public inbox for gcc-cvs@sourceware.org help / color / mirror / Atom feed
From: Richard Biener <rguenth@gcc.gnu.org> To: gcc-cvs@gcc.gnu.org Subject: [gcc r11-10481] tree-optimization/107254 - check and support live lanes from permutes Date: Tue, 24 Jan 2023 15:22:45 +0000 (GMT) [thread overview] Message-ID: <20230124152245.657423858D28@sourceware.org> (raw) https://gcc.gnu.org/g:fb335c938bbd54e83af2d281f0ccf79df1c342b3 commit r11-10481-gfb335c938bbd54e83af2d281f0ccf79df1c342b3 Author: Richard Biener <rguenther@suse.de> Date: Fri Oct 14 11:14:59 2022 +0200 tree-optimization/107254 - check and support live lanes from permutes The following fixes an omission from adding SLP permute nodes which is live lanes originating from those. We have to check that we can extract the lane and have to actually code generate them. PR tree-optimization/107254 * tree-vect-slp.c (vect_slp_analyze_node_operations_1): For permutes also analyze live lanes. (vect_schedule_slp_node): For permutes also code generate live lane extracts. * gfortran.dg/vect/pr107254.f90: New testcase. (cherry picked from commit 9ed4a849afb5b18b462bea311e7eee454c2c9f68) Diff: --- gcc/testsuite/gfortran.dg/vect/pr107254.f90 | 49 +++++++++++++++++++++++++++++ gcc/tree-vect-slp.c | 33 ++++++++++++++++--- 2 files changed, 77 insertions(+), 5 deletions(-) diff --git a/gcc/testsuite/gfortran.dg/vect/pr107254.f90 b/gcc/testsuite/gfortran.dg/vect/pr107254.f90 new file mode 100644 index 00000000000..85bcb5f3fa2 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/vect/pr107254.f90 @@ -0,0 +1,49 @@ +! { dg-do run } + +subroutine dlartg( f, g, s, r ) + implicit none + double precision :: f, g, r, s + double precision :: d, p + + d = sqrt( f*f + g*g ) + p = 1.d0 / d + if( abs( f ) > 1 ) then + s = g*sign( p, f ) + r = sign( d, f ) + else + s = g*sign( p, f ) + r = sign( d, f ) + end if +end subroutine + +subroutine dhgeqz( n, h, t ) + implicit none + integer n + double precision h( n, * ), t( n, * ) + integer jc + double precision c, s, temp, temp2, tempr + temp2 = 10d0 + call dlartg( 10d0, temp2, s, tempr ) + c = 0.9d0 + s = 1.d0 + do jc = 1, n + temp = c*h( 1, jc ) + s*h( 2, jc ) + h( 2, jc ) = -s*h( 1, jc ) + c*h( 2, jc ) + h( 1, jc ) = temp + temp2 = c*t( 1, jc ) + s*t( 2, jc ) + t( 2, jc ) = -s*t( 1, jc ) + c*t( 2, jc ) + t( 1, jc ) = temp2 + enddo +end subroutine dhgeqz + +program test + implicit none + double precision h(2,2), t(2,2) + h = 0 + t(1,1) = 1 + t(2,1) = 0 + t(1,2) = 0 + t(2,2) = 0 + call dhgeqz( 2, h, t ) + if (t(2,2).ne.0) STOP 1 +end program test diff --git a/gcc/tree-vect-slp.c b/gcc/tree-vect-slp.c index a034abf9628..7c75799f625 100644 --- a/gcc/tree-vect-slp.c +++ b/gcc/tree-vect-slp.c @@ -3824,7 +3824,23 @@ vect_slp_analyze_node_operations_1 (vec_info *vinfo, slp_tree node, /* Handle purely internal nodes. */ if (SLP_TREE_CODE (node) == VEC_PERM_EXPR) - return vectorizable_slp_permutation (vinfo, NULL, node, cost_vec); + { + if (!vectorizable_slp_permutation (vinfo, NULL, node, cost_vec)) + return false; + + stmt_vec_info slp_stmt_info; + unsigned int i; + FOR_EACH_VEC_ELT (SLP_TREE_SCALAR_STMTS (node), i, slp_stmt_info) + { + if (STMT_VINFO_LIVE_P (slp_stmt_info) + && !vectorizable_live_operation (vinfo, + slp_stmt_info, NULL, node, + node_instance, i, + false, cost_vec)) + return false; + } + return true; + } gcc_assert (STMT_SLP_TYPE (stmt_info) != loop_vect); if (is_a <bb_vec_info> (vinfo) @@ -6350,8 +6366,6 @@ vect_schedule_slp_node (vec_info *vinfo, } } - bool done_p = false; - /* Handle purely internal nodes. */ if (SLP_TREE_CODE (node) == VEC_PERM_EXPR) { @@ -6362,9 +6376,18 @@ vect_schedule_slp_node (vec_info *vinfo, but open-code it here (partly). */ bool done = vectorizable_slp_permutation (vinfo, &si, node, NULL); gcc_assert (done); - done_p = true; + stmt_vec_info slp_stmt_info; + unsigned int i; + FOR_EACH_VEC_ELT (SLP_TREE_SCALAR_STMTS (node), i, slp_stmt_info) + if (STMT_VINFO_LIVE_P (slp_stmt_info)) + { + done = vectorizable_live_operation (vinfo, + slp_stmt_info, &si, node, + instance, i, true, NULL); + gcc_assert (done); + } } - if (!done_p) + else vect_transform_stmt (vinfo, stmt_info, &si, node, instance); }
reply other threads:[~2023-01-24 15:22 UTC|newest] Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20230124152245.657423858D28@sourceware.org \ --to=rguenth@gcc.gnu.org \ --cc=gcc-cvs@gcc.gnu.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).