These tests rely on ADL for some functions, probably unintentionally.
The calls only work because the iterator wrappers derive from
std::iterator and so namespace std is an associated namespace.
libstdc++-v3/ChangeLog:
* testsuite/25_algorithms/inplace_merge/constrained.cc: Qualify
VARYING ranges are just normal ranges that span the entire domain. Such
ranges have had end-points for a few releases now, and the fact that the
legacy code was still treating all VR_VARYING the same was an oversight.
This patch fixes the oversight to match the multi-range behavior.
Tested on
On Sat, 22 May 2021, H.J. Lu via Gcc-patches wrote:
> 1. Update PUSH_ARGS to accept an argument. When the PUSH instruction
> usage is optional, pass the number of bytes to push to PUSH_ARGS so that
> the backend can decide if PUSH instructions should be generated.
> 2. Change x86 PUSH_ARGS to
Hi folks,
some more suggestions for corrections in the onlinedocs:
https://gcc.gnu.org/onlinedocs/gcc-11.1.0/gfortran/GERROR.html#GERROR
RESULT "Shall of type CHARACTER and of default"
--> "Shall BE of ... default KIND."
On Mon, May 24, 2021 at 12:37:30AM +0200, Bernhard Reutner-Fischer wrote:
> On 21 May 2021 22:56:09 CEST, Bill Schmidt via Gcc-patches
> wrote:
> >>> + char *buf = (char *) malloc (lastpos - pos + 2);
> >>> + memcpy (buf, [pos], lastpos - pos + 1);
> >>> + buf[lastpos - pos + 1] = '\0';
>
The attached patch introduces the get_no_warning(), set_no_warning(),
and copy_no_warning() APIs without making use of them in the rest of
GCC. They are in three files:
diagnostic-spec.{h,c}: Location-centric overloads.
warning-control.cc: Tree- and gimple*-centric overloads.
The
The attached patch replaces TREE_NO_WARNING, gimple_get_no_warning_p
and gimple_set_no_warning with the new APIs, get_no_warning,
set_no_warning, and copy_no_warning.
Add support for per-location warning groups.
gcc/ChangeLog:
* builtins.c (warn_string_no_nul): Replace uses of TREE_NO_WARNING,
The attached patch replaces the uses of TREE_NO_WARNING in libc11.
Add support for per-location warning groups.
libcc1/ChangeLog:
* libcp1plugin.cc (record_decl_address): Replace a direct use
of TREE_NO_WARNING with set_no_warning.
diff --git a/libcc1/libcp1plugin.cc b/libcc1/libcp1plugin.cc
In the testcase below, the initializer for C::b inside C's default
constructor is encoded as a TARGET_EXPR wrapping the CALL_EXPR f() in
C++17 mode. During massaging of this constexpr constructor,
build_target_expr_with_type called from bot_manip ends up trying to use
B's implicitly deleted copy
One last addendum to this. I discovered that that needs a "sort"
in front of "keys %logicals_addsub" because otherwise you may get
the operators in different orders sometimes which leads to fusion.md
having the patterns in different orders which isn't helpful for
sane debugging. Segher and I
On 5/21/21 4:35 PM, Patrick Palka wrote:
Here, during ahead of time access checking for the private member
EnumeratorRange::end_reached_ in the hidden friend f, we're triggering
the the assert in enforce_access that verifies we're not trying to add a
dependent access check to
I've committed a patch to the Go frontend that fixes this problem.
Thanks for the analysis.
Ian
On Thu, May 20, 2021 at 1:59 AM guojiufu wrote:
>
> On 2021-05-18 14:58, Richard Biener wrote:
> > On Mon, 17 May 2021, Ian Lance Taylor wrote:
> >
> >> On Mon, May 17, 2021 at 1:17 AM Richard Biener
The attached patch replaces the uses of TREE_NO_WARNING in the C
front end.
Add support for per-location warning groups.
gcc/c/ChangeLog:
* c-decl.c (pop_scope): Replace direct uses of TREE_NO_WARNING with
get_no_warning, set_no_warning, and copy_no_warning.
(diagnose_mismatched_decls):
The attached patch replaces the uses of TREE_NO_WARNING in the C
family common subset of the C and C++ front ends.
Add support for per-location warning groups.
gcc/c-family/ChangeLog:
* c-common.c (c_wrap_maybe_const): Remove TREE_NO_WARNING.
