>-----Original Message----- >From: gregkh@xxxxxxxxxxxxxxxxxxx [mailto:gregkh@xxxxxxxxxxxxxxxxxxx] >Sent: Friday, February 2, 2018 9:15 PM >To: Shaikh, Azhar <azhar.shaikh@xxxxxxxxx> >Cc: jarkko.sakkinen@xxxxxxxxxxxxxxx; stable@xxxxxxxxxxxxxxx >Subject: Re: patch "[PATCH] tpm_tis: Move ilb_base_addr to tpm_tis_data" >was seriously submitted to be applied to the 4.15-stable tree? > >On Fri, Feb 02, 2018 at 06:24:16PM +0000, Shaikh, Azhar wrote: >> Hi Greg, >> >> >> [1] is merged on the 4.15 stable tree. But this caused 2 issues: >> i. PS/2 keyboard and mouse were broken on an Intel Braswell system. The >fix for this is [4]. >> ii. There were some corner cases where TPM suspend was failing on >Braswell systems. The fix for this is [2] and [3]. >> >> The patch pointed out by you is [2]. >> [3] is dependent on [2]. So to fix the issue on stable tree we will need [2], >[3] and [4]. >> >> [4] is not there yet on Linus's branch. I think it is queued for 4.16. >> >> [5] fixes compile time warning introduced in [3]. > >This email was crazy difficult to read and understand :( > Really sorry for all the confusion :( >Please, footnotes in emails are for things you can go back and look at later, >after reading the body of a message, not to try to understand exactly what in >the world is going on and what I am supposed to do with something. > >In the future, just put the git sha1 in the email text, no need to link to >anything, we all have zillions of copies of the kernel source on our own >machines. And sometimes we do not have reliable web access (emails works >great as store and forward.) > >Just use the format: > c382babccba2 ("tpm_tis: Move ilb_base_addr to tpm_tis_data") to >describe a commit id, which can be easily obtained by doing: > git show -s --abbrev-commit --abbrev=12 --pretty=format:"%h >(\"%s\")%n"' >(hint, I have an alias for it so I don't have to ever type that again). > >So, with all of that, I still have no idea what I am supposed to do here. The >patch that triggered this email is obviously not a stable tree patch, as all it does >is move code around, right? Is there something else I should be doing here? >What action do you want me to take? > Yes you are right, the patch pointed out by you, just moves the code around. But this patch, alongwith other patches mentioned below are needed to fix a corner case issue. Also there is a fix for PS/2 keyboard and mouse broken on Intel Braswell systems. Both the issues were introduced by 5e572cab92f0("tpm: Enable CLKRUN protocol for Braswell systems") - Already present on Linux-stable branch Please pick below commits which are present on Linus's tree to fix corner case issue: 68021bf4734d("tpm: remove unused variables") 764325add6c2("tpm: delete the TPM_TIS_CLK_ENABLE flag") b3e958ce4c58("tpm: Keep CLKRUN enabled throughout the duration of transmit_cmd()") c382babccba2("tpm_tis: Move ilb_base_addr to tpm_tis_data") Below commit is still not present on Linus's tree but merged on TPM maintainer's (Jarkko Sakkinen) tree to fix PS/2 keyboard and mouse issue. git://git.infradead.org/users/jjs/linux-tpmdd.git master branch 6c9f0ce0dffe6("tpm: only attempt to disable the LPC CLKRUN if is already enabled") >totally confused, > Sorry again for the confusion and I hope this time it is more clear. >greg k-h Regards, Azhar Shaikh