* Development practices? @ 2015-09-03 19:34 Konrad Rzeszutek Wilk 2015-09-03 20:27 ` Colin Watson 2015-09-08 17:57 ` Konrad Rzeszutek Wilk 0 siblings, 2 replies; 13+ messages in thread From: Konrad Rzeszutek Wilk @ 2015-09-03 19:34 UTC (permalink / raw) To: grub-devel Hey, I've noticed a couple of things that the community seems to be doing different than other open source projects and was wondering why? And also how to conform to this (or perhaps it is time to change?) so that ideas/releases/patches can move forward. - Patches are not posted by committers. I see some patches that appear out of nowhere in the grub source code but they don't seem to be posted on the mailing list? Is that because there hadn't been enough reviewers on the mailing list? Which brings another question: - Some patches have been posted and hadn't much any architecture feedback or review. I am referring to Daniel's multboot2 extensions and the ARM multiboot implementation for example. Is that due to the tradition (not sure if that is the right word) that there aren't enough reviewers so folks shouldn't expect reviews? - Some patches that are in the git tree don't seem to have the Signed-Off-By which I find odd (*1). Patches that come from non-maintainers have them, but the maintainers/committers don't always? Is that because the committers have signed some form of 'implicit-Signed-off-by-when-I-check-in' document? I don't know enough about the community (or the history) to understand it but would very much appreciate input. And if I have offended somebody with my questions + feeble analysis: my deepest apologies - and please straighten me out! From what I have gathered so far the not enough reviewers is tied in folks being overworked - so there simply was no point of posting on the mailing list as nobody had the time to review it or test it properly? [*1]: My background is in Linux kernel and Xen Project. ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-03 19:34 Development practices? Konrad Rzeszutek Wilk @ 2015-09-03 20:27 ` Colin Watson 2015-09-08 17:57 ` Konrad Rzeszutek Wilk 1 sibling, 0 replies; 13+ messages in thread From: Colin Watson @ 2015-09-03 20:27 UTC (permalink / raw) To: grub-devel I won't speak for all of these, but: On Thu, Sep 03, 2015 at 03:34:29PM -0400, Konrad Rzeszutek Wilk wrote: > - Patches are not posted by committers. I see some patches that > appear out of nowhere in the grub source code but they > don't seem to be posted on the mailing list? Is that because > there hadn't been enough reviewers on the mailing list? > Which brings another question: Historically the point of being a committer was that you were trusted to exercise judgement, and would post things for review if you felt they required it but not necessarily in all cases. > - Some patches that are in the git tree don't seem to have the > Signed-Off-By which I find odd (*1). Patches that come from > non-maintainers have them, but the maintainers/committers don't > always? Is that because the committers have signed some form > of 'implicit-Signed-off-by-when-I-check-in' document? As far as I know S-O-B has never been standard in GRUB. It's a practice that many people who are familiar with Linux kernel development follow, and so it ends up in our commit logs, but that doesn't mean it's standard hereabouts nor in many other projects. -- Colin Watson [cjwatson@ubuntu.com] ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-03 19:34 Development practices? Konrad Rzeszutek Wilk 2015-09-03 20:27 ` Colin Watson @ 2015-09-08 17:57 ` Konrad Rzeszutek Wilk 2015-09-09 17:47 ` Felix Zielcke 1 sibling, 1 reply; 13+ messages in thread From: Konrad Rzeszutek Wilk @ 2015-09-08 17:57 UTC (permalink / raw) To: grub-devel, Andrei Borzenkov, Leif Lindholm, Vladimir Serbinenko, Felix Zielcke, Paulo Flabiano Smorigo On Thu, Sep 03, 2015 at 03:34:29PM -0400, Konrad Rzeszutek Wilk wrote: > Hey, Expanding the circle, including folks on the To list. Colin had answered some so snipping