This is on master-next now. From the follow ups, I assume that this is still needed. Once I get a bit of soak and runtime testing done on some other queued changes, this will be part of my next

In my world builds with ld-is-gold I've found only 2 more recipes which trigger this internal error. In case someone sees this Illegal instruction as well, then in my case it was caused by missing

I tend to agree. Until it lands in the Xen repo, we are still potentially introducing issues and are on our own for fixing them. Bruce -- - Thou shalt not follow the NULL pointer, for chaos and

Ross also mentioned on IRC, that fedora is cooking something up: https://src.fedoraproject.org/rpms/xen/blob/rawhide/f/xen.gcc11.fixes.patch But I have no idea about the upstream status,

When compiling xen-tools (4.14 and 4.15) with gcc-11 the following kind of errors are produced: linux.c:164:50: error: argument 7 of type 'const xen_pfn_t[]' {aka 'const long unsigned int[]'}

I'm ok with the workaround. The code hasn't changed, and we've lived with it so far, so I'm not really concerned at the technical level about the warning. And I'd rather no one spend the time to do a

Hello, We identified compilation errors for xen-tools (4.14 and 4.15) with GCC 11.1 due to `-Werror=vla-parameter`. There are a couple of source files with similar errors like

runc-opencontainers is the only recipe currently failing like this for me with gold, so instead of switching system linker I would prefer to force bfd only here. But there is also one more issue which

* just a work around for internal error in binutils-2.36 gold: http://errors.yoctoproject.org/Errors/Details/580099/ CGO_ENABLED=1 x86_64-oe-linux-go build -trimpath -tags "seccomp seccomp

<mark.hatle@...> wrote: Merged to the appropriate branches. Bruce -- - Thou shalt not follow the NULL pointer, for chaos and madness await thee at its end - "Use the force Harry" -

________________________________________ 发件人: Bruce Ashfield <bruce.ashfield@...> 发送时间: 2021年4月28日 22:12 收件人: Zhang, Qiang 抄送:

From: Mark Hatle <mark.hatle@...> As of the gatesgarth you are no longer allowed to remove setscene tasks as the eSDK generation expects for tasks to actually be executed. Signed-off-by: Mark

This change applies to gatesgarth, hardknott and master. The same patch applied to all three branches. Without this change, it's not possible to build an eSDK w/ Xen enabled. Mark Hatle (1): xen:

From: Mark Hatle <mark.hatle@...> As of the gatesgarth you are no longer allowed to remove setscene tasks as the eSDK generation expects for tasks to actually be executed. Signed-off-by: Mark

Allows the yocto cve-checker to flag CVEs, which would otherwise go unreported due to the package name not matching NIST NVD data. Signed-off-by: Ralph Siemsen <ralph.siemsen@...> ---

Allows the yocto cve-checker to flag CVEs, which would otherwise go unreported due to the package name not matching NIST NVD data. Signed-off-by: Ralph Siemsen <ralph.siemsen@...> ---

Allows the yocto cve-checker to flag CVEs, which would otherwise go unreported due to the package name not matching NIST NVD data. Signed-off-by: Ralph Siemsen <ralph.siemsen@...> ---

Let me try and summarize, DPDK has replaced a flag that used to point to the binary with a static/non-static flag and is using pkg-config to locate the binary. If the yocto dpdk package isn't

Hi, The serie looks good to me so you can add on the whole serie my: Reviewed-by: Bertrand Marquis <bertrand.marquis@...> I will not have the opportunity to actually test it this week