[git pull] amdkfd fixes for 4.16

: drm/amdkfd: Fix potential NULL pointer dereferences (2018-01-10 17:15:09 -0600) Gustavo A. R. Silva (1): drm/amdkfd: Fix potential NULL pointer dereferences Oded Gabbay (1): drm/amdkfd: add ull suffix to 64bit defines

Re: [PATCH v2] drm/amdkfd: Fix potential NULL pointer dereferences

ct kfd_dev > *dev) > mutex_lock(>mutex); > pdd = kfd_get_process_device_data(dev, p); > > + if (WARN_ON(!pdd)) { > + mutex_unlock(>mutex); > + continue; > + } > + >

[git pull] amdkfd next 4.16

Hi Dave, amdkfd stuff for 4.16: - Add CWSR (compute wave save restore) support for GFX8 (Carrizo) - Fix SDMA user-mode queues support for GFX7 (Kaveri) - Add SDMA user-mode queues support for GFX8 (Carrizo) - Allow HWS (hardware scheduling) to schedule multiple processes concurrently - Add

Re: [PATCH] drm: fix amdkfd use-after-free GP fault

It was sent to Dave Airle (drm maintainer) to be included in 4.15-rc2 or 4.15-rc3 (depends when Dave will send his drm fixes). Oded On Wed, Nov 29, 2017 at 12:41 AM, Randy Dunlap <rdun...@infradead.org> wrote: > On 11/13/2017 08:09 AM, Oded Gabbay wrote: >> On Sat, Nov 11, 2017 at

[git pull] amdkfd fixes 4.15

Hi Dave, This is amdkfd pull request for -rc2. It contains three small fixes to the CIK SDMA code, compilation error fix in kfd_ioctl.h and fix to accessing a pointer after it was released. Thanks, Oded The following changes since commit c209101fc1c91a318422733a3721ff6a9ff7899f: Merge tag

Re: [PATCH 1/1] drm/amdkfd: Do not ignore requested queue size during allocation

gt; > Regards, > Felix > > > On 2017-11-19 03:19 AM, Oded Gabbay wrote: >> On Thu, Nov 16, 2017 at 11:36 PM, Jan Vesely <jan.ves...@rutgers.edu> wrote: >>> Signed-off-by: Jan Vesely <jan.ves...@rutgers.edu> >>> --- >>> drivers/gpu/dr

Re: [PATCH 1/1] drm/amdkfd: Do not ignore requested queue size during allocation

On Thu, Nov 16, 2017 at 11:36 PM, Jan Vesely wrote: > Signed-off-by: Jan Vesely > --- > drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue_vi.c | 5 +++-- > 1 file changed, 3 insertions(+), 2 deletions(-) > > diff --git

Re: [git pull] amdkfd next 4.15 (fixes)

Hi Dave, Did you pick this pull request ? If not, please disregard it and I'll send you a new one for -rc1 with two additional fixes (total of 5 commits). Thanks, Oded On Thu, Nov 9, 2017 at 9:38 AM, Oded Gabbay <oded.gab...@gmail.com> wrote: > Hi Dave, > > Three small fixes to th

Re: [PATCH] uapi: fix linux/kfd_ioctl.h userspace compilation errors

On Mon, Nov 13, 2017 at 2:35 AM, Dmitry V. Levin wrote: > Consistently use types provided by via > to fix the following linux/kfd_ioctl.h userspace compilation errors: > > /usr/include/linux/kfd_ioctl.h:236:2: error: unknown type name 'uint64_t' > uint64_t va_addr; /* to

Re: [PATCH] drm: fix amdkfd use-after-free GP fault

50 4d 85 c0 74 2b 48 8b 86 88 00 00 00 48 85 c0 74 25 <48> > 8b 08 0f be 7b 01 48 c7 c2 96 0a aa 97 31 c0 83 ef 30 e8 7f > [ 947.650339] RIP: __dev_printk+0x29/0x90 RSP: a51e409c7e48 > [ 947.650388] ---[ end trace c41965e147ae98ae ]--- > > Signed-off-by: Randy Dunlap <rdun

[git pull] amdkfd next 4.15 (fixes)

Hi Dave, Three small fixes to the CIK SDMA code in amdkfd Thanks, Oded The following changes since commit d65d31388a23b14df9494135ad6c6549a59a3caa: Merge tag 'drm-misc-next-fixes-2017-11-07' of git://anongit.freedesktop.org/drm/drm-misc into drm-next (2017-11-08 05:22:49 +1000) are

Re: [gabbayo:amdkfd-next 5/8] drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c:540:3: error: dereferencing pointer to incomplete type 'struct cik_sdma_rlc_registers'

