Julien R Pierre - Sun Microsystems wrote:
Most intel OS's don't mark their binaries based on any special instruction needs. Instead application pretty regularly use the logic "if SSE2 is available, do this, otherwise do that". There is no architectural problem with this on windows (Thanks for drawing my attention to the patch, I see nothing inherently wrong and believe we should pick it up!).Nelson,Nelson B Bolyard wrote:Julien R Pierre - Sun Microsystems wrote, On 2009-01-21 15:03:I don't like much the way that we implemented SSE2 on Linux - together in a single freebl shared library with the non-SSE2 version. That stands in the way of us implementing SSE2 in Solaris. See bug 458480 about that. Also bug 439199 about the Windows implementation.
It's solaris that has the weirdness, and is why we added the ability to load different binaries for different instruction set on the same platform. Please do not misconstrue something we did to support a particular platform as the "way it must be done" for all platforms.
BTW, I don't have a problem with using that mechanism on Solaris, I do have a problem with forcing it on all other platforms which otherwise derive zero benefit from the cost of having to support multiple binaries and one is sufficient.
The Linux-like way is perfectly acceptable in Darwin. We have a proof of concept already.I am not sure what's the best answer for Darwin.
bob
smime.p7s
Description: S/MIME Cryptographic Signature
-- dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto
