Re: [gentoo-user] PERL_FEATURES
Many thanks, that helped a lot!
Now I have set only those features in the PERL_FEATURES entry that were
installed before -
the warning has gone and I hoped the correct version of perl is
installed now.
Helmut
On 05/07/2024 08:29:28 PM, Jack wrote:
On 2024.05.07 12:33, Helmut Jarausch wrote:
Hi,
according to the NEWS from today one has to set PERL_FEATURES in
/etc/portage/make.conf.
But how to do that?
I've tried
PERL_FEATURES="debug ithreads quadmath"
but emerging dev-lang/perl-5.38.2-r3 I get
* As of dev-lang/perl-5.38.2-r3, the useflags debug, ithreads,
quadmath move into
* a use-expand variable PERL_FEATURES, which should be set globally
in make.conf.
* It appears that you have not set this variable properly yet.
Many thanks for hint,
Helmut
The section of the ebuild that decides whether or not to issue that
warning is:
pkg_pretend() {
if \
( use perl_features_ithreads && has_version
' ( ! use perl_features_ithreads && has_version
' ( use perl_features_quadmath && has_version
' ( ! use perl_features_quadmath && has_version
' ( use perl_features_debug&& has_version
' ( ! use perl_features_debug&& has_version
'
then
(issue the warning)
As I read it (which may well be wrong) it is looking to see whether
the value for each of the three items is the same by the old use flag
and by the new PERL_FEATURES. Check how those three use flags are
set for -r2 and that the value for PERL_FEATURES for -r3 matches for
each.
Re: [gentoo-user] Hard drive and PWDIS or pin 3 power disable/reset.
Rich Freeman wrote: > On Tue, May 7, 2024 at 6:04 AM Michael wrote: >> On Tuesday, 7 May 2024 08:50:26 BST Dale wrote: >>> I'm aware of what it is and the cable part. I was curious what it looks >>> like to BIOS and the OS when one is connected and that pin has the drive >>> disabled. From what I've read in some places, the drive doesn't power >>> up at all. >> I don't have a drive like this, but as I understand it when the drive >> receives >> voltage on pin 3 it powers down. This requires a MoBo and firmware which >> supports such a function - probably unlikely to be found on consumer kit. > I have had these drives. If the drive is connected to many ATX power > supplies via a standard cable, the drive simply will not be detected > by the computer. With some power supplies it will work fine. It all > depends on whether the power supply follows the original SATA spec, or > was designed to be compatible with enterprise drives which use the > revised spec, which isn't backwards compatible (I don't know who the > genius was who had that idea). > > In order to actually toggle the reset line you need SOMETHING able to > switch the line in-between the drive and the PSU. That might be a > motherboard (especially with the newer trend towards running all the > power through the motherboard), or some other accessory card. Unless > the HBA provides the power it won't be there. > > However, you don't need any fancy hardware for the drive to just work > - that is only needed to send the hardware reset to the drive. All > you need is to not have that pin powered. That just means the right > power supply, the right cable, the right adapter, or some improvised > solution (tape over the pin is a common one). > > In any case, if the pin is the problem, the drive simply won't be > detected. Your SATA issues are due to something else. It might be a > bad drive, an incompatibility (maybe the drive isn't in the > smartmontools database yet), or maybe an issue with the HBA (for USB > HBAs in particular you often need to pass command line parameters as > there apparently isn't a standard way to pass these commands over > USB). I doubt the power line is your problem. > > As far as shucked drives go - that is typically indicated by the > label/model. If it isn't branded in any way it may have been shucked. > That shouldn't be a problem as long as you don't have the power issue > - the drive might simply be bad. > So, since the drive wasn't seen at all on my main rig or the NAS box, with a straight sata power connector, then it likely has the PWDIS pin. On those two rigs, the drive wasn't seen at all not even in the BIOS. Since it was seen on the old Dell rig with a molex sata adapter but SMART spit out a lot of errors, then the drive is bad. Now that makes sense and was kinda what I was thinking but not sure about. Not only did I get a drive with the PWDIS feature, I got a bad drive as well. While the box had no damage at all, it doesn't mean it didn't get hit or dropped and was damaged or went bad some other way. My take on this for future reference. If a drive isn't seen at all even by BIOS, either the drive is completely dead or it has the PWDIS feature. Oh, I wonder to about the genius who forgot to make this new feature backward compatible. I hope someone Gibbs smacks him real good. Maybe twice. Now to avoid buying another one of these drives again. I really wish sellers who should know would put in the description or list of features that the drive has PWDIS. After all, most buyers of small quantities of drives likely can't use that feature. The ones who do likely buy in bulk since they putting them in large systems. Thanks to all. I knew there would be someone on this list who actually had one of these things. Dale :-) :-)
Re: [gentoo-user] PERL_FEATURES
On 2024.05.07 12:33, Helmut Jarausch wrote:
Hi,
according to the NEWS from today one has to set PERL_FEATURES in
/etc/portage/make.conf.
