From: bugzilla-daemon@freedesktop.org
To: dri-devel@lists.freedesktop.org
Subject: [Bug 84232] PHINode containing itself causes segfault in LLVM when compiling Blender OpenCL kernel with R600 backend
Date: Fri, 10 Oct 2014 15:48:20 +0000 [thread overview]
Message-ID: <bug-84232-502-hStm2OPjRP@http.bugs.freedesktop.org/> (raw)
In-Reply-To: <bug-84232-502@http.bugs.freedesktop.org/>
[-- Attachment #1.1: Type: text/plain, Size: 3259 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=84232
--- Comment #7 from Vitaliy Filippov <vitalif@yourcmc.ru> ---
Another thought: it seems PHINode that contain itself are also a product of
StructurizeCFG! After running `opt-3.5 -S -structurizecfg kernel.ll >
kernel-scfg.ll` on the kernel from attached blender output, kernel-scfg.ll
contains phinode that reference themselves directly... While kernel.ll did not
contain such PHINodes.
grep -P '(%\S+) = phi.*\1' kernel-scfg.ll
%18919 = phi float [ %18919, %Flow196 ], [ %19044, %Flow162 ], [ %18839,
%18838 ]
%18920 = phi float [ %18920, %Flow196 ], [ %18915, %Flow162 ], [ %18840,
%18838 ]
%18921 = phi <4 x float> addrspace(1)* [ %18921, %Flow196 ], [ %18896,
%Flow162 ], [ %18841, %18838 ]
%18922 = phi i32 [ %18922, %Flow196 ], [ %18895, %Flow162 ], [ %18842, %18838
]
%18923 = phi i32 [ %18923, %Flow196 ], [ %18894, %Flow162 ], [ %18843, %18838
]
%19208 = phi i64 [ %19208, %Flow195 ], [ %indvars.iv.next.i.i, %Flow164 ], [
%18953, %Flow161 ]
%19209 = phi float [ %19209, %Flow195 ], [ %19056, %Flow164 ], [ %18954,
%Flow161 ]
%19210 = phi float [ %19210, %Flow195 ], [ %19055, %Flow164 ], [ %18955,
%Flow161 ]
%19211 = phi float [ %19211, %Flow195 ], [ %19054, %Flow164 ], [ %18956,
%Flow161 ]
%19212 = phi i32 [ %19212, %Flow195 ], [ %19053, %Flow164 ], [ %18957,
%Flow161 ]
%19213 = phi i32 [ %19213, %Flow195 ], [ %19052, %Flow164 ], [ %18958,
%Flow161 ]
%19227 = phi i1 [ %19227, %Flow195 ], [ %19059, %Flow164 ], [ true, %Flow161
]
%19590 = phi <4 x float> [ %19590, %Flow188 ], [ %20245, %Flow171 ], [
%19529, %Flow169 ]
%19591 = phi <4 x float> [ %19591, %Flow188 ], [ %20246, %Flow171 ], [
%19530, %Flow169 ]
%19592 = phi float [ %19592, %Flow188 ], [ %20247, %Flow171 ], [ %19531,
%Flow169 ]
%19593 = phi <4 x float> [ %19593, %Flow188 ], [ %20248, %Flow171 ], [
%19532, %Flow169 ]
%19594 = phi <4 x float> [ %19594, %Flow188 ], [ %20249, %Flow171 ], [
%19533, %Flow169 ]
%19595 = phi <4 x float> [ %19595, %Flow188 ], [ %20250, %Flow171 ], [
%19534, %Flow169 ]
%19596 = phi <4 x float> [ %19596, %Flow188 ], [ %20251, %Flow171 ], [
%19535, %Flow169 ]
%19597 = phi i32 [ %19597, %Flow188 ], [ %20252, %Flow171 ], [ %19536,
%Flow169 ]
%19598 = phi i32 [ %19598, %Flow188 ], [ %20253, %Flow171 ], [ %19537,
%Flow169 ]
%19599 = phi i32 [ %19599, %Flow188 ], [ %20254, %Flow171 ], [ %19538,
%Flow169 ]
%19600 = phi <4 x float> [ %19600, %Flow188 ], [ %20255, %Flow171 ], [
%19539, %Flow169 ]
%19601 = phi <4 x float> [ %19601, %Flow188 ], [ %20256, %Flow171 ], [
%19540, %Flow169 ]
%19609 = phi float [ %19609, %Flow188 ], [ undef, %Flow171 ], [ %19548,
%Flow169 ]
%19610 = phi float [ %19610, %Flow188 ], [ undef, %Flow171 ], [ %19549,
%Flow169 ]
%19611 = phi float [ %19611, %Flow188 ], [ undef, %Flow171 ], [ %19550,
%Flow169 ]
%19612 = phi i32 [ %19612, %Flow188 ], [ undef, %Flow171 ], [ %19551,
%Flow169 ]
%19613 = phi i32 [ %19613, %Flow188 ], [ undef, %Flow171 ], [ %19552,
%Flow169 ]
%19625 = phi i1 [ %19625, %Flow188 ], [ %20257, %Flow171 ], [ %19569,
%Flow169 ]
%19627 = phi i1 [ %19627, %Flow188 ], [ %20258, %Flow171 ], [ %19571,
%Flow169 ]
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #1.2: Type: text/html, Size: 4194 bytes --]
[-- Attachment #2: Type: text/plain, Size: 159 bytes --]
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2014-10-10 15:48 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-23 9:19 [Bug 84232] New: PHINode containing itself causes segfault in LLVM when compiling Blender OpenCL kernel with R600 backend bugzilla-daemon
2014-09-23 9:19 ` [Bug 84232] " bugzilla-daemon
2014-09-23 13:19 ` bugzilla-daemon
2014-10-07 21:23 ` bugzilla-daemon
2014-10-07 21:45 ` bugzilla-daemon
2014-10-07 22:23 ` bugzilla-daemon
2014-10-08 20:56 ` bugzilla-daemon
2014-10-09 8:22 ` bugzilla-daemon
2014-10-10 15:48 ` bugzilla-daemon [this message]
2014-10-10 20:31 ` bugzilla-daemon
2014-10-10 21:36 ` bugzilla-daemon
2014-10-22 23:07 ` bugzilla-daemon
2014-10-22 23:08 ` bugzilla-daemon
2014-10-22 23:16 ` bugzilla-daemon
2014-10-23 9:12 ` bugzilla-daemon
2014-12-05 15:44 ` bugzilla-daemon
2014-12-08 16:57 ` bugzilla-daemon
2014-12-08 21:11 ` bugzilla-daemon
2017-03-22 16:00 ` bugzilla-daemon
2017-03-22 16:00 ` bugzilla-daemon
2018-04-03 3:49 ` bugzilla-daemon
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=bug-84232-502-hStm2OPjRP@http.bugs.freedesktop.org/ \
--to=bugzilla-daemon@freedesktop.org \
--cc=dri-devel@lists.freedesktop.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: link
Be 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 NNTP newsgroup(s).