Messages in this thread | | | Date | Fri, 8 Sep 2017 11:28:27 +0200 | From | Pavel Machek <> | Subject | Re: [Intel-wired-lan] [PATCH] e1000e: changed some expensive calls of udelay to usleep_range |
| |
On Thu 2017-09-07 22:19:47, Brown, Aaron F wrote: > > From: Intel-wired-lan [mailto:intel-wired-lan-bounces@osuosl.org] On > > Behalf Of Pavel Machek > > Sent: Monday, September 4, 2017 9:26 AM > > To: Matthew Tan <matthew.tan_1@nxp.com> > > Cc: michael.kardonik@nxp.com; Williams, Mitch A > > <mitch.a.williams@intel.com>; linux-kernel@vger.kernel.org; > > john.ronciak@intel.com; intel-wired-lan@lists.osuosl.org; > > netdev@vger.kernel.org > > Subject: Re: [Intel-wired-lan] [PATCH] e1000e: changed some expensive calls > > of udelay to usleep_range > > > > Hi! > > > > > @@ -183,7 +183,7 @@ s32 e1000e_read_phy_reg_mdic(struct e1000_hw > > *hw, u32 offset, u16 *data) > > > * reading duplicate data in the next MDIC transaction. > > > */ > > > if (hw->mac.type == e1000_pch2lan) > > > - udelay(100); > > > + usleep_range(90, 100); > > > > > > return 0; > > > } > > > > Can you explain why shortening the delay is acceptable here? > > Maybe it's not. > > This patch is causing speed / duplex tests to fail on several of my test systems. Specifically a Lenova laptop with an 82577 and a NUC with an i218 (though that does not mean it is limited to those or that it's not related to the individual link partner.) >
Ok, this should be quite easy to verify -- just adjust all the ranges to be >= original ones.
Thanks, Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html [unhandled content-type:application/pgp-signature] | |