(c_common_truthvalue_conversion): Replace direct
On Tue, May 18, 2021 at 5:31 AM Iain Buclaw wrote:
>
> Excerpts from Lewis Hyatt via Gcc-patches's message of January 29, 2021 4:46
> pm:
> > Q1: What is the input charset?
> > A1:
> >
> > libcpp: Whatever was passed to -finput-charset (note, for Fortran,
> > -finput-charset is not
On 5/21/21 5:39 AM, Aldy Hernandez via Gcc-patches wrote:
This patch converts the remaining users of get_range_info and
get_ptr_nonnull to the range_query API.
No effort was made to move passes away from VR_ANTI_RANGE, or any other
use of deprecated methods. This was a straight up conversion
On 5/21/21 4:35 PM, Patrick Palka wrote:
Here, in C++17 mode, convert_nontype_argument_function is rejecting
binding a non-noexcept function reference template parameter to a
noexcept function (encoded as the template argument '*(int (&) (int)) ').
The first roadblock to making this work is
The attached patch replaces the uses of TREE_NO_WARNING in the C++
front end. For the set/get_no_warning calls I tried to find the most
appropriate option to disable and query. In some cases there are
helpful comments nearby that made this easy. In other cases it was
less straightforward, and
On 5/24/21 12:46 PM, Martin Sebor wrote:
Instantiating a hash_map on a number of integer types including,
for example, int or unsigned int (such as location_t), or
HOST_WIDE_INT, and using it with the garbage collector causes many
cryptic compilation errors due to incomplete support for such
The attached patch replaces the uses of TREE_NO_WARNING in
the Objective-C front end.
Add support for per-location warning groups.
gcc/objc/ChangeLog:
* objc-act.c (objc_maybe_build_modify_expr): Replace direct uses
of TREE_NO_WARNING with get_no_warning, and set_no_warning.
The attached patch replaces the uses of TREE_NO_WARNING in the Fortran
front end.
I don't know the Fortran front end and I made no effort to try to find
the right warning options in the calls to set/get_no_warning.
Add support for per-location warning groups.
gcc/fortran/ChangeLog:
*
The attached patch replaces the uses of TREE_NO_WARNING in the LTO
front end. The streaming support doesn't extend to multiple bits yet.
I plan to do that in a follow-up change.
Add support for per-location warning groups.
gcc/lto/ChangeLog:
* gimple-streamer-out.c (output_gimple_stmt): Same.
On 5/21/21 5:39 AM, Aldy Hernandez via Gcc-patches wrote:
This patch provides a generic API for accessing global ranges. It is
meant to replace get_range_info() and get_ptr_nonnull() with one
common interface. It uses the same API as the ranger (class
range_query), so there will now be one API
On Fri, 21 May 2021, Martin Liška wrote:
> CPUs:
> aarch64, alpha, alpha64, amdgcn, arc, arceb, arm, avr, bfin, bpf, cr16, cris,
> csky, epiphany, fido, fr30, frv, ft32, h8300, hppa, hppa2.0, hppa64, i486,
> i686, ia64, iq2000, lm32, m32c, m32r, m32rle, m68k, mcore, microblaze, mips,
> mips64,
This patch to the Go frontend marks global variables whose address is
taken. To implement this, change the backend to use flag bits for
variables. This fixes GCC PR 100537. Bootstrapped and ran Go
testsuite on x86_64-pc-linux-gnu. Committed to mainline.
Ian
PR go/100537
*
Having just one bit control whether an expression or statement should
be allowed to trigger warnings has been a source of bug reports about
false negatives for years. PR 74765 has a representative test case
that shows how by setting the bit to avoid -Wparentheses the C++ front
end also ends up
The attached patch replaces the uses of TREE_NO_WARNING in the rl78
back end.
Add support for per-location warning groups.
gcc/ChangeLog:
* config/rl78/rl78.c (rl78_handle_naked_attribute): Replace a direct
use of TREE_NO_WARNING with set_no_warning.
diff --git a/gcc/config/rl78/rl78.c
Instantiating a hash_map on a number of integer types including,
for example, int or unsigned int (such as location_t), or
HOST_WIDE_INT, and using it with the garbage collector causes many
cryptic compilation errors due to incomplete support for such types
(the PR shows a few examples). I ran
Please, get serious.
