| 1 |
Hans-Werner Hilse wrote: |
| 2 |
> Hi, |
| 3 |
> |
| 4 |
> On Mon, 20 Nov 2006 15:52:13 +0100 |
| 5 |
> Bo Ørsted Andresen <bo.andresen@××××.dk> wrote: |
| 6 |
> |
| 7 |
> |
| 8 |
>> Well, actually the mplayer ebuild don't care about your CFLAGS unless you |
| 9 |
>> enable the custom-cflags USE flag. Otherwise march is entirely unset. |
| 10 |
>> |
| 11 |
> |
| 12 |
> No, the mplayer configure script would autodetect optimization and set |
| 13 |
> the result as -march and -mcpu (and will also enable -O4). |
| 14 |
> |
| 15 |
> All such checking would of course be runtime only and your statement |
| 16 |
> about march would be true if the cpudetection USE flag is set. |
| 17 |
> |
| 18 |
> Don't ask me about what happens when both custom-cflags and |
| 19 |
> cpudetection are enabled... :-) |
| 20 |
> |
| 21 |
> -hwh |
| 22 |
> |
| 23 |
> |
| 24 |
|
| 25 |
I don't think there is such an option for gcc "-O4". AFAIK the shorthand |
| 26 |
optimization options are -O, -O0, -O1, -O2, -O3 and -Os" At least the |
| 27 |
man page reads: |
| 28 |
|
| 29 |
|
| 30 |
Optimization Options |
| 31 |
-falign-functions=n -falign-jumps=n -falign-labels=n -falign-loops=n |
| 32 |
-fbounds-check -fmudflap -fmudflapth -fmudflapir |
| 33 |
-fbranch-probabilities -fprofile-values -fvpt |
| 34 |
-fbranch-target-load-optimize -fbranch-target-load-optimize2 |
| 35 |
-fbtr-bb-exclusive -fcaller-saves -fcprop-registers |
| 36 |
-fcse-follow-jumps -fcse-skip-blocks -fcx-limited-range |
| 37 |
-fdata-sections -fdelayed-branch -fdelete-null-pointer-checks |
| 38 |
-fearly-inlining -fexpensive-optimizations -ffast-math -ffloat-store |
| 39 |
-fforce-addr -ffunction-sections -fgcse -fgcse-lm -fgcse-sm |
| 40 |
-fgcse-las -fgcse-after-reload -floop-optimize -fcrossjumping |
| 41 |
-fif-conversion -fif-conversion2 -finline-functions |
| 42 |
-finline-functions-called-once -finline-limit=n |
| 43 |
-fkeep-inline-functions -fkeep-static-consts -fmerge-constants |
| 44 |
-fmerge-all-constants -fmodulo-sched -fno-branch-count-reg |
| 45 |
-fno-default-inline -fno-defer-pop -floop-optimize2 |
| 46 |
-fmove-loop-invariants -fno-function-cse |
| 47 |
-fno-guess-branch-probability -fno-inline -fno-math-errno |
| 48 |
-fno-peephole -fno-peephole2 -funsafe-math-optimizations |
| 49 |
-funsafe-loop-optimizations -ffinite-math-only -fno-trapping-math |
| 50 |
-fno-zero-initialized-in-bss -fomit-frame-pointer |
| 51 |
-foptimize-register-move -foptimize-sibling-calls |
| 52 |
-fprefetch-loop-arrays -fprofile-generate -fprofile-use -fregmove |
| 53 |
-frename-registers -freorder-blocks -freorder-blocks-and-partition |
| 54 |
-freorder-functions -frerun-cse-after-loop -frerun-loop-opt |
| 55 |
-frounding-math -fschedule-insns -fschedule-insns2 |
| 56 |
-fno-sched-interblock -fno-sched-spec -fsched-spec-load |
| 57 |
-fsched-spec-load-dangerous -fsched-stalled-insns=n |
| 58 |
-fsched-stalled-insns-dep=n -fsched2-use-superblocks |
| 59 |
-fsched2-use-traces -freschedule-modulo-scheduled-loops |
| 60 |
-fsignaling-nans -fsingle-precision-constant -fstack-protector |
| 61 |
-fstack-protector-all -fstrength-reduce -fstrict-aliasing -ftracer |
| 62 |
-fthread-jumps -funroll-all-loops -funroll-loops -fpeel-loops |
| 63 |
-fsplit-ivs-in-unroller -funswitch-loops |
| 64 |
-fvariable-expansion-in-unroller -ftree-pre -ftree-ccp -ftree-dce |
| 65 |
-ftree-loop-optimize -ftree-loop-linear -ftree-loop-im |
| 66 |
-ftree-loop-ivcanon -fivopts -ftree-dominator-opts -ftree-dse |
| 67 |
-ftree-copyrename -ftree-sink -ftree-ch -ftree-sra -ftree-ter |
| 68 |
-ftree-lrs -ftree-fre -ftree-vectorize -ftree-vect-loop-version |
| 69 |
-ftree-salias -fweb -ftree-copy-prop -ftree-store-ccp |
| 70 |
-ftree-store-copy-prop -fwhole-program --param name=value -O -O0 -O1 |
| 71 |
-O2 -O3 -Os |
| 72 |
|
| 73 |
|
| 74 |
-- |
| 75 |
Best regards, |
| 76 |
Daniel |
| 77 |
|
| 78 |
|
| 79 |
-- |
| 80 |
gentoo-user@g.o mailing list |
Archives