Re: [Mageia-dev] Bugzilla mass ping
On 30/05/12 06:08, Sander Lepik wrote: 30.05.2012 05:14, blind Pete kirjutas: Would it be possible to have a bug report refer to multiple versions and simultaneously have X marked as OLD, Y marked as FIXED, and Z marked as BLOCKING? With new bugzilla we should be able to choose multiple versions. And we can use keywords to handle them. For example: tested_mgaX_arch - validate_update_mgaX - update_pushed_mgaX. -- Sander QA will still need two bugs though for updates (one for each release), we have no plans to incorporate those keywords into our workflow. It would make the validation process too confused and lead to errors. Claire
[Mageia-dev] Tonight's meeting
Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 Cheers -- http://mageia.org
[Mageia-dev] [RFC] radeon firmware in non-free
Hi I think we must discuss the following point for next release: currently radeon firmwares are in non-free. However this break havoc quite a lot of install b/c by default non-free is not enabled. 1) This makes the Mageia experience horrible on some machines (black screen, machines unusable by newbies (and worse unrecoverable by newbies), ...) Newer ATI/AMD cards cannot work w/o the needed firmware. That's also the issue with several network cards. (which makes network installation impossible by default) 2) this is unlike the live-cd where we've those firmwares (and the proprietary drivers as well) 3) This is unlike other pure FOSS distro such as Fedora which includes the firmwares. So there're two kinds of Linux distro: - those who provides free proprietary drivers - those who only provides free drivers, with the needed firmwares Mageia choose to provides either the first experience (the live cd case) or the worst one, the free drivers w/o the needed firmwares. A lot of thing broke havoc when firmwares get split from kernels. According to their licence (or their lack of license), we choose to distribute them in: - either kernel-firmware, always available and always installed - or kernel-firmware-nonfree, unavailable and not installed by default This make a lot of free drivers stop to work by default (at least with the classic installer, it's OK with the livecd) So we offer the choice between: - either be able to choose between working proprietary driver and working free driver - or to have non working free driver This is _INSANE_. I think It's time we reconsider this. Firmwares do not run on the main CPU. They're needed for additional chips where no one will ever ever write a program for. From the main CPU/memory view point they're data. I think we should offer the choice to: - either be able to choose between working proprietary driver and working free driver - or to have _working_ free driver (that is including the needed firmwares) See you
Re: [Mageia-dev] Tonight's meeting
On 30 May 2012 10:30, Anne Nicolas enn...@mageia.org wrote: Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 I suggest the following topic: - including all firmwares by default. other distro do it, we don't this break many free drivers (network, ati/amd, ...) See my previous mail on this ml
Re: [Mageia-dev] Tonight's meeting
Le 30/05/2012 11:00, Thierry Vignaud a écrit : On 30 May 2012 10:30, Anne Nicolasenn...@mageia.org wrote: Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 I suggest the following topic: - including all firmwares by default. other distro do it, we don't this break many free drivers (network, ati/amd, ...) See my previous mail on this ml ok but it would be nice if you can attend this meeting. -- http://mageia.org
[Mageia-dev] [RFC] x11-server-1.12
Hi x11-server-1.12 and rebuild drivers are in */update_testing. Please test. The following drivers failed to rebuild: - virtualbox - x11-driver-input-synaptics - x11-driver-video-chips - x11-driver-video-nouveau - x11-driver-video-rendition - x11-driver-video-sisimedia - x11-driver-video-sisusb - x11-driver-video-tseng - x11-driver-video-xgi Several of those have been ported to new x11-server-1.12 ABI in git but haven't yet been released (eg: chips, rendition, tseng) eg: we need to package a newer snapshot of nouveau Also I'ven't yet tried t orebuild several minor/obsolete input drivers. Please test. See you
Re: [Mageia-dev] [RFC] radeon firmware in non-free
30.05.2012 11:38, Thierry Vignaud skrev: Hi I think we must discuss the following point for next release: currently radeon firmwares are in non-free. However this break havoc quite a lot of install b/c by default non-free is not enabled. So its an installer issue. 1) This makes the Mageia experience horrible on some machines (black screen, machines unusable by newbies (and worse unrecoverable by newbies), ...) Newer ATI/AMD cards cannot work w/o the needed firmware. That's also the issue with several network cards. (which makes network installation impossible by default) So people should think twice about what hw they want to buy/use. 2) this is unlike the live-cd where we've those firmwares (and the proprietary drivers as well) This is because you cant install rpms on live media... (or you can, but as the change is gone with a reboot, there is not much point) 3) This is unlike other pure FOSS distro such as Fedora which includes the firmwares. Well, that's their choice, not ours... So there're two kinds of Linux distro: - those who provides free proprietary drivers - those who only provides free drivers, with the needed firmwares Mageia choose to provides either the first experience (the live cd case) or the worst one, the free drivers w/o the needed firmwares. A lot of thing broke havoc when firmwares get split from kernels. According to their licence (or their lack of license), we choose to distribute them in: - either kernel-firmware, always available and always installed - or kernel-firmware-nonfree, unavailable and not installed by default This make a lot of free drivers stop to work by default (at least with the classic installer, it's OK with the livecd) So we offer the choice between: - either be able to choose between working proprietary driver and working free driver - or to have non working free driver This is _INSANE_. I think It's time we reconsider this. Firmwares do not run on the main CPU. They're needed for additional chips where no one will ever ever write a program for. From the main CPU/memory view point they're data. But not from our perspective about free-only. I think we should offer the choice to: - either be able to choose between working proprietary driver and working free driver - or to have _working_ free driver (that is including the needed firmwares) As pointed out in earlier discussions about this issue... There is 2 points needing fix. - the installer should detect the need for nonfree, and ask if people want to enable it. (or simply ask if user want to add nonfree media) - the other thing to fix is to change the installer option about additional CD to additional media and allow for browsing for additional rpms on a cd/dvd/disk/usb/... media. -- Thomas
Re: [Mageia-dev] Tonight's meeting
On 30 May 2012 11:06, Anne Nicolas enn...@mageia.org wrote: Le 30/05/2012 11:00, Thierry Vignaud a écrit : On 30 May 2012 10:30, Anne Nicolasenn...@mageia.org wrote: Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 I suggest the following topic: - including all firmwares by default. other distro do it, we don't this break many free drivers (network, ati/amd, ...) See my previous mail on this ml ok but it would be nice if you can attend this meeting. I fear I've other social obligations. I'll try.
