From: Len Brown <lenb@kernel.org>
To: akpm@linux-foundation.org
Cc: linux-acpi@vger.kernel.org, astarikovskiy@suse.de,
andi@firstfloor.org, hmh@hmh.eng.br, maxi@daemonizer.de,
stable@kernel.org
Subject: Re: [patch 3/5] ACPI: EC: Don't degrade to poll mode at storm automatically
Date: Fri, 10 Oct 2008 13:02:53 -0400 (EDT) [thread overview]
Message-ID: <alpine.LFD.1.10.0810101300170.7059@localhost.localdomain> (raw)
In-Reply-To: <200809222137.m8MLbXl5031009@imap1.linux-foundation.org>
Alexey,
We want to drop this one, yes?
11089 is marked duplicate of 10919,
whichis fixed by "ACPI: EC: do transaction from interrupt context"
thanks,
-Len
On Mon, 22 Sep 2008, akpm@linux-foundation.org wrote:
> From: Alexey Starikovskiy <astarikovskiy@suse.de>
>
> Not all users of semi-broken EC devices want to degrade to poll mode, so
> give them right to choose.
>
> This fixes a regression in 2.6.26 (from 2.6.25.3). Initially reported as
> "Asus Eee PC hotkeys stop working if pressed quickly" in bugzilla
> <http://bugzilla.kernel.org/show_bug.cgi?id=11089>.
>
> The regression was caused by a recently added check for interrupt storms.
> The Eee PC triggers this check and switches to polling. When multiple
> events arrive between polling intervals, only one is fetched from the EC.
> This causes erroneous behaviour; ultimately events stop being delivered
> altogether when the EC buffer overflows.
>
> Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
> Cc: Alexey Starikovskiy <astarikovskiy@suse.de>
> Cc: Maximilian Engelhardt <maxi@daemonizer.de>
> Cc: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
> Cc: Andi Kleen <andi@firstfloor.org>
> Cc: <stable@kernel.org>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---
>
> Documentation/kernel-parameters.txt | 5 +++
> drivers/acpi/ec.c | 36 ++++++++++++++++++--------
> 2 files changed, 31 insertions(+), 10 deletions(-)
>
> diff -puN Documentation/kernel-parameters.txt~acpi-ec-dont-degrade-to-poll-mode-at-storm-automatically Documentation/kernel-parameters.txt
> --- a/Documentation/kernel-parameters.txt~acpi-ec-dont-degrade-to-poll-mode-at-storm-automatically
> +++ a/Documentation/kernel-parameters.txt
> @@ -736,6 +736,11 @@ and is between 256 and 4096 characters.
>
> eata= [HW,SCSI]
>
> + ec_intr= [HW,ACPI] ACPI Embedded Controller interrupt mode
> + Format: <int>
> + 0: polling mode
> + non-0: interrupt mode (default)
> +
> edd= [EDD]
> Format: {"off" | "on" | "skip[mbr]"}
>
> diff -puN drivers/acpi/ec.c~acpi-ec-dont-degrade-to-poll-mode-at-storm-automatically drivers/acpi/ec.c
> --- a/drivers/acpi/ec.c~acpi-ec-dont-degrade-to-poll-mode-at-storm-automatically
> +++ a/drivers/acpi/ec.c
> @@ -110,6 +110,8 @@ static struct acpi_ec {
> u8 handlers_installed;
> } *boot_ec, *first_ec;
>
> +int acpi_ec_intr = 1; /* Default is interrupt mode */
> +
> /*
> * Some Asus system have exchanged ECDT data/command IO addresses.
> */
> @@ -516,12 +518,14 @@ static u32 acpi_ec_gpe_handler(void *dat
> acpi_status status = AE_OK;
> struct acpi_ec *ec = data;
> u8 state = acpi_ec_read_status(ec);
> + static bool warn_done = false;
>
> pr_debug(PREFIX "~~~> interrupt\n");
> atomic_inc(&ec->irq_count);
> - if (atomic_read(&ec->irq_count) > 5) {
> - pr_err(PREFIX "GPE storm detected, disabling EC GPE\n");
> - ec_switch_to_poll_mode(ec);
> + if (!warn_done && atomic_read(&ec->irq_count) > 5) {
> + pr_warning(PREFIX "GPE storm detected, try to use ec_intr=0 "
> + "kernel option, if you see problems with keyboard.\n");
> + warn_done = 1;
> goto end;
> }
> clear_bit(EC_FLAGS_WAIT_GPE, &ec->flags);
> @@ -848,20 +852,21 @@ static int ec_install_handlers(struct ac
> acpi_status status;
> if (ec->handlers_installed)
> return 0;
> - status = acpi_install_gpe_handler(NULL, ec->gpe,
> + if (acpi_ec_intr) {
> + status = acpi_install_gpe_handler(NULL, ec->gpe,
> ACPI_GPE_EDGE_TRIGGERED,
> &acpi_ec_gpe_handler, ec);
> - if (ACPI_FAILURE(status))
> - return -ENODEV;
> -
> - acpi_set_gpe_type(NULL, ec->gpe, ACPI_GPE_TYPE_RUNTIME);
> - acpi_enable_gpe(NULL, ec->gpe, ACPI_NOT_ISR);
> + if (ACPI_FAILURE(status))
> + return -ENODEV;
>
> + acpi_set_gpe_type(NULL, ec->gpe, ACPI_GPE_TYPE_RUNTIME);
> + acpi_enable_gpe(NULL, ec->gpe, ACPI_NOT_ISR);
> + }
> status = acpi_install_address_space_handler(ec->handle,
> ACPI_ADR_SPACE_EC,
> &acpi_ec_space_handler,
> NULL, ec);
> - if (ACPI_FAILURE(status)) {
> + if (ACPI_FAILURE(status) && acpi_ec_intr) {
> acpi_remove_gpe_handler(NULL, ec->gpe, &acpi_ec_gpe_handler);
> return -ENODEV;
> }
> @@ -1047,3 +1052,14 @@ static void __exit acpi_ec_exit(void)
> return;
> }
> #endif /* 0 */
> +
> +static int __init acpi_ec_set_intr_mode(char *str)
> +{
> + if (!get_option(&str, &acpi_ec_intr)) {
> + acpi_ec_intr = 0;
> + return 0;
> + }
> + return 1;
> +}
> +
> +__setup("ec_intr=", acpi_ec_set_intr_mode);
> _
>
prev parent reply other threads:[~2008-10-10 17:03 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-22 21:37 [patch 3/5] ACPI: EC: Don't degrade to poll mode at storm automatically akpm
2008-09-23 1:03 ` Zhao Yakui
2008-09-23 1:39 ` Andrew Morton
2008-10-10 17:02 ` Len Brown [this message]
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=alpine.LFD.1.10.0810101300170.7059@localhost.localdomain \
--to=lenb@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=astarikovskiy@suse.de \
--cc=hmh@hmh.eng.br \
--cc=linux-acpi@vger.kernel.org \
--cc=maxi@daemonizer.de \
--cc=stable@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