Thanks for the pointer, I did that (these patches will be sent for next merge window) On Mon, Nov 6, 2017 at 6:35 PM, Felix Kuehling wrote: > I messed up while rebasing patches and didn't test every intermediate > patch as I should have. The next patch in the series fixes

[git pull] amdkfd next 4.15 (part 2)

cleanups Jay Cornwall (1): drm/amdkfd: Disable CP/SDMA ring/doorbell in MQD Oded Gabbay (1): drm/amdkfd: increase limit of signal events to 4096 per process Sean Keely (2): drm/amdkfd: Short cut for kfd_wait_on_events without waiting drm/amdkfd: Fix scheduler race

[git pull] amdkfd next 4.15

Hi Dave, This is the amdkfd pull request for 4.15 merge window. The patches here are relevant only for Kaveri and Carrizo. Still no dGPU patches. The main goal is to continue alignment with the internal AMD development tree. The following is a summary of the changes: - Improvements and fixes

[git pull] drm-amdkfd-fixes 4.14

Hi Dave, This is amdkfd pull request for 4.14-rc3. It contains the following fixes: - correct checking of return value - send correct parameter to function (According to the parameter type) - avoid spamming of dmesg log - fix queue wrapping calculations Thanks, Oded The following changes since

Re: [git pull] amdkfd next 4.14

On Sun, Sep 17, 2017 at 9:37 PM, Alex Deucher <alexdeuc...@gmail.com> wrote: > On Sat, Sep 16, 2017 at 11:58 AM, Oded Gabbay <oded.gab...@gmail.com> wrote: >> On Mon, Sep 4, 2017 at 10:50 AM, Daniel Vetter <dan...@ffwll.ch> wrote: >>> Hi Alex, >>

Re: [PATCH] drm/amdkfd: check for null dev to avoid a null pointer dereference

On Fri, Sep 8, 2017 at 5:13 PM, Colin King wrote: > From: Colin Ian King > > The call to kfd_device_by_id can potentially return null, so check that > dev is null and return with -EINVAL to avoid a null pointer dereference. > > Detected by

Re: [git pull] amdkfd next 4.14

ast partially). Oded > On Sat, Sep 02, 2017 at 04:29:38PM +0300, Oded Gabbay wrote: >> Hi Dave, >> >> Three more amdkfd commits for the 4.14 merge window, nothing too major: >> >> - Removing redundant memset before memcpy >> - Passing valid mqd pointer

[git pull] amdkfd next 4.14

memcpy Mikko Rapeli (1): uapi linux/kfd_ioctl.h: only use __u32 and __u64 Oded Gabbay (1): drm/amdkfd: pass queue's mqd when destroying mqd drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c | 2 +- .../gpu/drm/amd/amdkfd/kfd_process_queue_manager.c | 1 - include/uapi/linux

Re: [PATCH] drm/amdkfd: remove memset before memcpy

On Tue, Aug 29, 2017 at 10:03 PM, Himanshu Jha wrote: > calling memcpy immediately after memset with the same region of memory > makes memset redundant. > > Signed-off-by: Himanshu Jha > --- >

Re: [PATCH v06 36/36] uapi linux/kfd_ioctl.h: use __u32 and __u64 instead of uint32_t and uint64_t

ixes user space compilation errors like: >> >> linux/kfd_ioctl.h:33:2: error: unknown type name ‘uint32_t’ >> uint32_t major_version; /* from KFD */ >> ^~~~ >> >> Signed-off-by: Mikko Rapeli <mikko.rap...@iki.fi> >> Cc: Yair Shachar <yair.shac

[git pull] amdkfd next 4.14

Hi Dave, This is the amdkfd pull request for 4.14 merge window. AMD has started cleaning the pipe and sending patches from their internal development to the upstream community. The plan as I understand it is to first get all the non-dGPU patches to upstream and then move to upstream dGPU

[git pull] drm-amdkfd-fixes

Hi Dave, This is a pull request for 4.13-rc2/3. The first patch addresses a bug that appears when doing over-subscription of queues, due to a miscalculation of the number of available compute queues. The other three patches remove dead code that was created from the bugfix. Thanks, Oded The

[git pull] drm-amdkfd-next

Hi Dave, This is amdkfd's pull request for kernel 4.13. It contains a fix to a function which returned an incorrect error value and another spelling fix Thanks, Oded The following changes since commit 047b8e21e3bfa9faa4ed9a0c337fe0c687710251: Merge tag 'drm-misc-next-2017-06-19_0' of

