CARD_TEXT_(q->card, 5, "skbn%d", notification);
QETH_CARD_TEXT_(q->card, 5, "%lx", (long) skb);
- if (be16_to_cpu(skb->protocol) == ETH_P_AF_IUCV) {
- if (skb->sk) {
- struct iucv_sock *iucv = iucv_sk(skb->sk);
- iucv->sk_txnotify(skb, notification);
- }
- }
+ if (skb->protocol == htons(ETH_P_AF_IUCV) && skb->sk)
+ iucv_sk(skb->sk)->sk_txnotify(skb, notification);
}
}
static void qeth_release_skbs(struct qeth_qdio_out_buffer *buf)
{
- struct sk_buff *skb;
- struct iucv_sock *iucv;
- int notify_general_error = 0;
-
- if (atomic_read(&buf->state) == QETH_QDIO_BUF_PENDING)
- notify_general_error = 1;
-
/* release may never happen from within CQ tasklet scope */
WARN_ON_ONCE(atomic_read(&buf->state) == QETH_QDIO_BUF_IN_CQ);
- skb_queue_walk(&buf->skb_list, skb) {
- QETH_CARD_TEXT(buf->q->card, 5, "skbr");
- QETH_CARD_TEXT_(buf->q->card, 5, "%lx", (long) skb);
- if (notify_general_error &&
- be16_to_cpu(skb->protocol) == ETH_P_AF_IUCV) {
- if (skb->sk) {
- iucv = iucv_sk(skb->sk);
- iucv->sk_txnotify(skb, TX_NOTIFY_GENERALERROR);
- }
- }
- }
+ if (atomic_read(&buf->state) == QETH_QDIO_BUF_PENDING)
+ qeth_notify_skbs(buf->q, buf, TX_NOTIFY_GENERALERROR);
+
__skb_queue_purge(&buf->skb_list);
}
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 09/15] s390/qeth: pass card pointer in iob callback
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
This allows us to remove the CARD_FROM_CDEV calls in the iob callbacks.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core.h | 3 ++-
drivers/s390/net/qeth_core_main.c | 53 ++++++++++++++++++++++-----------------
2 files changed, 32 insertions(+), 24 deletions(-)
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 630a01b3212c..64bcb1237db0 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -582,7 +582,8 @@ struct qeth_cmd_buffer {
struct qeth_channel *channel;
unsigned char *data;
int rc;
- void (*callback) (struct qeth_channel *, struct qeth_cmd_buffer *);
+ void (*callback)(struct qeth_card *card, struct qeth_channel *channel,
+ struct qeth_cmd_buffer *iob);
};
static inline struct qeth_ipa_cmd *__ipa_cmd(struct qeth_cmd_buffer *iob)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 1b853f900720..86b9cce1f483 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -63,8 +63,9 @@ static struct kmem_cache *qeth_qdio_outbuf_cache;
static struct device *qeth_core_root_dev;
static struct lock_class_key qdio_out_skb_queue_key;
-static void qeth_send_control_data_cb(struct qeth_channel *,
- struct qeth_cmd_buffer *);
+static void qeth_send_control_data_cb(struct qeth_card *card,
+ struct qeth_channel *channel,
+ struct qeth_cmd_buffer *iob);
static struct qeth_cmd_buffer *qeth_get_buffer(struct qeth_channel *);
static void qeth_free_buffer_pool(struct qeth_card *);
static int qeth_qdio_establish(struct qeth_card *);
@@ -787,6 +788,13 @@ void qeth_release_buffer(struct qeth_channel *channel,
}
EXPORT_SYMBOL_GPL(qeth_release_buffer);
+static void qeth_release_buffer_cb(struct qeth_card *card,
+ struct qeth_channel *channel,
+ struct qeth_cmd_buffer *iob)
+{
+ qeth_release_buffer(channel, iob);
+}
+
static struct qeth_cmd_buffer *qeth_get_buffer(struct qeth_channel *channel)
{
struct qeth_cmd_buffer *buffer = NULL;
@@ -817,17 +825,16 @@ void qeth_clear_cmd_buffers(struct qeth_channel *channel)
}
EXPORT_SYMBOL_GPL(qeth_clear_cmd_buffers);
-static void qeth_send_control_data_cb(struct qeth_channel *channel,
- struct qeth_cmd_buffer *iob)
+static void qeth_send_control_data_cb(struct qeth_card *card,
+ struct qeth_channel *channel,
+ struct qeth_cmd_buffer *iob)
{
- struct qeth_card *card;
struct qeth_reply *reply, *r;
struct qeth_ipa_cmd *cmd;
unsigned long flags;
int keep_reply;
int rc = 0;
- card = CARD_FROM_CDEV(channel->ccwdev);
QETH_CARD_TEXT(card, 4, "sndctlcb");
rc = qeth_check_idx_response(card, iob->data);
switch (rc) {
@@ -1164,7 +1171,7 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
__qeth_issue_next_read(card);
if (iob && iob->callback)
- iob->callback(iob->channel, iob);
+ iob->callback(card, iob->channel, iob);
out:
wake_up(&card->wait_q);
@@ -1804,8 +1811,9 @@ static void qeth_init_func_level(struct qeth_card *card)
}
static int qeth_idx_activate_get_answer(struct qeth_channel *channel,
- void (*idx_reply_cb)(struct qeth_channel *,
- struct qeth_cmd_buffer *))
+ void (*reply_cb)(struct qeth_card *,
+ struct qeth_channel *,
+ struct qeth_cmd_buffer *))
{
struct qeth_cmd_buffer *iob;
int rc;
@@ -1816,7 +1824,7 @@ static int qeth_idx_activate_get_answer(struct qeth_channel *channel,
iob = qeth_get_buffer(channel);
if (!iob)
return -ENOMEM;
- iob->callback = idx_reply_cb;
+ iob->callback = reply_cb;
qeth_setup_ccw(channel->ccw, CCW_CMD_READ, QETH_BUFSIZE, iob->data);
wait_event(card->wait_q,
@@ -1847,8 +1855,9 @@ static int qeth_idx_activate_get_answer(struct qeth_channel *channel,
}
static int qeth_idx_activate_channel(struct qeth_channel *channel,
- void (*idx_reply_cb)(struct qeth_channel *,
- struct qeth_cmd_buffer *))
+ void (*reply_cb)(struct qeth_card *,
+ struct qeth_channel *,
+ struct qeth_cmd_buffer *))
{
struct qeth_card *card;
struct qeth_cmd_buffer *iob;
@@ -1864,7 +1873,7 @@ static int qeth_idx_activate_channel(struct qeth_channel *channel,
iob = qeth_get_buffer(channel);
if (!iob)
return -ENOMEM;
- iob->callback = idx_reply_cb;
+ iob->callback = reply_cb;
qeth_setup_ccw(channel->ccw, CCW_CMD_WRITE, IDX_ACTIVATE_SIZE,
iob->data);
