| Message ID | 4A666796.4020100@jp.fujitsu.com (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Headers | show |
On Wed, Jul 22, 2009 at 09:12:54AM +0800, Kenji Kaneshige wrote: > We must not clear bits in 'aspm_enabled' using 'aspm_support', or > 'aspm_enabled' and 'aspm_default' might be different from the actual > state. In addtion, 'aspm_default' should be intialized even if > 'aspm_support' is 0. In original implementation with aspm_default and aspm_support can help us check if BIOS wrongly enables aspm. Do we still have such check with this change? Thanks, Shaohua -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Shaohua Li wrote: > On Wed, Jul 22, 2009 at 09:12:54AM +0800, Kenji Kaneshige wrote: >> We must not clear bits in 'aspm_enabled' using 'aspm_support', or >> 'aspm_enabled' and 'aspm_default' might be different from the actual >> state. In addtion, 'aspm_default' should be intialized even if >> 'aspm_support' is 0. > In original implementation with aspm_default and aspm_support can help us check > if BIOS wrongly enables aspm. Do we still have such check with this change? Yes. This patch is to fixes this check. In the current implementation, I don't think this check works properly because aspm_enable bit is cleared using aspm_support. ASPM driver disables ASPM only when it is set in aspm_enabled and it is cleared in specified state. So we must not clear aspm_enable using aspm_support. In addition, we don't need to clear bits in aspm_default for this check. I think this just make confusion. Thanks, Kenji Kaneshige -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Index: 20090721/drivers/pci/pcie/aspm.c =================================================================== --- 20090721.orig/drivers/pci/pcie/aspm.c +++ 20090721/drivers/pci/pcie/aspm.c @@ -346,12 +346,12 @@ static void pcie_aspm_cap_init(struct pc link->latency.l0s = max_t(u32, link->latency.l0s, l0s); link->latency.l1 = max_t(u32, link->latency.l1, l1); + /* Save default state */ + link->aspm_default = link->aspm_enabled; + if (!link->aspm_support) return; - link->aspm_enabled &= link->aspm_support; - link->aspm_default = link->aspm_enabled; - /* ENDPOINT states*/ list_for_each_entry(child, &linkbus->devices, bus_list) { int pos;
We must not clear bits in 'aspm_enabled' using 'aspm_support', or 'aspm_enabled' and 'aspm_default' might be different from the actual state. In addtion, 'aspm_default' should be intialized even if 'aspm_support' is 0. Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com> --- drivers/pci/pcie/aspm.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html