thouse out. > > I've noticed a couple of things that the community seems > to be doing different than other open source projects and > was wondering why? And also how to conform to this (or perhaps > it is time to change?) so that ideas/releases/patches can move forward. > .. snip.. > > - Some patches have been posted and hadn't much any architecture > feedback or review. I am referring to Daniel's multboot2 > extensions and the ARM multiboot implementation for example. > > Is that due to the tradition (not sure if that is the right > word) that there aren't enough reviewers so folks shouldn't > expect reviews? > .. snip.. > > > I don't know enough about the community (or the history) to > understand it but would very much appreciate input. > And if I have offended somebody with my questions + feeble > analysis: my deepest apologies - and please straighten me out! > > > From what I have gathered so far the not enough reviewers > is tied in folks being overworked - so there simply was no > point of posting on the mailing list as nobody had the time > to review it or test it properly? > > [*1]: My background is in Linux kernel and Xen Project. ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-08 17:57 ` Konrad Rzeszutek Wilk @ 2015-09-09 17:47 ` Felix Zielcke 2015-09-11 14:34 ` Andrei Borzenkov 0 siblings, 1 reply; 13+ messages in thread From: Felix Zielcke @ 2015-09-09 17:47 UTC (permalink / raw) To: grub-devel Am Dienstag, den 08.09.2015, 13:57 -0400 schrieb Konrad Rzeszutek Wilk: > On Thu, Sep 03, 2015 at 03:34:29PM -0400, Konrad Rzeszutek Wilk > wrote: > I don't know enough about the community (or the history) to > > understand it but would very much appreciate input. > > And if I have offended somebody with my questions + feeble > > analysis: my deepest apologies - and please straighten me out! > > > > > > From what I have gathered so far the not enough reviewers > > is tied in folks being overworked - so there simply was no > > point of posting on the mailing list as nobody had the time > > to review it or test it properly? > > Hi Konrad, back in 2008/2009 (when Marco Gerards gave over Maintainance to Robert Millan) there were indeed not much people actively reviewing code. Active people on the mailing list was just given commit access. It was expected that they only commit stuff without posting which doesn't need a review and complies with the rules back at that time. Due to me missing a few years on the mailing list, I can't tell you unfortunately how it compares to today. ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-09 17:47 ` Felix Zielcke @ 2015-09-11 14:34 ` Andrei Borzenkov 2015-09-11 14:48 ` Konrad Rzeszutek Wilk 0 siblings, 1 reply; 13+ messages in thread From: Andrei Borzenkov @ 2015-09-11 14:34 UTC (permalink / raw) To: grub-devel 09.09.2015 20:47, Felix Zielcke пишет: > Am Dienstag, den 08.09.2015, 13:57 -0400 schrieb Konrad Rzeszutek Wilk: >> On Thu, Sep 03, 2015 at 03:34:29PM -0400, Konrad Rzeszutek Wilk >> wrote: > >> I don't know enough about the community (or the history) to >>> understand it but would very much appreciate input. >>> And if I have offended somebody with my questions + feeble >>> analysis: my deepest apologies - and please straighten me out! >>> >>> >>> From what I have gathered so far the not enough reviewers >>> is tied in folks being overworked - so there simply was no >>> point of posting on the mailing list as nobody had the time >>> to review it or test it properly? >>> > > Hi Konrad, > > back in 2008/2009 (when Marco Gerards gave over Maintainance to Robert > Millan) there were indeed not much people actively reviewing code. > > Active people on the mailing list was just given commit access. It was > expected that they only commit stuff without posting which doesn't need > a review and complies with the rules back at that time. > > Due to me missing a few years on the mailing list, I can't tell you > unfortunately how it compares to today. > Not much changes as far as I can tell. ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-11 14:34 ` Andrei Borzenkov @ 2015-09-11 14:48 ` Konrad Rzeszutek Wilk 2015-09-11 15:16 ` Andrei Borzenkov 0 siblings, 1 reply; 13+ messages in thread From: Konrad Rzeszutek Wilk @ 2015-09-11 14:48 UTC (permalink / raw) To: The development of GNU GRUB, arvidjaar On Fri, Sep 11, 2015 at 05:34:45PM +0300, Andrei Borzenkov wrote: > 09.09.2015 20:47, Felix Zielcke пишет: > >Am Dienstag, den 08.09.2015, 13:57 -0400 schrieb Konrad Rzeszutek Wilk: > >>On Thu, Sep 03, 2015 at 03:34:29PM -0400, Konrad Rzeszutek Wilk > >>wrote: > > > >>I don't know enough about the community (or the history) to > >>>understand it but would very much appreciate input. > >>>And if I have offended somebody with my questions + feeble > >>>analysis: my deepest apologies - and please straighten me out! > >>> > >>> > >>> From what I have gathered so far the not enough reviewers > >>>is tied in folks being overworked - so there simply was no > >>>point of posting on the mailing list as nobody had the time > >>>to review it or test it properly? > >>> > > > >Hi Konrad, > > > >back in 2008/2009 (when Marco Gerards gave over Maintainance to Robert > >Millan) there were indeed not much people actively reviewing code. > > > >Active people on the mailing list was just given commit access. It was > >expected that they only commit stuff without posting which doesn't need > >a review and complies with the rules back at that time. > > > >Due to me missing a few years on the mailing list, I can't tell you > >unfortunately how it compares to today. > > > > Not much changes as far as I can tell. OK. What qualifies as needing an review? Personal preference by the patch author? Thank you for answering my questions! ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-11 14:48 ` Konrad Rzeszutek Wilk @ 2015-09-11 15:16 ` Andrei Borzenkov 2015-09-11 15:34 ` Konrad Rzeszutek Wilk 0 siblings, 1 reply; 13+ messages in thread From: Andrei Borzenkov @ 2015-09-11 15:16 UTC (permalink / raw) To: Konrad Rzeszutek Wilk, The development of GNU GRUB 11.09.2015 17:48, Konrad Rzeszutek Wilk пишет: > On Fri, Sep 11, 2015 at 05:34:45PM +0300, Andrei Borzenkov wrote: >> 09.09.2015 20:47, Felix Zielcke пишет: >>> Am Dienstag, den 08.09.2015, 13:57 -0400 schrieb Konrad Rzeszutek Wilk: >>>> On Thu, Sep 03, 2015 at 03:34:29PM -0400, Konrad Rzeszutek Wilk >>>> wrote: >>> >>>> I don't know enough about the community (or the history) to >>>>> understand it but would very much appreciate input. >>>>> And if I have offended somebody with my questions + feeble >>>>> analysis: my deepest apologies - and please straighten me out! >>>>> >>>>> >>>>> From what I have gathered so far the not enough reviewers >>>>> is tied in folks being overworked - so there simply was no >>>>> point of posting on the mailing list as nobody had the time >>>>> to review it or test it properly? >>>>> >>> >>> Hi Konrad, >>> >>> back in 2008/2009 (when Marco Gerards gave over Maintainance to Robert >>> Millan) there were indeed not much people actively reviewing code. >>> >>> Active people on the mailing list was just given commit access. It was >>> expected that they only commit stuff without posting which doesn't need >>> a review and complies with the rules back at that time. >>> >>> Due to me missing a few years on the mailing list, I can't tell you >>> unfortunately how it compares to today. >>> >> >> Not much changes as far as I can tell. > > OK. > > What qualifies as needing an review? Personal preference by > the patch author? > I suppose, common sense. When I was given commit access, it was for "committing after review" so I still sent all patches to the list. Then it happened that Vladimir dropped off list for a long time and I tried to pick up obvious bug fixes from list or bug tracker to keep things going. I would say, any non-trivial bug fix or feature change needs to be posted first. I would love to have every patch posted and reviewed bug given current level of activity it is simply unrealistic. > Thank you for answering my questions! > ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-11 15:16 ` Andrei Borzenkov @ 2015-09-11 15:34 ` Konrad Rzeszutek Wilk 2015-09-22 18:28 ` Konrad Rzeszutek Wilk 0 siblings, 1 reply; 13+ messages in thread From: Konrad Rzeszutek Wilk @ 2015-09-11 15:34 UTC (permalink / raw) To: Andrei Borzenkov; +Cc: The development of GNU GRUB . snip.. > >>>>> From what I have gathered so far the not enough reviewers > >>>>>is tied in folks being overworked - so there simply was no > >>>>>point of posting on the mailing list as nobody had the time > >>>>>to review it or test it properly? > >>>>> > >>> > >>>Hi Konrad, > >>> > >>>back in 2008/2009 (when Marco Gerards gave over Maintainance to Robert > >>>Millan) there were indeed not much people actively reviewing code. > >>> > >>>Active people on the mailing list was just given commit access. It was > >>>expected that they only commit stuff without posting which doesn't need > >>>a review and complies with the rules back at that time. > >>> > >>>Due to me missing a few years on the mailing list, I can't tell you > >>>unfortunately how it compares to today. > >>> > >> > >>Not much changes as far as I can tell. > > > >OK. > > > >What qualifies as needing an review? Personal preference by > >the patch author? > > > > I suppose, common sense. When I was given commit access, it was for > "committing after review" so I still sent all patches to the list. Then it > happened that Vladimir dropped off list for a long time and I tried to pick > up obvious bug fixes from list or bug tracker to keep things going. > > I would say, any non-trivial bug fix or feature change needs to be posted > first. > > I would love to have every patch posted and reviewed bug given current level > of activity it is simply unrealistic. I see. From my perspective we are paid to work on the hobbies (Xen, Linux, etc) so the activity level is high since we have 8 hours a day to focus on it (minus bug activities, lunch, etc). While GRUB2 is all volunteer with whatever time can be spared? What if the companies that employ the committers allowed one day a week to focus on GRUB2 review/maintaince/etc? Would that help? Or is it unrealistic to expect that from committers employer's? > > >Thank you for answering my questions! > > > ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-11 15:34 ` Konrad Rzeszutek Wilk @ 2015-09-22 18:28 ` Konrad Rzeszutek Wilk 2015-09-24 18:42 ` Felix Zielcke 2015-09-24 19:09 ` Andrei Borzenkov 0 siblings, 2 replies; 13+ messages in thread From: Konrad Rzeszutek Wilk @ 2015-09-22 18:28 UTC (permalink / raw) To: Andrei Borzenkov, Leif Lindholm, Vladimir Serbinenko, Felix Zielcke, Paulo Flabiano Smorigo Cc: The development of GNU GRUB On Fri, Sep 11, 2015 at 11:34:53AM -0400, Konrad Rzeszutek Wilk wrote: > .. snip.. > > >>>>> From what I have gathered so far the not enough reviewers > > >>>>>is tied in folks being overworked - so there simply was no > > >>>>>point of posting on the mailing list as nobody had the time > > >>>>>to review it or test it properly? > > >>>>> > > >>> > > >>>Hi Konrad, > > >>> > > >>>back in 2008/2009 (when Marco Gerards gave over Maintainance to Robert > > >>>Millan) there were indeed not much people actively reviewing code. > > >>> > > >>>Active people on the mailing list was just given commit access. It was > > >>>expected that they only commit stuff without posting which doesn't need > > >>>a review and complies with the rules back at that time. > > >>> > > >>>Due to me missing a few years on the mailing list, I can't tell you > > >>>unfortunately how it compares to today. > > >>> > > >> > > >>Not much changes as far as I can tell. > > > > > >OK. > > > > > >What qualifies as needing an review? Personal preference by > > >the patch author? > > > > > > > I suppose, common sense. When I was given commit access, it was for > > "committing after review" so I still sent all patches to the