if (channel == &card->write) {
@@ -1916,7 +1925,7 @@ static int qeth_idx_activate_channel(struct qeth_channel *channel,
QETH_DBF_TEXT_(SETUP, 2, "2err%d", -ETIME);
return -ETIME;
}
- return qeth_idx_activate_get_answer(channel, idx_reply_cb);
+ return qeth_idx_activate_get_answer(channel, reply_cb);
}
static int qeth_peer_func_level(int level)
@@ -1928,10 +1937,10 @@ static int qeth_peer_func_level(int level)
return level;
}
-static void qeth_idx_write_cb(struct qeth_channel *channel,
- struct qeth_cmd_buffer *iob)
+static void qeth_idx_write_cb(struct qeth_card *card,
+ struct qeth_channel *channel,
+ struct qeth_cmd_buffer *iob)
{
- struct qeth_card *card;
__u16 temp;
QETH_DBF_TEXT(SETUP , 2, "idxwrcb");
@@ -1940,7 +1949,6 @@ static void qeth_idx_write_cb(struct qeth_channel *channel,
channel->state = CH_STATE_ACTIVATING;
goto out;
}
- card = CARD_FROM_CDEV(channel->ccwdev);
if (!(QETH_IS_IDX_ACT_POS_REPLY(iob->data))) {
if (QETH_IDX_ACT_CAUSE_CODE(iob->data) == QETH_IDX_ACT_ERR_EXCL)
@@ -1966,10 +1974,10 @@ static void qeth_idx_write_cb(struct qeth_channel *channel,
qeth_release_buffer(channel, iob);
}
-static void qeth_idx_read_cb(struct qeth_channel *channel,
- struct qeth_cmd_buffer *iob)
+static void qeth_idx_read_cb(struct qeth_card *card,
+ struct qeth_channel *channel,
+ struct qeth_cmd_buffer *iob)
{
- struct qeth_card *card;
__u16 temp;
QETH_DBF_TEXT(SETUP , 2, "idxrdcb");
@@ -1978,7 +1986,6 @@ static void qeth_idx_read_cb(struct qeth_channel *channel,
goto out;
}
- card = CARD_FROM_CDEV(channel->ccwdev);
if (qeth_check_idx_response(card, iob->data))
goto out;
@@ -2027,7 +2034,7 @@ void qeth_prepare_control_data(struct qeth_card *card, int len,
struct qeth_cmd_buffer *iob)
{
qeth_setup_ccw(iob->channel->ccw, CCW_CMD_WRITE, len, iob->data);
- iob->callback = qeth_release_buffer;
+ iob->callback = qeth_release_buffer_cb;
memcpy(QETH_TRANSPORT_HEADER_SEQ_NO(iob->data),
&card->seqno.trans_hdr, QETH_SEQ_NO_LENGTH);
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 12/15] s390/qeth: consume local address events
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
We have no code that is waiting for these events, so just drop them when
they arrive.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 5a2b90677235..880c15647442 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -688,10 +688,10 @@ static struct qeth_ipa_cmd *qeth_check_ipa_data(struct qeth_card *card,
return cmd;
case IPA_CMD_REGISTER_LOCAL_ADDR:
QETH_CARD_TEXT(card, 3, "irla");
- break;
+ return NULL;
case IPA_CMD_UNREGISTER_LOCAL_ADDR:
QETH_CARD_TEXT(card, 3, "urla");
- break;
+ return NULL;
default:
QETH_DBF_MESSAGE(2, "Received data is IPA "
"but not a reply!\n");
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 11/15] s390/qeth: remove various redundant code
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
1. tracing iob->rc makes no sense when it hasn't been modified by the
callback,
2. the qeth_dbf_list is declared with LIST_HEAD, which also initializes
the list,
3. the ccwgroup core only calls the thaw/restore callbacks if the gdev
is online, so we don't have to check for it again.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 7 -------
drivers/s390/net/qeth_l2_main.c | 5 +----
drivers/s390/net/qeth_l3_main.c | 5 +----
3 files changed, 2 insertions(+), 15 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index caa5d109841c..5a2b90677235 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -2167,7 +2167,6 @@ static int qeth_cm_enable_cb(struct qeth_card *card, struct qeth_reply *reply,
memcpy(&card->token.cm_filter_r,
QETH_CM_ENABLE_RESP_FILTER_TOKEN(iob->data),
QETH_MPC_TOKEN_LENGTH);
- QETH_DBF_TEXT_(SETUP, 2, " rc%d", iob->rc);
return 0;
}
@@ -2193,7 +2192,6 @@ static int qeth_cm_enable(struct qeth_card *card)
static int qeth_cm_setup_cb(struct qeth_card *card, struct qeth_reply *reply,
unsigned long data)
{
-
struct qeth_cmd_buffer *iob;
QETH_DBF_TEXT(SETUP, 2, "cmsetpcb");
@@ -2202,7 +2200,6 @@ static int qeth_cm_setup_cb(struct qeth_card *card, struct qeth_reply *reply,
memcpy(&card->token.cm_connection_r,
QETH_CM_SETUP_RESP_DEST_ADDR(iob->data),
QETH_MPC_TOKEN_LENGTH);
- QETH_DBF_TEXT_(SETUP, 2, " rc%d", iob->rc);
return 0;
}
@@ -2224,7 +2221,6 @@ static int qeth_cm_setup(struct qeth_card *card)
rc = qeth_send_control_data(card, CM_SETUP_SIZE, iob,
qeth_cm_setup_cb, NULL);
return rc;
-
}
static int qeth_update_max_mtu(struct qeth_card *card, unsigned int max_mtu)
@@ -2284,7 +2280,6 @@ static int qeth_get_mtu_outof_framesize(int framesize)
static int qeth_ulp_enable_cb(struct qeth_card *card, struct qeth_reply *reply,
unsigned long data)
{
-
__u16 mtu, framesize;
__u16 len;
__u8 link_type;
@@ -2312,7 +2307,6 @@ static int qeth_ulp_enable_cb(struct qeth_card *card, struct qeth_reply *reply,
} else
card->info.link_type = 0;
QETH_DBF_TEXT_(SETUP, 2, "link%d", card->info.link_type);
- QETH_DBF_TEXT_(SETUP, 2, " rc%d", iob->rc);
return 0;
}
@@ -6584,7 +6578,6 @@ static int __init qeth_core_init(void)
pr_info("loading core functions\n");
INIT_LIST_HEAD(&qeth_core_card_list.list);
- INIT_LIST_HEAD(&qeth_dbf_list);
rwlock_init(&qeth_core_card_list.rwlock);
qeth_wq = create_singlethread_workqueue("qeth_wq");
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 02566af7e23d..ed475b4ddd3e 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -1146,9 +1146,6 @@ static int qeth_l2_pm_resume(struct ccwgroup_device *gdev)