Re: [PATCH trivial] drm/amdkfd: Spelling s/apreture/aperture/

On Thu, Jun 1, 2017 at 1:28 PM, Geert Uytterhoeven wrote: > Signed-off-by: Geert Uytterhoeven > --- > drivers/gpu/drm/amd/amdkfd/kfd_process.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git

Re: [PATCH] drm/amdkfd: NULL dereference involving create_process()

On Wed, Jun 14, 2017 at 1:58 PM, Dan Carpenter wrote: > > We accidentally return ERR_PTR(0) which is NULL. The caller is not > expecting that and it leads to an Oops. > > Fixes: dd59239a9862 ("amdkfd: init aperture once per process") > Signed-off-by: Dan Carpenter

[PULL] drm-amdkfd-next

Hi Dave, Pull request for amdkfd. Just two patches for 4.11 merge window: - A small fix to return value of function in case of error - A spelling mistake Thanks, Oded The following changes since commit 282d0a35c8c42c3ac4dd86a7e408ed1b30db4897: Merge tag 'drm-misc-next-2017-01-09' of

[PATCH] amdkfd: fix spelling mistake in kfd_ioctl_dbg_unrgesiter

On Sat, Nov 12, 2016 at 7:33 PM, Colin King wrote: > From: Colin Ian King > > Trivial fix to spelling mistake, rename kfd_ioctl_dbg_unrgesiter > to kfd_ioctl_dbg_unregister > > Signed-off-by: Colin Ian King > --- > drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 6 +++--- > 1 file changed, 3

[PULL] drm-amdkfd-next

Hi Dave, Just wanted to make sure this pull request hasn't got lost in the mail. Thanks, Oded On Mon, Sep 19, 2016 at 10:49 PM, Oded Gabbay wrote: > Hi Dave, > > This is amdkfd's pull request for kernel 4.9. It contains a fix to a possible > infinite loop bug and a couple of

[PULL] drm-amdkfd-next

Hi Dave, This is amdkfd's pull request for kernel 4.9. It contains a fix to a possible infinite loop bug and a couple of other minor "cleaning" patches. Thanks, Oded The following changes since commit 9f8cf165c62913244479832f04c44cd77ffc9293: Merge tag 'topic/drm-misc-2016-09-19' of

[PULL] drm-amdkfd-fixes

Hi Dave, Just one patch for -rc that fixes a printk format error. Thanks, Oded The following changes since commit 36e9d08b58f44c3a02974c405ccaaa6ecfaf05b8: drm/cirrus: Fix NULL pointer dereference when registering the fbdev (2016-08-09 13:01:47 +1000) are available in the git

[PATCH] drm/amdkfd: print doorbell offset as a hex value

On Wed, Jul 13, 2016 at 10:36 AM, Colin King wrote: > From: Colin Ian King > > The doorbell offset is formatted with a 0x prefix to suggest it is > a hexadecimal value, when in fact %d is being used and this is confusing. > Use %X instead to match the proceeding 0x prefix. > > Signed-off-by:

[PULL] drm-amdkfd-next

to 7fd5e03ca6b41a591bd9fda083362b8a07cfb5f7: drm/amdkfd: destroy mutex if process creation fails (2016-07-03 08:05:45 +0300) Bhaktipriya Shridhar (1): drm/amdkfd: Remove create_workqueue() Oded Gabbay (1): drm/amdkfd: destroy mutex if process creation

[PATCH] drm/amdkfd: destroy mutex if process creation fails

Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 6482fee..771a18a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b

[PATCH 01/10] drm/amd-kfd: Clean up inline handling

On Tue, Jun 21, 2016 at 12:10 PM, Daniel Vetter wrote: > - inline functions need to be static inline, otherwise gcc can opt to > not inline and the linker gets unhappy. > - no forward decls for inline functions, just include the right headers. > > Cc: Oded Gabbay > Cc: Ben

[PATCH] drm/atomic-helpers: Clear up cleanup_done a bit

On Wed, Jun 15, 2016 at 1:08 PM, Daniel Vetter wrote: > It's not obvious at first sight that this is a fastpath, make that > clearer with a goto. Fallout from a discussion with Liviu on irc. > > Cc: Liviu.Dudau at arm.com > Acked-by: Liviu.Dudau at arm.com > Signed-off-by: Daniel Vetter > --- >

[PULL] drm-amdkfd-fixes

-amdkfd-fixes-2016-06-03 for you to fetch changes up to 0fbbbf8b599ff840ff1a3c0cc00dd67ba8a52c9c: drm/amdkfd: print once about mem_banks truncation (2016-06-03 08:50:40 +0300) Oded Gabbay (3): drm/amdkfd: unbind only existing

