This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: x86/MMX/SSE shift by immediate



On Dec 18, 2003, at 2:25 PM, Richard Henderson wrote:


On Fri, Dec 12, 2003 at 04:41:43PM -0800, Stuart Hastings wrote:
This looks like a typo that crept into several similar vector
instruction patterns.  Currently, MMX/SSE shifts by an immediate value
ICE.

Patch is ok.

Thank-you.


Test cases need to either be compile-only or have a runtime test
for no sse available.  There's code for the later in the testsuite
already -- see gcc.dg/20020523-2.c.

Very clever. I'll revise my testcase to differentiate between an MMX versus SSE versus non-vector host, and test as appropriate.


One question: would you prefer I cut&paste the CPU detection code from 20020523-2.c, or would you like to see it factored into a shared header file ? Suggestions, please.

I'll hold off committing the i386.md patch 'till the testcases are fixed, and commit them as a set.

stuart hastings
Apple Computer


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]