list. Then it > > happened that Vladimir dropped off list for a long time and I tried to pick > > up obvious bug fixes from list or bug tracker to keep things going. > > > > I would say, any non-trivial bug fix or feature change needs to be posted > > first. > > > > I would love to have every patch posted and reviewed bug given current level > > of activity it is simply unrealistic. > > I see. From my perspective we are paid to work on the hobbies (Xen, Linux, etc) > so the activity level is high since we have 8 hours a day to focus on it > (minus bug activities, lunch, etc). > > While GRUB2 is all volunteer with whatever time can be spared? > > What if the companies that employ the committers allowed one day a week > to focus on GRUB2 review/maintaince/etc? Would that help? > > Or is it unrealistic to expect that from committers employer's? > ping? > > > > >Thank you for answering my questions! > > > > > ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-22 18:28 ` Konrad Rzeszutek Wilk @ 2015-09-24 18:42 ` Felix Zielcke 2015-09-24 19:09 ` Andrei Borzenkov 1 sibling, 0 replies; 13+ messages in thread From: Felix Zielcke @ 2015-09-24 18:42 UTC (permalink / raw) To: grub-devel Am Dienstag, den 22.09.2015, 14:28 -0400 schrieb Konrad Rzeszutek Wilk: > > What if the companies that employ the committers allowed one day a > week > > to focus on GRUB2 review/maintaince/etc? Would that help? > > > > Or is it unrealistic to expect that from committers employer's? > > > > ping? Well my company only pays me for my in-firm training I make there. Due to the fact I'm not a full employee, I don't think I need to ask if I get some time paid for doing general Debian/GNU stuff. Instead of stuff with which the company earns its money ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-22 18:28 ` Konrad Rzeszutek Wilk 2015-09-24 18:42 ` Felix Zielcke @ 2015-09-24 19:09 ` Andrei Borzenkov 2015-09-24 19:27 ` Konrad Rzeszutek Wilk 1 sibling, 1 reply; 13+ messages in thread From: Andrei Borzenkov @ 2015-09-24 19:09 UTC (permalink / raw) To: Konrad Rzeszutek Wilk Cc: Vladimir Serbinenko, Paulo Flabiano Smorigo, Felix Zielcke, Leif Lindholm, The development of GNU GRUB Отправлено с iPhone > 22 сент. 2015 г., в 20:28, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> написал(а): > >> On Fri, Sep 11, 2015 at 11:34:53AM -0400, Konrad Rzeszutek Wilk wrote: >> .. snip.. >>>>>>>> From what I have gathered so far the not enough reviewers >>>>>>>> is tied in folks being overworked - so there simply was no >>>>>>>> point of posting on the mailing list as nobody had the time >>>>>>>> to review it or test it properly? >>>>>> >>>>>> Hi Konrad, >>>>>> >>>>>> back in 2008/2009 (when Marco Gerards gave over Maintainance to Robert >>>>>> Millan) there were indeed not much people actively reviewing code. >>>>>> >>>>>> Active people on the mailing list was just given commit access. It was >>>>>> expected that they only commit stuff without posting which doesn't need >>>>>> a review and complies with the rules back at that time. >>>>>> >>>>>> Due to me missing a few years on the mailing list, I can't tell you >>>>>> unfortunately how it compares to today. >>>>> >>>>> Not much changes as far as I can tell. >>>> >>>> OK. >>>> >>>> What qualifies as needing an review? Personal preference by >>>> the patch author? >>> >>> I suppose, common sense. When I was given commit access, it was for >>> "committing after review" so I still sent all patches to the list. Then it >>> happened that Vladimir dropped off list for a long time and I tried to pick >>> up obvious bug fixes from list or bug tracker to keep things going. >>> >>> I would say, any non-trivial bug fix or feature change needs to be posted >>> first. >>> >>> I would love to have every patch posted and reviewed bug given current level >>> of activity it is simply unrealistic. >> >> I see. From my perspective we are paid to work on the hobbies (Xen, Linux, etc) >> so the activity level is high since we have 8 hours a day to focus on it >> (minus bug activities, lunch, etc). >> >> While GRUB2 is all volunteer with whatever time can be spared? >> >> What if the companies that employ the committers allowed one day a week >> to focus on GRUB2 review/maintaince/etc? Would that help? >> >> Or is it unrealistic to expect that from committers employer's? > > ping? You realize that commiters' employers most likely do no read this list, right? >>> >>>> Thank you for answering my questions! >>> ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Development practices? 