Re: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-25 Thread Rainer Orth
Hi H.J., > On Tue, Oct 24, 2017 at 2:33 AM, Uros Bizjak wrote: >> On Tue, Oct 24, 2017 at 10:39 AM, Tsimbalist, Igor V >> wrote: >>> OK. >> >> +/* { dg-final { scan-assembler-times "endbr32" 2 { target ia32 } } } */ >> +/* { dg-final { scan-assembler-times "endbr64" 2 { target { ! ia32 } } } }

Re: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-24 Thread H.J. Lu
On Tue, Oct 24, 2017 at 2:33 AM, Uros Bizjak wrote: > On Tue, Oct 24, 2017 at 10:39 AM, Tsimbalist, Igor V > wrote: >> OK. > > +/* { dg-final { scan-assembler-times "endbr32" 2 { target ia32 } } } */ > +/* { dg-final { scan-assembler-times "endbr64" 2 { target { ! ia32 } } } } */ > > I think we c

Re: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-24 Thread Uros Bizjak
st, Igor V >> Cc: Uros Bizjak ; gcc-patches@gcc.gnu.org >> Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called >> directly >> >> On Mon, Oct 23, 2017 at 3:19 PM, Tsimbalist, Igor V >> wrote: >> > You are right. The functions in t

RE: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-24 Thread Tsimbalist, Igor V
OK. Igor > -Original Message- > From: H.J. Lu [mailto:hjl.to...@gmail.com] > Sent: Tuesday, October 24, 2017 1:01 AM > To: Tsimbalist, Igor V > Cc: Uros Bizjak ; gcc-patches@gcc.gnu.org > Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called

Re: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-23 Thread H.J. Lu
: Uros Bizjak ; gcc-patches@gcc.gnu.org >> Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called >> directly >> >> On Mon, Oct 23, 2017 at 3:01 PM, Tsimbalist, Igor V >> wrote: >> > Existing tests cet-label.c cet-switch-2.c cet-sjlj-1

RE: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-23 Thread Tsimbalist, Igor V
ilto:hjl.to...@gmail.com] > Sent: Tuesday, October 24, 2017 12:06 AM > To: Tsimbalist, Igor V > Cc: Uros Bizjak ; gcc-patches@gcc.gnu.org > Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called > directly > > On Mon, Oct 23, 2017 at 3:01 PM, Tsimbalist, Ig

Re: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-23 Thread H.J. Lu
rom: H.J. Lu [mailto:hjl.to...@gmail.com] >> Sent: Monday, October 23, 2017 11:50 PM >> To: Tsimbalist, Igor V >> Cc: Uros Bizjak ; gcc-patches@gcc.gnu.org >> Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called >> directly >> >> On Mon, O

RE: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-23 Thread Tsimbalist, Igor V
g > Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called > directly > > On Mon, Oct 23, 2017 at 2:44 PM, Tsimbalist, Igor V > wrote: > > The change will skip a whole function from endbr processing by > rest_of_insert_endbranch, > > which inser

Re: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-23 Thread H.J. Lu
u >> Cc: gcc-patches@gcc.gnu.org; Tsimbalist, Igor V >> >> Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called >> directly >> >> On Sun, Oct 22, 2017 at 4:13 PM, H.J. Lu wrote: >> > There is no need to insert ENDBR instruction if f

RE: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-23 Thread Tsimbalist, Igor V
ilto:ubiz...@gmail.com] > Sent: Monday, October 23, 2017 9:26 PM > To: H.J. Lu > Cc: gcc-patches@gcc.gnu.org; Tsimbalist, Igor V > > Subject: Re: [PATCH] i386: Don't generate ENDBR if function is only called > directly > > On Sun, Oct 22, 2017 at 4:13 PM, H.J. Lu

Re: [PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-23 Thread Uros Bizjak
On Sun, Oct 22, 2017 at 4:13 PM, H.J. Lu wrote: > There is no need to insert ENDBR instruction if function is only called > directly. > > OK for trunk if there is no regressions? Patch needs to be OK'd by Igor first. Uros. > H.J. > > gcc/ > > PR target/82659 > * config/i386

[PATCH] i386: Don't generate ENDBR if function is only called directly

2017-10-22 Thread H.J. Lu
There is no need to insert ENDBR instruction if function is only called directly. OK for trunk if there is no regressions? H.J. gcc/ PR target/82659 * config/i386/i386.c (pass_insert_endbranch::gate): Return false if function is only called directly. gcc/testsuite/