But how to do that?
I've tried
PERL_FEATURES="debug ithreads quadmath"
but emerging dev-lang/perl-5.38.2-r3 I get
* As of dev-lang/perl-5.38.2-r3, the useflags debug, ithreads,
quadmath move into
* a use-expand variable PERL_FEATURES, which should be set globally
in make.conf.
* It appears that you have not set this variable properly yet.
Many thanks for hint,
Helmut
The section of the ebuild that decides whether or not to issue that
warning is:
pkg_pretend() {
if \
( use perl_features_ithreads && has_version
' ( ! use perl_features_ithreads && has_version
' ( use perl_features_quadmath && has_version
' ( ! use perl_features_quadmath && has_version
' ( use perl_features_debug&& has_version
' ( ! use perl_features_debug&& has_version
'
then
(issue the warning)
As I read it (which may well be wrong) it is looking to see whether the
value for each of the three items is the same by the old use flag and
by the new PERL_FEATURES. Check how those three use flags are set for
-r2 and that the value for PERL_FEATURES for -r3 matches for each.
[gentoo-user] PERL_FEATURES
Hi, according to the NEWS from today one has to set PERL_FEATURES in /etc/portage/make.conf. But how to do that? I've tried PERL_FEATURES="debug ithreads quadmath" but emerging dev-lang/perl-5.38.2-r3 I get * As of dev-lang/perl-5.38.2-r3, the useflags debug, ithreads, quadmath move into * a use-expand variable PERL_FEATURES, which should be set globally in make.conf. * It appears that you have not set this variable properly yet. Many thanks for hint, Helmut
Re: [gentoo-user] bad $PATH
On Tue, 7 May 2024 at 15:37, Neil Bothwick wrote: > On Tue, 7 May 2024 13:22:47 +0100, Jorge Almeida wrote: > > > > Yes, of course. When I said "emerge -c doesn't clean it" I meant > > > "emerge > > -c" (without arguments). I know how to unmerge a package, which in this > > particular case I should have done years ago, but didn't, and forgot > > about it. > > Right, so we were talking at cross purposes. Apologies for the line noise. > > No problem. Cheers
Re: [gentoo-user] bad $PATH
On Tue, 7 May 2024 13:22:47 +0100, Jorge Almeida wrote: > > Yes, of course. When I said "emerge -c doesn't clean it" I meant > > "emerge > -c" (without arguments). I know how to unmerge a package, which in this > particular case I should have done years ago, but didn't, and forgot > about it. Right, so we were talking at cross purposes. Apologies for the line noise. -- Neil Bothwick without C people would code in Basi, Pasal and Obol pgpGh2LN46Zaj.pgp Description: OpenPGP digital signature
Re: [gentoo-user] bad $PATH
On Tue, 7 May 2024 at 13:10, Neil Bothwick wrote: > On Tue, 7 May 2024 11:32:43 +0100, Jorge Almeida wrote: > > > > Files in @world should only be wanted because you put them in there, so > you should be able to remove them. > > Yes, of course. When I said "emerge -c doesn't clean it" I meant "emerge -c" (without arguments). I know how to unmerge a package, which in this particular case I should have done years ago, but didn't, and forgot about it. Jorge
Re: [gentoo-user] bad $PATH
On Tue, 7 May 2024 11:32:43 +0100, Jorge Almeida wrote: > > > I seem to remember I installed a long time ago. It is in the world > > > file, hence emerge -c wouldn't clean it. I don't have csh but think > > > I had it > > > > emerge -c will remove packages that are in @world. It was probably > > also a dependency of something else when you tried before. > > > Hmmm... This would seem to contradict the man page. From man emerge --depclean, -c Cleans the system by removing packages that are not associated with explicitly merged packages. Depclean works by creating the full dependency tree from the @world set, then comparing it to installed packages. Packages installed, but not part of the dependency tree, will be uninstalled by depclean. It is a little ambiguous in that does the dependency tree from the world set include the world set itself? > How would emerge -c > decide which packages in the world file were unwanted? Maybe you're > thinking of slotted packages? No, just tried it. Pick a file from @world and emerge -cpv it. [root@phoucgh ~ 0]% grep x11vnc /var/lib/portage/world x11-misc/x11vnc [root@phoucgh ~ 0]% emerge -cpv x11vnc Calculating dependencies... done! >>> Calculating removal order... >>> These are the packages that would be unmerged: x11-misc/x11vnc selected: 0.9.16-r8 protected: none omitted: none All selected packages: =x11-misc/x11vnc-0.9.16-r8 >>> 'Selected' packages are slated for removal. >>> 'Protected' and 'omitted' packages will not be removed. Packages installed: 2083 Packages in world:350 Packages in system: 49 Required packages:2082 Number to remove: 1 Files in @world should only be wanted because you put them in there, so you should be able to remove them. -- Neil Bothwick "What I need is a list of specific unknown problems we will encounter." pgpJ1zRxuqJO9.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Hard drive and PWDIS or pin 3 power disable/reset.