=> this topic is closed for me, so STOP CC''ing ME, it is not welcome.
On Sun, 23 May 2021 at 19:03, Mike Stump wrote:
> This isn't a patch to gcc, please stop posting non-technical content to
> this list. Please review what this list is for and the rules for this list
>
On 18/05/21 00:53 -0400, Patrick Palka via Libstdc++ wrote:
On Mon, 17 May 2021, Tim Song wrote:
On Mon, May 17, 2021 at 2:59 PM Patrick Palka wrote:
>
> + constexpr _CachedPosition&
> + operator=(_CachedPosition&& __other) noexcept
> + {
> + if
[PATCH 2/11] use xxx_no_warning APIs in Ada.
Add support for per-location warning groups.
gcc/ada/ChangeLog:
* gcc-interface/trans.c (Handled_Sequence_Of_Statements_to_gnu):
Replace TREE_NO_WARNING with set_no_warning.
(gnat_gimplify_expr): Same.
* gcc-interface/utils.c (gnat_pushdecl):
On 5/24/21 1:48 PM, Patrick Palka wrote:
In the testcase below, the initializer for C::b inside C's default
constructor is encoded as a TARGET_EXPR wrapping the CALL_EXPR f() in
C++17 mode. During massaging of this constexpr constructor,
build_target_expr_with_type called from bot_manip ends up
On 5/24/21 5:08 PM, David Malcolm wrote:
On Mon, 2021-05-24 at 16:02 -0600, Martin Sebor wrote:
Having just one bit control whether an expression or statement should
be allowed to trigger warnings has been a source of bug reports about
false negatives for years. PR 74765 has a representative
On Mon, 2021-05-24 at 16:16 -0600, Martin Sebor via Gcc-patches wrote:
> The attached patch replaces TREE_NO_WARNING, gimple_get_no_warning_p
> and gimple_set_no_warning with the new APIs, get_no_warning,
> set_no_warning, and copy_no_warning.
Might be worth splitting this out into
(a) the
On 5/24/21 5:08 PM, David Malcolm via Gcc-patches wrote:
On Mon, 2021-05-24 at 16:16 -0600, Martin Sebor via Gcc-patches wrote:
The attached patch replaces TREE_NO_WARNING, gimple_get_no_warning_p
and gimple_set_no_warning with the new APIs, get_no_warning,
set_no_warning, and copy_no_warning.
Hi Thomas,
I reproduced this issue manually and it turns out this is a special case.
Script takes input from https://gcc.gnu.org/pipermail/gcc-regression/ and
it matches the exact error message in the triaging process. This failure
reported on gcc regression
On Mon, 2021-05-24 at 16:02 -0600, Martin Sebor wrote:
> Having just one bit control whether an expression or statement should
> be allowed to trigger warnings has been a source of bug reports about
> false negatives for years. PR 74765 has a representative test case
> that shows how by setting
Hi:
This patch is about to do transformation like below.
Bootstrapped and regtested on x86_64-linux-gnu{-m32,}.
Ok for trunk?
from
notl%edi
vpbroadcastd%edi, %xmm0
vpand %xmm1, %xmm0, %xmm0
to
vpbroadcastd%edi, %xmm0
vpandn %xmm1,
Hi Andre,
on 2021/5/24 下午2:17, Andre Vieira (lists) via Gcc-patches wrote:
> Hi,
>
> When vectorizing with --param vect-partial-vector-usage=1 the vectorizer uses
> an unpredicated (all-true predicate for SVE) main loop and a predicated tail
> loop. The way this was implemented seems to mean
On 5/21/21 2:39 PM, Marco Elver wrote:
On Fri, May 21, 2021 at 10:50AM +0200, Martin Liška wrote:
On 5/20/21 12:55 PM, Marco Elver wrote:
I think this came up with other no_sanitize [1] based on what I had
written to you last year [2].
Hi,
When vectorizing with --param vect-partial-vector-usage=1 the vectorizer
uses an unpredicated (all-true predicate for SVE) main loop and a
predicated tail loop. The way this was implemented seems to mean it
re-uses the same vector-mode for both loops, which means the tail loop
isn't an
On 5/23/21 2:37 PM, Tobias Burnus wrote:
As there was some confusion regarding when the ChangeLog is generated,
I propose the attached wwwdocs patch. Comments?