2015-09-24 19:09 ` Andrei Borzenkov @ 2015-09-24 19:27 ` Konrad Rzeszutek Wilk 2015-09-29 17:16 ` Sun, Ning 0 siblings, 1 reply; 13+ messages in thread From: Konrad Rzeszutek Wilk @ 2015-09-24 19:27 UTC (permalink / raw) To: Andrei Borzenkov Cc: Vladimir Serbinenko, Paulo Flabiano Smorigo, Felix Zielcke, Leif Lindholm, The development of GNU GRUB On September 24, 2015 3:09:20 PM EDT, Andrei Borzenkov <arvidjaar@gmail.com> wrote: > > >Отправлено с iPhone > >> 22 сент. 2015 г., в 20:28, Konrad Rzeszutek Wilk ><konrad.wilk@oracle.com> написал(а): >> >>> On Fri, Sep 11, 2015 at 11:34:53AM -0400, Konrad Rzeszutek Wilk >wrote: >>> .. snip.. >>>>>>>>> From what I have gathered so far the not enough reviewers >>>>>>>>> is tied in folks being overworked - so there simply was no >>>>>>>>> point of posting on the mailing list as nobody had the time >>>>>>>>> to review it or test it properly? >>>>>>> >>>>>>> Hi Konrad, >>>>>>> >>>>>>> back in 2008/2009 (when Marco Gerards gave over Maintainance to >Robert >>>>>>> Millan) there were indeed not much people actively reviewing >code. >>>>>>> >>>>>>> Active people on the mailing list was just given commit access. >It was >>>>>>> expected that they only commit stuff without posting which >doesn't need >>>>>>> a review and complies with the rules back at that time. >>>>>>> >>>>>>> Due to me missing a few years on the mailing list, I can't tell >you >>>>>>> unfortunately how it compares to today. >>>>>> >>>>>> Not much changes as far as I can tell. >>>>> >>>>> OK. >>>>> >>>>> What qualifies as needing an review? Personal preference by >>>>> the patch author? >>>> >>>> I suppose, common sense. When I was given commit access, it was for >>>> "committing after review" so I still sent all patches to the list. >Then it >>>> happened that Vladimir dropped off list for a long time and I tried >to pick >>>> up obvious bug fixes from list or bug tracker to keep things going. >>>> >>>> I would say, any non-trivial bug fix or feature change needs to be >posted >>>> first. >>>> >>>> I would love to have every patch posted and reviewed bug given >current level >>>> of activity it is simply unrealistic. >>> >>> I see. From my perspective we are paid to work on the hobbies (Xen, >Linux, etc) >>> so the activity level is high since we have 8 hours a day to focus >on it >>> (minus bug activities, lunch, etc). >>> >>> While GRUB2 is all volunteer with whatever time can be spared? >>> >>> What if the companies that employ the committers allowed one day a >week >>> to focus on GRUB2 review/maintaince/etc? Would that help? >>> >>> Or is it unrealistic to expect that from committers employer's? >> >> ping? > >You realize that commiters' employers most likely do no read this list, >right? > True, but I hoped that the commiters's would forward this to their managers. May I assume from your email that if you had one day a week it would allow you to make much more progress on reviews, commits and a release? > >>>> >>>>> Thank you for answering my questions! >>>> ^ permalink raw reply [flat|nested] 13+ messages in thread