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On 30 May 2012 10:38, Thierry Vignaud thierry.vign...@gmail.com wrote: I think we must discuss the following point for next release: currently radeon firmwares are in non-free. However this break havoc quite a lot of install b/c by default non-free is not enabled. 1) This makes the Mageia experience horrible on some machines (black screen, machines unusable by newbies (and worse unrecoverable by newbies), ...) Newer ATI/AMD cards cannot work w/o the needed firmware. That's also the issue with several network cards. (which makes network installation impossible by default) 2) this is unlike the live-cd where we've those firmwares (and the proprietary drivers as well) 3) This is unlike other pure FOSS distro such as Fedora which includes the firmwares. So there're two kinds of Linux distro: - those who provides free proprietary drivers - those who only provides free drivers, with the needed firmwares Mageia choose to provides either the first experience (the live cd case) or the worst one, the free drivers w/o the needed firmwares. A lot of thing broke havoc when firmwares get split from kernels. According to their licence (or their lack of license), we choose to distribute them in: - either kernel-firmware, always available and always installed - or kernel-firmware-nonfree, unavailable and not installed by default This make a lot of free drivers stop to work by default (at least with the classic installer, it's OK with the livecd) So we offer the choice between: - either be able to choose between working proprietary driver and working free driver - or to have non working free driver This is _INSANE_. I think It's time we reconsider this. Firmwares do not run on the main CPU. They're needed for additional chips where no one will ever ever write a program for. From the main CPU/memory view point they're data. I think we should offer the choice to: - either be able to choose between working proprietary driver and working free driver - or to have _working_ free driver (that is including the needed firmwares) See https://bugs.mageia.org/show_bug.cgi?id=3466 Likewise, fedora package AMD/Intel microcode for their CPUs whereas we put them in non-free, which makes microcode_ctl to not work by default See https://bugs.mageia.org/show_bug.cgi?id=4327
Re: [Mageia-dev] [RFC] x11-server-1.12
On Wed, May 30, 2012 at 11:08:27AM +0200, Thierry Vignaud wrote: eg: we need to package a newer snapshot of nouveau I have a nVidia card, so need nouveau to be able to test. -- Regards, Olav
Re: [Mageia-dev] [RFC] radeon firmware in non-free
30.05.2012 12:13, Thierry Vignaud skrev: Likewise, fedora package AMD/Intel microcode for their CPUs So if we are supposed to do everything fedora does, we might as well use fedora then... whereas we put them in non-free, which makes microcode_ctl to not work by default See https://bugs.mageia.org/show_bug.cgi?id=4327 So the real bug is installing a non-working package by default. and since it in reality has hard deps on nonfree stuff, the microcode_ctl should be in nonfree too, not core. -- Thomas
Re: [Mageia-dev] Bugzilla mass ping
On Wed, May 30, 2012 at 08:08:30AM +0300, Sander Lepik wrote: With new bugzilla we should be able to choose multiple versions. And we can use keywords to handle them. For example: tested_mgaX_arch - validate_update_mgaX - update_pushed_mgaX. Normally you'd use flags for this. You can setup flags that it automatically goes to a few people when it is requested, etc. The UI is confusing though (IMO). -- Regards, Olav
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On 30 May 2012 11:09, Thomas Backlund t...@mageia.org wrote: I think we must discuss the following point for next release: currently radeon firmwares are in non-free. However this break havoc quite a lot of install b/c by default non-free is not enabled. So its an installer issue. No it's a policy issue. media.cfg enforces not installing firmwares by default. But builders of live cds overrides this. Which is inconsistent Maybe should we split firmwares in another medium that should: - be enabled by default - be included in the install ISOs 1) This makes the Mageia experience horrible on some machines (black screen, machines unusable by newbies (and worse unrecoverable by newbies), ...) Newer ATI/AMD cards cannot work w/o the needed firmware. That's also the issue with several network cards. (which makes network installation impossible by default) So people should think twice about what hw they want to buy/use. Come on you're trolling. What's more, _YOU_ are the one building live cds with proprietary drivers... :-) That is totally inconsistent. Other distros either use proprietary drivers or free drivers with the needed firmwares. We do the first (live-cd) and the worst of all choices: free drivers w/o the needed firmwares (install ISO) 2) this is unlike the live-cd where we've those firmwares (and the proprietary drivers as well) This is because you cant install rpms on live media... (or you can, but as the change is gone with a reboot, there is not much point) Yet it is inconsistent. 3) This is unlike other pure FOSS distro such as Fedora which includes the firmwares. Well, that's their choice, not ours... Offering the choice between working proprietary drivers and NOT working free drivers is somewhat stupid don't you think? So we offer the choice between: - either be able to choose between working proprietary driver and working free driver - or to have non working free driver This is _INSANE_. I think It's time we reconsider this. Firmwares do not run on the main CPU. They're needed for additional chips where no one will ever ever write a program for. From the main CPU/memory view point they're data. But not from our perspective about free-only. Then why do you promote proprietary drivers on live cds :-) ? Again this is totally inconsistent... And nobody will ever wrote an alternative firmware anyway... And in most cases, nobody knows how the internal works. And for the AMD/ATI case, the free driver won't pick directly at the hardware but will use AtomBIOS to level differences between hw (remember the radeonhd fiasco). Who will rewrite the AMD firmwares for all radeon generations and/or make the driver work without Atom BIOS ?? Come on... What we're actually promoting to our end users is that free drivers suck whereas the live cds with proprietary drivers works smoothly. That's a total aberration regarding promoting FOSS I think we should offer the choice to: - either be able to choose between working proprietary driver and working free driver - or to have _working_ free driver (that is including the needed firmwares) As pointed out in earlier discussions about this issue... There is 2 points needing fix. - the installer should detect the need for nonfree, and ask if people want to enable it. (or simply ask if user want to add nonfree media) I'm actually planning to add a media screen in cauldron installer so that one can enable/disable media. But that needs to actually include a nonfree/release subset on the install ISO so that those w/o network can enable it. Detecting need for non-free could then be done (looking at modinfo output for needed modules) - the other thing to fix is to change the installer option about additional CD to additional media and allow for browsing for additional rpms on a cd/dvd/disk/usb/... media.
