Re: [PATCH] Fix dump message issue

On 2019/10/14 00:32, Jeff Law wrote: > On 10/8/19 4:45 AM, Martin Jambor wrote: >> Hi, >> >> On Tue, Oct 08 2019, luoxhu wrote: >>> '}' is missed at the end. >> >> heh, yeah, I wonder for how long. >> >> If it irritates you, I'd say the patch is obvious (though note that I >> cannot approve a

[Darwin, machopic 8/n, committed] Back out part of PR71767 fix.

We applied a conservative, but fairly large, hammer to fix PR71767. However, ideally, we want minimise the number of symbols visible to ld64 and to match the cases emitted by clang (since that's what ld64 is expecting). Now we've improved the handling of indirections, we can make the indirection

[Darwin, machopic 7/n, committed] Remove code that should be dead.

This code fragment was imported from the Apple branch (it was never applied to mainline). It is stated to fix up a problem sometimes created by reload (before that had been extended to have greater flexibility in assigning the pic registers). In any event, reload is no longer in use for the

Make ipa-reference bitmaps dense

Hi, this patch makes ipa-reference to assign sequential IDS to the static variables to make bitmaps more dense. DECL_UID is not very good choice since at WPA time just very small percentage of DECLs are static vars. The memory use for ipa-reference bitmaps after patch is 83MB compared to 197MB

Re: [PATCH] Fix dump message issue

On 10/8/19 4:45 AM, Martin Jambor wrote: > Hi, > > On Tue, Oct 08 2019, luoxhu wrote: >> '}' is missed at the end. > > heh, yeah, I wonder for how long. > > If it irritates you, I'd say the patch is obvious (though note that I > cannot approve a patch in this area). Looks obvious to me. And

Re: [patch] canonicalize unsigned [1,MAX] ranges into ~[0,0]

On 10/7/19 6:28 AM, Aldy Hernandez wrote: > > Fair enough.  I guess I don't care what we settle on, inasmuch as we > canonicalize into one true value.  For some reason, I thought the above > nonzero would cause you to cringe, I guess not :). :-) Takes more than that these days.. And just to

Re: [PATCH][MSP430] Add support for post increment addressing

On 10/7/19 2:28 PM, Jozef Lawrynowicz wrote: > MSP430 supports post increment addressing for the source operand only. The > attached patch enables post increment addressing for MSP430 in GCC. > > Regtested on trunk for the small and large memory models. > > Ok for trunk? > > >

Re: Add a constant_range_value_p function (PR 92033)

On 10/11/19 10:42 AM, Richard Sandiford wrote: The range-tracking code has a pretty hard-coded assumption that is_gimple_min_invariant is equivalent to "INTEGER_CST or invariant ADDR_EXPR". It seems better to add a predicate specifically for that rather than contiually fight cases in which it

Re: [patch, fortran] Fix PR 92004, restore Lapack compilation

OK, so here's the update. There was a problem with uninitialized variables, which for some reason was not detected on compilation. OK for trunk? 2019-10-13 Thomas Koenig PR fortran/92004 * array.c (expand_constructor): Set from_constructor on expression. *

Re: Add expr_callee_abi

On 10/11/19 8:39 AM, Richard Sandiford wrote: > This turned out to be useful for the SVE PCS support, and is a natural > tree-level analogue of insn_callee_abi. > > Tested on aarch64-linux-gnu and x86_64-linux-gnu. OK to install? > > Richard > > > 2019-10-11 Richard Sandiford > > gcc/ >

Re: [PATCH][GCC][ARM] Arm generates out of range conditional branches in Thumb2 (PR91816)

> > Patch bootstrapped and regression tested on arm-none-linux-gnueabihf, > however, on my native Aarch32 setup the test times out when run as part > of a big "make check-gcc" regression, but not when run individually. > > 2019-10-11 Stamatis Markianos-Wright > > * config/arm/arm.md:

Re: [PATCH] Fix -Wshadow=local warnings in elfos.h

On 10/5/19 12:16 AM, Bernd Edlinger wrote: > > > On 10/4/19 10:26 PM, Jeff Law wrote: >> On 10/4/19 12:24 PM, Bernd Edlinger wrote: >>> Hi, >>> >>> these macros don't use reserved names for local variables. >>> This caused -Wshadow=local warnings in varasm.c IIRC. >>> >>> Fixed by using

