| Message ID | 1516010110-11590-1-git-send-email-sergio.prado@e-labworks.com |
|---|---|
| State | Changes Requested |
| Headers | show |
| Series | wolfssl: fix build on ARMV8 | expand |
Hello, On Mon, 15 Jan 2018 07:55:10 -0200, Sergio Prado wrote: > diff --git a/package/wolfssl/wolfssl.mk b/package/wolfssl/wolfssl.mk > index c8ef50a9bc6e..3e661cd26ef4 100644 > --- a/package/wolfssl/wolfssl.mk > +++ b/package/wolfssl/wolfssl.mk > @@ -29,10 +29,4 @@ else > WOLFSSL_CONF_OPTS += --disable-sslv3 > endif > > -ifeq ($(BR2_ARM_CPU_ARMV8A),y) > -WOLFSSL_CONF_OPTS += --enable-armasm > -else > -WOLFSSL_CONF_OPTS += --disable-armasm > -endif So perhaps --disable-armasm should be passed unconditionally? Thomas
Hello, 2018-01-15 8:32 GMT-02:00 Thomas Petazzoni <thomas.petazzoni@free- electrons.com>: > Hello, > > On Mon, 15 Jan 2018 07:55:10 -0200, Sergio Prado wrote: > > > diff --git a/package/wolfssl/wolfssl.mk b/package/wolfssl/wolfssl.mk > > index c8ef50a9bc6e..3e661cd26ef4 100644 > > --- a/package/wolfssl/wolfssl.mk > > +++ b/package/wolfssl/wolfssl.mk > > @@ -29,10 +29,4 @@ else > > WOLFSSL_CONF_OPTS += --disable-sslv3 > > endif > > > > -ifeq ($(BR2_ARM_CPU_ARMV8A),y) > > -WOLFSSL_CONF_OPTS += --enable-armasm > > -else > > -WOLFSSL_CONF_OPTS += --disable-armasm > > -endif > > So perhaps --disable-armasm should be passed unconditionally? > It's disabled by default. Should we still pass it? > > Thomas > -- > Thomas Petazzoni, CTO, Free Electrons > Embedded Linux and Kernel engineering > http://free-electrons.com > <div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">Hello,</div><div class="gmail_quote"><br></div><div class="gmail_quote">2018-01-15 8:32 GMT-02:00 Thomas Petazzoni <span dir="ltr"><<a href="mailto:thomas.petazzoni@free-electrons.com" target="_blank">thomas.petazzoni@free-<wbr>electrons.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hello,<br> <span><br> On Mon, 15 Jan 2018 07:55:10 -0200, Sergio Prado wrote:<br> <br> > diff --git a/package/wolfssl/<a href="http://wolfssl.mk" rel="noreferrer" target="_blank">wolfssl.mk</a> b/package/wolfssl/<a href="http://wolfssl.mk" rel="noreferrer" target="_blank">wolfssl.mk</a><br> > index c8ef50a9bc6e..3e661cd26ef4 100644<br> > --- a/package/wolfssl/<a href="http://wolfssl.mk" rel="noreferrer" target="_blank">wolfssl.mk</a><br> > +++ b/package/wolfssl/<a href="http://wolfssl.mk" rel="noreferrer" target="_blank">wolfssl.mk</a><br> > @@ -29,10 +29,4 @@ else<br> > WOLFSSL_CONF_OPTS += --disable-sslv3<br> > endif<br> ><br> > -ifeq ($(BR2_ARM_CPU_ARMV8A),y)<br> > -WOLFSSL_CONF_OPTS += --enable-armasm<br> > -else<br> > -WOLFSSL_CONF_OPTS += --disable-armasm<br> > -endif<br> <br> </span>So perhaps --disable-armasm should be passed unconditionally?<br></blockquote><div><br></div><div>It's disabled by default. Should we still pass it?</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <span class="m_-3688480652619636346HOEnZb"><font color="#888888"><br> Thomas<br> --<br> Thomas Petazzoni, CTO, Free Electrons<br> Embedded Linux and Kernel engineering<br> <a href="http://free-electrons.com" rel="noreferrer" target="_blank">http://free-electrons.com</a><br> </font></span></blockquote></div><br></div></div>
Hello, On Mon, 15 Jan 2018 08:59:22 -0200, Sergio Prado wrote: > > So perhaps --disable-armasm should be passed unconditionally? > > > > It's disabled by default. Should we still pass it? Well, it's always good practice to be explicit about config options I believe. Thomas
diff --git a/package/wolfssl/wolfssl.mk b/package/wolfssl/wolfssl.mk index c8ef50a9bc6e..3e661cd26ef4 100644 --- a/package/wolfssl/wolfssl.mk +++ b/package/wolfssl/wolfssl.mk @@ -29,10 +29,4 @@ else WOLFSSL_CONF_OPTS += --disable-sslv3 endif -ifeq ($(BR2_ARM_CPU_ARMV8A),y) -WOLFSSL_CONF_OPTS += --enable-armasm -else -WOLFSSL_CONF_OPTS += --disable-armasm -endif - $(eval $(autotools-package))
Build on ARMV8 fails when hardware acceleration is enabled: /tmp/ccmGP5RI.s: Assembler messages: /tmp/ccmGP5RI.s:532: Error: invalid addressing mode at operand 2 -- `ld1 {v0.2d},[x0,256]' /tmp/ccmGP5RI.s:568: Error: invalid addressing mode at operand 2 -- `st1 {v0.2d},[x0,256]' /tmp/ccmGP5RI.s:581: Error: invalid addressing mode at operand 2 -- `ld1 {v0.2d},[x0,256]' /tmp/ccmGP5RI.s:621: Error: invalid addressing mode at operand 2 -- `st1 {v0.2d},[x0,256]' So let's disable ARMv8 hardware acceleration for now. Fixes: http://autobuild.buildroot.net/results/6080648394d09a0d382087831ee6f063c7638be9 Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com> --- package/wolfssl/wolfssl.mk | 6 ------ 1 file changed, 6 deletions(-)