struct qeth_card *card = dev_get_drvdata(&gdev->dev);
int rc = 0;
- if (gdev->state == CCWGROUP_OFFLINE)
- goto out;
-
if (card->state == CARD_STATE_RECOVER) {
rc = __qeth_l2_set_online(card->gdev, 1);
if (rc) {
@@ -1158,7 +1155,7 @@ static int qeth_l2_pm_resume(struct ccwgroup_device *gdev)
}
} else
rc = __qeth_l2_set_online(card->gdev, 0);
-out:
+
qeth_set_allowed_threads(card, 0xffffffff, 0);
netif_device_attach(card->dev);
if (rc)
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 7148ef71ac78..86c89338b1a2 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2714,9 +2714,6 @@ static int qeth_l3_pm_resume(struct ccwgroup_device *gdev)
struct qeth_card *card = dev_get_drvdata(&gdev->dev);
int rc = 0;
- if (gdev->state == CCWGROUP_OFFLINE)
- goto out;
-
if (card->state == CARD_STATE_RECOVER) {
rc = __qeth_l3_set_online(card->gdev, 1);
if (rc) {
@@ -2726,7 +2723,7 @@ static int qeth_l3_pm_resume(struct ccwgroup_device *gdev)
}
} else
rc = __qeth_l3_set_online(card->gdev, 0);
-out:
+
qeth_set_allowed_threads(card, 0xffffffff, 0);
netif_device_attach(card->dev);
if (rc)
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 13/15] s390/qeth: re-indent qeth_check_ipa_data()
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
Pull one level of checking up into qeth_send_control_data_cb(), and
clean up an else-after-return. No functional change.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 129 +++++++++++++++++---------------------
1 file changed, 58 insertions(+), 71 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 880c15647442..954dc6c688e8 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -626,80 +626,64 @@ static void qeth_issue_ipa_msg(struct qeth_ipa_cmd *cmd, int rc,
}
static struct qeth_ipa_cmd *qeth_check_ipa_data(struct qeth_card *card,
- struct qeth_cmd_buffer *iob)
+ struct qeth_ipa_cmd *cmd)
{
- struct qeth_ipa_cmd *cmd = NULL;
-
QETH_CARD_TEXT(card, 5, "chkipad");
- if (IS_IPA(iob->data)) {
- cmd = (struct qeth_ipa_cmd *) PDU_ENCAPSULATION(iob->data);
- if (IS_IPA_REPLY(cmd)) {
- if (cmd->hdr.command != IPA_CMD_SETCCID &&
- cmd->hdr.command != IPA_CMD_DELCCID &&
- cmd->hdr.command != IPA_CMD_MODCCID &&
- cmd->hdr.command != IPA_CMD_SET_DIAG_ASS)
- qeth_issue_ipa_msg(cmd,
- cmd->hdr.return_code, card);
- return cmd;
+
+ if (IS_IPA_REPLY(cmd)) {
+ if (cmd->hdr.command != IPA_CMD_SETCCID &&
+ cmd->hdr.command != IPA_CMD_DELCCID &&
+ cmd->hdr.command != IPA_CMD_MODCCID &&
+ cmd->hdr.command != IPA_CMD_SET_DIAG_ASS)
+ qeth_issue_ipa_msg(cmd, cmd->hdr.return_code, card);
+ return cmd;
+ }
+
+ /* handle unsolicited event: */
+ switch (cmd->hdr.command) {
+ case IPA_CMD_STOPLAN:
+ if (cmd->hdr.return_code == IPA_RC_VEPA_TO_VEB_TRANSITION) {
+ dev_err(&card->gdev->dev,
+ "Interface %s is down because the adjacent port is no longer in reflective relay mode\n",
+ QETH_CARD_IFNAME(card));
+ qeth_close_dev(card);
} else {
- switch (cmd->hdr.command) {
- case IPA_CMD_STOPLAN:
- if (cmd->hdr.return_code ==
- IPA_RC_VEPA_TO_VEB_TRANSITION) {
- dev_err(&card->gdev->dev,
- "Interface %s is down because the "
- "adjacent port is no longer in "
- "reflective relay mode\n",
- QETH_CARD_IFNAME(card));
- qeth_close_dev(card);
- } else {
- dev_warn(&card->gdev->dev,
- "The link for interface %s on CHPID"
- " 0x%X failed\n",
- QETH_CARD_IFNAME(card),
- card->info.chpid);
- qeth_issue_ipa_msg(cmd,
- cmd->hdr.return_code, card);
- }
- card->lan_online = 0;
- netif_carrier_off(card->dev);
- return NULL;
- case IPA_CMD_STARTLAN:
- dev_info(&card->gdev->dev,
- "The link for %s on CHPID 0x%X has"
- " been restored\n",
- QETH_CARD_IFNAME(card),
- card->info.chpid);
- netif_carrier_on(card->dev);
- card->lan_online = 1;
- if (card->info.hwtrap)
- card->info.hwtrap = 2;
- qeth_schedule_recovery(card);
- return NULL;
- case IPA_CMD_SETBRIDGEPORT_IQD:
- case IPA_CMD_SETBRIDGEPORT_OSA:
- case IPA_CMD_ADDRESS_CHANGE_NOTIF:
- if (card->discipline->control_event_handler
- (card, cmd))
- return cmd;
- else
- return NULL;
- case IPA_CMD_MODCCID:
- return cmd;
- case IPA_CMD_REGISTER_LOCAL_ADDR:
- QETH_CARD_TEXT(card, 3, "irla");
- return NULL;
- case IPA_CMD_UNREGISTER_LOCAL_ADDR:
- QETH_CARD_TEXT(card, 3, "urla");
- return NULL;
- default:
- QETH_DBF_MESSAGE(2, "Received data is IPA "
- "but not a reply!\n");
- break;
- }
+ dev_warn(&card->gdev->dev,
+ "The link for interface %s on CHPID 0x%X failed\n",
+ QETH_CARD_IFNAME(card), card->info.chpid);
+ qeth_issue_ipa_msg(cmd, cmd->hdr.return_code, card);
}
+ card->lan_online = 0;
+ netif_carrier_off(card->dev);
+ return NULL;
+ case IPA_CMD_STARTLAN:
+ dev_info(&card->gdev->dev,
+ "The link for %s on CHPID 0x%X has been restored\n",
+ QETH_CARD_IFNAME(card), card->info.chpid);
+ netif_carrier_on(card->dev);
+ card->lan_online = 1;
+ if (card->info.hwtrap)
+ card->info.hwtrap = 2;
+ qeth_schedule_recovery(card);
+ return NULL;
+ case IPA_CMD_SETBRIDGEPORT_IQD:
+ case IPA_CMD_SETBRIDGEPORT_OSA:
+ case IPA_CMD_ADDRESS_CHANGE_NOTIF:
+ if (card->discipline->control_event_handler(card, cmd))
+ return cmd;
+ return NULL;
+ case IPA_CMD_MODCCID:
+ return cmd;
+ case IPA_CMD_REGISTER_LOCAL_ADDR:
+ QETH_CARD_TEXT(card, 3, "irla");
+ return NULL;
+ case IPA_CMD_UNREGISTER_LOCAL_ADDR:
+ QETH_CARD_TEXT(card, 3, "urla");
+ return NULL;
+ default:
+ QETH_DBF_MESSAGE(2, "Received data is IPA but not a reply!\n");
+ return cmd;
}
- return cmd;
}
void qeth_clear_ipacmd_list(struct qeth_card *card)
