Message ID | cover.1357485747.git.thomas.petazzoni@free-electrons.com |
---|---|
State | Accepted |
Headers | show |
>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes:
Thomas> Hello,
Thomas> Here is a set of patches that add the elfutils and perf packages in
Thomas> Buildroot, based on previous efforts from Stefan Fröberg and Kaiwan
Thomas> Billimoria (thanks !).
Thomas> These packages are not trivial because:
Thomas> 1) perf is part of the kernel sources
Thomas> 2) elfutils is a nightmare to build against uClibc
Thomas> I've been working on this for about two days, trying different
Thomas> solutions, the one proposed here appeared to be the most reasonable
Thomas> one in terms of patch size, and maintenability.
Thanks everyone, committed series.
Thomas, please don't forget to (try to) send these patches upstream so
we don't have to maintain them ourselves in the future.
Dear Peter Korsgaard, On Mon, 07 Jan 2013 23:32:56 +0100, Peter Korsgaard wrote: > Thanks everyone, committed series. Thanks for merging them. Hopefully, it won't make the autobuilders explode too much. > Thomas, please don't forget to (try to) send these patches upstream so > we don't have to maintain them ourselves in the future. Which ones exactly? The big fat patches we are downloading from fedorahosted.org? Our patches? All our patches are really specific to uClibc support, and seeing the number of glibc-isms in elfutils, I am not sure the maintainers will be interested in making it uClibc friendly. But it's maybe worth trying. One nice thing that could be done is migrate the linux-kernel-module.c thing to use nftw(), so we could drop the fts() mess. Thomas
>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes: Hi, >> Thomas, please don't forget to (try to) send these patches upstream so >> we don't have to maintain them ourselves in the future. Thomas> Which ones exactly? The big fat patches we are downloading from Thomas> fedorahosted.org? Our patches? Thomas> All our patches are really specific to uClibc support, and Thomas> seeing the number of glibc-isms in elfutils, I am not sure the Thomas> maintainers will be interested in making it uClibc Thomas> friendly. But it's maybe worth trying. Our own. I would certainly say it's worth a try. Thomas> One nice thing that could be done is migrate the Thomas> linux-kernel-module.c thing to use nftw(), so we could drop the Thomas> fts() mess. Indeed.