I support the change. Moreover, I'm going to install the following
patch that adds a note about ChangeLog entries.
Martin
(Side
Hi Joseph,
I found some time to update this patch. The only real change
of the patch is the qualifier in the conditional expression for
pointer to arrays in C2X. All the rest are the warnings,
which were wrong in the last version.
I hope I got this correct this time in combination with
Hi,
On 20/5/2021 下午 9:08, Segher Boessenkool wrote:
Hi!
On Thu, May 20, 2021 at 04:29:07PM +0800, HAO CHEN GUI wrote:
On 19/5/2021 下午 9:20, Segher Boessenkool wrote:
On Wed, May 19, 2021 at 04:36:00PM +0800, HAO CHEN GUI wrote:
-/* Define this macro if it is advisable to hold scalars in
From: Xiong Hu Luo
vmrghb only accepts permute index {0, 16, 1, 17, 2, 18, 3, 19, 4, 20,
5, 21, 6, 22, 7, 23} no matter for BE or LE in ISA, similarly for vmrghlb.
Remove UNSPEC_VMRGH_DIRECT/UNSPEC_VMRGL_DIRECT pattern as vec_select
+ vec_concat as normal RTL.
Tested pass on P8LE, P9LE and
On 5/20/21 9:55 AM, Richard Biener wrote:
On Thu, May 20, 2021 at 12:29 AM Joern Wolfgang Rennecke
wrote:
We set default for some target options in TARGET_OPTION_OPTIMIZATION_TABLE,
but these can be overridden by specifying the corresponding explicit
-mXXX / -mno-XXX options.
When a function
ping?
On Mon, 17 May 2021 at 11:55, Christophe Lyon
wrote:
>
> ping?
>
> On Fri, 30 Apr 2021 at 16:09, Christophe Lyon
> wrote:
> >
> > This patch enables MVE vld2/vst2 instructions for auto-vectorization.
> > We move the existing expanders from neon.md and enable them for MVE,
> > calling the
ping?
On Mon, 17 May 2021 at 11:55, Christophe Lyon
wrote:
>
> ping?
>
> On Tue, 4 May 2021 at 16:57, Christophe Lyon
> wrote:
> >
> > On Tue, 4 May 2021 at 14:03, Andre Vieira (lists)
> > wrote:
> > >
> > > Hi Christophe,
> > >
> > > The series LGTM but you'll need the approval of an arm
On Sun, 23 May 2021, Tobias Burnus wrote:
> As there was some confusion regarding when the ChangeLog is generated,
> I propose the attached wwwdocs patch. Comments?
-Apply the patch to your local tree. ChangeLog entries will be
-automatically added to the corresponding ChangeLog files based
-on
On Mon, 24 May 2021 at 10:03, Martin Liška wrote:
> On 5/21/21 2:39 PM, Marco Elver wrote:
> > On Fri, May 21, 2021 at 10:50AM +0200, Martin Liška wrote:
> >> On 5/20/21 12:55 PM, Marco Elver wrote:
> >>> I think this came up with other no_sanitize [1] based on what I had
> >>> written to you
On 5/21/21 2:35 PM, David Edelsohn wrote:
Please remember that not all targets support LTO so a fallback to a
non-partial-LTO build needs to be provided and automatically invoked
for those targets.
Sure, for now it's definitely going to be a opt-in, enabled by something like:
make
On 5/21/21 12:25 PM, 王留帅 via Gcc-patches wrote:
|Hi, We, ByteDance, have seen an ICE on trunk which can be reproduced with the
following symbol testcase. The problem is caused by referencing a null pointer
when applying 'n_basic_blocks_for_fn' on a thunk symbol. For the thunk symbol,
Hi:
This patch is about to add expanders for vashl,
vlshr,
vashr and vashr.
Besides there's some assumption in expand_mult_const that mul and
add must be available at the same time, but for i386, addv8qi is
restricted under TARGET_64BIT, but mulv8qi not, that could cause ICE.
So restrict
On 24/05/21 11:31 +0200, François Dumont wrote:
On 20/05/21 6:44 pm, Jonathan Wakely wrote:
On 06/05/21 22:03 +0200, François Dumont via Libstdc++ wrote:
Hi
Considering your feedback on backtrace in debug mode is going
to take me some time so here is another one.