Re: [Mageia-dev] [RFC] radeon firmware in non-free
30.05.2012 12:23, Thierry Vignaud skrev: On 30 May 2012 11:09, Thomas Backlundt...@mageia.org wrote: I think we must discuss the following point for next release: currently radeon firmwares are in non-free. However this break havoc quite a lot of install b/c by default non-free is not enabled. So its an installer issue. No it's a policy issue. media.cfg enforces not installing firmwares by default. But builders of live cds overrides this. Which is inconsistent Well, it was a old design decision we followed for livecds... Maybe should we split firmwares in another medium that should: - be enabled by default - be included in the install ISOs 1) This makes the Mageia experience horrible on some machines (black screen, machines unusable by newbies (and worse unrecoverable by newbies), ...) Newer ATI/AMD cards cannot work w/o the needed firmware. That's also the issue with several network cards. (which makes network installation impossible by default) So people should think twice about what hw they want to buy/use. Come on you're trolling. What's more, _YOU_ are the one building live cds with proprietary drivers... :-) Well, I have no problem dropping them from livecds again as I can use the space... It did work for mga1 (as there were no prebuilt proprietary drivers then) That is totally inconsistent. Other distros either use proprietary drivers or free drivers with the needed firmwares. We do the first (live-cd) and the worst of all choices: free drivers w/o the needed firmwares (install ISO) 2) this is unlike the live-cd where we've those firmwares (and the proprietary drivers as well) This is because you cant install rpms on live media... (or you can, but as the change is gone with a reboot, there is not much point) Yet it is inconsistent. 3) This is unlike other pure FOSS distro such as Fedora which includes the firmwares. Well, that's their choice, not ours... Offering the choice between working proprietary drivers and NOT working free drivers is somewhat stupid don't you think? So we need to fix the hw/driver detection/switching and not just paper over it by pulling nonfree stuff by default So we offer the choice between: - either be able to choose between working proprietary driver and working free driver - or to have non working free driver This is _INSANE_. I think It's time we reconsider this. Firmwares do not run on the main CPU. They're needed for additional chips where no one will ever ever write a program for. From the main CPU/memory view point they're data. But not from our perspective about free-only. Then why do you promote proprietary drivers on live cds :-) ? Again this is totally inconsistent... So I guess we need to fix livecds by either dropping proprietary stuff, or make the display_driver_helper ask if user wants to boot with proprietary stuff or not... And nobody will ever wrote an alternative firmware anyway... And in most cases, nobody knows how the internal works. And for the AMD/ATI case, the free driver won't pick directly at the hardware but will use AtomBIOS to level differences between hw (remember the radeonhd fiasco). Who will rewrite the AMD firmwares for all radeon generations and/or make the driver work without Atom BIOS ?? Come on... What we're actually promoting to our end users is that free drivers suck whereas the live cds with proprietary drivers works smoothly. Again, this is hw specific. There is Intel, nouveau, sis,... that does not need fw/proprietary drivers. That's a total aberration regarding promoting FOSS Well, it's also a simple truth/fact. Not all hw is fully FOSS supported. I think we should offer the choice to: - either be able to choose between working proprietary driver and working free driver - or to have _working_ free driver (that is including the needed firmwares) As pointed out in earlier discussions about this issue... There is 2 points needing fix. - the installer should detect the need for nonfree, and ask if people want to enable it. (or simply ask if user want to add nonfree media) I'm actually planning to add a media screen in cauldron installer so that one can enable/disable media. Great. But that needs to actually include a nonfree/release subset on the install ISO so that those w/o network can enable it. Or ask for additional media pointed out below. Detecting need for non-free could then be done (looking at modinfo output for needed modules) - the other thing to fix is to change the installer option about additional CD to additional media and allow for browsing for additional rpms on a cd/dvd/disk/usb/... media. -- Thomas
Re: [Mageia-dev] [RFC] radeon firmware in non-free
30.05.2012 11:38, Thierry Vignaud skrev: [..] I think we should offer the choice to: - either be able to choose between working proprietary driver and working free driver - or to have _working_ free driver (that is including the needed firmwares) As pointed out in earlier discussions about this issue... There is 2 points needing fix. - the installer should detect the need for nonfree, and ask if people want to enable it. (or simply ask if user want to add nonfree media) of course this means that the DVD installer has nonfree on it's DVD (or have a separate media for that, which is less easy) - the other thing to fix is to change the installer option about additional CD to additional media and allow for browsing for additional rpms on a cd/dvd/disk/usb/... media. add here, also nfs/iSCSI/http/... in fact that screen should be redone and done differently: [X] this media (DVD or whatever) [ ] media from mirrors [dropdown with mirror] (by default automatic) [ ] (here the media added from advanced is added when closing advanced) [X] updates [ ] nonfree [ ] tainted [ ] backports [ ] testing [ ] debug + button advanced with custom media (for urpmi-proxy or local media for instance)
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On 30 May 2012 11:09, Thomas Backlund t...@mageia.org wrote: [...] Offering the choice between working proprietary drivers and NOT working free drivers is somewhat stupid don't you think? this is true, of course, except that iirc not all cards were non-working. i must add that even with my nvidia i had nouveau failing and had start failsafe to switch to proprietary driver maybe we should just be simple and have for all intel/AMD/NVidia cards just vesa driver for installation. at least it'll be working then. perhaps we can even ask the persons during installation, if we want to add awareness, perhaps this is the best solution for it. [...] I'm actually planning to add a media screen in cauldron installer so that one can enable/disable media. But that needs to actually include a nonfree/release subset on the install ISO so that those w/o network can enable it. Detecting need for non-free could then be done (looking at modinfo output for needed modules) [...] plz seem my email in this thread, it seems like a good redesign for mga3 for this.
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On 30 May 2012 12:21, AL13N al...@rmail.be wrote: Offering the choice between working proprietary drivers and NOT working free drivers is somewhat stupid don't you think? this is true, of course, except that iirc not all cards were non-working. i must add that even with my nvidia i had nouveau failing and had start failsafe to switch to proprietary driver maybe we should just be simple and have for all intel/AMD/NVidia cards just vesa driver for installation. at least it'll be working then. which would be dead slow, won't support gnome once they'll drop gnome classic and will make users flee and switch to other normal OSes.
