From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andy Shevchenko Date: Tue, 18 Feb 2020 11:36:55 +0200 Subject: [PATCH v1] Revert "x86: use invd instead of wbinvd in real mode start code" In-Reply-To: References: <20200217153012.14989-1-andriy.shevchenko@linux.intel.com> Message-ID: <20200218093655.GG10400@smile.fi.intel.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Tue, Feb 18, 2020 at 09:45:34AM +0900, Masahiro Yamada wrote: > On Tue, Feb 18, 2020 at 12:30 AM Andy Shevchenko > wrote: > > > > This reverts commit 0d67fac29f3187e67f4fd3ef15f73e91be2fad12. > > > > As real hardware testing (*) shows the above mentioned commit > > breaks U-Boot on it. Revert for the upcoming release. We may get > > more information in the future and optimize the code accordingly. > > > > (*) om Intel Edison board. > Reverting arch/x86/cpu/start.S is enough, isn't it? Had you tested on real hardware? Since Qemu ~= "no test" for the available hardware, I can't answer to this. > arch/x86/cpu/start16.S is not compiled for the > Edison board. > > start16.S is the start of the Real Mode, > and we are sure no program was running before > U-Boot. To be on the safer side I prefer to revert completely. Then you may come up with a new patch with better commit message and testing. To me it sounds fair. -- With Best Regards, Andy Shevchenko