]> git.ipfire.org Git - thirdparty/kernel/stable-queue.git/commitdiff
.35 patches
authorGreg Kroah-Hartman <gregkh@suse.de>
Thu, 23 Sep 2010 23:12:27 +0000 (16:12 -0700)
committerGreg Kroah-Hartman <gregkh@suse.de>
Thu, 23 Sep 2010 23:12:27 +0000 (16:12 -0700)
queue-2.6.35/alpha-fix-printk-format-errors.patch [new file with mode: 0644]
queue-2.6.35/drm-i915-ensure-that-the-crtcinfo-is-populated-during-mode_fixup.patch [new file with mode: 0644]
queue-2.6.35/series

diff --git a/queue-2.6.35/alpha-fix-printk-format-errors.patch b/queue-2.6.35/alpha-fix-printk-format-errors.patch
new file mode 100644 (file)
index 0000000..9dcba2a
--- /dev/null
@@ -0,0 +1,56 @@
+From 3e073367a57d41e506f20aebb98e308387ce3090 Mon Sep 17 00:00:00 2001
+From: Michael Cree <mcree@orcon.net.nz>
+Date: Wed, 1 Sep 2010 11:25:17 -0400
+Subject: alpha: Fix printk format errors
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+From: Michael Cree <mcree@orcon.net.nz>
+
+commit 3e073367a57d41e506f20aebb98e308387ce3090 upstream.
+
+When compiling alpha generic build get errors such as:
+arch/alpha/kernel/err_marvel.c: In function ‘marvel_print_err_cyc’:
+arch/alpha/kernel/err_marvel.c:119: error: format ‘%ld’ expects type ‘long int’, but argument 6 has type ‘u64’
+
+Replaced a number of %ld format specifiers with %lld since u64
+is unsigned long long.
+
+Signed-off-by: Michael Cree <mcree@orcon.net.nz>
+Signed-off-by: Matt Turner <mattst88@gmail.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
+
+---
+ arch/alpha/kernel/err_marvel.c |    6 +++---
+ 1 file changed, 3 insertions(+), 3 deletions(-)
+
+--- a/arch/alpha/kernel/err_marvel.c
++++ b/arch/alpha/kernel/err_marvel.c
+@@ -109,7 +109,7 @@ marvel_print_err_cyc(u64 err_cyc)
+ #define IO7__ERR_CYC__CYCLE__M        (0x7)
+       printk("%s        Packet In Error: %s\n"
+-             "%s        Error in %s, cycle %ld%s%s\n",
++             "%s        Error in %s, cycle %lld%s%s\n",
+              err_print_prefix, 
+              packet_desc[EXTRACT(err_cyc, IO7__ERR_CYC__PACKET)],
+              err_print_prefix,
+@@ -313,7 +313,7 @@ marvel_print_po7_ugbge_sym(u64 ugbge_sym
+       }
+       printk("%s      Up Hose Garbage Symptom:\n"
+-             "%s        Source Port: %ld - Dest PID: %ld - OpCode: %s\n", 
++             "%s        Source Port: %lld - Dest PID: %lld - OpCode: %s\n",
+              err_print_prefix,
+              err_print_prefix, 
+              EXTRACT(ugbge_sym, IO7__PO7_UGBGE_SYM__UPH_SRC_PORT),
+@@ -552,7 +552,7 @@ marvel_print_pox_spl_cmplt(u64 spl_cmplt
+ #define IO7__POX_SPLCMPLT__REM_BYTE_COUNT__M  (0xfff)
+       printk("%s      Split Completion Error:\n"      
+-             "%s         Source (Bus:Dev:Func): %ld:%ld:%ld\n",
++             "%s         Source (Bus:Dev:Func): %lld:%lld:%lld\n",
+              err_print_prefix,
+              err_print_prefix,
+              EXTRACT(spl_cmplt, IO7__POX_SPLCMPLT__SOURCE_BUS),
diff --git a/queue-2.6.35/drm-i915-ensure-that-the-crtcinfo-is-populated-during-mode_fixup.patch b/queue-2.6.35/drm-i915-ensure-that-the-crtcinfo-is-populated-during-mode_fixup.patch
new file mode 100644 (file)
index 0000000..431fc38
--- /dev/null
@@ -0,0 +1,45 @@
+From 2e217fac094a083b18cf69f76b09a56e71fb0a2b Mon Sep 17 00:00:00 2001
+From: Chris Wilson <chris@chris-wilson.co.uk>
+Date: Sun, 12 Sep 2010 18:25:19 +0100
+Subject: drm/i915: Ensure that the crtcinfo is populated during mode_fixup()
+
+From: Chris Wilson <chris@chris-wilson.co.uk>
+
+commit 897493504addc5609f04a2c4f73c37ab972c29b2 upstream.
+
+This should fix the mysterious mode setting failures reported during
+boot up and after resume, generally for i8xx class machines.
+
+Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=16478
+Reported-and-tested-by: Xavier Chantry <chantry.xavier@gmail.com>
+Buzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29413
+Tested-by: Daniel Vetter <daniel@ffwll.ch>
+Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
+Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
+
+---
+ drivers/gpu/drm/i915/intel_display.c |    8 ++++++++
+ 1 file changed, 8 insertions(+)
+
+--- a/drivers/gpu/drm/i915/intel_display.c
++++ b/drivers/gpu/drm/i915/intel_display.c
+@@ -2376,11 +2376,19 @@ static bool intel_crtc_mode_fixup(struct
+                                 struct drm_display_mode *adjusted_mode)
+ {
+       struct drm_device *dev = crtc->dev;
++
+       if (HAS_PCH_SPLIT(dev)) {
+               /* FDI link clock is fixed at 2.7G */
+               if (mode->clock * 3 > 27000 * 4)
+                       return MODE_CLOCK_HIGH;
+       }
++
++      /* XXX some encoders set the crtcinfo, others don't.
++       * Obviously we need some form of conflict resolution here...
++       */
++      if (adjusted_mode->crtc_htotal == 0)
++              drm_mode_set_crtcinfo(adjusted_mode, 0);
++
+       return true;
+ }
index 1d6d63a835edca60e79449734aeea044811b9099..53bd80fcfdce2d34d6ea6d526c19af7ddc3838cf 100644 (file)
@@ -76,3 +76,5 @@ mm-page-allocator-update-free-page-counters-after-pages-are-placed-on-the-free-l
 guard-page-for-stacks-that-grow-upwards.patch
 fix-unprotected-access-to-task-credentials-in-waitid.patch
 sctp-do-not-reset-the-packet-during-sctp_packet_config.patch
+drm-i915-ensure-that-the-crtcinfo-is-populated-during-mode_fixup.patch
+alpha-fix-printk-format-errors.patch