Re: [Mageia-dev] Tonight's meeting
mercoledì 30 maggio 2012 alle 10:30, Anne Nicolas ha scritto: Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 Can we talk also about backport? or we postpone the problem later again? TIA Angelo signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On 30 May 2012 12:21, AL13N al...@rmail.be wrote: Offering the choice between working proprietary drivers and NOT working free drivers is somewhat stupid don't you think? this is true, of course, except that iirc not all cards were non-working. i must add that even with my nvidia i had nouveau failing and had start failsafe to switch to proprietary driver maybe we should just be simple and have for all intel/AMD/NVidia cards just vesa driver for installation. at least it'll be working then. which would be dead slow, won't support gnome once they'll drop gnome classic and will make users flee and switch to other normal OSes. I meant to ask specifically before the configure step with a screen which drivers they want to use. So only for beginning of installer, if the installer is a problem. I don't believe the install would be terribly slow (a bit though).
Re: [Mageia-dev] Tonight's meeting
mercoledì 30 maggio 2012 alle 10:30, Anne Nicolas ha scritto: Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 Can we talk also about backport? or we postpone the problem later again? I would like to talk about backport too, but it's closely related to QA, and they have their meeting Thursday, i believe. I think it might be better to await the QA team meeting results. also, no need to have a longer meeting than 2 hours :-)
Re: [Mageia-dev] Tonight's meeting
2012/5/30 Angelo Naselli anase...@linux.it: mercoledì 30 maggio 2012 alle 10:30, Anne Nicolas ha scritto: Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 Can we talk also about backport? or we postpone the problem later again? No pb to add this topic but I would suggest to plan it for next week. QA is having meeting tomorrow and it's a big condition to manage backports. We do not forget it but postpone it for one week only TIA Angelo -- Anne http://www.mageia.org
Re: [Mageia-dev] Tonight's meeting
mercoledì 30 maggio 2012 alle 12:43, Anne nicolas ha scritto: No pb to add this topic but I would suggest to plan it for next week. QA is having meeting tomorrow and it's a big condition to manage backports. We do not forget it but postpone it for one week only One week more, sounds good to me :) Just avoid forgetting it ;) Angelo signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] [RFC] radeon firmware in non-free
mercoledì 30 maggio 2012 alle 10:38, Thierry Vignaud ha scritto: Hi I think we must discuss the following point for next release: currently radeon firmwares are in non-free. I'm in favour of having working installer either for newbies or for those people that does not read release notes, errata etc. OTOH, i think if we can reach a good installation result w/o non free firmware or by letting the user to decide during the installation (i mean no silent choice there) with additional media, on the network or local (usb key, non free driver cds, etc), it's ok as well. I recall anyway, that even our livecd is not that easy to get working in some cases we need to add xdriver=XXX parameter... right or not that is, the newbie does not understand that. Sure free is our choice, but from my experience i can say we have already at least two users less, they put our dvd, installation failed, they put (K)Ubuntu, installation went ok... should they have read, or asked to me for that... but they didn't... I believe they're not the only one. Cheers, Angelo signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] Tonight's meeting
Le 30/05/2012 10:30, Anne Nicolas a écrit : Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 Cheers What about the mentoring process ? there were a few requests in the last weeks
Re: [Mageia-dev] [changelog] [RPM] cauldron nonfree/updates_testing x11-driver-video-nouveau-0.0.16-0.20120530.1.mga3.nonfree
On Wed, May 30, 2012 at 12:26 PM, tv buildsystem-dae...@mageia.org wrote: Name : x11-driver-video-nouveau Relocations: (not relocatable) Version : 0.0.16 Vendor: Mageia.Org Release : 0.20120530.1.mga3.nonfree Build Date: Wed May 30 13:25:15 2012 nonfree? Same for the others
Re: [Mageia-dev] [RFC] x11-server-1.12
On 30 May 2012 11:18, Olav Vitters o...@vitters.nl wrote: eg: we need to package a newer snapshot of nouveau I have a nVidia card, so need nouveau to be able to test. I've just uploaded it. However I've mistakely uploaded the following packages in nonfree/updates_testing instead of in core/updates_testing. Can sysadmins move them back to core/updates_testing? Thx x11-driver-video-rendition-4.2.4-7.mga3 x11-driver-video-apm-1.2.4-2.mga3 x11-driver-input-void-1.4.0-4.mga3 x11-driver-input-mutouch-1.3.0-4.mga3 x11-driver-input-fpit-1.4.0-4.mga3 x11-driver-input-evtouch-0.8.8-10.mga3 x11-driver-input-aiptek-1.4.1-2.mga3 x11-driver-input-acecad-1.5.0-4.mga3 x11-driver-video-nouveau-0.0.16-0.20120530.1.mga3
Re: [Mageia-dev] Tonight's meeting
2012/5/30 Manuel Hiebel man...@hiebel.eu: Le 30/05/2012 10:30, Anne Nicolas a écrit : Hi there We could say on the road again :). Our meeting will happen as usual at 19hUTC on #mageia-dev. Here are the proposed topics: - Mageia 2 release and post-mortem - Development planning for Mageia 3 - Specifications for Mageia 3 Cheers What about the mentoring process ? there were a few requests in the last weeks ok if we have somebody to make review for tonight -- Anne http://www.mageia.org
Re: [Mageia-dev] [changelog] [RPM] cauldron nonfree/updates_testing x11-driver-video-nouveau-0.0.16-0.20120530.1.mga3.nonfree
On 30 May 2012 13:28, Pascal Terjan pter...@gmail.com wrote: Name : x11-driver-video-nouveau Relocations: (not relocatable) Version : 0.0.16 Vendor: Mageia.Org Release : 0.20120530.1.mga3.nonfree Build Date: Wed May 30 13:25:15 2012 nonfree? Same for the others Yeah my mistake. I just asked for them to be moved into core/updates_testing
Re: [Mageia-dev] Tonight's meeting
Hi Le 30/05/2012 10:30, Anne Nicolas a écrit : - Development planning for Mageia 3 - Specifications for Mageia 3 Could we talk about the Java dependency problem (don't know if it can be in one of this 2 points) ? Claire
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release perl-URPM-3.40-1.1.mga3
On 30 May 2012 13:38, Iurt the rebuild bot buildsystem-dae...@mageia.org wrote: tv tv 3.40-1.1.mga2: + Revision: 223706 - add URPM::traverse_tag_find() for urpme --env Please bump release and puts some changelog next time you rebuild :-( sg such as: rebuild for new perl
Re: [Mageia-dev] Tonight's meeting
2012/5/30 Claire REVILLET gren...@zarb.org: Hi Le 30/05/2012 10:30, Anne Nicolas a écrit : - Development planning for Mageia 3 - Specifications for Mageia 3 Could we talk about the Java dependency problem (don't know if it can be in one of this 2 points) ? We will not detail specs but rather the process to achieve it. Could be part of specs also it would be nice to have at least dmorgan around about that topic Claire -- Anne http://www.mageia.org
Re: [Mageia-dev] don't update perl modules yet
hi, On 12/05/29 18:48 +0200, Jerome Quelin wrote: blino will submit it soon using some special tricks - those are needed since there is a bootstrap problem between perl and urpmi. blino did his magic trick, and pascal then uploaded the result. thanks to both of them. as you can see, a lot of thing cannot be currently installed. this is the time where you should not proceed blindly, and don't force the upgrade - you've been warned. :-) i've started rebuilding the packages, but it may take some time... i'll let you know when things are finished on my side. jérôme
