public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
* [Bug middle-end/59777] New: Incorrect expansion of TLS arguments in a call @ 2014-01-12 18:10 danglin at gcc dot gnu.org 2014-01-12 18:13 ` [Bug middle-end/59777] " danglin at gcc dot gnu.org ` (2 more replies) 0 siblings, 3 replies; 4+ messages in thread From: danglin at gcc dot gnu.org @ 2014-01-12 18:10 UTC (permalink / raw) To: gcc-bugs http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59777 Bug ID: 59777 Summary: Incorrect expansion of TLS arguments in a call Product: gcc Version: 4.8.2 Status: UNCONFIRMED Severity: normal Priority: P3 Component: middle-end Assignee: unassigned at gcc dot gnu.org Reporter: danglin at gcc dot gnu.org Host: hppa-unknown-linux-gnu Target: hppa-unknown-linux-gnu Build: hppa-unknown-linux-gnu Created attachment 31813 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31813&action=edit Testcase. The attached testcase shows the problem. Argument expressions containing a TLS symbol reference need to be precomputed as a call may be needed to legitimize the address and this may clobber the setup for earlier arguments causing wrong code. For example, compilation of the testcase with: gcc-4.8 -fPIC -DPIC -W -Wall -Wextra -Wshadow -Wformat -Wundef -D_GNU_SOURCE -O0 cap-ng.c -fPIC -DPIC -o cap-ng results in the following output: $ ./cap-ng cc m.hdr1 = 0x400015c8 pidaaaa = 0 There seems to be some attempt to handle this in precompute_register_parameters(). ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug middle-end/59777] Incorrect expansion of TLS arguments in a call 2014-01-12 18:10 [Bug middle-end/59777] New: Incorrect expansion of TLS arguments in a call danglin at gcc dot gnu.org @ 2014-01-12 18:13 ` danglin at gcc dot gnu.org 2014-02-05 0:47 ` [Bug target/59777] " danglin at gcc dot gnu.org 2014-02-05 1:02 ` danglin at gcc dot gnu.org 2 siblings, 0 replies; 4+ messages in thread From: danglin at gcc dot gnu.org @ 2014-01-12 18:13 UTC (permalink / raw) To: gcc-bugs http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59777 --- Comment #1 from John David Anglin <danglin at gcc dot gnu.org> --- Created attachment 31814 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31814&action=edit Output from expand. One can see in .expand that TLS arguments to printf are not being precomputed causing wrong code. ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug target/59777] Incorrect expansion of TLS arguments in a call 2014-01-12 18:10 [Bug middle-end/59777] New: Incorrect expansion of TLS arguments in a call danglin at gcc dot gnu.org 2014-01-12 18:13 ` [Bug middle-end/59777] " danglin at gcc dot gnu.org @ 2014-02-05 0:47 ` danglin at gcc dot gnu.org 2014-02-05 1:02 ` danglin at gcc dot gnu.org 2 siblings, 0 replies; 4+ messages in thread From: danglin at gcc dot gnu.org @ 2014-02-05 0:47 UTC (permalink / raw) To: gcc-bugs http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59777 --- Comment #3 from John David Anglin <danglin at gcc dot gnu.org> --- Author: danglin Date: Wed Feb 5 00:47:25 2014 New Revision: 207494 URL: http://gcc.gnu.org/viewcvs?rev=207494&root=gcc&view=rev Log: PR target/59777 * config/pa/pa.c (legitimize_tls_address): Return original address if not passed a SYMBOL_REF rtx. (hppa_legitimize_address): Call legitimize_tls_address for all TLS addresses. (pa_emit_move_sequence): Simplify TLS source operands. (pa_legitimate_constant_p): Reject all TLS constants. * config/pa/pa.h (PA_SYMBOL_REF_TLS_P): Correct comment. (CONSTANT_ADDRESS_P): Reject TLS CONST addresses. Modified: branches/gcc-4_8-branch/gcc/ChangeLog branches/gcc-4_8-branch/gcc/config/pa/pa.c branches/gcc-4_8-branch/gcc/config/pa/pa.h ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug target/59777] Incorrect expansion of TLS arguments in a call 2014-01-12 18:10 [Bug middle-end/59777] New: Incorrect expansion of TLS arguments in a call danglin at gcc dot gnu.org 2014-01-12 18:13 ` [Bug middle-end/59777] " danglin at gcc dot gnu.org 2014-02-05 0:47 ` [Bug target/59777] " danglin at gcc dot gnu.org @ 2014-02-05 1:02 ` danglin at gcc dot gnu.org 2 siblings, 0 replies; 4+ messages in thread From: danglin at gcc dot gnu.org @ 2014-02-05 1:02 UTC (permalink / raw) To: gcc-bugs http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59777 John David Anglin <danglin at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution|--- |FIXED --- Comment #4 from John David Anglin <danglin at gcc dot gnu.org> --- Fixed on trunk and 4.8 branch. ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2014-02-05 1:02 UTC | newest] Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2014-01-12 18:10 [Bug middle-end/59777] New: Incorrect expansion of TLS arguments in a call danglin at gcc dot gnu.org 2014-01-12 18:13 ` [Bug middle-end/59777] " danglin at gcc dot gnu.org 2014-02-05 0:47 ` [Bug target/59777] " danglin at gcc dot gnu.org 2014-02-05 1:02 ` danglin at gcc dot gnu.org
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).