I want to dynamically assign values to variables using eval.
The following dummy example works:
var_name="fruit"
var_value="orange"
eval $(echo $var_name=$var_value)
echo $fruit
orange
However, when the variable value contains spaces, eval returns an error, even if $var_value is put between do...
@terdon Touche. Well, it was about the battle between good and evil, female empowerment and all that jazz. Or so the womens studies people would doubtless say.
I think there is a positive message or two buried in there somewhere. Though of course it was mostly about making money. Like pretty much all of US television.
At some point they had angels appearing I think. Dunno, watched it many years ago. I loved the first few seasons but then it devolved into a soap opera as I recall.
@FaheemMitha Not true. Watch some of the good stuff. Breaking Bad, Six Feet Under, Sopranos, The Wire. You know, series where the people actually act, not just sit there looking pretty.
@terdon I've never seen any of those. I hate crime stuff. Is any of that stuff actually not about crime? About, you know, normal people, who aren't criminals?
@FaheemMitha Well, like how a linear equation is y = m*x + b. 'm' and 'b' are the normal variables used there. Of course, the math works the same with y = a*x + b, but that's not the tradition.
@FaheemMitha Yeah. None of those are part of your kernel image, since distro kernels build almost all of them as modules. Try du -sh /lib/modules/$(uname -r)
@derobert I don't think there is really any rule here. And i've got two math degrees, and a statistics degree, and have seen too many equations to count.
@FaheemMitha The other one is taking the Debian config as a base, adding a patch, saying 'm' to any new questions, and building that. Patch due to bugzilla.kernel.org/show_bug.cgi?id=82961
@FaheemMitha The other one is taking the Debian config as a base, adding a patch, saying 'm' to any new questions, and building that. Patch due to https://bugzilla.kernel.org/show_bug.cgi?id=82961
@FaheemMitha No. No one understands the behavior my machine exhibits. So I keep building custom kernels. Someday I might get fed up and send a patch to add a boot flag instead...
@derobert Don't understand. You can get a patch to fix the issue, but nobody understands the behavior? Is this related to one of the questions you asked here?
@FaheemMitha Yep. Nobody understands the behavior. The short story is that Intel realized the hardware in this machine has a bug. Their workaround is to disable a feature in the chipset. But on this machine (and, apparently, only this machine) disabling that feature results in problems.
So the patch I have is just to ignore Intel, and leave the supposedly broken feature enabled.
Given, I don't know if anyone else with the same board has tested. Maybe its unique to the mobo.
Its odd too, because the vendor (who'd normally be responsible for disabling the feature in the BIOS in response to the Intel errata) instead puts out new BIOS versions enhancing the feature. So they act as if the errata is not there...
And they don't support Linux, of course, so are no help.
Bit of historical trivia. Once upon a time, a chap called Stallman was trying to write a free (as in freedom) C compiler. He saw Prof. Tanenbaum has written a compiler entitled Free University Compiler Kit, and apparently misunderstood the meaning of the word Free used there. As Stallman puts it, he derisively replied that the university was free, but the compiler was not, and then tried to recruit Stallman to work on the compiler.