[RFT v3] drm: use late_initcall() for amdkfd and radeon

On Tue, May 31, 2016 at 8:15 PM, Luis R. Rodriguez wrote: > On Sun, May 29, 2016 at 05:49:17PM +0300, Oded Gabbay wrote: >> On Fri, May 27, 2016 at 4:18 AM, Luis R. Rodriguez >> wrote: >> > diff --git a/drivers/gpu/drm/radeon/radeon_drv.c >> > b/drivers/gpu/d

[PATCH v3] gpu: drm: amd: amdkfd: Remove create_workqueue()

On Sun, May 29, 2016 at 6:44 PM, Bhaktipriya Shridhar wrote: > alloc_workqueue replaces deprecated create_workqueue(). > > create_workqueue has been replaced with alloc_workqueue with max_active > as 0 since there is no need for throttling the number of active work items. > > WQ_MEM_RECLAIM has

[PATCH v2] gpu: drm: amd: amdkfd: Remove create_workqueue()

On Thu, May 26, 2016 at 10:37 PM, Bhaktipriya Shridhar wrote: > alloc_workqueue replaces deprecated create_workqueue(). > > create_workqueue has been replaced with alloc_workqueue with max_active > as 0 since there is no need for throttling the number of active work items. > > WQ_MEM_RECLAIM has

[RFT v3] drm: use late_initcall() for amdkfd and radeon

On Fri, May 27, 2016 at 4:18 AM, Luis R. Rodriguez wrote: > To get KFD support in radeon we need the following > initialization to happen in this order, their > respective driver file that has its init routine > listed next to it: > > 0. AMD IOMMUv1:arch/x86/kernel/pci-dma.c > 1. AMD IOMMUv2:

[PATCH] drm/amdkfd: print once about mem_banks truncation

This print can really spam the kernel log in case we are truncating mem_banks, so just print this info once. It should also not be classified as warning. Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

[PATCH v2 2/2] drm/amdkfd: destroy dbgmgr in notifier release

amdkfd need to destroy the debug manager in case amdkfd's notifier function is called before the unbind function, because in that case, the unbind function will exit without destroying debug manager. Signed-off-by: Oded Gabbay CC: Stable --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 10

[PATCH v2 1/2] drm/amdkfd: unbind only existing processes

hash_for_each_rcu, we simply exit the function as well. Signed-off-by: Oded Gabbay CC: Stable --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 60 +++- 1 file changed, 35 insertions(+), 25 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu

[PATCH 2/2] drm/amdkfd: destroy dbgmgr in notifier release

amdkfd need to destroy the debug manager in case amdkfd's notifier function is called before the unbind function, because in that case, the unbind function will exit without destroying debug manager. Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 10 -- 1 file

[PATCH 1/2] drm/amdkfd: unbind only existing processes

removed the process from the process table. Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 8 +++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index ac00579..248deb7

[PULL] drm-amdkfd-next

Hi Dave, Here are a few amdkfd patches for 4.7, all of them fixes according to the Coccinelle tool. Thanks, Oded The following changes since commit b89359bdf0f1e95a4c5f92300594ba9dde323fc4: Merge branch 'for-next' of http://git.agner.ch/git/linux-drm-fsl-dcu into drm-next (2016-04-29

Misc DAL & amdkfd cleanups on top of drm-next-4.7-wip-dal

On Sat, Apr 30, 2016 at 5:06 PM, Edward O'Callaghan wrote: > The following series applies on top of the 'drm-next-4.7-wip-dal' > branch. One patch has already been reviewed, however I have included > it again here for completeness. The is the first of, I hope, further > work from me in helping

[PATCH] hsakmt: allow building with gcc 4.x v2

gt; +fi > + > AC_PROG_INSTALL > AC_PROG_LIBTOOL > AC_PROG_MAKE_SET > -- > 1.9.1 > > ___ > dri-devel mailing list > dri-devel at lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel Thanks John, Reviewed-by: Oded Gabbay Can you push it to the repo ? If you get permission error, we need to set-up permissions. Oded

[PULL] drm-amdkfd-next-fixes

Hi Dave, Just a small fix for uninitialized variable. Hoping you could squeeze it into the merge window's pull request, but -rc2 is perfectly fine as well. Thanks, Oded The following changes since commit 211afd577a186e18d3eece543c6767420d6f6737: Merge tag 'drm-vc4-next-2016-03-14' of

[RFC 1/7] drm/amdkfd: avoid fragile and inefficient snprintf use

On Tue, Mar 8, 2016 at 10:40 PM, Rasmus Villemoes wrote: > Passing overlapping source and destination buffers to snprintf > formally has undefined behaviour and is rather fragile. While the > rather special case of passing the output buffer as the argument > corresponding to a leading "%s" in the