Compared to latest
On Wed, Mar 10, 2021 at 11:55:43AM +0100, Tobias Burnus wrote:
> --- a/gcc/fortran/trans-openmp.c
> +++ b/gcc/fortran/trans-openmp.c
> @@ -360,6 +360,39 @@ gfc_has_alloc_comps (tree type, tree decl)
>return false;
> }
>
> +/* Return true if TYPE is polymorphic but not with pointer
On 20/05/21 6:44 pm, Jonathan Wakely wrote:
On 06/05/21 22:03 +0200, François Dumont via Libstdc++ wrote:
Hi
Considering your feedback on backtrace in debug mode is going to
take me some time so here is another one.
Compared to latest submission I've added a _Hash_arg_t partial
Hi Geng Qi,
I tested this patch but it got the following error when compiling libgcc,
during RTL pass:
reload/lhome/quxm/build-csky-upstream/source//gcc/libgcc/libgcc2.c: In
function
'__mulsc3':/lhome/quxm/build-csky-upstream/source//gcc/libgcc/libgcc2.c:2010:1:
internal compiler error:
> -Original Message-
> From: Gcc-patches On Behalf Of
> Christophe Lyon via Gcc-patches
> Sent: 30 April 2021 15:10
> To: gcc-patches@gcc.gnu.org
> Subject: [PATCH 9/9] arm: Auto-vectorization for MVE: vld4/vst4
>
> This patch enables MVE vld4/vst4 instructions for auto-vectorization.
gcc/ChangeLog:
* config/csky/csky.c (csky_option_override):
Init csky_arch_isa_features[] advanced, so TARGET_DSP
and TARGET_DIV can be set well.
---
gcc/config/csky/csky.c | 24
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git
gcc/ChangeLog:
* config/csky/csky.h
(FRAME_POINTER_REGNUM): Use HARD_FRAME_POINTER_REGNUM and
FRAME_POINTER_REGNUM instead of the signle definition. The
signle definition may not work well at simplify_subreg_regno().
(ELIMINABLE_REGS): Add for
gcc/ChangeLog:
* config/csky/constraints.md ("l", "h"): Delete.
* config/csky/csky.h (reg_class, REG_CLASS_NAMES,
REG_CLASS_CONTENTS): Delete LO_REGS and HI_REGS.
* config/csky/csky.c (regno_reg_classm,
csky_secondary_reload, csky_register_move_cost):
Hi:
Details described in PR.
Bootstrapped and regtest on
x86_64-linux-gnu{-m32,}/x86_64-linux-gnu{-m32\
-march=cascadelake,-march=cascadelake}
Ok for trunk?
gcc/ChangeLog:
PR tree-optimization/pr98365
* tree-if-conv.c (strip_nop_cond_scalar_reduction): New function.
On 5/20/21 10:03 AM, Richard Biener wrote:
Martin - was this intended? Can you fix it up please? (g:cb50701ec2c7)
It was not intentional. Fixed in the attached patch (where I did
a refactoring as well).
Patch can bootstrap on x86_64-linux-gnu and survives regression tests.
Ready to be
"Andre Vieira (lists)" writes:
> Hi,
>
> When vectorizing with --param vect-partial-vector-usage=1 the vectorizer
> uses an unpredicated (all-true predicate for SVE) main loop and a
> predicated tail loop. The way this was implemented seems to mean it
> re-uses the same vector-mode for both
> -Original Message-
> From: Gcc-patches On Behalf Of
> Christophe Lyon via Gcc-patches
> Sent: 30 April 2021 15:10
> To: gcc-patches@gcc.gnu.org
> Subject: [PATCH 8/9] arm: Auto-vectorization for MVE: vld2/vst2
>
> This patch enables MVE vld2/vst2 instructions for auto-vectorization.
On 22/05/21 18:35 +0200, François Dumont wrote:
This was indeed the right approach.
The only minor drawback is that __is_noexcept_invocable<> combines
noexcept qualification of the conversion and of the hash functor. So
if the hash functor is not noexcept we could end up creating
Version v2 uses movsi/di for GOT accesses until after reload as suggested. This
caused worse spilling, however improving the costs of GOT accesses resulted in
better codesize and performance gains:
Improve GOT addressing by treating the instructions as a pair. This reduces
register pressure and
gcc/ChangeLog:
* config/csky/constraints.md ("W"): New constriant for mem operand
with base reg, index register.