On Tue, May 7, 2024 at 6:04 AM Michael wrote: > > On Tuesday, 7 May 2024 08:50:26 BST Dale wrote: > > > > I'm aware of what it is and the cable part. I was curious what it looks > > like to BIOS and the OS when one is connected and that pin has the drive > > disabled. From what I've read in some places, the drive doesn't power > > up at all. > > I don't have a drive like this, but as I understand it when the drive receives > voltage on pin 3 it powers down. This requires a MoBo and firmware which > supports such a function - probably unlikely to be found on consumer kit. I have had these drives. If the drive is connected to many ATX power supplies via a standard cable, the drive simply will not be detected by the computer. With some power supplies it will work fine. It all depends on whether the power supply follows the original SATA spec, or was designed to be compatible with enterprise drives which use the revised spec, which isn't backwards compatible (I don't know who the genius was who had that idea). In order to actually toggle the reset line you need SOMETHING able to switch the line in-between the drive and the PSU. That might be a motherboard (especially with the newer trend towards running all the power through the motherboard), or some other accessory card. Unless the HBA provides the power it won't be there. However, you don't need any fancy hardware for the drive to just work - that is only needed to send the hardware reset to the drive. All you need is to not have that pin powered. That just means the right power supply, the right cable, the right adapter, or some improvised solution (tape over the pin is a common one). In any case, if the pin is the problem, the drive simply won't be detected. Your SATA issues are due to something else. It might be a bad drive, an incompatibility (maybe the drive isn't in the smartmontools database yet), or maybe an issue with the HBA (for USB HBAs in particular you often need to pass command line parameters as there apparently isn't a standard way to pass these commands over USB). I doubt the power line is your problem. As far as shucked drives go - that is typically indicated by the label/model. If it isn't branded in any way it may have been shucked. That shouldn't be a problem as long as you don't have the power issue - the drive might simply be bad. -- Rich
Re: [gentoo-user] bad $PATH
On Tue, 7 May 2024 at 10:36, Neil Bothwick wrote: > On Thu, 2 May 2024 16:37:24 +0100, Jorge Almeida wrote: > > > > > I seem to remember I installed a long time ago. It is in the world file, > > hence emerge -c wouldn't clean it. I don't have csh but think I had it > > emerge -c will remove packages that are in @world. It was probably also a > dependency of something else when you tried before. > > Hmmm... This would seem to contradict the man page. How would emerge -c decide which packages in the world file were unwanted? Maybe you're thinking of slotted packages? Jorge
Re: [gentoo-user] Hard drive and PWDIS or pin 3 power disable/reset.