[patch] drm/amdkfd: uninitialized variable in dbgdev_wave_control_set_registers()

On Fri, Mar 11, 2016 at 9:51 AM, Dan Carpenter wrote: > At the end of the function we expect "status" to be zero, but it's > either -EINVAL or unitialized. > > Fixes: 788bf83db301 ('drm/amdkfd: Add wave control operation to debugger') > Signed-off-by: Dan Carpenter > > diff --git

[PULL] drm-amdkfd-next

: Return -EPROBE_DEFER when amdkfd not loaded (2016-02-27 22:52:40 +0200) Oded Gabbay (3): drm/amdkfd: Track when module's init is complete drm/radeon: Return -EPROBE_DEFER when amdkfd not loaded drm/amdgpu: Return

[PATCH 0/3] Deferr load of radeon/amdgpu until amdkfd is loaded

On Tue, Feb 23, 2016 at 5:10 AM, Xinliang Liu wrote: > On 15 February 2016 at 19:04, Oded Gabbay wrote: >> On Sun, Feb 14, 2016 at 2:58 PM, Daniel Vetter wrote: >>> On Sun, Feb 14, 2016 at 11:16:52AM +0200, Oded Gabbay wrote: >>>> Following Daniel's request, I spe

[PATCH 0/3] Deferr load of radeon/amdgpu until amdkfd is loaded

On Sun, Feb 14, 2016 at 2:58 PM, Daniel Vetter wrote: > On Sun, Feb 14, 2016 at 11:16:52AM +0200, Oded Gabbay wrote: >> Following Daniel's request, I spent some time removing the hard requirement >> that radeon and amdgpu will always appear _after_ amdkfd in the drm Makefile. >

[PATCH 3/3] drm/amdgpu: Return -EPROBE_DEFER when amdkfd not loaded

loading by propagating -EPROBE_DEFER to the kernel's drivers loading infrastructure. That will put amdgpu into the pending drivers list (see description in dd.c). Once amdkfd is loaded, a call to kgd2kfd_init() will return successfully and amdgpu will be able to load. Signed-off-by: Oded Gabbay

[PATCH 2/3] drm/radeon: Return -EPROBE_DEFER when amdkfd not loaded

loading by propagating -EPROBE_DEFER to the kernel's drivers loading infrastructure. That will put radeon into the pending drivers list (see description in dd.c). Once amdkfd is loaded, a call to kgd2kfd_init() will return successfully and radeon will be able to load. Signed-off-by: Oded Gabbay

[PATCH 1/3] drm/amdkfd: Track when module's init is complete

nal radeon/amdgpu they need to defer their loading until amdkfd is loaded. Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 4 ++-- drivers/gpu/drm/amd/amdkfd/kfd_module.c | 15 --- drivers/gpu/drm/amd/include/kgd_kfd_interface.h | 2 +- driv

[PATCH 0/3] Deferr load of radeon/amdgpu until amdkfd is loaded

. amdgpu (CIK enabled) + amdkfd inside the kernel image (radeon not compiled) 4. amdgpu (CIK enabled) inside the kernel image (radeon + amdkfd not compiled) 5. radeon + amdgpu (CIK disabled) as kernel modules (amdkfd not compiled) Thanks, Oded Oded Gabbay (3): drm/amdkfd: Track when

[PATCH 27/29] drm/amd/dal: Correctly interpret rotation as bit set

On Thu, Feb 11, 2016 at 7:20 PM, Harry Wentland wrote: > Signed-off-by: Harry Wentland > Reviewed-by: Mykola Lysenko > --- > drivers/gpu/drm/amd/dal/amdgpu_dm/amdgpu_dm_types.c | 8 > 1 file changed, 4 insertions(+), 4 deletions(-) > > diff --git

[RESEND PATCH v1] drm: Clean up drm Makefile

On Tue, Feb 9, 2016 at 11:29 AM, Daniel Vetter wrote: > On Sat, Jan 30, 2016 at 01:11:01PM +0200, Oded Gabbay wrote: >> On Sat, Jan 30, 2016 at 11:56 AM, Xinliang Liu >> wrote: >> > This patch cleans up the Makefile of drm root directory. >> > Make core and

[PATCH 1/3] drm: add helper to check for wc memory support

On Wed, Feb 3, 2016 at 3:53 PM, Alex Deucher wrote: > On Wed, Feb 3, 2016 at 3:17 AM, Oded Gabbay wrote: >> On Tue, Feb 2, 2016 at 5:07 PM, Alex Deucher >> wrote: >>> On Tue, Feb 2, 2016 at 7:01 AM, Oded Gabbay >>> wrote: >>>> On Mon, Feb

