All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org,
	Alexandru-Mihai Maftei <amaftei@solarflare.com>,
	Martin Habets <mhabets@solarflare.com>,
	"David S. Miller" <davem@davemloft.net>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.4 50/60] sfc: fix timestamp reconstruction at 16-bit rollover points
Date: Thu, 19 Mar 2020 14:04:28 +0100	[thread overview]
Message-ID: <20200319123935.284186646@linuxfoundation.org> (raw)
In-Reply-To: <20200319123919.441695203@linuxfoundation.org>

From: Alex Maftei (amaftei) <amaftei@solarflare.com>

[ Upstream commit 23797b98909f34b75fd130369bde86f760db69d0 ]

We can't just use the top bits of the last sync event as they could be
off-by-one every 65,536 seconds, giving an error in reconstruction of
65,536 seconds.

This patch uses the difference in the bottom 16 bits (mod 2^16) to
calculate an offset that needs to be applied to the last sync event to
get to the current time.

Signed-off-by: Alexandru-Mihai Maftei <amaftei@solarflare.com>
Acked-by: Martin Habets <mhabets@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/sfc/ptp.c | 38 +++++++++++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index af15a737c6756..59b4f16896a81 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -560,13 +560,45 @@ efx_ptp_mac_nic_to_ktime_correction(struct efx_nic *efx,
 				    u32 nic_major, u32 nic_minor,
 				    s32 correction)
 {
+	u32 sync_timestamp;
 	ktime_t kt = { 0 };
+	s16 delta;
 
 	if (!(nic_major & 0x80000000)) {
 		WARN_ON_ONCE(nic_major >> 16);
-		/* Use the top bits from the latest sync event. */
-		nic_major &= 0xffff;
-		nic_major |= (last_sync_timestamp_major(efx) & 0xffff0000);
+
+		/* Medford provides 48 bits of timestamp, so we must get the top
+		 * 16 bits from the timesync event state.
+		 *
+		 * We only have the lower 16 bits of the time now, but we do
+		 * have a full resolution timestamp at some point in past. As
+		 * long as the difference between the (real) now and the sync
+		 * is less than 2^15, then we can reconstruct the difference
+		 * between those two numbers using only the lower 16 bits of
+		 * each.
+		 *
+		 * Put another way
+		 *
+		 * a - b = ((a mod k) - b) mod k
+		 *
+		 * when -k/2 < (a-b) < k/2. In our case k is 2^16. We know
+		 * (a mod k) and b, so can calculate the delta, a - b.
+		 *
+		 */
+		sync_timestamp = last_sync_timestamp_major(efx);
+
+		/* Because delta is s16 this does an implicit mask down to
+		 * 16 bits which is what we need, assuming
+		 * MEDFORD_TX_SECS_EVENT_BITS is 16. delta is signed so that
+		 * we can deal with the (unlikely) case of sync timestamps
+		 * arriving from the future.
+		 */
+		delta = nic_major - sync_timestamp;
+
+		/* Recover the fully specified time now, by applying the offset
+		 * to the (fully specified) sync time.
+		 */
+		nic_major = sync_timestamp + delta;
 
 		kt = ptp->nic_to_kernel_time(nic_major, nic_minor,
 					     correction);
-- 
2.20.1




