From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Ben Hutchings <bhutchings@solarflare.com>
Subject: [ 20/72] sfc: Work-around flush timeout when flushes have completed
Date: Tue, 26 Mar 2013 15:51:03 -0700 [thread overview]
Message-ID: <20130326224921.801486051@linuxfoundation.org> (raw)
In-Reply-To: <20130326224919.675227837@linuxfoundation.org>
3.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Pieczko <dpieczko@solarflare.com>
[ Upstream commit 525d9e824018cd7cc8d8d44832ddcd363abfe6e1 ]
We sometimes hit a "failed to flush" timeout on some TX queues, but the
flushes have completed and the flush completion events seem to go missing.
In this case, we can check the TX_DESC_PTR_TBL register and drain the
queues if the flushes had finished.
[bwh: Minor fixes to coding style]
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/net/ethernet/sfc/net_driver.h | 1
drivers/net/ethernet/sfc/nic.c | 56 +++++++++++++++++++++++++++++++---
2 files changed, 53 insertions(+), 4 deletions(-)
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -194,6 +194,7 @@ struct efx_tx_queue {
/* Members shared between paths and sometimes updated */
unsigned int empty_read_count ____cacheline_aligned_in_smp;
#define EFX_EMPTY_COUNT_VALID 0x80000000
+ atomic_t flush_outstanding;
};
/**
--- a/drivers/net/ethernet/sfc/nic.c
+++ b/drivers/net/ethernet/sfc/nic.c
@@ -73,6 +73,8 @@
_EFX_CHANNEL_MAGIC(_EFX_CHANNEL_MAGIC_TX_DRAIN, \
(_tx_queue)->queue)
+static void efx_magic_event(struct efx_channel *channel, u32 magic);
+
/**************************************************************************
*
* Solarstorm hardware access
@@ -495,6 +497,9 @@ static void efx_flush_tx_queue(struct ef
struct efx_nic *efx = tx_queue->efx;
efx_oword_t tx_flush_descq;
+ WARN_ON(atomic_read(&tx_queue->flush_outstanding));
+ atomic_set(&tx_queue->flush_outstanding, 1);
+
EFX_POPULATE_OWORD_2(tx_flush_descq,
FRF_AZ_TX_FLUSH_DESCQ_CMD, 1,
FRF_AZ_TX_FLUSH_DESCQ, tx_queue->queue);
@@ -670,6 +675,47 @@ static bool efx_flush_wake(struct efx_ni
&& atomic_read(&efx->rxq_flush_pending) > 0));
}
+static bool efx_check_tx_flush_complete(struct efx_nic *efx)
+{
+ bool i = true;
+ efx_oword_t txd_ptr_tbl;
+ struct efx_channel *channel;
+ struct efx_tx_queue *tx_queue;
+
+ efx_for_each_channel(channel, efx) {
+ efx_for_each_channel_tx_queue(tx_queue, channel) {
+ efx_reado_table(efx, &txd_ptr_tbl,
+ FR_BZ_TX_DESC_PTR_TBL, tx_queue->queue);
+ if (EFX_OWORD_FIELD(txd_ptr_tbl,
+ FRF_AZ_TX_DESCQ_FLUSH) ||
+ EFX_OWORD_FIELD(txd_ptr_tbl,
+ FRF_AZ_TX_DESCQ_EN)) {
+ netif_dbg(efx, hw, efx->net_dev,
+ "flush did not complete on TXQ %d\n",
+ tx_queue->queue);
+ i = false;
+ } else if (atomic_cmpxchg(&tx_queue->flush_outstanding,
+ 1, 0)) {
+ /* The flush is complete, but we didn't
+ * receive a flush completion event
+ */
+ netif_dbg(efx, hw, efx->net_dev,
+ "flush complete on TXQ %d, so drain "
+ "the queue\n", tx_queue->queue);
+ /* Don't need to increment drain_pending as it
+ * has already been incremented for the queues
+ * which did not drain
+ */
+ efx_magic_event(channel,
+ EFX_CHANNEL_MAGIC_TX_DRAIN(
+ tx_queue));
+ }
+ }
+ }
+
+ return i;
+}
+
/* Flush all the transmit queues, and continue flushing receive queues until
* they're all flushed. Wait for the DRAIN events to be recieved so that there
* are no more RX and TX events left on any channel. */
@@ -730,7 +776,8 @@ int efx_nic_flush_queues(struct efx_nic
timeout);
}
- if (atomic_read(&efx->drain_pending)) {
+ if (atomic_read(&efx->drain_pending) &&
+ !efx_check_tx_flush_complete(efx)) {
netif_err(efx, hw, efx->net_dev, "failed to flush %d queues "
"(rx %d+%d)\n", atomic_read(&efx->drain_pending),
atomic_read(&efx->rxq_flush_outstanding),
@@ -1017,9 +1064,10 @@ efx_handle_tx_flush_done(struct efx_nic
if (qid < EFX_TXQ_TYPES * efx->n_tx_channels) {
tx_queue = efx_get_tx_queue(efx, qid / EFX_TXQ_TYPES,
qid % EFX_TXQ_TYPES);
-
- efx_magic_event(tx_queue->channel,