[PATCH 1/3] drm: add helper to check for wc memory support

On Tue, Feb 2, 2016 at 5:07 PM, Alex Deucher wrote: > On Tue, Feb 2, 2016 at 7:01 AM, Oded Gabbay wrote: >> On Mon, Feb 1, 2016 at 7:01 PM, Alex Deucher >> wrote: >>> >>> On Sat, Jan 30, 2016 at 4:57 AM, Christian König >>> wrote: >&g

[PATCH 1/3] drm: add helper to check for wc memory support

On Mon, Feb 1, 2016 at 7:01 PM, Alex Deucher wrote: > > On Sat, Jan 30, 2016 at 4:57 AM, Christian König > wrote: > > Am 30.01.2016 um 06:59 schrieb Oded Gabbay: > >> > >> From: Dave Airlie > >> > >> Signed-off-by: Dave Airlie > >> S

[RESEND PATCH v1] drm: Clean up drm Makefile

On Sat, Jan 30, 2016 at 11:56 AM, Xinliang Liu wrote: > This patch cleans up the Makefile of drm root directory. > Make core and device drivers configuration list sorted Alphabetically. > > Signed-off-by: Xinliang Liu > Reviewed-by: Xinwei Kong > Reviewed-by: Yifan Liu > --- >

[PATCH 2/3] drm/radeon: mask out WC from BO on unsupported arches

On Sat, Jan 30, 2016 at 10:11 AM, Michel Dänzer wrote: > On 30.01.2016 14:59, Oded Gabbay wrote: >> Signed-off-by: Oded Gabbay >> --- >> drivers/gpu/drm/radeon/radeon_object.c | 8 >> 1 file changed, 8 insertions(+) >> >> diff --git a/drive

[PATCH 3/3] drm/amdgpu: mask out WC from BO on unsupported arches

Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 9 + 1 file changed, 9 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index c3ce103..7b963e6 100644 --- a/drivers/gpu/drm/amd/amdgpu

[PATCH 2/3] drm/radeon: mask out WC from BO on unsupported arches

Signed-off-by: Oded Gabbay --- drivers/gpu/drm/radeon/radeon_object.c | 8 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index 84d4563..255faf6 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b

[PATCH 1/3] drm: add helper to check for wc memory support

From: Dave Airlie <airl...@redhat.com> Signed-off-by: Dave Airlie Signed-off-by: Oded Gabbay --- include/drm/drm_cache.h | 9 + 1 file changed, 9 insertions(+) diff --git a/include/drm/drm_cache.h b/include/drm/drm_cache.h index 7bfb063..461a055 100644 --- a/include/drm/drm_c

ongoing writecombine on ppc

On Fri, Jan 22, 2016 at 4:32 AM, Michel Dänzer wrote: > On 22.01.2016 02:10, Alex Deucher wrote: >> On Thu, Jan 21, 2016 at 10:39 AM, Oded Gabbay >> wrote: >>> +Alex >> >> No objections from me. Care to respin with amdgpu support and signed >>

[PULL] drm-amdkfd-fixes

Hi Dave, Just two small static analyzer code fixes for 4.5 Thanks, Oded The following changes since commit 92e963f50fc74041b5e9e744c330dca48e04f08d: Linux 4.5-rc1 (2016-01-24 13:06:47 -0800) are available in the git repository at: git://people.freedesktop.org/~gabbayo/linux

[PATCH] gpu: drm: amd: amdkfd: Remove unnecessary cast in kfree

On Mon, Jan 25, 2016 at 7:33 PM, Amitoj Kaur Chawla wrote: > Remove an unnecassary cast in the argument to kfree. > > Found using Coccinelle. The semantic patch used to find this is as follows: > > // > @@ > type T; > expression *f; > @@ > > - kfree((T *)(f)); > + kfree(f); > // > >

[PATCH] drm/amdgpu: fix non-ANSI declaration of amdgpu_amdkfd_gfx_*_get_functions()

On Fri, Jan 22, 2016 at 7:35 PM, Colin King wrote: > From: Colin Ian King > > amdgpu_amdkfd_gfx_7_get_functions and amdgpu_amdkfd_gfx_8_0_get_functions > have no parameters, so use the normal void parameter convention to make > them match their prototypes in the header file >

ongoing writecombine on ppc

+Alex On Thu, Jan 21, 2016 at 5:24 PM, Oded Gabbay wrote: > On Fri, Oct 2, 2015 at 7:56 AM, Benjamin Herrenschmidt > wrote: >> On Fri, 2015-10-02 at 14:53 +1000, Dave Airlie wrote: >>> On 2 October 2015 at 14:45, Benjamin Herrenschmidt >>> wrote: >>> &g

ongoing writecombine on ppc

hink that's a given :) > >> So we have to keep the AGP trapdoor in place, and at the moment only >> the drivers know if they are AGP, hence why I had to add this in the >> driver instead of in the drm core. > Cheers, > Ben. > > > Bumping this and adding my r-b: Reviewed-by: Oded Gabbay

