February 2018 - Linux-stable-mirror

Patch "extcon: int3496: process id-pin first so that we start with the right status" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled extcon: int3496: process id-pin first so that we start with the right status to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: extcon-int3496-process-id-pin-first-so-that-we-start-with-the-right-status.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 0434352d3d2e950cf5e743f6062abd87de22f960 Mon Sep 17 00:00:00 2001 From: Hans de Goede <hdegoede(a)redhat.com> Date: Tue, 13 Feb 2018 20:25:50 +0100 Subject: extcon: int3496: process id-pin first so that we start with the right status From: Hans de Goede <hdegoede(a)redhat.com> commit 0434352d3d2e950cf5e743f6062abd87de22f960 upstream. Some other drivers may be waiting for our extcon to show-up, exiting their probe methods with -EPROBE_DEFER until we show up. These drivers will typically get the cable state directly after getting the extcon, this commit changes the int3496 code to wait for the initial processing of the id-pin to complete before exiting probe() with 0, which will cause devices waiting on the defered probe to get reprobed. This fixes a race where the initial work might still be running while other drivers were already calling extcon_get_state(). Fixes: 2f556bdb9f2e ("extcon: int3496: Add Intel INT3496 ACPI ... driver") Cc: stable(a)vger.kernel.org Signed-off-by: Hans de Goede <hdegoede(a)redhat.com> Signed-off-by: Chanwoo Choi <cw00.choi(a)samsung.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/extcon/extcon-intel-int3496.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) --- a/drivers/extcon/extcon-intel-int3496.c +++ b/drivers/extcon/extcon-intel-int3496.c @@ -153,8 +153,9 @@ static int int3496_probe(struct platform return ret; } - /* queue initial processing of id-pin */ + /* process id-pin so that we start with the right status */ queue_delayed_work(system_wq, &data->work, 0); + flush_delayed_work(&data->work); platform_set_drvdata(pdev, data); Patches currently in stable-queue which might be from hdegoede(a)redhat.com are queue-4.15/extcon-int3496-process-id-pin-first-so-that-we-start-with-the-right-status.patch