@@ -820,8 +804,8 @@ static void qeth_send_control_data_cb(struct qeth_card *card,
struct qeth_channel *channel,
struct qeth_cmd_buffer *iob)
{
+ struct qeth_ipa_cmd *cmd = NULL;
struct qeth_reply *reply, *r;
- struct qeth_ipa_cmd *cmd;
unsigned long flags;
int keep_reply;
int rc = 0;
@@ -839,7 +823,10 @@ static void qeth_send_control_data_cb(struct qeth_card *card,
goto out;
}
- cmd = qeth_check_ipa_data(card, iob);
+ if (IS_IPA(iob->data)) {
+ cmd = (struct qeth_ipa_cmd *) PDU_ENCAPSULATION(iob->data);
+ cmd = qeth_check_ipa_data(card, cmd);
+ }
if ((cmd == NULL) && (card->state != CARD_STATE_DOWN))
goto out;
/*in case of OSN : check if cmd is set */
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 15/15] s390/qeth: remove duplicated carrier state tracking
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
The netdevice is always available, apply any carrier state changes to it
without caching them.
On a STARTLAN event (ie. carrier-up), defer updating the state to
qeth_core_hardsetup_card() in the subsequent recovery action.
Also remove the carrier-state checks from the xmit routines. Stopping
transmission on carrier-down is the responsibility of upper-level code
(eg see dev_direct_xmit()).
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core.h | 1 -
drivers/s390/net/qeth_core_main.c | 12 +++++-------
drivers/s390/net/qeth_core_sys.c | 7 +++----
drivers/s390/net/qeth_l2_main.c | 6 +-----
drivers/s390/net/qeth_l3_main.c | 6 +-----
5 files changed, 10 insertions(+), 22 deletions(-)
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 64bcb1237db0..cd44ff2df6fe 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -763,7 +763,6 @@ struct qeth_switch_info {
struct qeth_card {
struct list_head list;
enum qeth_card_states state;
- int lan_online;
spinlock_t lock;
struct ccwgroup_device *gdev;
struct qeth_channel read;
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 9cbdc6760aba..0078b5d217cc 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -652,16 +652,13 @@ static struct qeth_ipa_cmd *qeth_check_ipa_data(struct qeth_card *card,
"The link for interface %s on CHPID 0x%X failed\n",
QETH_CARD_IFNAME(card), card->info.chpid);
qeth_issue_ipa_msg(cmd, cmd->hdr.return_code, card);
+ netif_carrier_off(card->dev);
}
- card->lan_online = 0;
- netif_carrier_off(card->dev);
return NULL;
case IPA_CMD_STARTLAN:
dev_info(&card->gdev->dev,
"The link for %s on CHPID 0x%X has been restored\n",
QETH_CARD_IFNAME(card), card->info.chpid);
- netif_carrier_on(card->dev);
- card->lan_online = 1;
if (card->info.hwtrap)
card->info.hwtrap = 2;
qeth_schedule_recovery(card);
@@ -5133,13 +5130,14 @@ int qeth_core_hardsetup_card(struct qeth_card *card)
if (rc == IPA_RC_LAN_OFFLINE) {
dev_warn(&card->gdev->dev,
"The LAN is offline\n");
- card->lan_online = 0;
+ netif_carrier_off(card->dev);
} else {
rc = -ENODEV;
goto out;
}
- } else
- card->lan_online = 1;
+ } else {
+ netif_carrier_on(card->dev);
+ }
card->options.ipa4.supported_funcs = 0;
card->options.ipa6.supported_funcs = 0;
diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c
index 970f6c71a66e..30f61608fa22 100644
--- a/drivers/s390/net/qeth_core_sys.c
+++ b/drivers/s390/net/qeth_core_sys.c
@@ -31,10 +31,9 @@ static ssize_t qeth_dev_state_show(struct device *dev,
case CARD_STATE_SOFTSETUP:
return sprintf(buf, "SOFTSETUP\n");
case CARD_STATE_UP:
- if (card->lan_online)
- return sprintf(buf, "UP (LAN ONLINE)\n");
- else
- return sprintf(buf, "UP (LAN OFFLINE)\n");
+ return sprintf(buf, "UP (LAN %s)\n",
+ netif_carrier_ok(card->dev) ? "ONLINE" :
+ "OFFLINE");
case CARD_STATE_RECOVER:
return sprintf(buf, "RECOVER\n");
default:
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index ed475b4ddd3e..c810d53fff51 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -694,7 +694,7 @@ static netdev_tx_t qeth_l2_hard_start_xmit(struct sk_buff *skb,
int tx_bytes = skb->len;
int rc;
- if ((card->state != CARD_STATE_UP) || !card->lan_online) {
+ if (card->state != CARD_STATE_UP) {
card->stats.tx_carrier_errors++;
goto tx_drop;
}
@@ -997,10 +997,6 @@ static int __qeth_l2_set_online(struct ccwgroup_device *gdev, int recovery_mode)
goto out_remove;
}
card->state = CARD_STATE_SOFTSETUP;
- if (card->lan_online)
- netif_carrier_on(card->dev);
- else
- netif_carrier_off(card->dev);
qeth_set_allowed_threads(card, 0xffffffff, 0);
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 86c89338b1a2..80893481bb85 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2233,7 +2233,7 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
goto tx_drop;
}
- if (card->state != CARD_STATE_UP || !card->lan_online) {
+ if (card->state != CARD_STATE_UP) {
card->stats.tx_carrier_errors++;
goto tx_drop;
}
@@ -2573,10 +2573,6 @@ static int __qeth_l3_set_online(struct ccwgroup_device *gdev, int recovery_mode)
qeth_set_allowed_threads(card, 0xffffffff, 0);
qeth_l3_recover_ip(card);
- if (card->lan_online)
- netif_carrier_on(card->dev);
- else
- netif_carrier_off(card->dev);
qeth_enable_hw_features(card->dev);
if (recover_flag == CARD_STATE_RECOVER) {
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 10/15] s390/qeth: remove CARD_FROM_CDEV helper
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
The cdev-to-card translation walks through two layers of drvdata,
with no locking or refcounting (where eg. the ccwgroup core only
accesses a cdev's drvdata while holding the ccwlock).