[PATCH 2/3] radeon: Fix VCE ring test for Big-Endian systems

On Mon, Dec 7, 2015 at 9:51 AM, Michel Dänzer wrote: > On 05.12.2015 06:09, Oded Gabbay wrote: >> This patch fixes the VCE ring test when running on Big-Endian machines. >> Every write to the ring needs to be translated to little-endian. >> >> Signed-off-by:

[PATCH 1/3] radeon/cik: Fix GFX IB test on Big-Endian

OK, I'll take a look Oded > Yeah, CIK support in amdgpu was only for debugging and bringup. But since > it's only shared code you touch when it works with CIK it should work with VI > as well and that's rather interesting to us. Regards, Christian.

[PATCH 1/3] radeon/cik: Fix GFX IB test on Big-Endian

On Sun, Dec 6, 2015 at 8:45 PM, Christian König wrote: > On 06.12.2015 08:29, Oded Gabbay wrote: >> >> On Sat, Dec 5, 2015 at 12:23 PM, Christian König >> wrote: >>> >>> Patch #1 & #2 are Reviewed-by: Christian König >> amd.com> >>

[PATCH 1/3] radeon/cik: Fix GFX IB test on Big-Endian

atively a helper function adding DW to an IB with swapping could do it > as well. > Don't you think its an overkill ? It's just a few places in the code. > Regards, > Christian. > > > On 04.12.2015 22:09, Oded Gabbay wrote: >> >> This patch makes the IB test on th

[PATCH 3/3] radeon: Fix VCE IB test on Big-Endian systems

This patch makes the VCE IB test pass on Big-Endian systems. It converts to little-endian the contents of the VCE message. Signed-off-by: Oded Gabbay Cc: stable at vger.kernel.org --- drivers/gpu/drm/radeon/radeon_vce.c | 68 ++--- 1 file changed, 34 insertions

[PATCH 2/3] radeon: Fix VCE ring test for Big-Endian systems

This patch fixes the VCE ring test when running on Big-Endian machines. Every write to the ring needs to be translated to little-endian. Signed-off-by: Oded Gabbay Cc: stable at vger.kernel.org --- drivers/gpu/drm/radeon/radeon_vce.c | 32 1 file changed, 16

[PATCH 1/3] radeon/cik: Fix GFX IB test on Big-Endian

This patch makes the IB test on the GFX ring pass for CI-based cards installed in Big-Endian machines. Signed-off-by: Oded Gabbay Cc: stable at vger.kernel.org --- drivers/gpu/drm/radeon/cik.c | 6 +- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/gpu/drm/radeon/cik.c

[ANNOUNCE] hsakmt release 1.0.0 now available

.tar.gz SHA1: fd3561dec3d88956d756a561c8d907694a70a03d hsakmt-1.0.0.tar.bz2 GPG signature: http://xorg.freedesktop.org/archive/individual/lib/hsakmt-1.0.0.tar.gz.sha1.asc (signed by Oded Gabbay ) Git: git://anongit.freedesktop.org/amd/hsakmt ta

[hsakmt] hsakmt organization and formal releases

no problem removing these, but I feared it might be too radical for you... If you want to see how it will look like without the include folder, check out http://cgit.freedesktop.org/amd/hsakmt/tree/?id=e8a6286922e9add2d34ca29fd2b3c6b3ace35f69 Thanks, Oded > Everything else looks goo

[hsakmt] hsakmt organization and formal releases

On Fri, Oct 9, 2015 at 7:46 PM, Alex Deucher wrote: > On Tue, Oct 6, 2015 at 8:00 AM, Oded Gabbay wrote: >> Hi, >> >> I had some time during the recent local holidays, so I thought I >> improve the hsakmt library in terms of releases: >> >> 1. I added

[hsakmt] hsakmt organization and formal releases

Hi, I had some time during the recent local holidays, so I thought I improve the hsakmt library in terms of releases: 1. I added automake/autoconf files to standardize the package to be created using configure/make/make install. 2. I created a very simple scheme of numbering so we could track

