--- Day changed Mon Dec 22 2014 | ||
mor1kx | [mor1kx] skristiansson pushed 2 new commits to master: https://github.com/openrisc/mor1kx/compare/0ebb2db56cc6...adce0adf164f | 01:28 |
---|---|---|
mor1kx | mor1kx/master 3c39f03 Stefan Kristiansson: dmmu: add support for multiway tlb... | 01:28 |
mor1kx | mor1kx/master adce0ad Stefan Kristiansson: immu: add support for multiway tlb... | 01:28 |
stekern | my 'random' replacement routine in Linux is appalling... | 01:53 |
stekern | low bits of the tick timer | 01:53 |
stekern | wallento isn't around, I wanted to ask him if he's planning on adding support for the LRU bits together with the dirty and accessed bits | 01:54 |
stekern | anyway, appalling sw or not, the hw seems to do it's job correctly now at least | 01:56 |
squaresinewave | . | 06:27 |
stekern | wallento: I pushed the multiway changes last night, I'm kind of missing the LRU field though. Was that something you planned to include in your dirty&accessed changes? | 12:35 |
blueCmd | olofk: single snapshot | 13:06 |
* olofk also feels dirty&accessed sometimes | 13:56 | |
stekern | olofk: what asus laptop was it that you had? | 14:06 |
wallento | stekern: I think the accessed and dirty are not relevant for replacement, as they file if multiple pages have been accessed | 14:21 |
wallento | we need an extra LRU field | 14:21 |
wallento | not sure how this fits here | 14:21 |
wallento | is something like this in the spec? | 14:21 |
stekern | wallento: the LRU field is already there | 14:25 |
stekern | and I agree that it's not related on a conceptional level, implementation wise I think it might be related though | 14:27 |
stekern | since you can probably reuse the same read-modify-write logic you'll need for the dirty&accessed | 14:28 |
olofk | stekern: I think it's a Zenbook Touch UX31A. Is there somewehere in /sys I can read this out? | 14:28 |
stekern | not sure, I finally got around my decision agony and ordered one of these: http://www.verkkokauppa.com/fi/product/0840/dtxds/Asus-UX32LN-13-3-i7-4510U-8-Gt-128-Gt-SSD-GT-840M-Windows-8 | 14:29 |
olofk | stekern: Looks quite similar, but with a Nvidia card and without the touch screen | 14:30 |
olofk | Touch screen is the worst fucking thing ever with a 1 1/2 year old kid | 14:31 |
stekern | yeah, I decided against touch screen | 14:31 |
stekern | I was looking at a lenovo yoga2 13 for a long while | 14:32 |
stekern | similar specs, but without ssd and with touch and flipable screen | 14:35 |
olofk | When I bought this one, I looked at three different brands with similar specs. I went for this one since it had the least compromises when it came to the keyboard | 14:37 |
stekern | mmm, the lenovo had a pretty decent keyboard as well | 14:38 |
olofk | The Acer only had four rows, so function keys were shared with number keys | 14:38 |
olofk | There was something with the Lenovo that I didn't like. When I tried it in the store I missed some of the keys | 14:39 |
stekern | oh, that sucks big time | 14:39 |
wallento | stekern: yes, this sounds good, I will have a look at it | 14:39 |
wallento | on lenovo the function keys and F-keys are overlayed | 14:39 |
wallento | and insert and end share a key | 14:40 |
stekern | the default behaviour of lenovo and hp that you have to press fn key to get f1-f10 sucks too | 14:40 |
stekern | luckily you can revert that behaviour | 14:40 |
wallento | you can change this in the bios | 14:40 |
wallento | and lenovo keyboards have backlight, which I don't want to miss again | 14:40 |
olofk | My biggest problem with this keyboard is that Page up/down home end are on the arrow keys, and my hand isn't big enough to reach between fn and the keys | 14:41 |
olofk | Makes onehanded operation annoying | 14:41 |
olofk | Yeah. Backlit keyboard was surprisingly nice | 14:41 |
stekern | yes, that was another factor, both the yoga and the asus has that | 14:41 |
stekern | the lenovo had dedicated full-size page-up and pagedown | 14:42 |
stekern | *but* no free room next to the arrow keys, which annoys the hell out of me | 14:42 |
stekern | ...because I have a habit of taping that empty space when I'm thinking ;) | 14:43 |
olofk | haha | 14:43 |
olofk | Yeah, I came to realize that the keyboard is one of the most important things when buying a new laptop. Thankfully I had a store nearby where I could try out all options before ordering from somewhere else :) | 14:43 |
wallento | yes, thats a problem. imo they can remove this silly pad or make it smaller and add the extra row again | 14:44 |
wallento | lenovo at least has the keys where you expect them, arrows etc. | 14:44 |
wallento | on dell keyboards they are scattered through all free spaces.. | 14:44 |
stekern | another important thing is that the ctrl is located on the left side of the fn key | 14:46 |
olofk | stekern: Yes! Switching those is incredibly annoying | 14:46 |
wallento | this is not the case for lenovo I think | 14:46 |
olofk | I think that was more common a few years ago | 14:47 |
olofk | Has anyone used the Altera ALTDQ/ALTDQS IPs? | 14:48 |
stekern | nope | 14:53 |
olofk | Time to explore Vaarainjärvi now! | 15:00 |
stekern | did they pick names of lakes for their releases because they are anticipating thousands of releases? | 15:07 |
olofk | lol | 15:07 |
_franck__ | olofk: I want to compile some software before starting a simulation in fusesoc. What do you thinnk of introducing a way to start a script before anything (this script would run make or/and something else) | 17:52 |
olofk | _franck__: Yes, I think that's a good idea. I want to be able to execute custom scripts between all stages | 18:07 |
_franck__ | ok great | 18:07 |
olofk | There's probably a few more env vars that would be nice to export to the scripts as well. Please let me know if you can think of any | 18:08 |
_franck__ | olofk: we already have pre_build_script, I just enabled it for cores: https://github.com/fjullien/fusesoc/commit/3a3a9362650c4183003501137b4568e737c78999 | 18:31 |
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!