> Am 02.02.2025 um 08:00 schrieb H.J. Lu :
>
> Don't increase callee-saved register cost by 1000x, which leads to that
> callee-saved registers aren't used to preserve local variable values
> across calls, by capping the scale to 300.
>PR rtl-optimization/111673
>PR rtl-optimization/1
Don't increase callee-saved register cost by 1000x, which leads to that
callee-saved registers aren't used to preserve local variable values
across calls, by capping the scale to 300.
PR rtl-optimization/111673
PR rtl-optimization/115932
PR rtl-optimization/116028
P
On Sat, Feb 01, 2025 at 09:49:17PM +0100, Harald Anlauf wrote:
> Am 01.02.25 um 21:03 schrieb Steve Kargl:
> > On Sat, Feb 01, 2025 at 07:25:51PM +0100, Harald Anlauf wrote:
> > >
> > > the attached patch downgrades different constant character lengths in an
> > > array constructor from a GNU to a
Add a test for the target independent bug, PR rtl-optimization/111673.
PR rtl-optimization/111673
* gcc.target/i386/pr111673.c: New file.
--
H.J.
From 149da4e8927509a4e72eb01ee8277b6952757c7c Mon Sep 17 00:00:00 2001
From: "H.J. Lu"
Date: Sun, 2 Feb 2025 06:46:29 +0800
Subject: [PATCH] x86: Add
Am 01.02.25 um 21:49 schrieb Harald Anlauf:
So why must we be so tough?
Here's what I get from other compilers:
Intel, Nvidia, AMD flang: silent by default.
ifx -stand f18:
char_array_constructor_5.f90(11): warning #8208: If type specification
is omitted, each ac-value expression in the arr
Update
commit dd6247cb8fc11a15e23e949092f89d24ff329209
Author: H.J. Lu
Date: Fri Jan 31 12:29:04 2025 +0800
x86: Handle TARGET_INDIRECT_BRANCH_REGISTER for -fno-plt
to change "if (TARGET_X32 ...)" back to "else if (TARGET_X32 ...)".
PR target/118713
* config/i386/i386-expand.cc (ix86_exp
Am 01.02.25 um 21:03 schrieb Steve Kargl:
On Sat, Feb 01, 2025 at 07:25:51PM +0100, Harald Anlauf wrote:
the attached patch downgrades different constant character lengths in an
array constructor from a GNU to a legacy extension, so that users get a
warning with -std=gnu. We continue to genera
On Sat, Feb 01, 2025 at 07:25:51PM +0100, Harald Anlauf wrote:
>
> the attached patch downgrades different constant character lengths in an
> array constructor from a GNU to a legacy extension, so that users get a
> warning with -std=gnu. We continue to generate an error when standard
> conforman
On 2/1/25 10:25 AM, Harald Anlauf wrote:
Dear all,
the attached patch downgrades different constant character lengths in an
array constructor from a GNU to a legacy extension, so that users get a
warning with -std=gnu. We continue to generate an error when standard
conformance is requested.
Re
Dear all,
the attached patch downgrades different constant character lengths in an
array constructor from a GNU to a legacy extension, so that users get a
warning with -std=gnu. We continue to generate an error when standard
conformance is requested.
Regtested on x86_64-pc-linux-gnu (found one
On Sat, Feb 1, 2025 at 6:33 PM H.J. Lu wrote:
>
> On Sat, Feb 1, 2025 at 5:52 PM Uros Bizjak wrote:
> >
> > On Sat, Feb 1, 2025 at 9:51 AM H.J. Lu wrote:
> > >
> > > If TARGET_INDIRECT_BRANCH_REGISTER is true, indirect call and jump should
> > > use register, not memory. Update Bs, Bw and Bz co
Some SARIF files offload most of the properties within logical locations
in the results to an array of "cached" instances in
theRun.logicalLocations, so the information can be consolidated (and to
support the "parentIndex" property, which is PR 116176).
Support such files in sarif-replay.
Success
[ Returning to an old discussion... ]
On 8/19/24 3:50 AM, Richard Sandiford wrote:
Jeff Law writes:
On 8/12/24 3:50 PM, Jeff Law wrote:
On 8/12/24 1:49 PM, Richard Sandiford wrote:
- regno = subreg_regno (x);
+ /* A paradoxical should always be REGNO (y) + 0. Using
subreg_re
Regtested on x86_64-pc-linux-gnu (so far just "dg.exp=contract*
modules.exp=contract*"), OK for trunk if full bootstrap+regtest passes?
-- >8 --
Modules makes some assumptions about types that currently aren't
fulfilled by the types created in contracts logic. This patch ensures
that exporting i
This is a regression present on the mainline, 14 and 13 branches: the compiler
segfaults when warning about an uninitialized variable as operand of primitive
operator of derived real type, because of a small internal adjustment.
Tested on x86-64/Linux, applied on the mainline, 14 and 13 branches
On Sat, Feb 1, 2025 at 5:52 PM Uros Bizjak wrote:
>
> On Sat, Feb 1, 2025 at 9:51 AM H.J. Lu wrote:
> >
> > If TARGET_INDIRECT_BRANCH_REGISTER is true, indirect call and jump should
> > use register, not memory. Update Bs, Bw and Bz constraints to disable
> > indirect call over memmory if TARGET
On Sat, Feb 1, 2025 at 11:14 AM H.J. Lu wrote:
>
> Verify that -mstack-protector-guard=global works on x86. Default stack
> protector uses TLS. -mstack-protector-guard=global uses a global variable,
> __stack_chk_guard, instead of TLS.
>
> * gcc.target/i386/ssp-global.c: New file.
OK.
Thanks,
Bootstrapped and regtested on x86_64-pc-linux-gnu, OK for trunk?
-- >8 --
'start_cleanup_fn' is not currently viable in modules, due to generating
functions relying on the 'start_cleanup_cnt' counter which is reset to 0
with each new TU. This means that cleanup functions declared in a TU
will co
Verify that -mstack-protector-guard=global works on x86. Default stack
protector uses TLS. -mstack-protector-guard=global uses a global variable,
__stack_chk_guard, instead of TLS.
* gcc.target/i386/ssp-global.c: New file.
--
H.J.
From 520d249938561639ec5c99f36947fd6795a9f2ed Mon Sep 17 00:00
On Sat, Feb 1, 2025 at 9:51 AM H.J. Lu wrote:
>
> If TARGET_INDIRECT_BRANCH_REGISTER is true, indirect call and jump should
> use register, not memory. Update Bs, Bw and Bz constraints to disable
> indirect call over memmory if TARGET_INDIRECT_BRANCH_REGISTER true, change
> x32 call over GOT slot
See PR rtl-optimization/117327 for a similar issue.
> gcc/
> * reorg.cc (label_with_barrier_p): New function.
> (skip_consecutive_labels): Use it. Don't skip the label if an
> empty block is found.
> (first_active_target_insn): Likewise. Don't ignore the empty
> blo
If TARGET_INDIRECT_BRANCH_REGISTER is true, indirect call and jump should
use register, not memory. Update Bs, Bw and Bz constraints to disable
indirect call over memmory if TARGET_INDIRECT_BRANCH_REGISTER true, change
x32 call over GOT slot to call over register and also disable sibcall
over memo
22 matches
Mail list logo