* RE: Development practices? 2015-09-24 19:27 ` Konrad Rzeszutek Wilk @ 2015-09-29 17:16 ` Sun, Ning 0 siblings, 0 replies; 13+ messages in thread From: Sun, Ning @ 2015-09-29 17:16 UTC (permalink / raw) To: The development of GNU GRUB, Andrei Borzenkov Cc: Leif Lindholm, Vladimir Serbinenko, Felix Zielcke, Paulo Flabiano Smorigo We need better documentation so as to let more engineers understand the status quo quickly before they can jump into as reviewer. -----Original Message----- From: grub-devel-bounces+ning.sun=intel.com@gnu.org [mailto:grub-devel-bounces+ning.sun=intel.com@gnu.org] On Behalf Of Konrad Rzeszutek Wilk Sent: Thursday, September 24, 2015 12:28 PM To: Andrei Borzenkov Cc: Vladimir Serbinenko; Paulo Flabiano Smorigo; Felix Zielcke; Leif Lindholm; The development of GNU GRUB Subject: Re: Development practices? On September 24, 2015 3:09:20 PM EDT, Andrei Borzenkov <arvidjaar@gmail.com> wrote: > > >Отправлено с iPhone > >> 22 сент. 2015 г., в 20:28, Konrad Rzeszutek Wilk ><konrad.wilk@oracle.com> написал(а): >> >>> On Fri, Sep 11, 2015 at 11:34:53AM -0400, Konrad Rzeszutek Wilk >wrote: >>> .. snip.. >>>>>>>>> From what I have gathered so far the not enough reviewers is >>>>>>>>> tied in folks being overworked - so there simply was no point >>>>>>>>> of posting on the mailing list as nobody had the time to >>>>>>>>> review it or test it properly? >>>>>>> >>>>>>> Hi Konrad, >>>>>>> >>>>>>> back in 2008/2009 (when Marco Gerards gave over Maintainance to >Robert >>>>>>> Millan) there were indeed not much people actively reviewing >code. >>>>>>> >>>>>>> Active people on the mailing list was just given commit access. >It was >>>>>>> expected that they only commit stuff without posting which >doesn't need >>>>>>> a review and complies with the rules back at that time. >>>>>>> >>>>>>> Due to me missing a few years on the mailing list, I can't tell >you >>>>>>> unfortunately how it compares to today. >>>>>> >>>>>> Not much changes as far as I can tell. >>>>> >>>>> OK. >>>>> >>>>> What qualifies as needing an review? Personal preference by the >>>>> patch author? >>>> >>>> I suppose, common sense. When I was given commit access, it was for >>>> "committing after review" so I still sent all patches to the list. >Then it >>>> happened that Vladimir dropped off list for a long time and I tried >to pick >>>> up obvious bug fixes from list or bug tracker to keep things going. >>>> >>>> I would say, any non-trivial bug fix or feature change needs to be >posted >>>> first. >>>> >>>> I would love to have every patch posted and reviewed bug given >current level >>>> of activity it is simply unrealistic. >>> >>> I see. From my perspective we are paid to work on the hobbies (Xen, >Linux, etc) >>> so the activity level is high since we have 8 hours a day to focus >on it >>> (minus bug activities, lunch, etc). >>> >>> While GRUB2 is all volunteer with whatever time can be spared? >>> >>> What if the companies that employ the committers allowed one day a >week >>> to focus on GRUB2 review/maintaince/etc? Would that help? >>> >>> Or is it unrealistic to expect that from committers employer's? >> >> ping? > >You realize that commiters' employers most likely do no read this list, >right? > True, but I hoped that the commiters's would forward this to their managers. May I assume from your email that if you had one day a week it would allow you to make much more progress on reviews, commits and a release? > >>>> >>>>> Thank you for answering my questions! >>>> _______________________________________________ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel ^ permalink raw reply [flat|nested] 13+ messages in thread
end of thread, other threads:[~2015-10-08 6:02 UTC | newest] Thread overview: 13+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2015-09-03 19:34 Development practices? Konrad Rzeszutek Wilk 2015-09-03 20:27 ` Colin Watson 2015-09-08 17:57 ` Konrad Rzeszutek Wilk 2015-09-09 17:47 ` Felix Zielcke 2015-09-11 14:34 ` Andrei Borzenkov 2015-09-11 14:48 ` Konrad Rzeszutek Wilk 2015-09-11 15:16 ` Andrei Borzenkov 2015-09-11 15:34 ` Konrad Rzeszutek Wilk 2015-09-22 18:28 ` Konrad Rzeszutek Wilk 2015-09-24 18:42 ` Felix Zielcke 2015-09-24 19:09 ` Andrei Borzenkov 2015-09-24 19:27 ` Konrad Rzeszutek Wilk 2015-09-29 17:16 ` Sun, Ning
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.