This might be safe for now, but any careless usage of the helper has the
potential for subtle races and use-after-free's. Luckily there's only
one occurrence where we _really_ need it (in qeth_irq()), for any other
user we can just pass through an appropriate card pointer.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 73 ++++++++++++++++-----------------------
1 file changed, 29 insertions(+), 44 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 86b9cce1f483..caa5d109841c 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -746,18 +746,10 @@ static int qeth_check_idx_response(struct qeth_card *card,
return 0;
}
-static struct qeth_card *CARD_FROM_CDEV(struct ccw_device *cdev)
-{
- struct qeth_card *card = dev_get_drvdata(&((struct ccwgroup_device *)
- dev_get_drvdata(&cdev->dev))->dev);
- return card;
-}
-
static struct qeth_cmd_buffer *__qeth_get_buffer(struct qeth_channel *channel)
{
__u8 index;
- QETH_CARD_TEXT(CARD_FROM_CDEV(channel->ccwdev), 6, "getbuff");
index = channel->io_buf_no;
do {
if (channel->iob[index].state == BUF_STATE_FREE) {
@@ -778,7 +770,6 @@ void qeth_release_buffer(struct qeth_channel *channel,
{
unsigned long flags;
- QETH_CARD_TEXT(CARD_FROM_CDEV(channel->ccwdev), 6, "relbuff");
spin_lock_irqsave(&channel->iob_lock, flags);
iob->state = BUF_STATE_FREE;
iob->callback = qeth_send_control_data_cb;
@@ -980,16 +971,15 @@ void qeth_schedule_recovery(struct qeth_card *card)
}
EXPORT_SYMBOL_GPL(qeth_schedule_recovery);
-static int qeth_get_problem(struct ccw_device *cdev, struct irb *irb)
+static int qeth_get_problem(struct qeth_card *card, struct ccw_device *cdev,
+ struct irb *irb)
{
int dstat, cstat;
char *sense;
- struct qeth_card *card;
sense = (char *) irb->ecw;
cstat = irb->scsw.cmd.cstat;
dstat = irb->scsw.cmd.dstat;
- card = CARD_FROM_CDEV(cdev);
if (cstat & (SCHN_STAT_CHN_CTRL_CHK | SCHN_STAT_INTF_CTRL_CHK |
SCHN_STAT_CHN_DATA_CHK | SCHN_STAT_CHAIN_CHECK |
@@ -1029,14 +1019,11 @@ static int qeth_get_problem(struct ccw_device *cdev, struct irb *irb)
return 0;
}
-static long __qeth_check_irb_error(struct ccw_device *cdev,
- unsigned long intparm, struct irb *irb)
+static long qeth_check_irb_error(struct qeth_card *card,
+ struct ccw_device *cdev, unsigned long intparm,
+ struct irb *irb)
{
- struct qeth_card *card;
-
- card = CARD_FROM_CDEV(cdev);
-
- if (!card || !IS_ERR(irb))
+ if (!IS_ERR(irb))
return 0;
switch (PTR_ERR(irb)) {
@@ -1073,10 +1060,13 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
int rc;
int cstat, dstat;
struct qeth_cmd_buffer *iob = NULL;
+ struct ccwgroup_device *gdev;
struct qeth_channel *channel;
struct qeth_card *card;
- card = CARD_FROM_CDEV(cdev);
+ /* while we hold the ccwdev lock, this stays valid: */
+ gdev = dev_get_drvdata(&cdev->dev);
+ card = dev_get_drvdata(&gdev->dev);
if (!card)
return;
@@ -1096,7 +1086,7 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
if (qeth_intparm_is_iob(intparm))
iob = (struct qeth_cmd_buffer *) __va((addr_t)intparm);
- if (__qeth_check_irb_error(cdev, intparm, irb)) {
+ if (qeth_check_irb_error(card, cdev, intparm, irb)) {
/* IO was terminated, free its resources. */
if (iob)
qeth_release_buffer(iob->channel, iob);
@@ -1151,7 +1141,7 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
channel->state = CH_STATE_DOWN;
goto out;
}
- rc = qeth_get_problem(cdev, irb);
+ rc = qeth_get_problem(card, cdev, irb);
if (rc) {
card->read_or_write_problem = 1;
qeth_clear_ipacmd_list(card);
@@ -1514,12 +1504,11 @@ static struct qeth_card *qeth_alloc_card(struct ccwgroup_device *gdev)
return NULL;
}
-static int qeth_clear_channel(struct qeth_channel *channel)
+static int qeth_clear_channel(struct qeth_card *card,
+ struct qeth_channel *channel)
{
- struct qeth_card *card;
int rc;
- card = CARD_FROM_CDEV(channel->ccwdev);
QETH_CARD_TEXT(card, 3, "clearch");
spin_lock_irq(get_ccwdev_lock(channel->ccwdev));
rc = ccw_device_clear(channel->ccwdev, QETH_CLEAR_CHANNEL_PARM);
@@ -1537,12 +1526,11 @@ static int qeth_clear_channel(struct qeth_channel *channel)
return 0;
}
-static int qeth_halt_channel(struct qeth_channel *channel)
+static int qeth_halt_channel(struct qeth_card *card,
+ struct qeth_channel *channel)
{
- struct qeth_card *card;
int rc;
- card = CARD_FROM_CDEV(channel->ccwdev);
QETH_CARD_TEXT(card, 3, "haltch");
spin_lock_irq(get_ccwdev_lock(channel->ccwdev));
rc = ccw_device_halt(channel->ccwdev, QETH_HALT_CHANNEL_PARM);
@@ -1564,9 +1552,9 @@ static int qeth_halt_channels(struct qeth_card *card)
int rc1 = 0, rc2 = 0, rc3 = 0;
QETH_CARD_TEXT(card, 3, "haltchs");
- rc1 = qeth_halt_channel(&card->read);
- rc2 = qeth_halt_channel(&card->write);
- rc3 = qeth_halt_channel(&card->data);
+ rc1 = qeth_halt_channel(card, &card->read);
+ rc2 = qeth_halt_channel(card, &card->write);
+ rc3 = qeth_halt_channel(card, &card->data);
if (rc1)
return rc1;
if (rc2)
@@ -1579,9 +1567,9 @@ static int qeth_clear_channels(struct qeth_card *card)
int rc1 = 0, rc2 = 0, rc3 = 0;