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release perl-Glib-1.251.0-3.mga3
On 30 May 2012 14:36, jquelin buildsystem-dae...@mageia.org wrote: jquelin jquelin 1.251.0-3.mga3: + Revision: 250648 - rebuild for perl 5.16 You can resubmit all of perl-Gtk2, perl-Gnome2 now. Preferably starting with perl-Gtk2, then everything else. There's a couple interdepedancies left but that would be a beginning
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On Wednesday 30 May 2012 10:38, Thierry Vignaud wrote: So there're two kinds of Linux distro: I beg to differ, there are three kinds, using your way of counting. - those who provides free proprietary drivers - those who only provides free drivers, with the needed firmwares - those who only provides free drivers, without proprietary firmwares You have Ututo, Blag, gNewSense and the other FSF approved distros which takes every care in ensuring that No proprietary drivers of any kind, including firmware, is not available. They even remove the few proprietary drivers which are included in the upstream kernel. (Debian also does this as of version 6, in the main tree.) If some hardware does not work without having a proprietary firmware or driver, then that particular hardware does not work. Period. The user must install a proprietary driver them selves afterwards if they want to use it, and the distro does not help them in any way to achieve this. -- Johnny A. Solbu PGP key ID: 0xFA687324 signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release xfprint-4.6.1-6.mga3
On 30.05.2012 07:51, Charles A Edwards wrote: On Wed, 30 May 2012 06:58:50 +0300 Jani Välimaa jani.vali...@gmail.com wrote: Libxfcegui4 is deprecated, but it's not entirely dead and buried, there's still some plugins which aren't ported to libxfce4ui and needs libxfcegui4. Also, new version 4.10.0 was released about a month ago when Xfce 4.10.0 came out. While true that there is a version 4.10.0 it was purposely not included by Nick in the xfce4-4.10 source pkg (he will not let it officially die until xfce ports to GTK3). There are 26 apps, some of which are not packaged in Mageia, which still need libxfcegui4. To build Without xfcegui4 patches for these plugins can be found here cpufreq: https://bugzilla.xfce.org/show_bug.cgi?id=8051 datetime: https://bugzilla.xfce.org/show_bug.cgi?id=8064 eyes: https://bugzilla.xfce.org/show_bug.cgi?id=8074 diskperf: https://bugzilla.xfce.org/show_bug.cgi?id=8099 Thx for the links, I'll check them out. I'm OK to consider dropping libxfcegui4 after all pkgs depending on it are ported to libxfce4ui.
[Mageia-dev] rpmbuild -ba --sign , does nt work any more
I see that rpmbuild does n't ask me any more (after upgrading to mga2) for the password when i add the --sign. With the bm it works fine. I don't know if it is dropped the --sign from rpmbuild or is something that needs to have a look. -- Dimitrios Glentadakis
Re: [Mageia-dev] [soft-commits] [4714] rename internal function to mark themas private
On 29 May 2012 19:29, r...@mageia.org wrote:
-sub sqr { $_[0] * $_[0] }
-sub round { int($_[0] + 0.5) }
-sub group_by2 {
+sub _sqr { $_[0] * $_[0] }
+sub _round { int($_[0] + 0.5) }
+sub _group_by2 {
Please add a comment that all these functions came from MDK::Common.
So that we think about syncing if there's ever a bugfix
Re: [Mageia-dev] rpmbuild -ba --sign , does nt work any more
On 30 May 2012 16:28, Dimitrios Glentadakis dgl...@gmail.com wrote: I see that rpmbuild does n't ask me any more (after upgrading to mga2) for the password when i add the --sign. With the bm it works fine. I don't know if it is dropped the --sign from rpmbuild or is something that needs to have a look. Use rpmsign
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On 30 May 2012 15:32, Johnny A. Solbu coo...@solbu.net wrote: So there're two kinds of Linux distro: I beg to differ, there are three kinds, using your way of counting. - those who provides free proprietary drivers - those who only provides free drivers, with the needed firmwares - those who only provides free drivers, without proprietary firmwares You have Ututo, Blag, gNewSense and the other FSF approved distros which takes every care in ensuring that No proprietary drivers of any kind, including firmware, is not available. They even remove the few proprietary drivers which are included in the upstream kernel. (Debian also does this as of version 6, in the main tree.) If some hardware does not work without having a proprietary firmware or driver, then that particular hardware does not work. Period. The user must install a proprietary driver them selves afterwards if they want to use it, and the distro does not help them in any way to achieve this. Please... We're speaking about those who actually have (a decent number of) users
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On Wednesday 30 May 2012 16:20, Angelo Naselli wrote: If you really mean that you should also remove non-free repositories etc... I don't think being a don Quijote is that good... always imho of course. They do exactly that. They do not have any non-free repo. They have a policy that do not allow any non-free software (programs, drivers, firmware etc.) in their distributions, and they don't have any exceptions. Some might find this silly and stupid. But they live by the rule that every software should be Free. If it is not, they will not use it because they can't do what they want with it. My point is just that there really do exist several distributions that do not allow any non-free software in their repositories. if they don't have any non-free software there is no point in having an empty non-free repo. The way Thierry listed the two categories he seemed to suggest that there where no such distributions. -- Johnny A. Solbu PGP key ID: 0xFA687324 signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] packages pushed to cauldron updates_testing before release
On Tue, 29 May 2012, David Walser wrote: Thomas Backlund tmb@... writes: 29.05.2012 19:48, David Walser skrev: Packagers, If you pushed a package to updates_testing in Cauldron before Mageia 2 was released, please make sure you re-push it if you want the fix/update included in Mageia 2. The change should already be in the SVN, so all you need to do is resubmit it to the build system (and file a bug with an advisory and assign it to QA of course). Nope. SVN was branched according on what was tagged and uploaded on mirrors as that's the only way to make sure svn updates tree matches release tree, meaning no unwanted/unplanned changes get queued in updates tree. So if yo want to submit anything to updates_testing, you need to commit the changes in updates tree and then submit. Thomas, Are you sure about that? I had pushed taglib to updates_testing in Cauldron before the release and after SVN was branched the update was already in SVN. I think the branch was generated from releases/ but that includes anything that was built, whether it went to release or updates_testing. Ah it seems your are right, the branch was created from the commit of the latest markrelease. And markrelease is also done for packages in cauldron updates_testing repository.