- EFX_CHANNEL_MAGIC_TX_DRAIN(tx_queue));
+ if (atomic_cmpxchg(&tx_queue->flush_outstanding, 1, 0)) {
+ efx_magic_event(tx_queue->channel,
+ EFX_CHANNEL_MAGIC_TX_DRAIN(tx_queue));
+ }
}
}
next prev parent reply other threads:[~2013-03-26 22:51 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-26 22:50 [ 00/72] 3.4.38-stable review Greg Kroah-Hartman
2013-03-26 22:50 ` [ 01/72] Revert "USB: EHCI: dont check DMA values in QH overlays" Greg Kroah-Hartman
2013-03-26 22:50 ` [ 02/72] sunsu: Fix panic in case of nonexistent port at "console=ttySY" cmdline option Greg Kroah-Hartman
2013-03-26 22:50 ` [ 03/72] net/ipv4: Ensure that location of timestamp option is stored Greg Kroah-Hartman
2013-03-26 22:50 ` [ 04/72] netconsole: dont call __netpoll_cleanup() while atomic Greg Kroah-Hartman
2013-03-26 22:50 ` [ 05/72] bonding: dont call update_speed_duplex() under spinlocks Greg Kroah-Hartman
2013-03-26 22:50 ` [ 06/72] tg3: 5715 does not link up when autoneg off Greg Kroah-Hartman
2013-03-26 22:50 ` [ 07/72] sctp: Use correct sideffect command in duplicate cookie handling Greg Kroah-Hartman
2013-03-26 22:50 ` [ 08/72] sctp: dont break the loop while meeting the active_path so as to find the matched transport Greg Kroah-Hartman
2013-03-26 22:50 ` [ 09/72] ipv4: fix definition of FIB_TABLE_HASHSZ Greg Kroah-Hartman
2013-03-26 22:50 ` [ 10/72] tcp: fix skb_availroom() Greg Kroah-Hartman
2013-03-26 22:50 ` [ 11/72] rtnetlink: Mask the rta_type when range checking Greg Kroah-Hartman
2013-03-26 22:50 ` [ 12/72] vhost/net: fix heads usage of ubuf_info Greg Kroah-Hartman
2013-03-26 22:50 ` [ 13/72] bnx2x: fix occasional statistics off-by-4GB error Greg Kroah-Hartman
2013-03-26 22:50 ` [ 14/72] inet: limit length of fragment queue hash table bucket lists Greg Kroah-Hartman
2013-03-26 22:50 ` [ 15/72] sfc: Do not attempt to flush queues if DMA is disabled Greg Kroah-Hartman
2013-03-26 22:50 ` [ 16/72] sfc: Convert firmware subtypes to native byte order in efx_mcdi_get_board_cfg() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 17/72] sfc: Add parentheses around use of bitfield macro arguments Greg Kroah-Hartman
2013-03-26 22:51 ` [ 18/72] sfc: Fix MCDI structure field lookup Greg Kroah-Hartman
2013-03-26 22:51 ` [ 19/72] sfc: Really disable flow control while flushing Greg Kroah-Hartman
2013-03-26 22:51 ` Greg Kroah-Hartman [this message]
2013-03-26 22:51 ` [ 21/72] sfc: lock TX queues when calling netif_device_detach() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 22/72] sfc: Fix timekeeping in efx_mcdi_poll() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 23/72] sfc: Disable VF queues during register self-test Greg Kroah-Hartman
2013-03-26 22:51 ` [ 24/72] sfc: Avoid generating over-length MC_CMD_FLUSH_RX_QUEUES request Greg Kroah-Hartman
2013-03-26 22:51 ` [ 25/72] sfc: Correctly initialise reset_method in siena_test_chip() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 26/72] sfc: Properly sync RX DMA buffer when it is not the last in the page Greg Kroah-Hartman
2013-03-26 22:51 ` [ 27/72] sfc: Fix efx_rx_buf_offset() in the presence of swiotlb Greg Kroah-Hartman
2013-03-26 22:51 ` [ 28/72] sfc: Detach net device when stopping queues for reconfiguration Greg Kroah-Hartman
2013-03-26 22:51 ` [ 29/72] sfc: Disable soft interrupt handling during efx_device_detach_sync() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 30/72] sfc: Only use TX push if a single descriptor is to be written Greg Kroah-Hartman
2013-03-26 22:51 ` [ 31/72] ALSA: hda/cirrus - Fix the digital beep registration Greg Kroah-Hartman
2013-03-26 22:51 ` [ 32/72] ALSA: hda - Fix typo in checking IEC958 emphasis bit Greg Kroah-Hartman
2013-03-26 22:51 ` [ 33/72] ALSA: snd-usb: mixer: propagate errors up the call chain Greg Kroah-Hartman
2013-03-26 22:51 ` [ 34/72] ALSA: snd-usb: mixer: ignore -EINVAL in snd_usb_mixer_controls() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 35/72] drm/i915: restrict kernel address leak in debugfs Greg Kroah-Hartman
