Linaro GCC 4.5-2011.08

Milestone information

Linaro GCC
Michael Hope
Release registered:
No. Drivers cannot target bugs and blueprints to this milestone.  

Download RDF metadata


Assigned to you:
No blueprints or bugs assigned to you.
1 Ramana Radhakrishnan
No blueprints are targeted to this milestone.
1 Fix Released

Download files for this release

After you've downloaded a file, you can verify its authenticity using its MD5 sum or signature. (How do I verify a download?)

File Description Downloads
download icon gcc-linaro-4.5-2011.08.tar.bz2 (md5, sig) Linaro GCC Source 22,329
last downloaded 2 weeks ago
Total downloads: 22,329

Release notes 

Linaro GCC 4.5 2011.08 is the thirteenth release in the 4.5
series. Based off the latest GCC 4.5.3+svn177552, the release is
focused on maintenance.

Interesting changes in 4.5 include:
 * Updates to 4.5.3+r177552
 * Now builds for PowerPC

 * LP: #736007 ICE immed_double_const at emit-rtl.c
 * LP: #809768 ICE when compiling bionic's libm
 * LP: #815435 ICE: insn does not satisfy its constraints

The source tarball is available from:

Downloads are available from the Linaro GCC page on Launchpad:

Mailing list:



Interested in commercial support? inquire at


View the full changelog

2011-08-16 Matthias Klose <email address hidden>

 Backport from FSF:

 2010-12-20 Joseph Myers <email address hidden>

 * config/rs6000/freebsd.h (SVR4_ASM_SPEC): Don't define.
 * config/rs6000/lynx.h (DBX_REGISTER_NUMBER): Define.
 * config/rs6000/netbsd.h (DBX_REGISTER_NUMBER): Define.
 * config/rs6000/sysv4.h (SIZE_TYPE): Define.
 (ASM_SPEC): Define without using SVR4_ASM_SPEC.
 * config.gcc (powerpc-*-eabispe*, powerpc-*-eabisimaltivec*,
 powerpc-*-eabisim*, powerpc-*-elf*, powerpc-*-eabialtivec*,
 powerpc-xilinx-eabi*, powerpc-*-eabi*, powerpc-*-rtems*,
 powerpc-*-linux* | powerpc64-*-linux*, powerpc64-*-gnu*,
 powerpc-*-gnu-gnualtivec*, powerpc-*-gnu*,
 powerpc-wrs-vxworks|powerpc-wrs-vxworksae, powerpcle-*-elf*,
 powerpcle-*-eabisim*, powerpcle-*-eabi*): Don't use svr4.h.

2011-08-15 Richard Sandiford <email address hidden>

 * config/rs6000/rs6000.c (paired_expand_vector_init): Don't create
 CONST_VECTORs with symbolic elements.
 (rs6000_expand_vector_init): Likewise.

2011-08-11 Andrew Stubbs <email address hidden>

 Merge from FSF 4.5 branch r177552 (pre 4.5.4).

2011-08-09 Revital Eres <email address hidden>

 Backport from trunk -r176972:

 * ddg.c (create_ddg_dep_from_intra_loop_link): Remove
 the creation of anti-dep edge from a branch.
 Create anti-dep edge from a branch.

2011-08-09 Revital Eres <email address hidden>

 Backport from trunk -r176970:

 * modulo-sched.c: Change comment.
 (reset_sched_times): Fix print message.
 (print_partial_schedule): Add print

2011-08-03 Revital Eres <email address hidden>

 * modulo-sched.c (calculate_stage_count,
 calculate_must_precede_follow, get_sched_window,
 try_scheduling_node_in_cycle, remove_node_from_ps): Add
 (update_node_sched_params, set_must_precede_follow, optimize_sc):
 New functions.
 (reset_sched_times): Call update_node_sched_params.
 (sms_schedule): Call optimize_sc.
 (get_sched_window): Change function arguments.
 (sms_schedule_by_order): Update call to get_sched_window.
 Call set_must_precede_follow.
 (calculate_stage_count): Add function argument.

2011-07-29 Ramana Radhakrishnan <email address hidden>

 Backport only bits marked as [*] from mainline:
        2010-12-19 Chung-Lin Tang <email address hidden>

 * config/arm/arm.c ([*]arm_legitimate_index_p): Add VFP load/store
 index range case. Change to SF/DFmode tests to avoid capturing HFmode.
 (thumb2_legitimate_index_p): Same.

2011-07-21 Richard Sandiford <email address hidden>

 Backport from mainline:

 2011-07-21 Richard Sandiford <email address hidden>

 * regcprop.c (maybe_mode_change): Check HARD_REGNO_MODE_OK.

2011-07-21 Richard Sandiford <email address hidden>

 PR middle-end/49736
 * expr.c (all_zeros_p): Undo bogus part of last change.

2011-07-21 Richard Sandiford <email address hidden>

 Backport from mainline:
 2011-07-13 Richard Sandiford <email address hidden>

 * typeck2.c (split_nonconstant_init_1): Pass the initializer directly,
 rather than a pointer to it. Return true if the whole of the value
 was initialized by the generated statements. Use
 complete_ctor_at_level_p instead of count_type_elements.

 2011-07-13 Richard Sandiford <email address hidden>

 * tree.h (categorize_ctor_elements): Remove comment. Fix long line.
 (count_type_elements): Delete.
 (complete_ctor_at_level_p): Declare.
 * expr.c (flexible_array_member_p): New function, split out from...
 (count_type_elements): Make static. Replace allow_flexarr
 parameter with for_ctor_p. When for_ctor_p is true, return the
 number of elements that should appear in the top-level constructor,
 otherwise return an estimate of the number of scalars.
 (categorize_ctor_elements): Replace p_must_clear with p_complete.
 (categorize_ctor_elements_1): Likewise. Use complete_ctor_at_level_p.
 (complete_ctor_at_level_p): New function, borrowing union logic
 from old categorize_ctor_elements_1.
 (mostly_zeros_p): Return true if the constructor is not complete.
 (all_zeros_p): Update call to categorize_ctor_elements.
 * gimplify.c (gimplify_init_constructor): Update call to
 categorize_ctor_elements. Don't call count_type_elements.
 Unconditionally prevent clearing for variable-sized types,
 otherwise rely on categorize_ctor_elements to detect
 incomplete initializers.

 2011-07-13 Chung-Lin Tang <email address hidden>

 * New test.

2011-07-19 Revital Eres <email address hidden>

 Backport from mainline -r175090
 * ddg.c (add_intra_loop_mem_dep): New function.
 (build_intra_loop_deps): Call it.

 * gcc.dg/sms-9.c: New file.

0 blueprints and 1 bug targeted

Bug report Importance Assignee Status
815435 #815435 ICE: insn does not satisfy its constraints 3 High Ramana Radhakrishnan  10 Fix Released
This milestone contains Public information
Everyone can see this information.