QETH_CARD_TEXT(card, 3, "clearchs");
- rc1 = qeth_clear_channel(&card->read);
- rc2 = qeth_clear_channel(&card->write);
- rc3 = qeth_clear_channel(&card->data);
+ rc1 = qeth_clear_channel(card, &card->read);
+ rc2 = qeth_clear_channel(card, &card->write);
+ rc3 = qeth_clear_channel(card, &card->data);
if (rc1)
return rc1;
if (rc2)
@@ -1810,17 +1798,16 @@ static void qeth_init_func_level(struct qeth_card *card)
}
}
-static int qeth_idx_activate_get_answer(struct qeth_channel *channel,
+static int qeth_idx_activate_get_answer(struct qeth_card *card,
+ struct qeth_channel *channel,
void (*reply_cb)(struct qeth_card *,
struct qeth_channel *,
struct qeth_cmd_buffer *))
{
struct qeth_cmd_buffer *iob;
int rc;
- struct qeth_card *card;
QETH_DBF_TEXT(SETUP, 2, "idxanswr");
- card = CARD_FROM_CDEV(channel->ccwdev);
iob = qeth_get_buffer(channel);
if (!iob)
return -ENOMEM;
@@ -1854,20 +1841,18 @@ static int qeth_idx_activate_get_answer(struct qeth_channel *channel,
return rc;
}
-static int qeth_idx_activate_channel(struct qeth_channel *channel,
+static int qeth_idx_activate_channel(struct qeth_card *card,
+ struct qeth_channel *channel,
void (*reply_cb)(struct qeth_card *,
struct qeth_channel *,
struct qeth_cmd_buffer *))
{
- struct qeth_card *card;
struct qeth_cmd_buffer *iob;
__u16 temp;
__u8 tmp;
int rc;
struct ccw_dev_id temp_devid;
- card = CARD_FROM_CDEV(channel->ccwdev);
-
QETH_DBF_TEXT(SETUP, 2, "idxactch");
iob = qeth_get_buffer(channel);
@@ -1925,7 +1910,7 @@ static int qeth_idx_activate_channel(struct qeth_channel *channel,
QETH_DBF_TEXT_(SETUP, 2, "2err%d", -ETIME);
return -ETIME;
}
- return qeth_idx_activate_get_answer(channel, reply_cb);
+ return qeth_idx_activate_get_answer(card, channel, reply_cb);
}
static int qeth_peer_func_level(int level)
@@ -5131,7 +5116,7 @@ int qeth_core_hardsetup_card(struct qeth_card *card)
qeth_determine_capabilities(card);
qeth_init_tokens(card);
qeth_init_func_level(card);
- rc = qeth_idx_activate_channel(&card->read, qeth_idx_read_cb);
+ rc = qeth_idx_activate_channel(card, &card->read, qeth_idx_read_cb);
if (rc == -ERESTARTSYS) {
QETH_DBF_TEXT(SETUP, 2, "break2");
return rc;
@@ -5142,7 +5127,7 @@ int qeth_core_hardsetup_card(struct qeth_card *card)
else
goto retry;
}
- rc = qeth_idx_activate_channel(&card->write, qeth_idx_write_cb);
+ rc = qeth_idx_activate_channel(card, &card->write, qeth_idx_write_cb);
if (rc == -ERESTARTSYS) {
QETH_DBF_TEXT(SETUP, 2, "break3");
return rc;
--
2.16.4
^ permalink raw reply related
* [PATCH net-next 14/15] s390/qeth: clean up drop conditions for received cmds
From: Julian Wiedmann @ 2018-09-26 16:29 UTC (permalink / raw)
To: David Miller
Cc: netdev, linux-s390, Martin Schwidefsky, Heiko Carstens,
Stefan Raspl, Ursula Braun, Julian Wiedmann
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
If qeth_check_ipa_data() consumed an event, there's no point in
processing it further. So drop it early, and make the surrounding code
a tiny bit more readable.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
drivers/s390/net/qeth_core_main.c | 21 +++++++++++----------
1 file changed, 11 insertions(+), 10 deletions(-)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 954dc6c688e8..9cbdc6760aba 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -826,16 +826,17 @@ static void qeth_send_control_data_cb(struct qeth_card *card,
if (IS_IPA(iob->data)) {
cmd = (struct qeth_ipa_cmd *) PDU_ENCAPSULATION(iob->data);
cmd = qeth_check_ipa_data(card, cmd);
- }
- if ((cmd == NULL) && (card->state != CARD_STATE_DOWN))
- goto out;
- /*in case of OSN : check if cmd is set */
- if (card->info.type == QETH_CARD_TYPE_OSN &&
- cmd &&
- cmd->hdr.command != IPA_CMD_STARTLAN &&
- card->osn_info.assist_cb != NULL) {
- card->osn_info.assist_cb(card->dev, cmd);
- goto out;
+ if (!cmd)
+ goto out;
+ if (IS_OSN(card) && card->osn_info.assist_cb &&
+ cmd->hdr.command != IPA_CMD_STARTLAN) {
+ card->osn_info.assist_cb(card->dev, cmd);
+ goto out;
+ }
+ } else {
+ /* non-IPA commands should only flow during initialization */
+ if (card->state != CARD_STATE_DOWN)
+ goto out;
}
spin_lock_irqsave(&card->lock, flags);
--
2.16.4
^ permalink raw reply related
* Re: [PATCH net-next v6 07/23] zinc: ChaCha20 ARM and ARM64 implementations
From: Ard Biesheuvel @ 2018-09-26 16:54 UTC (permalink / raw)
To: Jason A. Donenfeld
Cc: Herbert Xu, Thomas Gleixner, Linux Kernel Mailing List,
<netdev@vger.kernel.org>,
open list:HARDWARE RANDOM NUMBER GENERATOR CORE, David S. Miller,
Greg Kroah-Hartman, Samuel Neves, Andy Lutomirski,
Jean-Philippe Aumasson, Russell King, linux-arm-kernel
In-Reply-To: <CAHmME9p5b=L0FSL72gCszhvut-kr=aD4ZniY9qsJxiBnZk8qNQ@mail.gmail.com>
On Wed, 26 Sep 2018 at 17:41, Jason A. Donenfeld <Jason@zx2c4.com> wrote:
>
> On Wed, Sep 26, 2018 at 4:02 PM Ard Biesheuvel
> <ard.biesheuvel@linaro.org> wrote:
> > I don't think it makes sense to keep
> > it simple now and add the complexity later (and the same concern
> > applies to async support btw).