Re: [Mageia-dev] [RFC] radeon firmware in non-free
And here is the opinion of a user. I read little English and I'm not sure that I understand the totality of interventions. Anyway, the basic user requests a distribution that works, without needing to get their hands dirty. If this can illuminate your discussion! Païou (Georges)
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On Wednesday 30 May 2012 16:36, Thierry Vignaud wrote: We're speaking about those who actually have (a decent number of) users Then you should say so, and not pretend that the other ones don't exist. -- Johnny A. Solbu PGP key ID: 0xFA687324 signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] [RFC] radeon firmware in non-free
mercoledì 30 maggio 2012 alle 16:38, Johnny A. Solbu ha scritto: On Wednesday 30 May 2012 16:20, Angelo Naselli wrote: If you really mean that you should also remove non-free repositories etc... I don't think being a don Quijote is that good... always imho of course. They do exactly that. They do not have any non-free repo. They have a policy that do not allow any non-free software (programs, drivers, firmware etc.) in their distributions, and they don't have any exceptions. Some might find this silly and stupid. But they live by the rule that every software should be Free. If it is not, they will not use it because they can't do what they want with it. My point is just that there really do exist several distributions that do not allow any non-free software in their repositories. if they don't have any non-free software there is no point in having an empty non-free repo. The way Thierry listed the two categories he seemed to suggest that there where no such distributions. That's sounds different of course, i thought you referred to us ;) signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] rpmbuild -ba --sign , does nt work any more
Στις 30/05/2012 16:35:29 Thierry Vignaud έγραψε: On 30 May 2012 16:28, Dimitrios Glentadakis dgl...@gmail.com wrote: I see that rpmbuild does n't ask me any more (after upgrading to mga2) for the password when i add the --sign. With the bm it works fine. I don't know if it is dropped the --sign from rpmbuild or is something that needs to have a look. Use rpmsign Thanks, i did n't knew this one, i used the rpm --sign too -- Dimitrios Glentadakis
Re: [Mageia-dev] rpmbuild -ba --sign , does nt work any more
On Wednesday 30 May 2012 16:35, Thierry Vignaud wrote: dropped the --sign from rpmbuild Use rpmsign Why is --sign no longer supported? -- Johnny A. Solbu PGP key ID: 0xFA687324 signature.asc Description: This is a digitally signed message part.
Re: [Mageia-dev] rpmbuild -ba --sign , does nt work any more
On 30 May 2012 17:25, Johnny A. Solbu coo...@solbu.net wrote: dropped the --sign from rpmbuild Use rpmsign Why is --sign no longer supported? http://lists.rpm.org/pipermail/rpm-announce/2011-March/27.html * New command line utilities for spec queries (with much enhanced functionality), package signing, keyring operations and database maintenance http://rpm.org/wiki/Releases/4.9.0 New command-line utilities - The keyring related operations of importing public keys and verifying package signatures have been split into a new rpmkeys cli utility - Package signing and signature deletion operations have been split into a new rpmsign cli utility (related to RhBug:624585) (...) Removed features - Support for --sign during rpmbuild has been removed. Signing must be done as a separate step using the new rpmsign command.
Re: [Mageia-dev] rpmbuild -ba --sign , does nt work any more
Le 30 mai 2012 18:42, Thierry Vignaud thierry.vign...@gmail.com a écrit : On 30 May 2012 17:25, Johnny A. Solbu coo...@solbu.net wrote: dropped the --sign from rpmbuild Use rpmsign Why is --sign no longer supported? http://lists.rpm.org/pipermail/rpm-announce/2011-March/27.html * New command line utilities for spec queries (with much enhanced functionality), package signing, keyring operations and database maintenance http://rpm.org/wiki/Releases/4.9.0 New command-line utilities - The keyring related operations of importing public keys and verifying package signatures have been split into a new rpmkeys cli utility - Package signing and signature deletion operations have been split into a new rpmsign cli utility (related to RhBug:624585) (...) Removed features - Support for --sign during rpmbuild has been removed. Signing must be done as a separate step using the new rpmsign command. It is still present in rpmbuilt --help, may be is forgotten to be removed
Re: [Mageia-dev] [RFC] x11-server-1.12
On 30 May 2012 11:08, Thierry Vignaud thierry.vign...@gmail.com wrote:
Several of those have been ported to new x11-server-1.12 ABI in git
but haven't yet been released (eg: chips, rendition, tseng)
I've pushed those
eg: we need to package a newer snapshot of nouveau
Also I'ven't yet tried t orebuild several minor/obsolete input drivers.
I've pushed those too.
The following drivers failed to rebuild:
- virtualbox
- x11-driver-input-synaptics
- x11-driver-video-sisimedia
- x11-driver-video-sisusb
- x11-driver-video-xgi
synaptics needs rediffing some patches.