  parent reply	other threads:[~2020-03-19 13:24 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-19 13:03 [PATCH 5.4 00/60] 5.4.27-rc1 review Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 01/60] netfilter: hashlimit: do not use indirect calls during gc Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 02/60] netfilter: xt_hashlimit: unregister proc file before releasing mutex Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 03/60] drm/amdgpu: Fix TLB invalidation request when using semaphore Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 04/60] mmc: core: Default to generic_cmd6_time as timeout in __mmc_switch() Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 05/60] mmc: core: Allow host controllers to require R1B for CMD6 Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 06/60] mmc: sdhci-tegra: Fix busy detection by enabling MMC_CAP_NEED_RSP_BUSY Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 07/60] mmc: sdhci-omap: " Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 08/60] mmc: core: Respect MMC_CAP_NEED_RSP_BUSY for eMMC sleep command Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 09/60] mmc: core: Respect MMC_CAP_NEED_RSP_BUSY for erase/trim/discard Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 10/60] ACPI: watchdog: Allow disabling WDAT at boot Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 11/60] HID: apple: Add support for recent firmware on Magic Keyboards Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 12/60] ACPI: watchdog: Set default timeout in probe Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 13/60] HID: i2c-hid: add Trekstor Surfbook E11B to descriptor override Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 14/60] mips: vdso: fix jalr t9 crash in vdso code Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 15/60] MIPS: Disable VDSO time functionality on microMIPS Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 16/60] mips: vdso: add build time check that no jalr t9 calls left Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 17/60] HID: hid-bigbenff: fix general protection fault caused by double kfree Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 18/60] HID: hid-bigbenff: call hid_hw_stop() in case of error Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 19/60] HID: hid-bigbenff: fix race condition for scheduled work during removal Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 20/60] MIPS: vdso: Wrap -mexplicit-relocs in cc-option Greg Kroah-Hartman
2020-03-19 13:03 ` [PATCH 5.4 21/60] selftests/rseq: Fix out-of-tree compilation Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 22/60] tracing: Fix number printing bug in print_synth_event() Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 23/60] cfg80211: check reg_rule for NULL in handle_channel_custom() Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 24/60] scsi: libfc: free response frame from GPN_ID Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 25/60] net: usb: qmi_wwan: restore mtu min/max values after raw_ip switch Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 26/60] net: ks8851-ml: Fix IRQ handling and locking Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 27/60] mac80211: rx: avoid RCU list traversal under mutex Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 28/60] net: ll_temac: Fix race condition causing TX hang Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 29/60] net: ll_temac: Add more error handling of dma_map_single() calls Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 30/60] net: ll_temac: Fix RX buffer descriptor handling on GFP_ATOMIC pressure Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 31/60] net: ll_temac: Handle DMA halt condition caused by buffer underrun Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 32/60] blk-mq: insert passthrough request into hctx->dispatch directly Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 33/60] drm/amdgpu: fix memory leak during TDR test(v2) Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 34/60] kbuild: add dtbs_check to PHONY Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 35/60] kbuild: add dt_binding_check to PHONY in a correct place Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 36/60] signal: avoid double atomic counter increments for user accounting Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 37/60] slip: not call free_netdev before rtnl_unlock in slip_open Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 38/60] net: phy: mscc: fix firmware paths Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 39/60] hinic: fix a irq affinity bug Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 40/60] hinic: fix a bug of setting hw_ioctxt Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 41/60] hinic: fix a bug of rss configuration Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 42/60] net: rmnet: fix NULL pointer dereference in rmnet_newlink() Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 43/60] net: rmnet: fix NULL pointer dereference in rmnet_changelink() Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 44/60] net: rmnet: fix suspicious RCU usage Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 45/60] net: rmnet: remove rcu_read_lock in rmnet_force_unassociate_device() Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 46/60] net: rmnet: do not allow to change mux id if mux id is duplicated Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 47/60] net: rmnet: use upper/lower device infrastructure Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 48/60] net: rmnet: fix bridge mode bugs Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 49/60] net: rmnet: fix packet forwarding in rmnet bridge mode Greg Kroah-Hartman
2020-03-19 13:04 ` Greg Kroah-Hartman [this message]
2020-03-19 13:04 ` [PATCH 5.4 51/60] jbd2: fix data races at struct journal_head Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 52/60] blk-mq: insert flush request to the front of dispatch queue Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 53/60] net: qrtr: fix len of skb_put_padto in qrtr_node_enqueue Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 54/60] ARM: 8957/1: VDSO: Match ARMv8 timer in cntvct_functional() Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 55/60] ARM: 8958/1: rename missed uaccess .fixup section Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 56/60] mm: slub: add missing TID bump in kmem_cache_alloc_bulk() Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 57/60] HID: google: add moonball USB id Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 58/60] HID: add ALWAYS_POLL quirk to lenovo pixart mouse Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 59/60] ARM: 8961/2: Fix Kbuild issue caused by per-task stack protector GCC plugin Greg Kroah-Hartman
2020-03-19 13:04 ` [PATCH 5.4 60/60] ipv4: ensure rcu_read_lock() in cipso_v4_error() Greg Kroah-Hartman
2020-03-19 21:59 ` [PATCH 5.4 00/60] 5.4.27-rc1 review Naresh Kamboju
2020-03-20  7:12   ` Greg Kroah-Hartman
2020-03-21 18:18     ` Naresh Kamboju
2020-03-19 23:55 ` Guenter Roeck
2020-03-20 10:55   ` Greg Kroah-Hartman
2020-03-20 11:23     ` Naresh Kamboju
2020-03-20 11:31       ` Greg Kroah-Hartman

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=20200319123935.284186646@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=amaftei@solarflare.com \
    --cc=davem@davemloft.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mhabets@solarflare.com \
    --cc=sashal@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.