On Tuesday, 7 May 2024 08:50:26 BST Dale wrote: > William Kenworthy wrote: > > See https://www.disctech.com/powerdisable > > > > BillK > > I'm aware of what it is and the cable part. I was curious what it looks > like to BIOS and the OS when one is connected and that pin has the drive > disabled. From what I've read in some places, the drive doesn't power > up at all. I've seen some claim it shows up but you can't access it. > I've read different explanations of what the drive does and maybe it > varies from one maker to another. What I was hoping for, someone on > this list has connected a drive with that pin disabling the drive and > can confirm what I posted is what it looks like or explain what the > system does when one is connected. > > When I connected that drive to my NAS box, it was very slow to go > through the BIOS post and where it usually lists the connected drives, > only the drive with the OS showed up. When connected to the older Dell > rig, it booted up normally. It doesn't list connected drives like the > NAS box does. It did show up in /proc/partitions. The SMART command > showed a different story. It makes it look like the drive has power but > may not be spinning up. It can get the info such as model number, > serial number, capacity and other info but can't read the SMART data, > which I assume is written to the platter part of the drive. I'm not > sure on that tho. What I don't know, is that because of the PWDIS pin > or has nothing to do with it. > > As it is, the seller agreed to take the drive back. If the seller says > the drive works for them, I guess it has that PWDIS feature. If it > doesn't work for them, just a bad drive, perhaps damaged in shipping or > something. I'm just curious as to how a drive behaves when it has that > PWDIS pin and the drive is disabled. > > I did notice in the pic on your link that the new power connector has 5 > wires. It has the PWDIS ability. Mine has 4 wires. However, I've read > that on the older systems, the pins inside the connector apply power to > pin 3 which disables the drives. Mine doesn't have that extra wire but > I don't think that really matters. Maybe I need to put a voltmeter on > the connector and see if there is anything on pin 3 or not. I suspect > there is tho. > > Has anyone had one of these drives connected and remember what the > system reports when disabled?? What clues it gives that shows it is > disabled if any?? > > Dale > > :-) :-) I don't have a drive like this, but as I understand it when the drive receives voltage on pin 3 it powers down. This requires a MoBo and firmware which supports such a function - probably unlikely to be found on consumer kit. https://documents.westerndigital.com/content/dam/doc-library/en_us/assets/ public/western-digital/collateral/tech-brief/tech-brief-western-digital-power- disable-pin.pdf If the drive does not spin up, you could try to insulate pin 3 with electrician's tape and see if the disk spins up, or use a molex power adaptor as already discussed. Usually you can feel the disk vibrating when powering up. Since you managed to make it spin but smartclt barfs, I would think there is something wrong with it. Either way, life is too short to bother with disks which do not work as you reasonably expect them to work. RMA it and/or buy a different disk after you confirm its replacement does not come with a PWDIS feature. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] bad $PATH
On Thu, 2 May 2024 16:37:24 +0100, Jorge Almeida wrote: > So, it seems I have at least one package which I wouldn't expect to > have: app-admin/xstow > > I seem to remember I installed a long time ago. It is in the world file, > hence emerge -c wouldn't clean it. I don't have csh but think I had it > once upon a time. Removing... emerge -c will remove packages that are in @world. It was probably also a dependency of something else when you tried before. -- Neil Bothwick "Mmmm, trouble with grammer have I, yes?" - Yoda pgp4VKqIbtUUd.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Hard drive and PWDIS or pin 3 power disable/reset.
William Kenworthy wrote: > See https://www.disctech.com/powerdisable > > BillK > I'm aware of what it is and the cable part. I was curious what it looks like to BIOS and the OS when one is connected and that pin has the drive disabled. From what I've read in some places, the drive doesn't power up at all. I've seen some claim it shows up but you can't access it. I've read different explanations of what the drive does and maybe it varies from one maker to another. What I was hoping for, someone on this list has connected a drive with that pin disabling the drive and can confirm what I posted is what it looks like or explain what the system does when one is connected. When I connected that drive to my NAS box, it was very slow to go through the BIOS post and where it usually lists the connected drives, only the drive with the OS showed up. When connected to the older Dell rig, it booted up normally. It doesn't list connected drives like the NAS box does. It did show up in /proc/partitions. The SMART command showed a different story. It makes it look like the drive has power but may not be spinning up. It can get the info such as model number, serial number, capacity and other info but can't read the SMART data, which I assume is written to the platter part of the drive. I'm not sure on that tho. What I don't know, is that because of the PWDIS pin or has nothing to do with it. As it is, the seller agreed to take the drive back. If the seller says the drive works for them, I guess it has that PWDIS feature. If it doesn't work for them, just a bad drive, perhaps damaged in shipping or something. I'm just curious as to how a drive behaves when it has that PWDIS pin and the drive is disabled. I did notice in the pic on your link that the new power connector has 5 wires. It has the PWDIS ability. Mine has 4 wires. However, I've read that on the older systems, the pins inside the connector apply power to pin 3 which disables the drives. Mine doesn't have that extra wire but I don't think that really matters. Maybe I need to put a voltmeter on the connector and see if there is anything on pin 3 or not. I suspect there is tho. Has anyone had one of these drives connected and remember what the system reports when disabled?? What clues it gives that shows it is disabled if any?? Dale :-) :-)
