projects / mesa.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: dd5c580) | patch
gallivm: Altivec floating-point rounding
authorAdhemerval Zanella <azanella@linux.vnet.ibm.com>
Thu, 22 Nov 2012 17:37:18 +0000 (11:37 -0600)
committerJosé Fonseca <jfonseca@vmware.com>
Thu, 29 Nov 2012 11:52:00 +0000 (11:52 +0000)
commit43ce9efdbf8c8e1200fe54744820ca2523f36e6b
tree2e2f8f978a4584599de7e52196db49352b58122etree
parentdd5c58081672bd495e0ffef1c3cc1229620f0f88commit | diff
gallivm: Altivec floating-point rounding

This patch adds Altivec intrinsics for float vector types. It changes
the SSE specific definitions to a platform neutral and adds the calls
to Altivec intrinsic builder.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
src/gallium/auxiliary/gallivm/lp_bld_arit.c diff | blob | history