6 years, 10 months

1
0
0 0

Patch "drm: Handle unexpected holes in color-eviction" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm: Handle unexpected holes in color-eviction to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-handle-unexpected-holes-in-color-eviction.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From b8ff1802815913aad52695898cccbc9f77b7e726 Mon Sep 17 00:00:00 2001 From: Chris Wilson <chris(a)chris-wilson.co.uk> Date: Mon, 19 Feb 2018 11:35:43 +0000 Subject: drm: Handle unexpected holes in color-eviction From: Chris Wilson <chris(a)chris-wilson.co.uk> commit b8ff1802815913aad52695898cccbc9f77b7e726 upstream. During eviction, the driver may free more than one hole in the drm_mm due to the side-effects in evicting the scanned nodes. However, drm_mm_scan_color_evict() expects that the scan result is the first available hole (in the mru freed hole_stack list): kernel BUG at drivers/gpu/drm/drm_mm.c:844! invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI Dumping ftrace buffer: (ftrace buffer empty) Modules linked in: i915 snd_hda_codec_analog snd_hda_codec_generic coretemp snd_hda_intel snd_hda_codec snd_hwdep snd_hda_core lpc_ich snd_pcm e1000e mei_me prime_numbers mei CPU: 1 PID: 1490 Comm: gem_userptr_bli Tainted: G U 4.16.0-rc1-g740f57c54ecf-kasan_6+ #1 Hardware name: Dell Inc. OptiPlex 755 /0PU052, BIOS A08 02/19/2008 RIP: 0010:drm_mm_scan_color_evict+0x2b8/0x3d0 RSP: 0018:ffff880057a573f8 EFLAGS: 00010287 RAX: ffff8800611f5980 RBX: ffff880057a575d0 RCX: dffffc0000000000 RDX: 00000000029d5000 RSI: 1ffff1000af4aec1 RDI: ffff8800611f5a10 RBP: ffff88005ab884d0 R08: ffff880057a57600 R09: 000000000afff000 R10: 1ffff1000b5710b5 R11: 0000000000001000 R12: 1ffff1000af4ae82 R13: ffff8800611f59b0 R14: ffff8800611f5980 R15: ffff880057a57608 FS: 00007f2de0c2e8c0(0000) GS:ffff88006ac40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f2ddde1e000 CR3: 00000000609b2000 CR4: 00000000000006e0 Call Trace: ? drm_mm_scan_remove_block+0x330/0x330 ? drm_mm_scan_remove_block+0x151/0x330 i915_gem_evict_something+0x711/0xbd0 [i915] ? igt_evict_contexts+0x50/0x50 [i915] ? nop_clear_range+0x10/0x10 [i915] ? igt_evict_something+0x90/0x90 [i915] ? i915_gem_gtt_reserve+0x1a1/0x320 [i915] i915_gem_gtt_insert+0x237/0x400 [i915] __i915_vma_do_pin+0xc25/0x1a20 [i915] eb_lookup_vmas+0x1c63/0x3790 [i915] ? i915_gem_check_execbuffer+0x250/0x250 [i915] ? trace_hardirqs_on_caller+0x33f/0x590 ? _raw_spin_unlock_irqrestore+0x39/0x60 ? __pm_runtime_resume+0x7d/0xf0 i915_gem_do_execbuffer+0x86a/0x2ff0 [i915] ? __kmalloc+0x132/0x340 ? i915_gem_execbuffer2_ioctl+0x10f/0x760 [i915] ? drm_ioctl_kernel+0x12e/0x1c0 ? drm_ioctl+0x662/0x980 ? eb_relocate_slow+0xa90/0xa90 [i915] ? i915_gem_execbuffer2_ioctl+0x10f/0x760 [i915] ? __might_fault+0xea/0x1a0 i915_gem_execbuffer2_ioctl+0x3cc/0x760 [i915] ? i915_gem_execbuffer_ioctl+0xba0/0xba0 [i915] ? lock_acquire+0x3c0/0x3c0 ? i915_gem_execbuffer_ioctl+0xba0/0xba0 [i915] drm_ioctl_kernel+0x12e/0x1c0 drm_ioctl+0x662/0x980 ? i915_gem_execbuffer_ioctl+0xba0/0xba0 [i915] ? drm_getstats+0x20/0x20 ? debug_check_no_obj_freed+0x2a6/0x8c0 do_vfs_ioctl+0x170/0xe70 ? ioctl_preallocate+0x170/0x170 ? task_work_run+0xbe/0x160 ? lock_acquire+0x3c0/0x3c0 ? trace_hardirqs_on_caller+0x33f/0x590 ? _raw_spin_unlock_irq+0x2f/0x50 SyS_ioctl+0x36/0x70 ? do_vfs_ioctl+0xe70/0xe70 do_syscall_64+0x18c/0x5d0 entry_SYSCALL_64_after_hwframe+0x26/0x9b RIP: 0033:0x7f2ddf13b587 RSP: 002b:00007fff15c4f9d8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f2ddf13b587 RDX: 00007fff15c4fa20 RSI: 0000000040406469 RDI: 0000000000000003 RBP: 00007fff15c4fa20 R08: 0000000000000000 R09: 00007f2ddf3fe120 R10: 0000000000000073 R11: 0000000000000246 R12: 0000000040406469 R13: 0000000000000003 R14: 00007fff15c4fa20 R15: 00000000000000c7 Code: 00 00 00 4a c7 44 22 08 00 00 00 00 42 c7 44 22 10 00 00 00 00 48 81 c4 b8 00 00 00 5b 5d 41 5c 41 5d 41 5e 41 5f c3 0f 0b 0f 0b <0f> 0b 31 c0 eb c0 4c 89 ef e8 9a 09 41 ff e9 1e fe ff ff 4c 89 RIP: drm_mm_scan_color_evict+0x2b8/0x3d0 RSP: ffff880057a573f8 We can trivially relax this assumption by searching the hole_stack for the scan result and warn instead if the driver called us without any result. Fixes: 3fa489dabea9 ("drm: Apply tight eviction scanning to color_adjust") Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen(a)linux.intel.com> Cc: <stable(a)vger.kernel.org> # v4.11+ Reviewed-by: Joonas Lahtinen <joonas.lahtinen(a)linux.intel.com> Reviewed-by: Daniel Vetter <daniel.vetter(a)ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20180219113543.8010-1-chris@c… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/drm_mm.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) --- a/drivers/gpu/drm/drm_mm.c +++ b/drivers/gpu/drm/drm_mm.c @@ -836,9 +836,24 @@ struct drm_mm_node *drm_mm_scan_color_ev if (!mm->color_adjust) return NULL; - hole = list_first_entry(&mm->hole_stack, typeof(*hole), hole_stack); - hole_start = __drm_mm_hole_node_start(hole); - hole_end = hole_start + hole->hole_size; + /* + * The hole found during scanning should ideally be the first element + * in the hole_stack list, but due to side-effects in the driver it + * may not be. + */ + list_for_each_entry(hole, &mm->hole_stack, hole_stack) { + hole_start = __drm_mm_hole_node_start(hole); + hole_end = hole_start + hole->hole_size; + + if (hole_start <= scan->hit_start && + hole_end >= scan->hit_end) + break; + } + + /* We should only be called after we found the hole previously */ + DRM_MM_BUG_ON(&hole->hole_stack == &mm->hole_stack); + if (unlikely(&hole->hole_stack == &mm->hole_stack)) + return NULL; DRM_MM_BUG_ON(hole_start > scan->hit_start); DRM_MM_BUG_ON(hole_end < scan->hit_end); Patches currently in stable-queue which might be from chris(a)chris-wilson.co.uk are queue-4.15/drm-handle-unexpected-holes-in-color-eviction.patch

6 years, 10 months

1
0
0 0

Patch "drm/edid: Add 6 bpc quirk for CPT panel in Asus UX303LA" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm/edid: Add 6 bpc quirk for CPT panel in Asus UX303LA to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-edid-add-6-bpc-quirk-for-cpt-panel-in-asus-ux303la.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 06998a756a3865817b87a129a7e5d5bb66dc1ec3 Mon Sep 17 00:00:00 2001 From: Kai-Heng Feng <kai.heng.feng(a)canonical.com> Date: Sun, 18 Feb 2018 16:53:59 +0800 Subject: drm/edid: Add 6 bpc quirk for CPT panel in Asus UX303LA From: Kai-Heng Feng <kai.heng.feng(a)canonical.com> commit 06998a756a3865817b87a129a7e5d5bb66dc1ec3 upstream. Similar to commit e10aec652f31 ("drm/edid: Add 6 bpc quirk for display AEO model 0."), the EDID reports "DFP 1.x compliant TMDS" but it support 6bpc instead of 8 bpc. Hence, use 6 bpc quirk for this panel. Fixes: 196f954e2509 ("drm/i915/dp: Revert "drm/i915/dp: fall back to 18 bpp when sink capability is unknown"") BugLink: https://bugs.launchpad.net/bugs/1749420 Signed-off-by: Kai-Heng Feng <kai.heng.feng(a)canonical.com> Reviewed-by: Mario Kleiner <mario.kleiner.de(a)gmail.com> Cc: <stable(a)vger.kernel.org> # v4.8+ Signed-off-by: Daniel Vetter <daniel.vetter(a)ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20180218085359.7817-1-kai.hen… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/drm_edid.c | 3 +++ 1 file changed, 3 insertions(+) --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -113,6 +113,9 @@ static const struct edid_quirk { /* AEO model 0 reports 8 bpc, but is a 6 bpc panel */ { "AEO", 0, EDID_QUIRK_FORCE_6BPC }, + /* CPT panel of Asus UX303LA reports 8 bpc, but is a 6 bpc panel */ + { "CPT", 0x17df, EDID_QUIRK_FORCE_6BPC }, + /* Belinea 10 15 55 */ { "MAX", 1516, EDID_QUIRK_PREFER_LARGE_60 }, { "MAX", 0x77e, EDID_QUIRK_PREFER_LARGE_60 }, Patches currently in stable-queue which might be from kai.heng.feng(a)canonical.com are queue-4.15/drm-amdgpu-add-new-device-to-use-atpx-quirk.patch queue-4.15/drm-edid-add-6-bpc-quirk-for-cpt-panel-in-asus-ux303la.patch

6 years, 10 months

1
0
0 0

Patch "drm/cirrus: Load lut in crtc_commit" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm/cirrus: Load lut in crtc_commit to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-cirrus-load-lut-in-crtc_commit.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 745fd50f3b044db6a3922e1718306555613164b0 Mon Sep 17 00:00:00 2001 From: Daniel Vetter <daniel.vetter(a)ffwll.ch> Date: Wed, 31 Jan 2018 12:04:50 +0100 Subject: drm/cirrus: Load lut in crtc_commit From: Daniel Vetter <daniel.vetter(a)ffwll.ch> commit 745fd50f3b044db6a3922e1718306555613164b0 upstream. In the past the ast driver relied upon the fbdev emulation helpers to call ->load_lut at boot-up. But since commit b8e2b0199cc377617dc238f5106352c06dcd3fa2 Author: Peter Rosin <peda(a)axentia.se> Date: Tue Jul 4 12:36:57 2017 +0200 drm/fb-helper: factor out pseudo-palette that's cleaned up and drivers are expected to boot into a consistent lut state. This patch fixes that. Fixes: b8e2b0199cc3 ("drm/fb-helper: factor out pseudo-palette") Cc: Peter Rosin <peda(a)axenita.se> Cc: Daniel Vetter <daniel.vetter(a)ffwll.ch> Cc: <stable(a)vger.kernel.org> # v4.14+ References: https://bugzilla.kernel.org/show_bug.cgi?id=198123 Signed-off-by: Daniel Vetter <daniel.vetter(a)intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20180131110450.22153-1-daniel.… Signed-off-by: Gerd Hoffmann <kraxel(a)redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/cirrus/cirrus_mode.c | 40 ++++++++++++++++++++--------------- 1 file changed, 23 insertions(+), 17 deletions(-) --- a/drivers/gpu/drm/cirrus/cirrus_mode.c +++ b/drivers/gpu/drm/cirrus/cirrus_mode.c @@ -294,22 +294,7 @@ static void cirrus_crtc_prepare(struct d { } -/* - * This is called after a mode is programmed. It should reverse anything done - * by the prepare function - */ -static void cirrus_crtc_commit(struct drm_crtc *crtc) -{ -} - -/* - * The core can pass us a set of gamma values to program. We actually only - * use this for 8-bit mode so can't perform smooth fades on deeper modes, - * but it's a requirement that we provide the function - */ -static int cirrus_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green, - u16 *blue, uint32_t size, - struct drm_modeset_acquire_ctx *ctx) +static void cirrus_crtc_load_lut(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; struct cirrus_device *cdev = dev->dev_private; @@ -317,7 +302,7 @@ static int cirrus_crtc_gamma_set(struct int i; if (!crtc->enabled) - return 0; + return; r = crtc->gamma_store; g = r + crtc->gamma_size; @@ -330,6 +315,27 @@ static int cirrus_crtc_gamma_set(struct WREG8(PALETTE_DATA, *g++ >> 8); WREG8(PALETTE_DATA, *b++ >> 8); } +} + +/* + * This is called after a mode is programmed. It should reverse anything done + * by the prepare function + */ +static void cirrus_crtc_commit(struct drm_crtc *crtc) +{ + cirrus_crtc_load_lut(crtc); +} + +/* + * The core can pass us a set of gamma values to program. We actually only + * use this for 8-bit mode so can't perform smooth fades on deeper modes, + * but it's a requirement that we provide the function + */ +static int cirrus_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green, + u16 *blue, uint32_t size, + struct drm_modeset_acquire_ctx *ctx) +{ + cirrus_crtc_load_lut(crtc); return 0; } Patches currently in stable-queue which might be from daniel.vetter(a)ffwll.ch are queue-4.15/drm-handle-unexpected-holes-in-color-eviction.patch queue-4.15/drm-cirrus-load-lut-in-crtc_commit.patch queue-4.15/drm-edid-add-6-bpc-quirk-for-cpt-panel-in-asus-ux303la.patch

6 years, 10 months

1
0
0 0

Patch "drm/atomic: Fix memleak on ERESTARTSYS during non-blocking commits" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm/atomic: Fix memleak on ERESTARTSYS during non-blocking commits to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-atomic-fix-memleak-on-erestartsys-during-non-blocking-commits.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 54f809cfbd6b4a43959039f5d33596ed3297ce16 Mon Sep 17 00:00:00 2001 From: "Leo (Sunpeng) Li" <sunpeng.li(a)amd.com> Date: Wed, 17 Jan 2018 12:51:08 +0100 Subject: drm/atomic: Fix memleak on ERESTARTSYS during non-blocking commits From: Leo (Sunpeng) Li <sunpeng.li(a)amd.com> commit 54f809cfbd6b4a43959039f5d33596ed3297ce16 upstream. During a non-blocking commit, it is possible to return before the commit_tail work is queued (-ERESTARTSYS, for example). Since a reference on the crtc commit object is obtained for the pending vblank event when preparing the commit, the above situation will leave us with an extra reference. Therefore, if the commit_tail worker has not consumed the event at the end of a commit, release it's reference. Changes since v1: - Also check for state->event->base.completion being set, to handle the case where stall_checks() fails in setup_crtc_commit(). Changes since v2: - Add a flag to drm_crtc_commit, to prevent dereferencing a freed event. i915 may unreference the state in a worker. Fixes: 24835e442f28 ("drm: reference count event->completion") Cc: <stable(a)vger.kernel.org> # v4.11+ Signed-off-by: Leo (Sunpeng) Li <sunpeng.li(a)amd.com> Acked-by: Harry Wentland <harry.wentland(a)amd.com> #v1 Signed-off-by: Maarten Lankhorst <maarten.lankhorst(a)linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180117115108.29608-1-maarte… Reviewed-by: Sean Paul <seanpaul(a)chromium.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/drm_atomic_helper.c | 15 +++++++++++++++ include/drm/drm_atomic.h | 9 +++++++++ 2 files changed, 24 insertions(+) --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -1778,6 +1778,8 @@ int drm_atomic_helper_setup_commit(struc new_crtc_state->event->base.completion = &commit->flip_done; new_crtc_state->event->base.completion_release = release_crtc_commit; drm_crtc_commit_get(commit); + + commit->abort_completion = true; } for_each_oldnew_connector_in_state(state, conn, old_conn_state, new_conn_state, i) { @@ -3327,8 +3329,21 @@ EXPORT_SYMBOL(drm_atomic_helper_crtc_dup void __drm_atomic_helper_crtc_destroy_state(struct drm_crtc_state *state) { if (state->commit) { + /* + * In the event that a non-blocking commit returns + * -ERESTARTSYS before the commit_tail work is queued, we will + * have an extra reference to the commit object. Release it, if + * the event has not been consumed by the worker. + * + * state->event may be freed, so we can't directly look at + * state->event->base.completion. + */ + if (state->event && state->commit->abort_completion) + drm_crtc_commit_put(state->commit); + kfree(state->commit->event); state->commit->event = NULL; + drm_crtc_commit_put(state->commit); } --- a/include/drm/drm_atomic.h +++ b/include/drm/drm_atomic.h @@ -134,6 +134,15 @@ struct drm_crtc_commit { * &drm_pending_vblank_event pointer to clean up private events. */ struct drm_pending_vblank_event *event; + + /** + * @abort_completion: + * + * A flag that's set after drm_atomic_helper_setup_commit takes a second + * reference for the completion of $drm_crtc_state.event. It's used by + * the free code to remove the second reference if commit fails. + */ + bool abort_completion; }; struct __drm_planes_state { Patches currently in stable-queue which might be from sunpeng.li(a)amd.com are queue-4.15/drm-atomic-fix-memleak-on-erestartsys-during-non-blocking-commits.patch

6 years, 10 months

1
0
0 0

Patch "drm/amdgpu: fix VA hole handling on Vega10 v3" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm/amdgpu: fix VA hole handling on Vega10 v3 to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-amdgpu-fix-va-hole-handling-on-vega10-v3.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From bb7939b2030ab55acd203c86160c37db22f5796a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig(a)amd.com> Date: Mon, 6 Nov 2017 15:37:01 +0100 Subject: drm/amdgpu: fix VA hole handling on Vega10 v3 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit From: Christian König <christian.koenig(a)amd.com> commit bb7939b2030ab55acd203c86160c37db22f5796a upstream. Similar to the CPU address space the VA on Vega10 has a hole in it. v2: use dev_dbg instead of dev_err v3: add some more comments to explain how the hw works Signed-off-by: Christian König <christian.koenig(a)amd.com> Reviewed-by: Alex Deucher <alexander.deucher(a)amd.com> CC: stable(a)vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 10 +++++----- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 11 +++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 4 +++- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 13 +++++++++++++ 4 files changed, 32 insertions(+), 6 deletions(-) --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -865,8 +865,8 @@ static int amdgpu_cs_ib_vm_chunk(struct struct amdgpu_bo_va_mapping *m; struct amdgpu_bo *aobj = NULL; struct amdgpu_cs_chunk *chunk; + uint64_t offset, va_start; struct amdgpu_ib *ib; - uint64_t offset; uint8_t *kptr; chunk = &p->chunks[i]; @@ -876,14 +876,14 @@ static int amdgpu_cs_ib_vm_chunk(struct if (chunk->chunk_id != AMDGPU_CHUNK_ID_IB) continue; - r = amdgpu_cs_find_mapping(p, chunk_ib->va_start, - &aobj, &m); + va_start = chunk_ib->va_start & AMDGPU_VA_HOLE_MASK; + r = amdgpu_cs_find_mapping(p, va_start, &aobj, &m); if (r) { DRM_ERROR("IB va_start is invalid\n"); return r; } - if ((chunk_ib->va_start + chunk_ib->ib_bytes) > + if ((va_start + chunk_ib->ib_bytes) > (m->last + 1) * AMDGPU_GPU_PAGE_SIZE) { DRM_ERROR("IB va_start+ib_bytes is invalid\n"); return -EINVAL; @@ -896,7 +896,7 @@ static int amdgpu_cs_ib_vm_chunk(struct } offset = m->start * AMDGPU_GPU_PAGE_SIZE; - kptr += chunk_ib->va_start - offset; + kptr += va_start - offset; memcpy(ib->ptr, kptr, chunk_ib->ib_bytes); amdgpu_bo_kunmap(aobj); --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -563,6 +563,17 @@ int amdgpu_gem_va_ioctl(struct drm_devic return -EINVAL; } + if (args->va_address >= AMDGPU_VA_HOLE_START && + args->va_address < AMDGPU_VA_HOLE_END) { + dev_dbg(&dev->pdev->dev, + "va_address 0x%LX is in VA hole 0x%LX-0x%LX\n", + args->va_address, AMDGPU_VA_HOLE_START, + AMDGPU_VA_HOLE_END); + return -EINVAL; + } + + args->va_address &= AMDGPU_VA_HOLE_MASK; + if ((args->flags & ~valid_flags) && (args->flags & ~prt_flags)) { dev_err(&dev->pdev->dev, "invalid flags combination 0x%08X\n", args->flags); --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -586,7 +586,9 @@ static int amdgpu_info_ioctl(struct drm_ if (amdgpu_sriov_vf(adev)) dev_info.ids_flags |= AMDGPU_IDS_FLAGS_PREEMPTION; dev_info.virtual_address_offset = AMDGPU_VA_RESERVED_SIZE; - dev_info.virtual_address_max = (uint64_t)adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE; + dev_info.virtual_address_max = + min(adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE, + AMDGPU_VA_HOLE_START); dev_info.virtual_address_alignment = max((int)PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE); dev_info.pte_fragment_size = (1 << adev->vm_manager.fragment_size) * AMDGPU_GPU_PAGE_SIZE; dev_info.gart_page_size = AMDGPU_GPU_PAGE_SIZE; --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -96,6 +96,19 @@ struct amdgpu_bo_list_entry; /* hardcode that limit for now */ #define AMDGPU_VA_RESERVED_SIZE (8ULL << 20) +/* VA hole for 48bit addresses on Vega10 */ +#define AMDGPU_VA_HOLE_START 0x0000800000000000ULL +#define AMDGPU_VA_HOLE_END 0xffff800000000000ULL + +/* + * Hardware is programmed as if the hole doesn't exists with start and end + * address values. + * + * This mask is used to remove the upper 16bits of the VA and so come up with + * the linear addr value. + */ +#define AMDGPU_VA_HOLE_MASK 0x0000ffffffffffffULL + /* max vmids dedicated for process */ #define AMDGPU_VM_MAX_RESERVED_VMID 1 Patches currently in stable-queue which might be from christian.koenig(a)amd.com are queue-4.15/drm-amdgpu-add-dpm-quirk-for-jet-pro-v2.patch queue-4.15/drm-amdgpu-fix-va-hole-handling-on-vega10-v3.patch queue-4.15/drm-amdgpu-add-atpx-quirk-handling-v2.patch

6 years, 10 months

1
0
0 0

Patch "drm/amdgpu: only check mmBIF_IOV_FUNC_IDENTIFIER on tonga/fiji" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm/amdgpu: only check mmBIF_IOV_FUNC_IDENTIFIER on tonga/fiji to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-amdgpu-only-check-mmbif_iov_func_identifier-on-tonga-fiji.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 57ad33a307bf85cafda3a77c03a555c9f9ee4139 Mon Sep 17 00:00:00 2001 From: Alex Deucher <alexander.deucher(a)amd.com> Date: Tue, 19 Dec 2017 09:52:31 -0500 Subject: drm/amdgpu: only check mmBIF_IOV_FUNC_IDENTIFIER on tonga/fiji From: Alex Deucher <alexander.deucher(a)amd.com> commit 57ad33a307bf85cafda3a77c03a555c9f9ee4139 upstream. We only support SR-IOV on tonga/fiji. Don't check this register on other VI parts. Fixes: 048765ad5af7c89 (amdgpu: fix asic initialization for virtualized environments (v2)) Reviewed-by: Xiangliang Yu <Xiangliang.Yu(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/amd/amdgpu/vi.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -449,14 +449,19 @@ static bool vi_read_bios_from_rom(struct static void vi_detect_hw_virtualization(struct amdgpu_device *adev) { - uint32_t reg = RREG32(mmBIF_IOV_FUNC_IDENTIFIER); - /* bit0: 0 means pf and 1 means vf */ - /* bit31: 0 means disable IOV and 1 means enable */ - if (reg & 1) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF; + uint32_t reg = 0; - if (reg & 0x80000000) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV; + if (adev->asic_type == CHIP_TONGA || + adev->asic_type == CHIP_FIJI) { + reg = RREG32(mmBIF_IOV_FUNC_IDENTIFIER); + /* bit0: 0 means pf and 1 means vf */ + /* bit31: 0 means disable IOV and 1 means enable */ + if (reg & 1) + adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF; + + if (reg & 0x80000000) + adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV; + } if (reg == 0) { if (is_virtual_machine()) /* passthrough mode exclus sr-iov mode */ Patches currently in stable-queue which might be from alexander.deucher(a)amd.com are queue-4.15/drm-amdgpu-add-new-device-to-use-atpx-quirk.patch queue-4.15/drm-amdgpu-add-dpm-quirk-for-jet-pro-v2.patch queue-4.15/drm-amdgpu-only-check-mmbif_iov_func_identifier-on-tonga-fiji.patch queue-4.15/drm-amdgpu-disable-mmhub-power-gating-on-raven.patch queue-4.15/drm-amdgpu-fix-va-hole-handling-on-vega10-v3.patch queue-4.15/drm-amdgpu-avoid-leaking-pm-domain-on-driver-unbind-v2.patch queue-4.15/drm-amdgpu-add-atpx-quirk-handling-v2.patch

6 years, 10 months

1
0
0 0

Patch "drm/amdgpu: disable MMHUB power gating on raven" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm/amdgpu: disable MMHUB power gating on raven to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-amdgpu-disable-mmhub-power-gating-on-raven.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 400b6afbaa949914460e5fd1d769c5e26ef1f6b8 Mon Sep 17 00:00:00 2001 From: Huang Rui <ray.huang(a)amd.com> Date: Thu, 14 Dec 2017 13:47:16 +0800 Subject: drm/amdgpu: disable MMHUB power gating on raven From: Huang Rui <ray.huang(a)amd.com> commit 400b6afbaa949914460e5fd1d769c5e26ef1f6b8 upstream. MMHUB power gating still has issue, and doesn't work on raven at current. So disable it for the moment. Signed-off-by: Huang Rui <ray.huang(a)amd.com> Acked-by: Hawking Zhang <Hawking.Zhang(a)amd.com> Acked-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/amd/amdgpu/soc15.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -659,8 +659,8 @@ static int soc15_common_early_init(void AMD_CG_SUPPORT_MC_LS | AMD_CG_SUPPORT_SDMA_MGCG | AMD_CG_SUPPORT_SDMA_LS; - adev->pg_flags = AMD_PG_SUPPORT_SDMA | - AMD_PG_SUPPORT_MMHUB; + adev->pg_flags = AMD_PG_SUPPORT_SDMA; + adev->external_rev_id = 0x1; break; default: Patches currently in stable-queue which might be from ray.huang(a)amd.com are queue-4.15/drm-amdgpu-disable-mmhub-power-gating-on-raven.patch

6 years, 10 months

1
0
0 0

Patch "drm/amdgpu: Avoid leaking PM domain on driver unbind (v2)" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled drm/amdgpu: Avoid leaking PM domain on driver unbind (v2) to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: drm-amdgpu-avoid-leaking-pm-domain-on-driver-unbind-v2.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 458d876eb869d5a88b53074c6c271b8b9adc0f07 Mon Sep 17 00:00:00 2001 From: Alex Deucher <alexander.deucher(a)amd.com> Date: Mon, 22 Jan 2018 23:13:32 -0500 Subject: drm/amdgpu: Avoid leaking PM domain on driver unbind (v2) From: Alex Deucher <alexander.deucher(a)amd.com> commit 458d876eb869d5a88b53074c6c271b8b9adc0f07 upstream. We only support vga_switcheroo and runtime pm on PX/HG systems so forcing runpm to 1 doesn't do anything useful anyway. Only call vga_switcheroo_init_domain_pm_ops() for PX/HG so that the cleanup path is correct as well. This mirrors what radeon does as well. v2: rework the patch originally sent by Lukas (Alex) Acked-by: Lukas Wunner <lukas(a)wunner.de> Reported-by: Lukas Wunner <lukas(a)wunner.de> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Lukas Wunner <lukas(a)wunner.de> (v1) Cc: stable(a)vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 -- 1 file changed, 2 deletions(-) --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2228,8 +2228,6 @@ int amdgpu_device_init(struct amdgpu_dev * ignore it */ vga_client_register(adev->pdev, adev, NULL, amdgpu_vga_set_decode); - if (amdgpu_runtime_pm == 1) - runtime = true; if (amdgpu_device_is_px(ddev)) runtime = true; if (!pci_is_thunderbolt_attached(adev->pdev)) Patches currently in stable-queue which might be from alexander.deucher(a)amd.com are queue-4.15/drm-amdgpu-add-new-device-to-use-atpx-quirk.patch queue-4.15/drm-amdgpu-add-dpm-quirk-for-jet-pro-v2.patch queue-4.15/drm-amdgpu-only-check-mmbif_iov_func_identifier-on-tonga-fiji.patch queue-4.15/drm-amdgpu-disable-mmhub-power-gating-on-raven.patch queue-4.15/drm-amdgpu-fix-va-hole-handling-on-vega10-v3.patch queue-4.15/drm-amdgpu-avoid-leaking-pm-domain-on-driver-unbind-v2.patch queue-4.15/drm-amdgpu-add-atpx-quirk-handling-v2.patch

6 years, 10 months

1
0
0 0

Patch "Kbuild: always define endianess in kconfig.h" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled Kbuild: always define endianess in kconfig.h to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: kbuild-always-define-endianess-in-kconfig.h.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 101110f6271ce956a049250c907bc960030577f8 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann <arnd(a)arndb.de> Date: Wed, 21 Feb 2018 14:45:20 -0800 Subject: Kbuild: always define endianess in kconfig.h From: Arnd Bergmann <arnd(a)arndb.de> commit 101110f6271ce956a049250c907bc960030577f8 upstream. Build testing with LTO found a couple of files that get compiled differently depending on whether asm/byteorder.h gets included early enough or not. In particular, include/asm-generic/qrwlock_types.h is affected by this, but there are probably others as well. The symptom is a series of LTO link time warnings, including these: net/netlabel/netlabel_unlabeled.h:223: error: type of 'netlbl_unlhsh_add' does not match original declaration [-Werror=lto-type-mismatch] int netlbl_unlhsh_add(struct net *net, net/netlabel/netlabel_unlabeled.c:377: note: 'netlbl_unlhsh_add' was previously declared here include/net/ipv6.h:360: error: type of 'ipv6_renew_options_kern' does not match original declaration [-Werror=lto-type-mismatch] ipv6_renew_options_kern(struct sock *sk, net/ipv6/exthdrs.c:1162: note: 'ipv6_renew_options_kern' was previously declared here net/core/dev.c:761: note: 'dev_get_by_name_rcu' was previously declared here struct net_device *dev_get_by_name_rcu(struct net *net, const char *name) net/core/dev.c:761: note: code may be misoptimized unless -fno-strict-aliasing is used drivers/gpu/drm/i915/i915_drv.h:3377: error: type of 'i915_gem_object_set_to_wc_domain' does not match original declaration [-Werror=lto-type-mismatch] i915_gem_object_set_to_wc_domain(struct drm_i915_gem_object *obj, bool write); drivers/gpu/drm/i915/i915_gem.c:3639: note: 'i915_gem_object_set_to_wc_domain' was previously declared here include/linux/debugfs.h:92:9: error: type of 'debugfs_attr_read' does not match original declaration [-Werror=lto-type-mismatch] ssize_t debugfs_attr_read(struct file *file, char __user *buf, fs/debugfs/file.c:318: note: 'debugfs_attr_read' was previously declared here include/linux/rwlock_api_smp.h:30: error: type of '_raw_read_unlock' does not match original declaration [-Werror=lto-type-mismatch] void __lockfunc _raw_read_unlock(rwlock_t *lock) __releases(lock); kernel/locking/spinlock.c:246:26: note: '_raw_read_unlock' was previously declared here include/linux/fs.h:3308:5: error: type of 'simple_attr_open' does not match original declaration [-Werror=lto-type-mismatch] int simple_attr_open(struct inode *inode, struct file *file, fs/libfs.c:795: note: 'simple_attr_open' was previously declared here All of the above are caused by include/asm-generic/qrwlock_types.h failing to include asm/byteorder.h after commit e0d02285f16e ("locking/qrwlock: Use 'struct qrwlock' instead of 'struct __qrwlock'") in linux-4.15. Similar bugs may or may not exist in older kernels as well, but there is no easy way to test those with link-time optimizations, and kernels before 4.14 are harder to fix because they don't have Babu's patch series We had similar issues with CONFIG_ symbols in the past and ended up always including the configuration headers though linux/kconfig.h. This works around the issue through that same file, defining either __BIG_ENDIAN or __LITTLE_ENDIAN depending on CONFIG_CPU_BIG_ENDIAN, which is now always set on all architectures since commit 4c97a0c8fee3 ("arch: define CPU_BIG_ENDIAN for all fixed big endian archs"). Link: http://lkml.kernel.org/r/20180202154104.1522809-2-arnd@arndb.de Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> Cc: Babu Moger <babu.moger(a)amd.com> Cc: Andi Kleen <ak(a)linux.intel.com> Cc: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Cc: Masahiro Yamada <yamada.masahiro(a)socionext.com> Cc: Nicolas Pitre <nico(a)linaro.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Thomas Gleixner <tglx(a)linutronix.de> Cc: Will Deacon <will.deacon(a)arm.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- include/linux/kconfig.h | 6 ++++++ 1 file changed, 6 insertions(+) --- a/include/linux/kconfig.h +++ b/include/linux/kconfig.h @@ -3,6 +3,12 @@ #include <generated/autoconf.h> +#ifdef CONFIG_CPU_BIG_ENDIAN +#define __BIG_ENDIAN 4321 +#else +#define __LITTLE_ENDIAN 1234 +#endif + #define __ARG_PLACEHOLDER_1 0, #define __take_second_arg(__ignored, val, ...) val Patches currently in stable-queue which might be from arnd(a)arndb.de are queue-4.9/x86-oprofile-fix-bogus-gcc-8-warning-in-nmi_setup.patch queue-4.9/cfg80211-fix-cfg80211_beacon_dup.patch queue-4.9/kbuild-always-define-endianess-in-kconfig.h.patch

6 years, 10 months

3
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror February 2018