>
> Ugh, no. I don't want to add needless complexity, period. Zinc is
> synchronous, not asynchronous. It provides software implementations.
> That's what it does. While many of your reviews have been useful, many
> of your comments indicate some desire to change and mold the purpose
> and focus of Zinc away from Zinc's intents. Stop that. It's not going
> to become a bloated mess of "things Ard wanted and quipped about on
> LKML." Things like these only serve to filibuster the patchset
> indefinitely. But maybe that's what you'd like all along? Hard to
> tell, honestly. So, no, sorry, Zinc isn't gaining an async interface
> right now.
Framing it as /needless/ complexity does not help at all. The changes
you are proposing are very useful, but nobody wants two crypto
subsystems with two different maintainers in the kernel, so I would
like to understand where this is going in the future. I am not saying
it should block these patches though.
Also, I have spent a *lot* of time looking at your code, and trying to
make it better, especially for use cases that weren't on your radar to
begin with (e.g., 'pet projects' [your words] like the Cortex-A7 which
will be in almost every new 32-bit Android phone). So characterizing
my feedback as some kind of sabotage is not very productive either.
Contrary to what you seem to think, I am not deeply invested in the
crypto API. What I do care about is that the ARM crypto pieces in the
kernel are maintained, supported and improved by someone who
understands the use cases Linaro's members care about, and is willing
to make an effort to gain such understanding if he doesn't. I have no
doubt that your involvement in the kernel's crypto subsystem will have
a significant positive impact when it comes to code quality,
robustness and usability. I'd just like to see a bit more
consideration for other aspects of kernel programming, e.g.,
preemption under -rt, stack size constraints, coding style, importing
code from other projects etc. - please try to be less dismissive of
feedback first time around, but try to understand why people are
raising these issues; I'm sure you will appreciate it when future
contributors to zinc will do the same.
^ permalink raw reply
* Re: [PATCH net-next 00/15] s390/net: updates 2018-09-26
From: David Miller @ 2018-09-26 17:02 UTC (permalink / raw)
To: jwi; +Cc: netdev, linux-s390, schwidefsky, heiko.carstens, raspl, ubraun
In-Reply-To: <20180926162916.102720-1-jwi@linux.ibm.com>
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 26 Sep 2018 18:29:01 +0200
> please apply one more series of cleanups and small improvements for qeth
> to net-next. Note that one patch needs to touch both af_iucv and qeth, in
> order to untangle their receive paths.
Series applied, thank you.
^ permalink raw reply
* Re: [PATCH net-next v6 07/23] zinc: ChaCha20 ARM and ARM64 implementations
From: Jason A. Donenfeld @ 2018-09-26 17:03 UTC (permalink / raw)
To: Andy Lutomirski
Cc: Ard Biesheuvel, Herbert Xu, Thomas Gleixner, LKML, Netdev,
Linux Crypto Mailing List, David Miller, Greg Kroah-Hartman,
Samuel Neves, Andrew Lutomirski, Jean-Philippe Aumasson,
Russell King - ARM Linux, linux-arm-kernel
In-Reply-To: <D93C2C15-DC70-4A95-9350-AD4036C9556A@amacapital.net>
On Wed, Sep 26, 2018 at 6:21 PM Andy Lutomirski <luto@amacapital.net> wrote:
> Are, is what you’re saying that the Zinc chacha20 functions should call simd_relax() every n bytes automatically for some reasonable value of n? If so, seems sensible, except that some care might be needed to make sure they interact with preemption correctly.
>
> What I mean is: the public Zinc entry points should either be callable in an atomic context or they should not be. I think this should be checked at runtime in an appropriate place with an __might_sleep or similar. Or simd_relax should learn to *not* schedule if the result of preempt_enable() leaves it atomic. (And the latter needs to be done in a way that works even on non-preempt kernels, and I don’t remember whether that’s possible.). And this should happen regardless of how many bytes are processed. IOW, calling into Zinc should be equally not atomic-safe for 100 bytes and for 10 MB.
I'm not sure this is actually a problem. Namely:
preempt_disable();
kernel_fpu_begin();
kernel_fpu_end();
schedule(); <--- bug!
Calling kernel_fpu_end() disables preemption, but AFAIK, preemption
enabling/disabling is recursive, so kernel_fpu_end's use of
preempt_disable won't actually do anything until the outer preempt
enable is called:
preempt_disable();
kernel_fpu_begin();
kernel_fpu_end();
preempt_enable();
schedule(); <--- works!
Or am I missing some more subtle point?
