arm64-force-serial-to-be-active-consdev.patch

force serial consoles to be the primary console device instead of
defaulting to tty0. No changes to drivers outside of ARM-land.

also update kernel-arm64.patch.
This commit is contained in:
Kyle McMartin 2014-12-04 09:55:03 -05:00
parent 6494397039
commit 831ed82ff5
3 changed files with 4808 additions and 1485 deletions

View File

@ -0,0 +1,84 @@
From rharm-kernel-patches-bounces@redhat.com Fri Nov 21 18:21:38 2014
From: Mark Salter <msalter@redhat.com>
To: rharm-kernel-patches@redhat.com
Subject: [PATCH] tty/pl011: make ttyAMA0 the active console device
Date: Fri, 21 Nov 2014 18:21:30 -0500
Message-Id: <1416612090-31594-1-git-send-email-msalter@redhat.com>
Upstream: No. Temporary workaround to avoid console=
Testing: Seattle booting with devicetree
The pl011 uart driver doesn't register itself as a console
until device_initcall time. This allows the virtual console
driver to register the active console if no console= is
given on the cmdline. This patch allows ttyAMA0 to take
over the active console device role from any existing
console device if no console= is given on the cmdline.
Signed-off-by: Mark Salter <msalter@redhat.com>
---
drivers/tty/serial/amba-pl011.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/drivers/tty/serial/amba-pl011.c b/drivers/tty/serial/amba-pl011.c
index 8572f2a..454aa26 100644
--- a/drivers/tty/serial/amba-pl011.c
+++ b/drivers/tty/serial/amba-pl011.c
@@ -2182,7 +2182,15 @@ static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
}
}
+ /*
+ * temp hack to avoid need for console= on cmdline
+ * this can go away when we switch completely to acpi
+ */
+ if (amba_reg.cons && !console_set_on_cmdline && uap->port.line == 0)
+ amba_reg.cons->flags |= CON_CONSDEV;
ret = uart_add_one_port(&amba_reg, &uap->port);
+ if (amba_reg.cons && !console_set_on_cmdline && uap->port.line == 0)
+ amba_reg.cons->flags &= ~CON_CONSDEV;
if (ret) {
amba_ports[i] = NULL;
uart_unregister_driver(&amba_reg);
--
1.9.3
From rharm-kernel-patches-bounces@redhat.com Tue Nov 18 22:09:42 2014
From: Mark Salter <msalter@redhat.com>
To: rharm-kernel-patches@redhat.com
Subject: [PATCH v2] tty/sbsauart: make ttySBSA the active console device
Date: Tue, 18 Nov 2014 22:09:04 -0500
Message-Id: <1416366544-31116-2-git-send-email-msalter@redhat.com>
The sbsauart driver doesn't register itself as a console
until module_initcall time. This allows the virtual console
driver to register the active console if no console= is
given on the cmdline. This patch allows ttySBSA to take
over the active console device role from any existing
console device if no console= is given on the cmdline.
Upstream: No
Testing: Seattle and Foundation Model
Signed-off-by: Mark Salter <msalter@redhat.com>
---
drivers/tty/sbsauart.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/tty/sbsauart.c b/drivers/tty/sbsauart.c
index 3a3ff88..e7d35d5 100644
--- a/drivers/tty/sbsauart.c
+++ b/drivers/tty/sbsauart.c
@@ -280,6 +280,9 @@ static int sbsa_tty_probe(struct platform_device *pdev)
qtty->console.device = sbsa_tty_console_device;
qtty->console.setup = sbsa_tty_console_setup;
qtty->console.flags = CON_PRINTBUFFER;
+ /* if no console= on cmdline, make this the console device */
+ if (!console_set_on_cmdline)
+ qtty->console.flags |= CON_CONSDEV;
qtty->console.index = pdev->id;
register_console(&qtty->console);
--
1.9.3

File diff suppressed because it is too large Load Diff

View File

@ -42,7 +42,7 @@ Summary: The Linux kernel
# For non-released -rc kernels, this will be appended after the rcX and
# gitX tags, so a 3 here would become part of release "0.rcX.gitX.3"
#
%global baserelease 301
%global baserelease 302
%global fedora_build %{baserelease}
# base_sublevel is the kernel version we're starting with and patching
@ -571,6 +571,7 @@ Patch15000: watchdog-Disable-watchdog-on-virtual-machines.patch
# PPC
Patch18000: ppc64-fixtools.patch
# ARM64
Patch20000: arm64-force-serial-to-be-active-consdev.patch
# ARMv7
Patch21020: ARM-tegra-usb-no-reset.patch
@ -1375,6 +1376,10 @@ ApplyPatch drm-i915-Ignore-long-hpds-on-eDP-ports.patch
ApplyPatch kernel-arm64.patch
%ifnarch aarch64 # this is stupid, but i want to notice before secondary koji does.
ApplyPatch kernel-arm64.patch -R
%else
# arm64-force-serial-to-be-active-consdev.patch: not for upstream
# solved with SPCR in future
ApplyPatch arm64-force-serial-to-be-active-consdev.patch
%endif
%endif
@ -2239,6 +2244,12 @@ fi
# ||----w |
# || ||
%changelog
* Thu Dec 04 2014 Kyle McMartin <kyle@fedoraproject.org> - 3.17.4-302
- kernel-arm64.patch: update.
- arm64-force-serial-to-be-active-consdev.patch: force serial consoles
to be the primary console device instead of defaulting to tty0. No
changes to drivers outside of ARM-land.
* Mon Dec 01 2014 Josh Boyer <jwboyer@fedoraproject.org>
- Add patch to quiet i915 driver on long hdps
- Add patch to fix oops when using xpad (rhbz 1094048)