[PATCH] teach gengtype about 'mutable'

In constifying some more of line-map I discovered gengtype didn't know mutable. Added thusly. nathan -- Nathan Sidwell 2019-10-13 Nathan Sidwell * gengtype-lex.l (CXX_KEYWORD): Add 'mutable'. Index: gcc/gengtype-lex.l ===

[patch, fortran, committed] Fix PR 92017

Hello world, I have committed the attached patch as obvious and simple after regression-testing. It fixes an ICE on valid for a corner case, so I don't really feel that it needs to be backported. If anybody disagrees, please speak up (or do it yourself :-) Regards Thomas 2019-10-13

Re: [patch, fortran] Fix PR 92004, restore Lapack compilation

Hm, my trunk is doing strange things (debugging not working), and I think I have found an additional problem. I'll need some time to work this out, and will resubmit. Regards Thomas

Re: Make C2X imply -fno-fp-int-builtin-inexact

Rainer Orth writes: I’m not quite sure what you’re proposing here (probably missing something obvious). >>> >>> At the moment, gcc/testsuite/lib/target-supports.exp >>> (add_options_for_c99_runtime) adds -mmacosx-version-min=10.3 to the >>> testcase flags on powerpc-*-darwin*. Since,

Grow GGC after cgraph and summary streaming

Hi, we use ggc_grow to prevent garbage collector from triggering at WPA time. After streaming in global trees we know that basically all of them will stay reachable until end of WPA compilation. This no longer works because tree memory use got down to about 123MB out of 700MB of GGC memory needed

Fix duplicated renumbering of statements during streaming

Hi, currently we renumber statements during streaming twice - once using renumber_gimple_stmt_uids and second inside output_function. The numbering only differs by fact that first one does mix normal and virtual PHI sets while the second doesn't. This patch reduces renumbering to only one per

[PATCH 4/5] PRU: Remove TARGET_HARD_REGNO_CALL_PART_CLOBBERED

Per clarification in [1], macro is supposed to check for partial clobbering of single HW registers. Since PRU declares only 8-bit HW registers, and ABI does not define individual bit clobbering, it is safe to remove the implementation. [1] https://gcc.gnu.org/ml/gcc-patches/2019-09/msg00778.html

[PATCH 3/5] PRU: Fix comment about R3/RA

Comment had a typo. Fix it and clarify. gcc/ChangeLog: 2019-10-13 Dimitar Dimitrov * config/pru/pru.h: Clarify R3/RA ABI. Signed-off-by: Dimitar Dimitrov --- gcc/config/pru/pru.h | 8 +++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/gcc/config/pru/pru.h

[PATCH 5/5] Add pru to compare-all-tests

contrib/ChangeLog: 2019-10-13 Dimitar Dimitrov * compare-all-tests (all_targets): Add pru target. Signed-off-by: Dimitar Dimitrov --- contrib/compare-all-tests | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/compare-all-tests b/contrib/compare-all-tests

[PATCH 1/5] PRU: Fix comment to avoid fall through warning

gcc/ChangeLog: 2019-10-13 Dimitar Dimitrov * config/pru/pru.c (pru_print_operand): Fix comment. Signed-off-by: Dimitar Dimitrov --- gcc/config/pru/pru.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/gcc/config/pru/pru.c b/gcc/config/pru/pru.c index

[PATCH 2/5] PRU: Simplify machine description

Use the new @insn syntax for simpler gen_* invocation. gcc/ChangeLog: 2019-10-13 Dimitar Dimitrov * config/pru/pru.c (pru_emit_doloop): Use new gen_doloop_end_internal and gen_doloop_begin_internal. (pru_reorg_loop): Use gen_pruloop with mode. *

[PATCH 0/5] Assorted minor cleanups for PRU backend

Apart from the last change, these are all minor cleanups to the PRU backend. Dimitar Dimitrov (5): PRU: Fix comment to avoid fall through warning PRU: Simplify machine description PRU: Fix comment about R3/RA PRU: Remove TARGET_HARD_REGNO_CALL_PART_CLOBBERED Add pru to compare-all-tests