On Sat, May 26, 2012 at 3:26 PM, Peter Stuge <pe...@stuge.se> wrote: > Xiaofan Chen wrote: >> The problem is > .. >> There is another problem > > I believe I pointed out these issues during review of the new libusb > code.
I know the configure problem long time ago but to me auto-tools are anyway hard and difficult and ugly and ... But I know the alternatives (eg: CMake) are also not that easy and not that beautiful anyway. So I do not really consider configure hassle to be a big problem. The configure problem is not introduced by the libusb-1.0 inclusion anyway, it was there always. We certainly have different opinions when it comes to whether to require pkg-config as compulsory. http://openocd.zylin.com/#/c/491/ But even you mentioned that using CLAGS and LDFLAGS is okay in the above review. "I think it's perfectly fine to require that pkg-config is installed, but regardless it is always possible to manually specify CFLAGS and LDFLAGS to make configure find libusb-1.0." > I'm glad that these problems are clear to you! Too bad that they > weren't during the review, maybe more review would have avoid it's > premature inclusion, and maybe, just maybe, the code would then have > been fixed already. Hey I was just one of the reviewers at that time and I think it is good to clear up the J-Link driver codes anyway at that time. I do not see it as a regression anyway. You were part of the review process anyway. And you did not strongly protest at that time. http://openocd.zylin.com/#/c/33/ Since you know libusb-1.0 quite well, why not fix it. If you just keep complaining the "premature inclusion" of the patch and not coming out with a patch, the thing will always be there. In the following thread, you even advocate the revert of the libusb-1.0 patch and you think you are the only one to fix it. http://comments.gmane.org/gmane.comp.debugging.openocd.devel/20586 "You know just as well as I do that the only patch for this that "we" will get will have to come from me, since noone else seems to really care much about it." In that case, why not spend some time to come out with some proposal to fix it. It was May 9 when you said the above, today it is May 26, if you really really care, you may already have some proposal on the table already... Actually if we count from the following review where you mentioned that "Unfortunately, the libusb1 code in OpenOCD is completely fubar", it is nearly 3 months... http://openocd.zylin.com/#/c/491/ -- Xiaofan ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ OpenOCD-devel mailing list OpenOCD-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openocd-devel
