diff options
| author | Lv Zheng <lv.zheng@intel.com> | 2015-05-15 09:16:34 +0300 |
|---|---|---|
| committer | Rafael J. Wysocki <rafael.j.wysocki@intel.com> | 2015-05-16 02:51:17 +0300 |
| commit | 373783e6e9394b0dd5050eba152f436e08577d69 (patch) | |
| tree | 0d57092fc5bb08777c87d8633ab4fd8fe1c016d3 /drivers/acpi/ec.c | |
| parent | 5ab82a11e58b6af704976f5d50ce098472f8abbd (diff) | |
| download | linux-373783e6e9394b0dd5050eba152f436e08577d69.tar.xz | |
ACPI / EC: Remove irqs_disabled() check.
The following commit merges polling and interrupt modes for EC driver:
Commit: 2a84cb9852f52c0cd1c48bca41a8792d44ad06cc Mon Sep 17 00:00:00 2001
Subject: ACPI: EC: Merge IRQ and POLL modes
The irqs_disabled() check introduced in it tries to fall into busy polling
mode when the context of ec_poll() cannot sleep.
Actually ec_poll() is ensured to be invoked in the contexts that can sleep
(from a sysfs /sys/kernel/debug/ec/ec0/io access, or from
acpi_evaluate_object(), or from acpi_ec_gpe_poller()). Without the MSI
quirk, we never saw the udelay() logic invoked. Thus this check is useless
and can be removed.
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Diffstat (limited to 'drivers/acpi/ec.c')
| -rw-r--r-- | drivers/acpi/ec.c | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 170d74387800..20bd43f29853 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -504,8 +504,7 @@ static int ec_poll(struct acpi_ec *ec) msecs_to_jiffies(ec_delay); unsigned long usecs = ACPI_EC_UDELAY_POLL; do { - /* don't sleep with disabled interrupts */ - if (EC_FLAGS_MSI || irqs_disabled()) { + if (EC_FLAGS_MSI) { usecs = ACPI_EC_MSI_UDELAY; udelay(usecs); if (ec_transaction_completed(ec)) |