Skip to contentSkip to conversation

Debian maintainer does not forward bug reports

DISCUSSION Bjarni Ingi Gislason 3 participants 4 messages Apr 21 – Apr 25 PCI

lore
Share & tools
lore
#0 lore raw

See

https://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=pciutils;dist=testing

-.-

Additional remarks.

Mails from me to "[email protected]" are no longer acknowledged. A
Debian maintainer told me, that he would contact the mail administrator about me not wanting to send bugs upstream.

-.-

From "/usr/share/doc/debian/bug-reporting.txt.gz":

Don't file bugs upstream

If you file a bug in Debian, don't send a copy to the upstream software maintainers yourself, as it is possible that the bug exists only in Debian. If necessary, the maintainer of the package will forward the bug upstream.

-.-

For forwarding bug reports to upstream see:

https://www.debian.org/Bugs/Developer#forward

-.-

I do not send reports upstream if I have to get an account there.
The Debian maintainers have one already.

If I get a negative (or no) response from upstream, I send henceforth
bugs to Debian.

#1 lore raw

Cc: Guillem who is the Debian maintainer of pciutils; see below for context.

Let us go through all outstanding bugs filed against Debian pciutils:

Quoted 2 lines

#887113 pciutils: lspci does not detect sound for nvidia MPC67 on Pavilion DV9720us
#775453 /usr/bin/lspci: "lspci does not detect nvidia geforce 210M on Asus UL80Vt"

These are likely kernel issues. Since they are a decade old, we should ask the submitter for testing on a recent kernel.

Quoted 1 line

#656925 failure to detect PCIe Gigabit LAN

Not a pciutils problem, probably a missing kernel driver. It's from 2012, so it should be tested with a recent kernel.

Quoted 1 line

#774063 pciutils: update-pciids changes md5sum of pci.ids

This is specific to Debian. Another alternative would be to make update-pciids store the downloaded file to /usr/local/share and let libpci prefer this file over the one if /usr/share if it exists.

Quoted 1 line

#595328 /usr/bin/lspci: lspci reports wrong video ram size

It's from 2010, so I am not sure if it's worth investigating.

When a bug like this is submitted, please ask the author to provide the output of "lspci -vvxxx".

Quoted 1 line

#640239 /usr/bin/lspci: lspci -t (-v) doesn't show bridges

We have rewritten the whole tree display code since this version, I hope that this problem does not exist any more.

Quoted 1 line

#870000 pciutils: Debian Stretch freeze when I use "lspci"

Very likely a hardware bug. If it should be worked around, the right place is the kernel.

Quoted 1 line

#924237 lspci shows wrong/misleading string instead of "Device xxxx"

Interesting... can you ask the submitter if they can reproduce it with current pciutils?

Quoted 3 lines

#1087569 pcilmr.8: Some remarks and editorial changes for this man page
#1087678 pcilib.7: Some remarks and editorial changes for this man page
#1099040 setpci.8: Some remarks and a patch with editorial changes for this man page

Bjarni, I assume that these are the bugs that triggered your message. I don't see any Debian patches to pciutils man pages, so I will apply your patches upstream.

Quoted 1 line

#585146 lspci: query the database for specific vendor/device IDs

This could be useful. I will consider adding this feature.

Quoted 1 line

#929984 lspci: please make -nn the default

I admit that it would help in some cases, but on the other hand, it would make the default output more cluttered, so I don't think it's a clear win.

Have a nice fortnight

Signature · 4 lines
Martin `MJ' Mareš                        <[email protected]>   http://mj.ucw.cz/
United Computer Wizards, Prague, Czech Republic, Europe, Earth, Universe

---
Quoted 36 lines

See

https://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=pciutils;dist=testing

-.-

Additional remarks.

Mails from me to "[email protected]" are no longer acknowledged. A
Debian maintainer told me, that he would contact the mail administrator about me not wanting to send bugs upstream.

-.-

From "/usr/share/doc/debian/bug-reporting.txt.gz":

Don't file bugs upstream

If you file a bug in Debian, don't send a copy to the upstream software maintainers yourself, as it is possible that the bug exists only in Debian. If necessary, the maintainer of the package will forward the bug upstream.

-.-

For forwarding bug reports to upstream see:

https://www.debian.org/Bugs/Developer#forward

-.-

I do not send reports upstream if I have to get an account there.
The Debian maintainers have one already.

If I get a negative (or no) response from upstream, I send henceforth
bugs to Debian.

#2 lore raw

Hi!

Quoted 4 lines · Martin Mareš

Cc: Guillem who is the Debian maintainer of pciutils; see below for context.

Let us go through all outstanding bugs filed against Debian pciutils:

Ah, thanks for that. When I took over the package in Debian in 2019, I had a brief look, but then never found the moment or forgot to really go over all of them. :/

Quoted 5 lines
Quoted 2 lines

#887113 pciutils: lspci does not detect sound for nvidia MPC67 on Pavilion DV9720us
#775453 /usr/bin/lspci: "lspci does not detect nvidia geforce 210M on Asus UL80Vt"

These are likely kernel issues. Since they are a decade old, we should ask the submitter for testing on a recent kernel.

Ack, I've asked the reporters.

Quoted 4 lines
Quoted 1 line

#656925 failure to detect PCIe Gigabit LAN

Not a pciutils problem, probably a missing kernel driver. It's from 2012, so it should be tested with a recent kernel.

Ack, I've asked the reporter.

Quoted 5 lines
Quoted 1 line

#774063 pciutils: update-pciids changes md5sum of pci.ids

This is specific to Debian. Another alternative would be to make update-pciids store the downloaded file to /usr/local/share and let libpci prefer this file over the one if /usr/share if it exists.

I've had a couple of packaging changes for pciutils and pci.ids to switch them to use alternatives:

https://git.hadrons.org/cgit/debian/pkgs/pci.ids.git/log/?h=pu/alternatives
https://git.hadrons.org/cgit/debian/pkgs/pciutils.git/log/?h=pu/alternatives

But never merged those, because then /usr/share/misc/pci.ids would stop being known by dpkg, and then users cannot search for its owner. Which is arguably a problem in dpkg itself, but…

At least the nice thing about the alternatives solution is that it defaults to use the latest version of the file, where using a set of directories (/var/lib/ comes to mind as perhaps better than /usr/local/share) with fixed preferred order, would not have that property.

Quoted 3 lines
Quoted 1 line

#595328 /usr/bin/lspci: lspci reports wrong video ram size

It's from 2010, so I am not sure if it's worth investigating.

I've closed this one as it was affecting a motherboard for 32-bit x86, which Debian does not support anymore on real hardware.

Quoted 2 lines

When a bug like this is submitted, please ask the author to provide the output of "lspci -vvxxx".

Thanks, I'll add a package script for reportbug to print a note to that effect so that the submitter can see it in the template report to fill.

Quoted 4 lines
Quoted 1 line

#640239 /usr/bin/lspci: lspci -t (-v) doesn't show bridges

We have rewritten the whole tree display code since this version, I hope that this problem does not exist any more.

I've closed this now, on your request.

Quoted 4 lines
Quoted 1 line

#870000 pciutils: Debian Stretch freeze when I use "lspci"

Very likely a hardware bug. If it should be worked around, the right place is the kernel.

Ack, I've asked the reported whether this can still be reproduced.

Quoted 4 lines
Quoted 1 line

#924237 lspci shows wrong/misleading string instead of "Device xxxx"

Interesting... can you ask the submitter if they can reproduce it with current pciutils?

Ack, done.

Quoted 7 lines
Quoted 3 lines

#1087569 pcilmr.8: Some remarks and editorial changes for this man page
#1087678 pcilib.7: Some remarks and editorial changes for this man page
#1099040 setpci.8: Some remarks and a patch with editorial changes for this man page

Bjarni, I assume that these are the bugs that triggered your message. I don't see any Debian patches to pciutils man pages, so I will apply your patches upstream.

Thanks, I think these just slipped through the cracks on my side.

Quoted 3 lines
Quoted 1 line

#585146 lspci: query the database for specific vendor/device IDs

This could be useful. I will consider adding this feature.

Ah, great, let me know if you'd like a report filed upstream for tracking.

Quoted 5 lines
Quoted 1 line

#929984 lspci: please make -nn the default

I admit that it would help in some cases, but on the other hand, it would make the default output more cluttered, so I don't think it's a clear win.

Ack. I've closed this now.

Thanks,
Guillem

#3 lore raw
Quoted 6 lines
Quoted 3 lines
Quoted 1 line

#585146 lspci: query the database for specific vendor/device IDs

This could be useful. I will consider adding this feature.

Ah, great, let me know if you'd like a report filed upstream for tracking.

Yes please.

Martin