2013-03-26 22:51 ` [ 36/72] tracing: Fix race in snapshot swapping Greg Kroah-Hartman
2013-03-26 22:51 ` [ 37/72] tracing: Fix free of probe entry by calling call_rcu_sched() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 38/72] rtlwifi: rtl8192cu: Fix schedule while atomic bug splat Greg Kroah-Hartman
2013-03-26 22:51 ` [ 39/72] rtlwifi: rtl8192cu: Fix problem that prevents reassociation Greg Kroah-Hartman
2013-03-26 22:51 ` [ 40/72] mwifiex: fix potential out-of-boundary access to ibss rate table Greg Kroah-Hartman
2013-03-26 22:51 ` [ 41/72] drm/i915: bounds check execbuffer relocation count Greg Kroah-Hartman
2013-03-26 22:51 ` [ 42/72] KMS: fix EDID detailed timing vsync parsing Greg Kroah-Hartman
2013-03-26 22:51 ` [ 43/72] KMS: fix EDID detailed timing frame rate Greg Kroah-Hartman
2013-03-26 22:51 ` [ 44/72] mm/hugetlb: fix total hugetlbfs pages count when using memory overcommit accouting Greg Kroah-Hartman
2013-03-26 22:51 ` [ 45/72] target/iscsi: Fix mutual CHAP auth on big-endian arches Greg Kroah-Hartman
2013-03-26 22:51 ` [ 46/72] drm/radeon: add Richland pci ids Greg Kroah-Hartman
2013-03-26 22:51 ` [ 47/72] drm/radeon: add support for Richland APUs Greg Kroah-Hartman
2013-03-26 22:51 ` [ 48/72] drm/radeon/benchmark: make sure bo blit copy exists before using it Greg Kroah-Hartman
2013-03-26 22:51 ` [ 49/72] cifs: ignore everything in SPNEGO blob after mechTypes Greg Kroah-Hartman
2013-03-26 22:51 ` [ 50/72] jbd2: fix use after free in jbd2_journal_dirty_metadata() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 51/72] ext4: fix the wrong number of the allocated blocks in ext4_split_extent() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 52/72] usb-storage: add unusual_devs entry for Samsung YP-Z3 mp3 player Greg Kroah-Hartman
2013-03-26 22:51 ` [ 53/72] ext4: fix data=journal fast mount/umount hang Greg Kroah-Hartman
2013-03-26 22:51 ` [ 54/72] IPoIB: Fix send lockup due to missed TX completion Greg Kroah-Hartman
2013-03-26 22:51 ` [ 55/72] clockevents: Dont allow dummy broadcast timers Greg Kroah-Hartman
2013-03-26 22:51 ` [ 56/72] x86-64: Fix the failure case in copy_user_handle_tail() Greg Kroah-Hartman
2013-03-26 22:51 ` [ 57/72] USB: xhci - fix bit definitions for IMAN register Greg Kroah-Hartman
2013-03-26 22:51 ` [ 58/72] USB: xhci: correctly enable interrupts Greg Kroah-Hartman
2013-03-26 22:51 ` [ 59/72] USB: cdc-acm: fix device unregistration Greg Kroah-Hartman
2013-03-26 22:51 ` [ 60/72] USB: serial: fix interface refcounting Greg Kroah-Hartman
2013-03-26 22:51 ` [ 61/72] nohz: Make tick_nohz_irq_exit() irq safe Greg Kroah-Hartman
2013-03-26 22:51 ` [ 62/72] udf: Fix bitmap overflow on large filesystems with small block size Greg Kroah-Hartman
2013-03-26 22:51 ` [ 63/72] USB: garmin_gps: fix memory leak on disconnect Greg Kroah-Hartman
2013-03-26 22:51 ` [ 64/72] USB: io_ti: fix get_icount for two port adapters Greg Kroah-Hartman
2013-03-26 22:51 ` [ 65/72] key: Fix resource leak Greg Kroah-Hartman
2013-03-26 22:51 ` [ 66/72] isofs: avoid info leak on export Greg Kroah-Hartman
2013-03-26 22:51 ` [ 67/72] udf: " Greg Kroah-Hartman
2013-03-26 22:51 ` [ 68/72] tools: hv: Netlink source address validation allows DoS Greg Kroah-Hartman
2013-03-26 22:51 ` [ 69/72] i915: initialize CADL in opregion Greg Kroah-Hartman
2013-03-26 22:51 ` [ 70/72] exec: use -ELOOP for max recursion depth Greg Kroah-Hartman
2013-03-26 22:51 ` [ 71/72] rt2x00: error in configurations with mesh support disabled Greg Kroah-Hartman
2013-03-26 22:51 ` [ 72/72] asus-laptop: Do not call HWRS on init Greg Kroah-Hartman
2013-03-27 18:33 ` [ 00/72] 3.4.38-stable review Shuah Khan
2013-03-28 14:17 ` Satoru Takeuchi
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=20130326224921.801486051@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=bhutchings@solarflare.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).