| Message ID | 1450237773-1685-1-git-send-email-ike.pan@canonical.com (mailing list archive) |
|---|---|
| State | Changes Requested, archived |
| Headers | show |
On Wed, Dec 16, 2015 at 11:49:33AM +0800, Ike Panhc wrote: Hi Ike, +David Woodhouse (MODULE_AUTHOR) > Hardware radio switch is rare on recently ideapad and some of them > reported radio hardware blocked by error. With more and more ideapads > available in market to maintain the dmi table becomes a > never-finished job. Indeed, inverting this logic or eliminating it would be an improvement. To be clear, the only platforms which need to be listed in the DMI table are those without a hw radio switch AND which will report hw_blocked when the ACPI VPCCMD_R_RF call is made (instead of always returning 0 as they should)? As such, the DMI list is a list of laptops with buggy firmware - correct? > Therefore I am thinking an easy way to detect by response from > hardware. This patch will make driver says hardware switch is not > blocked if the response from ACPI is always radio blocked. > > For an ideapad without radio switch, no matter what ACPI says, driver > will report false on hardware blocked. > > For an ideapad with radio switch, if driver loaded with radio on, no > behavior is changed. > > For an ideapad with radio switch and driver loaded with radio off, > driver will report unblocked falsely and network manager might not > scan if wireless driver reports blocked. Once the switch is on, > driver will report correct information. This would be a regression for existing platforms though. Do we have the ASL for some of the offending models compared with the good ones so we can inspect and look for a deterministic detection mechanism? > > Signed-off-by: Ike Panhc <ike.pan@canonical.com> > --- > drivers/platform/x86/ideapad-laptop.c | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c > index a313dfc..91ccb4e 100644 > --- a/drivers/platform/x86/ideapad-laptop.c > +++ b/drivers/platform/x86/ideapad-laptop.c > @@ -482,11 +482,16 @@ static void ideapad_sync_rfk_state(struct ideapad_private *priv) > { > unsigned long hw_blocked = 0; > int i; > + static int hw_unblock_once; > > if (priv->has_hw_rfkill_switch) { > if (read_ec_data(priv->adev->handle, VPCCMD_R_RF, &hw_blocked)) > return; > + if (hw_blocked) > + hw_unblock_once = 1; > hw_blocked = !hw_blocked; > + if (!hw_unblock_once) > + hw_blocked = 0; > } > > for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++) > -- > 1.9.1 > >
On 12/19/2015 06:56 AM, Darren Hart wrote: > On Wed, Dec 16, 2015 at 11:49:33AM +0800, Ike Panhc wrote: > > Hi Ike, > > +David Woodhouse (MODULE_AUTHOR) > >> Hardware radio switch is rare on recently ideapad and some of them >> reported radio hardware blocked by error. With more and more ideapads >> available in market to maintain the dmi table becomes a >> never-finished job. > > Indeed, inverting this logic or eliminating it would be an improvement. > > To be clear, the only platforms which need to be listed in the DMI table are > those without a hw radio switch AND which will report hw_blocked when the ACPI > VPCCMD_R_RF call is made (instead of always returning 0 as they should)? > > As such, the DMI list is a list of laptops with buggy firmware - correct? > >> Therefore I am thinking an easy way to detect by response from >> hardware. This patch will make driver says hardware switch is not >> blocked if the response from ACPI is always radio blocked. >> >> For an ideapad without radio switch, no matter what ACPI says, driver >> will report false on hardware blocked. >> >> For an ideapad with radio switch, if driver loaded with radio on, no >> behavior is changed. >> >> For an ideapad with radio switch and driver loaded with radio off, >> driver will report unblocked falsely and network manager might not >> scan if wireless driver reports blocked. Once the switch is on, >> driver will report correct information. > > > This would be a regression for existing platforms though. > > Do we have the ASL for some of the offending models compared with the good ones > so we can inspect and look for a deterministic detection mechanism? Thanks for the suggestion. It makes sense to me. Unfortunately I do not have one yet. Let me ask for acpidump for them. > > >> >> Signed-off-by: Ike Panhc <ike.pan@canonical.com> >> --- >> drivers/platform/x86/ideapad-laptop.c | 5 +++++ >> 1 file changed, 5 insertions(+) >> >> diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c >> index a313dfc..91ccb4e 100644 >> --- a/drivers/platform/x86/ideapad-laptop.c >> +++ b/drivers/platform/x86/ideapad-laptop.c >> @@ -482,11 +482,16 @@ static void ideapad_sync_rfk_state(struct ideapad_private *priv) >> { >> unsigned long hw_blocked = 0; >> int i; >> + static int hw_unblock_once; >> >> if (priv->has_hw_rfkill_switch) { >> if (read_ec_data(priv->adev->handle, VPCCMD_R_RF, &hw_blocked)) >> return; >> + if (hw_blocked) >> + hw_unblock_once = 1; >> hw_blocked = !hw_blocked; >> + if (!hw_unblock_once) >> + hw_blocked = 0; >> } >> >> for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++) >> -- >> 1.9.1 >> >> > -- To unsubscribe from this list: send the line "unsubscribe platform-driver-x86" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index a313dfc..91ccb4e 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -482,11 +482,16 @@ static void ideapad_sync_rfk_state(struct ideapad_private *priv) { unsigned long hw_blocked = 0; int i; + static int hw_unblock_once; if (priv->has_hw_rfkill_switch) { if (read_ec_data(priv->adev->handle, VPCCMD_R_RF, &hw_blocked)) return; + if (hw_blocked) + hw_unblock_once = 1; hw_blocked = !hw_blocked; + if (!hw_unblock_once) + hw_blocked = 0; } for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++)
Hardware radio switch is rare on recently ideapad and some of them reported radio hardware blocked by error. With more and more ideapads available in market to maintain the dmi table becomes a never-finished job. Therefore I am thinking an easy way to detect by response from hardware. This patch will make driver says hardware switch is not blocked if the response from ACPI is always radio blocked. For an ideapad without radio switch, no matter what ACPI says, driver will report false on hardware blocked. For an ideapad with radio switch, if driver loaded with radio on, no behavior is changed. For an ideapad with radio switch and driver loaded with radio off, driver will report unblocked falsely and network manager might not scan if wireless driver reports blocked. Once the switch is on, driver will report correct information. Signed-off-by: Ike Panhc <ike.pan@canonical.com> --- drivers/platform/x86/ideapad-laptop.c | 5 +++++ 1 file changed, 5 insertions(+)