IRC logs for #openrisc Tuesday, 2016-11-29

--- Log opened Tue Nov 29 00:00:04 2016
_franck_shorne: men, there's a lot of work ahead to get GDB upstream !05:22
shorne_franck_: yeah, I am really happy Yao replied with all the detail he did05:51
shorneIts all doable.  Wnat to help on any of it?05:52
_franck_at least you know what needs to be corrected.05:52
_franck_well if I can with the little time I have05:52
shorne:) Also, he mention Franck Jullien is missing copyright assignment, I assume thats you? Can you confirm?05:53
_franck_it's me05:53
shorneDid you ever fill out the fsf forms for copyright assignment? are you willing to?05:54
shorneIt might be related to the email address in the particular part of code05:54
_franck_but at some point I think jeremybennett said that as he is the principal author of the port and because he has an assignment we could push code on behalf of him05:54
_franck_you should change my email in the copyright section05:55
_franck_and yes, I'm willing to sign papers05:55
shorne_franck_: ok, understood, Ill mention that in my reply.  But I am not sure it will be sufficient.  I just sent a mail to fsf requesting copyright assignment forms.06:12
shorneOnce I figure out the process Ill let you know06:12
shorneDoes anyone know how to write a multiple author changeLog?06:12
olofkshorne: jeremybennett is a bit of a ChangeLog fanatic06:13
olofk :)06:13
shorneolofk: I could understand why they are used, Answering why on commit messages is important just as much as changelogs.06:20
stekern"But they cannot show you why it was changed." That is what you should write in the commit msg.06:20
shorneI figured out how to do mutli author changelog entries06:20
shornesomething like:06:21
shorne2016-11-29  Stafford Horne  <email@blah>06:21
shorne(tabs?)     Someone Else  <>06:21
stekernisn't there any examples on how to do that in the pre-existing changelogs?06:22
shorne        * file.c: added xyz, because blah.06:22
shorneI couldnt find Examples, I found a program that someone wrote to convert ChangeLog to HTML which had a "new feature" to support multi author changelogs. It had an example06:23
olofkshorne: Is the emacs change-log-mode of any help?06:56
shorneolofk: I think it would be if I used emacs. I should try it ouf for editing these gdb files.07:24
olofkLooking at the review now. It's a really helpful review07:42
olofkWhat's the deal with matchpoints and gdbserver? Are we ok with dropping that? At least for now07:43
olofkAnd is the readspr deprecated? I have some vague memory that _franck_ did a job to use standard commands for reading out SPRs07:46
_franck_olofk: I also have this vague memory07:48
_franck_I changed this in openocd but there is maybe some left in gdb code07:49
olofkah.. true.07:49
shorneolofk: with gdbserver I think we need to have a build of gdb that runs on openrisc linux. I dont think we have what is needed for that.08:31
shorneDo you know differently?08:32
_franck_shorne: as for the current status you should check our documentation:
shorne_franck_: as per your comment 2013: "blueCmd: you have no chance to compile gdbserver, we don't have or1k-linux-tdec.c ...."08:46
shornebut... it seems blueCmd got it working08:46
shorneBut I dont find the or1k-linux-tdep anywhere08:46
olofkBut we can live without gdbserver for the first patchset I guess?08:46
_franck_we get some WIP from jonas and blueCmd worked from it08:47
shorneolofk: yea, I am hoping so08:47
shorne_franck_: do you know of any repo? I didnt see in the "docs" :)08:48
_franck_I'm looking for it08:48
_franck_I think it was on which seems to be down08:49
shornealright, if anyone finds it let us know, not too important for the first patches.  Also, I figure it can be rebuilt since it runs on ptrace which is pretty stable in our linux port.08:50
_franck_shorne: git://
_franck_that's what I had08:53
shorne_franck_: I see, year that domain is gone, do you have it on any branches you can push somewhere?08:55
_franck_I don't, maybe be blueCmd08:56
shorneno worries08:58
shorneolofk: re matchpoints I am removing, cant see what its for, maybe related to the debug control registers (there are 8 which matches the max matchpoint value)09:13
olofkshorne: I was thinking it could be something like watchpoints15:30
olofkAnyone got Chisel stuff sorted out and can generate the verilog files for me in
olofkWould be a great starting point until I have managed to install the Chisel stuff myself16:06
ZipCPUolofk: You might consider asking someone in the #riscv forum ... ?16:10
shorneolofk_: right thats what I mean, those debug control registers are for setting watchpoints, there are 8.  The code in question also says 8 matchpoints22:21
--- Log closed Wed Nov 30 00:00:06 2016

Generated by 2.15.2 by Marius Gedminas - find it at!