fat-binaries

CUDA: How to use -arch and -code and SM vs COMPUTE

喜夏-厌秋 提交于 2019-11-27 08:01:49
I am still not sure how to properly specify the architectures for code generation when building with nvcc. I am aware that there is machine code as well as PTX code embedded in my binary and that this can be controlled via the controller switches -code and -arch (or a combination of both using -gencode ). Now, according to this apart from the two compiler flags there are also two ways of specifying architectures: sm_XX and compute_XX , where compute_XX refers to a virtual and sm_XX to a real architecture. The flag -arch only takes identifiers for virtual architectures (such as compute_XX )

How do i compile a static library (fat) for armv6, armv7 and i386

做~自己de王妃 提交于 2019-11-27 06:50:37
I know this question has been posed several times, but my goal is slightly different with regard to what I have found searching the web. Specifically, I am already able to build a static library for iPhone, but the final fat file I am able to build only contains arm and i386 architectures (and I am not sure to what arm refers: is v6 or v7?). I am not able to compile specifically for armv6 and armv7 and them merge both architectures using lipo. The lipo tool complains that the same architecture (arm, not armv6 or armv7) is present in both the armv6 and armv7 libraries. Can someone explain

How do i compile a static library (fat) for armv6, armv7 and i386

余生颓废 提交于 2019-11-26 12:12:40
问题 I know this question has been posed several times, but my goal is slightly different with regard to what I have found searching the web. Specifically, I am already able to build a static library for iPhone, but the final fat file I am able to build only contains arm and i386 architectures (and I am not sure to what arm refers: is v6 or v7?). I am not able to compile specifically for armv6 and armv7 and them merge both architectures using lipo. The lipo tool complains that the same