Jason
^ permalink raw reply
* Re: [PATCH net-next v2 0/9] net: bridge: convert bool options to bits
From: David Miller @ 2018-09-26 17:04 UTC (permalink / raw)
To: nikolay; +Cc: netdev, roopa, stephen, bridge, idosch
In-Reply-To: <2ab2aeb3-7667-816c-a1cd-97dbe723db33@cumulusnetworks.com>
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Wed, 26 Sep 2018 17:04:03 +0300
> On 26/09/18 17:00, Nikolay Aleksandrov wrote:
>> Hi,
>> A lot of boolean bridge options have been added around the net_bridge
>> structure resulting in holes and more importantly different cache lines
>> that need to be fetched in the fast path. This set moves all of those
>> to bits in a bitfield which resides in a hot cache line thus reducing
>> the size of net_bridge, the number of holes and the number of cache
>> lines needed for the fast path.
>> The set is also sent in preparation for new boolean options to avoid
>> spreading them in the structure and making new holes.
>> One nice side-effect is that we avoid potential race conditions by using
>> the bitops since some of the options were bits being directly set in
>> parallel risking hard to debug issues (has_ipv6_addr).
>>
>> Before:
>> size: 1184, holes: 8, sum holes: 30
>> After:
>> size: 1160, holes: 3, sum holes: 7
>>
>> Patch 01 is a trivial style fix
>> Patch 02 adds the new options bitfield and converts the vlan boolean
>> options to bits
>> Patches 03-08 convert the rest of the boolean options to bits
>> Patch 09 re-arranges a few fields in net_bridge to further reduce size
>>
>> v2: patch 09: remove the comment about offload_fwd_mark in net_bridge and
>> leave it in the last 4 bytes, thanks to Ido for spotting it
>
> And obviously it's not in the last 4 bytes, but that's fine. That
> must've said just: "leave it where it is now".
I fixed up the wording and applied this series, looks great!
Thanks!
^ permalink raw reply
* RE: [PATCH] hv_netvsc: Make sure out channel is fully opened on send
From: Haiyang Zhang @ 2018-09-26 17:13 UTC (permalink / raw)
To: Mohammed Gamal, Stephen Hemminger, netdev@vger.kernel.org
Cc: KY Srinivasan, vkuznets, otubo@redhat.com, cavery,
linux-kernel@vger.kernel.org, devel@linuxdriverproject.org
In-Reply-To: <1537979659-26979-1-git-send-email-mgamal@redhat.com>
> -----Original Message-----
> From: Mohammed Gamal <mgamal@redhat.com>
> Sent: Wednesday, September 26, 2018 12:34 PM
> To: Stephen Hemminger <sthemmin@microsoft.com>; netdev@vger.kernel.org
> Cc: KY Srinivasan <kys@microsoft.com>; Haiyang Zhang
> <haiyangz@microsoft.com>; vkuznets <vkuznets@redhat.com>;
> otubo@redhat.com; cavery <cavery@redhat.com>; linux-
> kernel@vger.kernel.org; devel@linuxdriverproject.org; Mohammed Gamal
> <mgamal@redhat.com>
> Subject: [PATCH] hv_netvsc: Make sure out channel is fully opened on send
>
> Dring high network traffic changes to network interface parameters such as
> number of channels or MTU can cause a kernel panic with a NULL pointer
> dereference. This is due to netvsc_device_remove() being called and
> deallocating the channel ring buffers, which can then be accessed by
> netvsc_send_pkt() before they're allocated on calling
> netvsc_device_add()
>
> The patch fixes this problem by checking the channel state and returning
> ENODEV if not yet opened. We also move the call to hv_ringbuf_avail_percent()
> which may access the uninitialized ring buffer.
>
> Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
> ---
> drivers/net/hyperv/netvsc.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index
> fe01e14..75f1b31 100644
> --- a/drivers/net/hyperv/netvsc.c
> +++ b/drivers/net/hyperv/netvsc.c
> @@ -825,7 +825,12 @@ static inline int netvsc_send_pkt(
> struct netdev_queue *txq = netdev_get_tx_queue(ndev, packet->q_idx);
> u64 req_id;
> int ret;
> - u32 ring_avail = hv_get_avail_to_write_percent(&out_channel-
> >outbound);
> + u32 ring_avail;
> +
> + if (out_channel->state != CHANNEL_OPENED_STATE)
> + return -ENODEV;
> +
> + ring_avail = hv_get_avail_to_write_percent(&out_channel->outbound);
When you reproducing the NULL ptr panic, does your kernel include the following patch?
hv_netvsc: common detach logic
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=7b2ee50c0cd513a176a26a71f2989facdd75bfea
We call netif_tx_disable(ndev) and netif_device_detach(ndev) before doing the changes
on MTU or #channels. So there should be no call to start_xmit() when channel is not ready.
If you see the check for CHANNEL_OPENED_STATE is still necessary on upstream kernel (including
the patch " common detach logic "), we should debug further on the code and find out the
root cause.
Thanks,
- Haiyang
^ permalink raw reply
* Re: [PATCH] net/ncsi: Add NCSI OEM command for FB Tiogapass
From: Vijay Khemka @ 2018-09-26 17:07 UTC (permalink / raw)
To: Samuel Mendoza-Jonas, Joel Stanley
Cc: linux-aspeed@lists.ozlabs.org, OpenBMC Maillist, Sai Dasari,
Amithash Prasad, netdev@vger.kernel.org, Justin.Lee1@Dell.com
In-Reply-To: <7cd8a49a057f1032e126b1e104fcf61e4956e06d.camel@mendozajonas.com>
> Hi Vijay,
> Thanks for the patch; before I get too into a review though I'd like to
> loop in Justin (cc'd) who I know is also working on an OEM command patch.
> The changes here are very specific (eg. a command specific config option
> "CONFIG_NCSI_OEM_CMD_GET_MAC"), which is ok on a small scale but if we
> start to add an increasing amount of commands could get out of hand.
> As I understand Justin's version adds a generic handler, using the NCSI
> Netlink interface to pass OEM commands and responses to and from
> userspace, which does the actual packet handling.
> It would be good to compare these two approaches first before committing
> to any one path
Hi Sam,
My oem command handler is generic and can be used for any oem commands and oem response handler can be made more generic. We can certainly write a wrapper to support netlink oem command to receive form user space. There are Mellanox specific functions which sends Mellanox specific request. These needed as a part of initial configuration. We can remove Kconfig option with more generic approach.
-Vijay
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox