linux-kernel.vger.kernel.org archive mirror
 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, Johannes Berg <johannes.berg@intel.com>,
	Mordechay Goodstein <mordechay.goodstein@intel.com>,
	Luca Coelho <luciano.coelho@intel.com>,
	Kalle Valo <kvalo@codeaurora.org>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.4 03/36] iwlwifi: pcie: limit memory read spin time
Date: Mon, 14 Dec 2020 18:27:47 +0100	[thread overview]
Message-ID: <20201214172543.478587794@linuxfoundation.org> (raw)
In-Reply-To: <20201214172543.302523401@linuxfoundation.org>

From: Johannes Berg <johannes.berg@intel.com>

[ Upstream commit 04516706bb99889986ddfa3a769ed50d2dc7ac13 ]

When we read device memory, we lock a spinlock, write the address we
want to read from the device and then spin in a loop reading the data
in 32-bit quantities from another register.

As the description makes clear, this is rather inefficient, incurring
a PCIe bus transaction for every read. In a typical device today, we
want to read 786k SMEM if it crashes, leading to 192k register reads.
Occasionally, we've seen the whole loop take over 20 seconds and then
triggering the soft lockup detector.

Clearly, it is unreasonable to spin here for such extended periods of
time.

To fix this, break the loop down into an outer and an inner loop, and
break out of the inner loop if more than half a second elapsed. To
avoid too much overhead, check for that only every 128 reads, though
there's no particular reason for that number. Then, unlock and relock
to obtain NIC access again, reprogram the start address and continue.

This will keep (interrupt) latencies on the CPU down to a reasonable
time.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/iwlwifi.20201022165103.45878a7e49aa.I3b9b9c5a10002915072312ce75b68ed5b3dc6e14@changeid
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 .../net/wireless/intel/iwlwifi/pcie/trans.c   | 36 ++++++++++++++-----
 1 file changed, 27 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index c76d26708e659..ef5a8ecabc60a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -2178,18 +2178,36 @@ static int iwl_trans_pcie_read_mem(struct iwl_trans *trans, u32 addr,
 				   void *buf, int dwords)
 {
 	unsigned long flags;
-	int offs, ret = 0;
+	int offs = 0;
 	u32 *vals = buf;
 
-	if (iwl_trans_grab_nic_access(trans, &flags)) {
-		iwl_write32(trans, HBUS_TARG_MEM_RADDR, addr);
-		for (offs = 0; offs < dwords; offs++)
-			vals[offs] = iwl_read32(trans, HBUS_TARG_MEM_RDAT);
-		iwl_trans_release_nic_access(trans, &flags);
-	} else {
-		ret = -EBUSY;
+	while (offs < dwords) {
+		/* limit the time we spin here under lock to 1/2s */
+		ktime_t timeout = ktime_add_us(ktime_get(), 500 * USEC_PER_MSEC);
+
+		if (iwl_trans_grab_nic_access(trans, &flags)) {
+			iwl_write32(trans, HBUS_TARG_MEM_RADDR,
+				    addr + 4 * offs);
+
+			while (offs < dwords) {
+				vals[offs] = iwl_read32(trans,
+							HBUS_TARG_MEM_RDAT);
+				offs++;
+
+				/* calling ktime_get is expensive so
+				 * do it once in 128 reads
+				 */
+				if (offs % 128 == 0 && ktime_after(ktime_get(),
+								   timeout))
+					break;
+			}
+			iwl_trans_release_nic_access(trans, &flags);
+		} else {
+			return -EBUSY;
+		}
 	}
-	return ret;
+
+	return 0;
 }
 
 static int iwl_trans_pcie_write_mem(struct iwl_trans *trans, u32 addr,
-- 
2.27.0




