Re: Does TYPE_MAX_VALUE (size_type_node) overflow?

2011-01-21 Thread Richard Guenther
On Thu, Jan 20, 2011 at 9:50 PM, Florian Weimer wrote: > * Richard Guenther: > >> On Wed, Jan 19, 2011 at 10:53 PM, Florian Weimer wrote: >>> I get strange warnings when I do arithmetic involving TYPE_MAX_VALUE >>> (size_type_node), in particular this code: >>> >>> /* Multiplies MUL1 with MUL2, a

Turn on -funroll-loops at -O3?

2011-01-21 Thread H.J. Lu
Hi, SInce -O3 turns on vectorizer, should it also turn on -funroll-loops? -- H.J.

Re: Turn on -funroll-loops at -O3?

2011-01-21 Thread Richard Guenther
On Fri, Jan 21, 2011 at 7:43 PM, H.J. Lu wrote: > Hi, > > SInce -O3 turns on vectorizer, should it also turn on > -funroll-loops? We need to split the tree and RTL effects of -funroll-loops (and -fpeel-loops). -funroll-loops can at the moment lead to unneccessary code bloat. But yes, something

Backporting the 4.5 plugin framework to 3.4.5

2011-01-21 Thread Kyle Girard
Hello, I currently have a plugin for gcc 4.5 that works great. However, the need has arisen to have the same plugin run on gcc 3.4.5. Knowing that the plugin api wasn't added until 4.5 I was wondering if anyone could tell me how much pain i would be in for if I attempt to backport the plugin api

Re: Strangeness with SSE(1)

2011-01-21 Thread David Mathog
Can somebody please explain the behavior of the following program to me? cat >test.c < #include #include int main(void){ register __m128 var; fprintf(stdout,"pre %X\n",var); var = _mm_setzero_ps(); fprintf(stdout,"post %X\n",var); fprintf(stdout,"zerof %X\n",0.0f); exit(EXIT_SU

Re: Backporting the 4.5 plugin framework to 3.4.5

2011-01-21 Thread Benjamin Smedberg
On 1/21/2011 4:12 PM, Kyle Girard wrote: Hello, I currently have a plugin for gcc 4.5 that works great. However, the need has arisen to have the same plugin run on gcc 3.4.5. Knowing that the plugin api wasn't added until 4.5 I was wondering if anyone could tell me how much pain i would be in f

Re: Turn on -funroll-loops at -O3?

2011-01-21 Thread Tim Prince
On 1/21/2011 10:43 AM, H.J. Lu wrote: Hi, SInce -O3 turns on vectorizer, should it also turn on -funroll-loops? Only if a conservative default value for max-unroll-times is set 2<= value <= 4 -- Tim Prince

Re: Strangeness with SSE(1)

2011-01-21 Thread Ian Lance Taylor
"David Mathog" writes: > Can somebody please explain the behavior of the following program > to me? This question is not appropriate for the gcc@gcc.gnu.org mailing list, which is for the development of gcc itself. It would be appropriate on gcc-h...@gcc.gnu.org. Please take any followups to g

[pph] Merge from trunk r169107

2011-01-21 Thread Diego Novillo
I've merged trunk rev 169107 into the pph branch. No new failures. Lawrence, you will probably find little/no merge conflicts with your timevar patch. Diego.