The only other driver we miss is virtualbox which fails to compile:
/usr/include/xorg/os.h:496:25: error: expected identifier or '('
before '__extension__'
Drakx installer works fine with new xserver (just need a little adaptation
b/c a xkb file has disappeared)
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release perl-URPM-3.40-1.1.mga3
Thierry Vignaud thierry.vign...@gmail.com writes: On 30 May 2012 13:38, Iurt the rebuild bot buildsystem-dae...@mageia.org wrote: tv tv 3.40-1.1.mga2: + Revision: 223706 - add URPM::traverse_tag_find() for urpme --env Please bump release and puts some changelog next time you rebuild :-( sg such as: rebuild for new perl Yes, it would be nice, but that's difficult to automate from a iurt shell... (since the automated perl bootstrap rebuild takes place in a iurt shell) Anyway, jq is supposed to be rebuild it for the new perl (in the perlapi-5.14 rebuild batch), so this is transient. -- Olivier Blin - blino
Re: [Mageia-dev] [RFC] x11-server-1.12
'Twas brillig, and Thierry Vignaud at 30/05/12 17:53 did gyre and gimble:
On 30 May 2012 11:08, Thierry Vignaud thierry.vign...@gmail.com wrote:
Several of those have been ported to new x11-server-1.12 ABI in git
but haven't yet been released (eg: chips, rendition, tseng)
I've pushed those
eg: we need to package a newer snapshot of nouveau
Also I'ven't yet tried t orebuild several minor/obsolete input drivers.
I've pushed those too.
The following drivers failed to rebuild:
- virtualbox
- x11-driver-input-synaptics
- x11-driver-video-sisimedia
- x11-driver-video-sisusb
- x11-driver-video-xgi
synaptics needs rediffing some patches.
The only other driver we miss is virtualbox which fails to compile:
/usr/include/xorg/os.h:496:25: error: expected identifier or '('
before '__extension__'
Drakx installer works fine with new xserver (just need a little adaptation
b/c a xkb file has disappeared)
FWIW, gdm can now be compiled with without the --without-systemd switch
to enable the automatic multi-seat stuff (as Xorg binary should not
accept a --seat argument). But this can probably wait until it's in main :)
Col
--
Colin Guthrie
colin(at)mageia.org
http://colin.guthr.ie/
Day Job:
Tribalogic Limited http://www.tribalogic.net/
Open Source:
Mageia Contributor http://www.mageia.org/
PulseAudio Hacker http://www.pulseaudio.org/
Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] Mageia 3 specifications
Le mercredi 30 mai 2012 23:43:03, Anne Nicolas a écrit : Hi there Mageia 2 is out, it's now time to work on next version, Mageia 3. We tried to improve the specifications process so that it can be more efficient and less frustrating :). Some conclusions on Mageia 2 specifications: - lots of proposals, very few achieved - lots of incomplete proposals - specifications that are not really specifications like update to xxx version for minor packages - some kind of catalog of items without logic - lots of proposals without any resources ... So in order to improve all this, we looked at what were doing other distributions, especially in Fedora. Then we worked on first version for Features policy that will be used for Mageia. Here is the process: - until 15/06: make proposals for Mageia 3 specifications (features) following our policy - 16 to 18/06: choose features that will be added for official specifications of Mageia 3 - publish and communicate on it Mageia features policy: https://wiki.mageia.org/en/Features_policy Template for feature proposal: https://wiki.mageia.org/en/Feature_Template Please read carefully this policy and use the template. Once it's done, feel free to mail -dev if you think you need to discuss your feature and find people so that you can propose resources to work on it. Cheers Maybe this would deserve a blog post so that people who don't follow the mageia-dev mailing list have a chance to propose new features for Mageia 3, provided they follow the features policy? Samuel
Re: [Mageia-dev] rpmbuild -ba --sign , does nt work any more
On Wed, 30 May 2012 18:49:19 +0200 Dimitrios Glentadakis dgl...@gmail.com wrote: - The keyring related operations of importing public keys and verifying package signatures have been split into a new rpmkeys cli utility - Package signing and signature deletion operations have been split into a new rpmsign cli utility (related to RhBug:624585) (...) Removed features - Support for --sign during rpmbuild has been removed. Signing must be done as a separate step using the new rpmsign command. It is still present in rpmbuilt --help, may be is forgotten to be removed It still working here on this system. Instead of asking for pass phrase at start of build and signing the rpms as they are built, it waits till the rpms are built, asks for the pass prase and then signs the rpms. Example: rpmbuild -ba --clean --sign imagemagick.spec + umask 022 + cd /home/charles/rpm/BUILD big snip + exit 0 + umask 022 + cd /home/charles/rpm/BUILD + cd ImageMagick-6.7.7-4 + rm -rf /home/charles/rpm/BUILDROOT/imagemagick-6.7.7.4-1zae.x86_64 + exit 0 + umask 022 + cd /home/charles/rpm/BUILD + rm -rf ImageMagick-6.7.7-4 + exit 0 Enter pass phrase: Pass phrase is good. /home/charles/rpm/SRPMS/imagemagick-6.7.7.4-1zae.src.rpm: /home/charles/rpm/RPMS/x86_64/imagemagick-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/imagemagick-desktop-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/lib64magick5-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/lib64magick-devel-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/perl-Image-Magick-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/imagemagick-doc-6.7.7.4-1zae.x86_64.rpm: [charles@SuperSize SPECS]$ cd /home/charles/rpm/RPMS/x86_64/ [charles@SuperSize x86_64]$ rpm -K *rpm imagemagick-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK imagemagick-desktop-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK imagemagick-doc-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK lib64magick5-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK lib64magick-devel-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK perl-Image-Magick-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK Charles -- BOFH excuse #341: HTTPD Error 666 : BOFH was here -- Mageia release 3 (Cauldron) for x86_64$ On SuperSizehttp://www.eslrahc.com Registered Linux user #182463 3.3.6-tmb-server-2.mga2 x86_64 -- signature.asc Description: PGP signature
Re: [Mageia-dev] [RFC] radeon firmware in non-free
Hi, I can see two objectives here. First objective is to have a high installation success rate. And second objective is to have happy users from both camps; the GNU-purists-camp and the don't-care-camp. I think the GNU-purists-camp is a minority here, but I could be wrong, I never counted them. For them I would make a pure installation DVDs of all editions, like Desktop, Server, Live, etc. I bet all purists are expert Linux users and they know enough about their hardware and if there are GNU drivers for it. On the other hand, the don't-care-camp needs stuff that works. They don't care where it comes from, and they never read the messages. They just click through. For them I would make a mixed DVD with pre-selected components that have higher success rate. For example, if we know that non-free radeon drivers install better than the free ones, don't ask the user at all just go ahead and install the non-free version. This way, we are making the users choose early way before they start installing. I would also call those editions like this: 1g) Mageia 3 Server (G) Installation DVD i686 - Includes GNU drivers 1v) Mageia 3 Server (V) Installation DVD i686 - Includes Vendor drivers We would also say If you don't know which to choose, go for Vendor editions I think providing free and non-free installation media will increase Mageia's quality and reputation. But, everything must be in a single DVD because there are laptops, servers, and desktops with Broadcom PHY's and they can't get supplemental DVD's or packages without a working network card or wifi. Putting them all together in a single non-free DVD will increase the installation success rate for sure. And at last let me introduce my self. My name is Agron Selimaj. I am an experienced Linux user and programmer. I have 11 years of Linux experience and I am a Mandriva Powerpack user since 2006 or 2007. I am a professional C++ programmer for Windows, Linux embedded and Linux Desktops at my day job. I like the spirit of Mageia creators and I would like to join you. At the moment, I am an apprentice candidate waiting for a mentor to guide me through package creation process, bug fixing, etc. //Agron On Wed, May 30, 2012 at 11:11 AM, Angelo Naselli anase...@linux.it wrote: mercoledì 30 maggio 2012 alle 16:38, Johnny A. Solbu ha scritto: On Wednesday 30 May 2012 16:20, Angelo Naselli wrote: If you really mean that you should also remove non-free repositories etc... I don't think being a don Quijote is that good... always imho of course. They do exactly that. They do not have any non-free repo. They have a policy that do not allow any non-free software (programs, drivers, firmware etc.) in their distributions, and they don't have any exceptions. Some might find this silly and stupid. But they live by the rule that every software should be Free. If it is not, they will not use it because they can't do what they want with it. My point is just that there really do exist several distributions that do not allow any non-free software in their repositories. if they don't have any non-free software there is no point in having an empty non-free repo. The way Thierry listed the two categories he seemed to suggest that there where no such distributions. That's sounds different of course, i thought you referred to us ;)
Re: [Mageia-dev] [RFC] radeon firmware in non-free
On 30/05/12 09:38, Thierry Vignaud wrote: Hi I think we must discuss the following point for next release: currently radeon firmwares are in non-free. However this break havoc quite a lot of install b/c by default non-free is not enabled. 1) This makes the Mageia experience horrible on some machines (black screen, machines unusable by newbies (and worse unrecoverable by newbies), ...) Newer ATI/AMD cards cannot work w/o the needed firmware. That's also the issue with several network cards. (which makes network installation impossible by default) 2) this is unlike the live-cd where we've those firmwares (and the proprietary drivers as well) 3) This is unlike other pure FOSS distro such as Fedora which includes the firmwares. So there're two kinds of Linux distro: - those who provides free proprietary drivers - those who only provides free drivers, with the needed firmwares Mageia choose to provides either the first experience (the live cd case) or the worst one, the free drivers w/o the needed firmwares. A lot of thing broke havoc when firmwares get split from kernels. According to their licence (or their lack of license), we choose to distribute them in: - either kernel-firmware, always available and always installed - or kernel-firmware-nonfree, unavailable and not installed by default This make a lot of free drivers stop to work by default (at least with the classic installer, it's OK with the livecd) So we offer the choice between: - either be able to choose between working proprietary driver and working free driver - or to have non working free driver This is _INSANE_. I think It's time we reconsider this. Firmwares do not run on the main CPU. They're needed for additional chips where no one will ever ever write a program for. From the main CPU/memory view point they're data. I think we should offer the choice to: - either be able to choose between working proprietary driver and working free driver - or to have _working_ free driver (that is including the needed firmwares) See you I personally think people should read before doing anything, cant expect people to throw a coaster into the computer and hit the go button computers are sophisticated complicated tools, not toasters Lemonzest
Re: [Mageia-dev] rpmbuild -ba --sign , does nt work any more
Στις 30/05/2012 20:44:12 Charles A Edwards έγραψε: On Wed, 30 May 2012 18:49:19 +0200 Dimitrios Glentadakis dgl...@gmail.com wrote: - The keyring related operations of importing public keys and verifying package signatures have been split into a new rpmkeys cli utility - Package signing and signature deletion operations have been split into a new rpmsign cli utility (related to RhBug:624585) (...) Removed features - Support for --sign during rpmbuild has been removed. Signing must be done as a separate step using the new rpmsign command. It is still present in rpmbuilt --help, may be is forgotten to be removed It still working here on this system. Instead of asking for pass phrase at start of build and signing the rpms as they are built, it waits till the rpms are built, asks for the pass prase and then signs the rpms. Example: rpmbuild -ba --clean --sign imagemagick.spec + umask 022 + cd /home/charles/rpm/BUILD big snip + exit 0 + umask 022 + cd /home/charles/rpm/BUILD + cd ImageMagick-6.7.7-4 + rm -rf /home/charles/rpm/BUILDROOT/imagemagick-6.7.7.4-1zae.x86_64 + exit 0 + umask 022 + cd /home/charles/rpm/BUILD + rm -rf ImageMagick-6.7.7-4 + exit 0 Enter pass phrase: Pass phrase is good. /home/charles/rpm/SRPMS/imagemagick-6.7.7.4-1zae.src.rpm: /home/charles/rpm/RPMS/x86_64/imagemagick-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/imagemagick-desktop-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/lib64magick5-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/lib64magick-devel-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/perl-Image-Magick-6.7.7.4-1zae.x86_64.rpm: /home/charles/rpm/RPMS/x86_64/imagemagick-doc-6.7.7.4-1zae.x86_64.rpm: [charles@SuperSize SPECS]$ cd /home/charles/rpm/RPMS/x86_64/ [charles@SuperSize x86_64]$ rpm -K *rpm imagemagick-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK imagemagick-desktop-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK imagemagick-doc-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK lib64magick5-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK lib64magick-devel-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK perl-Image-Magick-6.7.7.4-1zae.x86_64.rpm: (sha1) dsa sha1 md5 gpg OK Charles Yes, you are right it asks at the end, as bm does. -- Dimitrios Glentadakis