[PULL] drm-amdkfd-next-fixes

Hi Dave, Just one small fix before 4.3 merge window: - Use linux/mman.h instead of uapi's mman-common.h inside the driver. Thanks, Oded The following changes since commit 92cffd56b21c825579f3b37bc7803e4c37073076: drm/nouveau/dispnv04: fix build on powerpc (2015-08-28 20:33:58 +1000)

[PULL] drm-amdkfd-next-fixes

On Wed, Aug 5, 2015 at 10:48 AM, Daniel Vetter wrote: > Forwarding -fixes pull to Linus since Dave is on vacation for 2 weeks. > -Daniel > Daniel, This is for 4.3 merge window... This is not for 4.2-rcX I don't think we need to forward this to Linus. Oded

[PULL] drm-amdkfd-next-fixes

initializing sdma vm (2015-07-30 09:26:15 +0300) Oded Gabbay (2): drm/amdgpu: fix bug when amdkfd destroys hqd drm/amdkfd: fix bug when initializing sdma vm drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 2

[PATCH v2 1/2] drm/amdgpu: fix bug when amdkfd destroys hqd

The wrong define was used to check if the hqd is still active v2: Don't use SHIFT as the MASK is already shifted Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu

[PATCH 1/2] drm/amdgpu: fix bug when amdkfd destroys hqd

On Wed, Jul 29, 2015 at 11:49 AM, Ilia Mirkin wrote: > On Wed, Jul 29, 2015 at 3:44 AM, Oded Gabbay wrote: >> The wrong define was used to check if the hqd is still active >> >> Signed-off-by: Oded Gabbay >> --- >> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v

[PATCH 2/2] drm/amdkfd: fix bug when initializing sdma vm

A logical AND operation was used during mask and shift, instead of a bitwise AND operation. This patch fixes this bug by changing the operation to bitwise AND. Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c | 2 +- drivers/gpu/drm/amd/amdkfd

[PATCH 1/2] drm/amdgpu: fix bug when amdkfd destroys hqd

The wrong define was used to check if the hqd is still active Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu

drm/amdkfd: Add support for VI in DQM

On Tue, Jul 28, 2015 at 7:00 PM, Dan Carpenter wrote: > Hello Ben Goz, > > The patch 914bea6329b2: "drm/amdkfd: Add support for VI in DQM" from > Jan 12, 2015, leads to the following static checker warning: > > drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c:158 >

drm/amdgpu: add amdgpu <--> amdkfd gfx7 interface

On Tue, Jul 28, 2015 at 6:56 PM, Dan Carpenter wrote: > Hello Oded Gabbay, > > The patch 32c22e994f44: "drm/amdgpu: add amdgpu <--> amdkfd gfx7 > interface" from Jun 12, 2015, leads to the following static checker > warning: > > drivers/gpu/drm/

[PULL] drm-amdkfd-next

packet headers for VI drm/amdkfd: add support for VI in MQD manager drm/amdkfd: Add support for VI in DQM drm/amdkfd: fix runlist length calculation drm/amdkfd: Implement create_map_queues() for Carrizo drm/amdkfd: Set correct doorbell packet type for Carrizo Oded Gab

[PULL] drm-amdkfd-fixes

Hi Dave, A single fix so far for 4.2: - checking a pointer is not null before using it Thanks, Oded The following changes since commit d770e558e21961ad6cfdf0ff7df0eb5d7d4f0754: Linux 4.2-rc1 (2015-07-05 11:01:52 -0700) are available in the git repository at:

[PATCH v2] drm/amdkfd: validate pdd where it acquired first

On Thu, Jul 9, 2015 at 12:11 PM, Maninder Singh wrote: > Currently pdd is validate after dereferencing it, which is > not correct, Thus validate pdd before its first use. > > Signed-off-by: Maninder Singh > --- > v1: remove validation of pdd after its usage > v2: do validation at first place

[Nouveau] CUDA fixed VA allocations and sparse mappings

On Tue, Jul 7, 2015 at 8:27 PM, Jerome Glisse wrote: > On Tue, Jul 07, 2015 at 11:29:38AM -0400, Ilia Mirkin wrote: >> On Mon, Jul 6, 2015 at 8:42 PM, Andrew Chew wrote: >> > Hello, >> > >> > I am currently looking into ways to support fixed virtual address >> > allocations >> > and sparse

[RFC][PATCH 1/1] drm/amdkfd: Remove redundant pdd validation

On Mon, Jun 29, 2015 at 7:33 AM, Maninder Singh wrote: > > pdd is already dereferenced before this check. > So it is redundtant to validate pdd here. > > Signed-off-by: Maninder Singh > --- > drivers/gpu/drm/amd/amdkfd/kfd_process.c |3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-)

<    1   2   3   4   5   6   7   8   9   10   >