Hi Christophe, > -----Original Message----- > From: Christophe Lyon <[email protected]> > Sent: Monday, March 4, 2019 13:32 > To: Tamar Christina <[email protected]> > Cc: James Greenhalgh <[email protected]>; Jakub Jelinek > <[email protected]>; Kyrill Tkachov <[email protected]>; gcc- > [email protected]; nd <[email protected]>; Richard Earnshaw > <[email protected]>; Marcus Shawcroft > <[email protected]> > Subject: Re: [PATCH][GCC][AArch64] Have empty HWCAPs string ignored > during native feature detection > > On Wed, 27 Feb 2019 at 18:32, Tamar Christina <[email protected]> > wrote: > > > > Hi James, > > > > > -----Original Message----- > > > From: James Greenhalgh <[email protected]> > > > Sent: Wednesday, February 27, 2019 17:22 > > > To: Tamar Christina <[email protected]> > > > Cc: Jakub Jelinek <[email protected]>; Kyrill Tkachov > > > <[email protected]>; [email protected]; nd > > > <[email protected]>; Richard Earnshaw <[email protected]>; > Marcus > > > Shawcroft <[email protected]> > > > Subject: Re: [PATCH][GCC][AArch64] Have empty HWCAPs string ignored > > > during native feature detection > > > > > > On Thu, Feb 07, 2019 at 04:43:24AM -0600, Tamar Christina wrote: > > > > Hi All, > > > > > > > > Since this hasn't been reviewed yet anyway I've updated this patch > > > > to also > > > fix the memory leaks etc. > > > > > > > > -- > > > > > > > > This patch makes the feature detection code for AArch64 GCC not > > > > add features automatically when the feature had no hwcaps string > > > > to match > > > against. > > > > > > > > This means that -mcpu=native no longer adds feature flags such as > +profile. > > > > The behavior wasn't noticed before because at the time +profile > > > > was added a bug was preventing any feature bits from being added > > > > by native > > > detections. > > > > > > > > The loop has also been changed as Jakub specified in order to > > > > avoid a memory leak that was present in the existing code and to > > > > be slightly more > > > efficient. > > > > > > > > Bootstrapped Regtested on aarch64-none-linux-gnu and no issues. > > > > > > > > Ok for trunk? > > > > > > OK. Is this also desirable for a backport? > > > > Yes I believe we have this problem in GCC8 as well the profile extensions. > > > > Kind regards, > > Tamar > > > > Hi Tamar, > > The new test fails with a cross-compiler, because: > FAIL: gcc.target/aarch64/options_set_10.c (test for excess errors) Excess > errors: > cc1: error: unknown value 'native' for -mcpu > > I don't know how to restrict tests to native compilers only.
Ah, thanks, I tested only the elf builds cross. I'll fix it up with a new testsuite directive that checks if -mcpu=native compiles something. Regards, Tamar > > Christophe > > > > > > > Thanks, > > > James > > > > > > > > > > > Thanks, > > > > Tamar > > > > > > > > gcc/ChangeLog: > > > > > > > > 2019-02-07 Tamar Christina <[email protected]> > > > > > > > > PR target/88530 > > > > * config/aarch64/aarch64-option-extensions.def: Document it. > > > > * config/aarch64/driver-aarch64.c (host_detect_local_cpu): > > > > Skip > > > feature > > > > if empty hwcaps. > > > > > > > > gcc/testsuite/ChangeLog: > > > > > > > > 2019-02-07 Tamar Christina <[email protected]> > > > > > > > > PR target/88530 > > > > * gcc.target/aarch64/options_set_10.c: New test. > > > >
