Hi. Thanks for reply. For the record (I forgot to mention this), the internal mouse works fine after installation though. Re -5, I believe the kernel is requesting it even after loading -6 (which appears to be sufficient). I can get the actual log entry when I get home. Don't worry about spending any time fixing the firmware issues -- I'd rather have you work on stuff related to free software and getting the release out. Thought I'd just send an install report so my experience was documented, in case the documentation of my problem is useful for anyone else. /Simon > Hi Simon, > > and thanks for your report! > > Simon Josefsson <simon@josefsson.org> (2015-01-27): > > Installation went fine, but some nits: > > > > 1) I tried a graphical install, but the builtin mouse did not work > > well: moving the cursor didn't work, but clicking buttons did. > > I really should get my hands on some such laptop to debug this issue. > > Since we have several of those already, I'm not sure cloning a > separate bug report is needed here. > > > 2) It queries for IWLWIFI firmware and mentions the following > > firmwares: > > > > iwlwifi-6000g2b-5.ucode iwlwifi-6000g2b-6.ucode > > iwlwifi-6000g2b-5.ucode iwlwifi-6000g2b-6.ucode > > > > Note how the filenames are duplicated. This seems like a cosmetic > > problem? > > Deduplicating shouldn't cost much, yeah. $files and $modules in the > check_missing() function of check-missing-firmware.sh (package > hw-detect), if somebody wants to take a look. Can be cloned away. > > > 3) After inserting a USB stick with firmware, it loads > > iwlwifi-6000g2b-6.ucode fine (and wifi works). However, it still > > complains about missing firmware files: > > > > iwlwifi-6000g2b-5.ucode iwlwifi-6000g2b-5.ucode > > > > I click 'No' and continue installation. The -5 firmware file is > > not included in the firmware-iwlwifi package -- should it be? > > ISTR there's some kind of versioning here, and without knowing about > it much, I'd guess -6 obsoleted -5 and is sufficient there. > > > Wifi appears to be working fine without it, though. Or should > > the installer be tought that the firmware isn't needed? > > I'm not familiar with these components, but: maybe the firmware check > doesn't distinguish between early and later requests? I'd expect the > kernel not to complain anymore about a missing -5 if it's presented > with -6 in the meanwile. That should be easy to check in your > installer logs (/var/log/installer/syslog). > > If the kernel indeed complains about -5 on the second attempt, I'm not > sure how we can deal with that; if it doesn't, chaning check_missing() > to only look at later kernel logs shouldn't be hard; in which case > that can be cloned away. > > > 4) The machine also requires firmware for the builtin NIC > > (rtl_nic/rtl8168e-2.fw) but I don't get a question about it. > > While this may be intention (there is just one global > > load-non-free-firmware-yes-no?), it is a bit sub-optimal. Maybe > > I only want to load firmware for one of the hardware pieces in my > > machine and not the others? This doesn't seem possible. So this > > is a feature request to prompt with explicit filenames for every > > different hardware separately. > > Again, not familiar with these components, so I'll stick "don't know" > for now, sorry. > > Mraw, > KiBi.
Attachment:
pgpk8safXsn_L.pgp
Description: OpenPGP digital signatur