* [stable 4.9-4.19] Fix arm64 build regression in xen-netback
@ 2021-03-01 19:56 Ben Hutchings
2021-03-04 13:49 ` Greg KH
0 siblings, 1 reply; 2+ messages in thread
From: Ben Hutchings @ 2021-03-01 19:56 UTC (permalink / raw)
To: stable
[-- Attachment #1.1: Type: text/plain, Size: 1092 bytes --]
The arm64 implementations of some atomic operations had incorrect
assembly constraints. Depending on the compiler version and
options, this can result in a build failure for some parameter
values:
/tmp/ccDOb5nB.s: Assembler messages:
/tmp/ccDOb5nB.s:2214: Error: immediate out of range at operand 3 -- `bic w1,w0,5'
This has specifically been seen when building a 4.9 stable kernel with
gcc 6.3.0, since commit 23025393dbeb "xen/netback: use lateeoi irq
binding" was applied. I can also reproduce it with 4.14.
I cannot reproduce it with 4.19, but the same fixes are applicable and
the issue presumably could occur when using different compiler
options.
I haven't done anything about the 4.4 branch since it does not have
that xen-netback fix and it has significantly different definitions
for arm64 atomic ops.
I've attached a mailbox of patches for each of the 4.9, 4.14, and 4.19
branches.
Ben.
--
Ben Hutchings
I'm always amazed by the number of people who take up solipsism because
they heard someone else explain it. - E*Borg on alt.fan.pratchett
[-- Attachment #1.2: arm64-atomic-4.9.mbox --]
[-- Type: application/mbox, Size: 27817 bytes --]
[-- Attachment #1.3: arm64-atomic-4.14.mbox --]
[-- Type: application/mbox, Size: 25979 bytes --]
[-- Attachment #1.4: arm64-atomic-4.19.mbox --]
[-- Type: application/mbox, Size: 26055 bytes --]
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [stable 4.9-4.19] Fix arm64 build regression in xen-netback
2021-03-01 19:56 [stable 4.9-4.19] Fix arm64 build regression in xen-netback Ben Hutchings
@ 2021-03-04 13:49 ` Greg KH
0 siblings, 0 replies; 2+ messages in thread
From: Greg KH @ 2021-03-04 13:49 UTC (permalink / raw)
To: Ben Hutchings; +Cc: stable
On Mon, Mar 01, 2021 at 08:56:07PM +0100, Ben Hutchings wrote:
> The arm64 implementations of some atomic operations had incorrect
> assembly constraints. Depending on the compiler version and
> options, this can result in a build failure for some parameter
> values:
>
> /tmp/ccDOb5nB.s: Assembler messages:
> /tmp/ccDOb5nB.s:2214: Error: immediate out of range at operand 3 -- `bic w1,w0,5'
>
> This has specifically been seen when building a 4.9 stable kernel with
> gcc 6.3.0, since commit 23025393dbeb "xen/netback: use lateeoi irq
> binding" was applied. I can also reproduce it with 4.14.
>
> I cannot reproduce it with 4.19, but the same fixes are applicable and
> the issue presumably could occur when using different compiler
> options.
>
> I haven't done anything about the 4.4 branch since it does not have
> that xen-netback fix and it has significantly different definitions
> for arm64 atomic ops.
>
> I've attached a mailbox of patches for each of the 4.9, 4.14, and 4.19
> branches.
All now queued up, thanks.
greg k-h
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-03-04 13:51 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-03-01 19:56 [stable 4.9-4.19] Fix arm64 build regression in xen-netback Ben Hutchings
2021-03-04 13:49 ` Greg KH
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.