  parent reply	other threads:[~2020-12-14 17:32 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-14 17:27 [PATCH 5.4 00/36] 5.4.84-rc1 review Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 01/36] Kbuild: do not emit debug info for assembly with LLVM_IAS=1 Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 02/36] x86/lib: Change .weak to SYM_FUNC_START_WEAK for arch/x86/lib/mem*_64.S Greg Kroah-Hartman
2020-12-14 17:27 ` Greg Kroah-Hartman [this message]
2020-12-14 17:27 ` [PATCH 5.4 04/36] arm64: dts: rockchip: Assign a fixed index to mmc devices on rk3399 boards Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 05/36] iwlwifi: pcie: set LTR to avoid completion timeout Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 06/36] iwlwifi: mvm: fix kernel panic in case of assert during CSA Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 07/36] powerpc: Drop -me200 addition to build flags Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 08/36] arm64: dts: broadcom: clear the warnings caused by empty dma-ranges Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 09/36] ARC: stack unwinding: dont assume non-current task is sleeping Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 10/36] scsi: ufs: Make sure clk scaling happens only when HBA is runtime ACTIVE Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 11/36] interconnect: qcom: qcs404: Remove GPU and display RPM IDs Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 12/36] ibmvnic: skip tx timeout reset while in resetting Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 13/36] irqchip/gic-v3-its: Unconditionally save/restore the ITS state on suspend Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 14/36] spi: spi-nxp-fspi: fix fspi panic by unexpected interrupts Greg Kroah-Hartman
2020-12-14 17:27 ` [PATCH 5.4 15/36] soc: fsl: dpio: Get the cpumask through cpumask_of(cpu) Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 16/36] arm64: tegra: Disable the ACONNECT for Jetson TX2 Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 17/36] platform/x86: thinkpad_acpi: Do not report SW_TABLET_MODE on Yoga 11e Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 18/36] platform/x86: thinkpad_acpi: Add BAT1 is primary battery quirk for Thinkpad Yoga 11e 4th gen Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 19/36] platform/x86: acer-wmi: add automatic keyboard background light toggle key as KEY_LIGHTS_TOGGLE Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 20/36] platform/x86: intel-vbtn: Support for tablet mode on HP Pavilion 13 x360 PC Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 21/36] platform/x86: touchscreen_dmi: Add info for the Irbis TW118 tablet Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 22/36] can: m_can: m_can_dev_setup(): add support for bosch mcan version 3.3.0 Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 23/36] ktest.pl: Fix incorrect reboot for grub2bls Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 24/36] Input: cm109 - do not stomp on control URB Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 25/36] Input: i8042 - add Acer laptops to the i8042 reset list Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 26/36] pinctrl: amd: remove debounce filter setting in IRQ type setting Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 27/36] mmc: block: Fixup condition for CMD13 polling for RPMB requests Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 28/36] drm/i915/display/dp: Compute the correct slice count for VDSC on DP Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 29/36] kbuild: avoid static_assert for genksyms Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 30/36] proc: use untagged_addr() for pagemap_read addresses Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 31/36] scsi: be2iscsi: Revert "Fix a theoretical leak in beiscsi_create_eqs()" Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 32/36] x86/mm/mem_encrypt: Fix definition of PMD_FLAGS_DEC_WP Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 33/36] x86/membarrier: Get rid of a dubious optimization Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 34/36] x86/apic/vector: Fix ordering in vector assignment Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 35/36] mm/zsmalloc.c: drop ZSMALLOC_PGTABLE_MAPPING Greg Kroah-Hartman
2020-12-14 17:28 ` [PATCH 5.4 36/36] compiler.h: fix barrier_data() on clang Greg Kroah-Hartman
2020-12-14 22:21 ` [PATCH 5.4 00/36] 5.4.84-rc1 review Shuah Khan
2020-12-15  6:51 ` Naresh Kamboju
2020-12-15 20:31 ` Guenter Roeck

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=20201214172543.478587794@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=johannes.berg@intel.com \
    --cc=kvalo@codeaurora.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luciano.coelho@intel.com \
    --cc=mordechay.goodstein@intel.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 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).