("Q"): Renamed and modified "csky_valid_fpuv2_mem_operand" to
"csky_valid_mem_constraint_operand" to deal with both "Q" and "W"
constraint.
Ping patch:
| Subject: [PATCH 1/2] Add IEEE 128-bit min/max support on PowerPC.
| Message-ID: <20210518202606.ga14...@ibm-toto.the-meissners.org>
--
Michael Meissner, IBM
IBM, M/S 2506R, 550 King Street, Littleton, MA 01460-6245, USA
email: meiss...@linux.ibm.com, phone: +1 (978) 899-4797
Ping patch.
| Subject: [PATCH 2/2] Add IEEE 128-bit fp conditional move on PowerPC.
| Message-ID: <20210518202827.gb14...@ibm-toto.the-meissners.org>
Note this patch needs the following patch before it can be applied.
| Subject: [PATCH 1/2] Add IEEE 128-bit min/max support on PowerPC.
|
Ping patch. This is independent of the other patches.
| Date: Tue, 18 May 2021 16:32:33 -0400
| Subject: [PATCH] Fix long double tests when default long double is not IBM.
| Message-ID: <20210518203233.ga15...@ibm-toto.the-meissners.org>
--
Michael Meissner, IBM
IBM, M/S 2506R, 550 King
Ping patch. This is independent of the other patches.
| Date: Tue, 18 May 2021 16:36:32 -0400
| Subject: [PATCH] Allow __ibm128 on older PowerPC systems.
| Message-ID: <20210518203632.ga15...@ibm-toto.the-meissners.org>
--
Michael Meissner, IBM
IBM, M/S 2506R, 550 King Street, Littleton, MA
Ping patch. This is independent of the other patches.
| Date: Tue, 18 May 2021 16:39:28 -0400
| Subject: [PATCH] Change rs6000_const_f32_to_i32 return type.
| Message-ID: <20210518203928.ga15...@ibm-toto.the-meissners.org>
--
Michael Meissner, IBM
IBM, M/S 2506R, 550 King Street, Littleton, MA
Ping patch. This patch is a set of 2 patches. The second patch will need this
patch applied. In addition, the next 3 patches that I will be submitting (to
add support for generating XXSPLTIW, XXSPLTIDP, and XXSPLTI32DX) will need this
patch applied (or else I would just have to reformulate the
Ping patch.
| Date: Tue, 18 May 2021 16:47:58 -0400
| Subject: [PATCH 2/2] Fix xxeval predicates.
| Message-ID: <20210518204758.gb16...@ibm-toto.the-meissners.org>
This needs the following patch to have been applied:
| Date: Tue, 18 May 2021 16:46:47 -0400
| Subject: [PATCH 1/2] Move xx*
Ping patch. This is independent of the other patches.
| Date: Tue, 18 May 2021 16:57:59 -0400
| Subject: [PATCH 1/2] Deal with prefixed loads/stores in tests, PR
testsuite/100166
| Message-ID: <20210518205759.ga18...@ibm-toto.the-meissners.org>
--
Michael Meissner, IBM
IBM, M/S 2506R, 550
Ping patch.
| Date: Tue, 18 May 2021 16:59:12 -0400
| Subject: [PATCH 2/2] Fix tests when running on power10, PR testsuite/100166
| Message-ID: <20210518205912.gb18...@ibm-toto.the-meissners.org>
--
Michael Meissner, IBM
IBM, M/S 2506R, 550 King Street, Littleton, MA 01460-6245, USA
email:
On 5/21/21 1:39 PM, Aldy Hernandez wrote:
This patch provides a generic API for accessing global ranges. It is
meant to replace get_range_info() and get_ptr_nonnull() with one
common interface. It uses the same API as the ranger (class
range_query), so there will now be one API for
Ping https://gcc.gnu.org/pipermail/gcc-patches/2021-May/570139.html
On Tue, May 11, 2021 at 8:29 PM Fangrui Song wrote:
>
> This was introduced in 2014-12 to use local binding for external symbols
> for -fPIE. Now that we have H.J. Lu's GOTPCRELX for years which mostly
> nullify the benefit of
79 matches
Mail list logo
