LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2021-08-02 15:10 Mark Brown
  2021-08-02 15:48 ` Mark Brown
  0 siblings, 1 reply; 29+ messages in thread
From: Mark Brown @ 2021-08-02 15:10 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linux Kernel Mailing List, Linux Next Mailing List,
	Eric Huang

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdkfd/kfd_chardev.c

between commits:

  f5cc09acece4 ("Revert "drm/amdkfd: Add memory sync before TLB flush on unmap"")
  c37387c354c8 ("Revert "drm/amdkfd: Make TLB flush conditional on mapping"")
  5adcd7458a78 ("Revert "drm/amdkfd: Add heavy-weight TLB flush after unmapping"")

from Linus' tree and commits:

  b928ecfbe369 ("Revert "Revert "drm/amdkfd: Add memory sync before TLB flush on unmap""")
  b0228fa4868b ("drm/amdkfd: Only apply heavy-weight TLB flush on Aldebaran")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

diff --cc drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index e48acdd03c1a,86afd37b098d..000000000000
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@@ -1565,10 -1568,29 +1565,33 @@@ static int kfd_ioctl_unmap_memory_from_
  		}
  		args->n_success = i+1;
  	}
++
+ 	mutex_unlock(&p->mutex);
+ 
+ 	if (dev->device_info->asic_family == CHIP_ALDEBARAN) {
+ 		err = amdgpu_amdkfd_gpuvm_sync_memory(dev->kgd,
+ 				(struct kgd_mem *) mem, true);
+ 		if (err) {
+ 			pr_debug("Sync memory failed, wait interrupted by user signal\n");
+ 			goto sync_memory_failed;
+ 		}
+ 
+ 		/* Flush TLBs after waiting for the page table updates to complete */
+ 		for (i = 0; i < args->n_devices; i++) {
+ 			peer = kfd_device_by_id(devices_arr[i]);
+ 			if (WARN_ON_ONCE(!peer))
+ 				continue;
+ 			peer_pdd = kfd_get_process_device_data(peer, p);
+ 			if (WARN_ON_ONCE(!peer_pdd))
+ 				continue;
+ 			kfd_flush_tlb(peer_pdd, TLB_FLUSH_HEAVYWEIGHT);
+ 		}
+ 	}
++
  	kfree(devices_arr);
  
 +	mutex_unlock(&p->mutex);
 +
  	return 0;
  
  bind_process_to_device_failed:

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2021-08-02 15:10 linux-next: manual merge of the amdgpu tree with Linus' tree Mark Brown
@ 2021-08-02 15:48 ` Mark Brown
  0 siblings, 0 replies; 29+ messages in thread
From: Mark Brown @ 2021-08-02 15:48 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linux Kernel Mailing List, Linux Next Mailing List,
	Eric Huang

[-- Attachment #1: Type: text/plain, Size: 423 bytes --]

On Mon, Aug 02, 2021 at 04:10:53PM +0100, Mark Brown wrote:
> + 	if (dev->device_info->asic_family == CHIP_ALDEBARAN) {
> + 		err = amdgpu_amdkfd_gpuvm_sync_memory(dev->kgd,
> + 				(struct kgd_mem *) mem, true);
> + 		if (err) {
> + 			pr_debug("Sync memory failed, wait interrupted by user signal\n");
> + 			goto sync_memory_failed;

Ugh, no.  This is broken because that label is gone.  I'll stack some
more fixups on.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2022-01-26 23:32 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2022-01-26 23:32 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linus Torvalds, Linux Kernel Mailing List,
	Linux Next Mailing List, Wenjing Liu

[-- Attachment #1: Type: text/plain, Size: 1632 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c

between commit:

  75b950ef6166 ("Revert "drm/amd/display: Fix for otg synchronization logic"")

from Linus' tree and commit:

  d9eb8fea6862 ("drm/amd/display: Drop DCN for DP2.x logic")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 78192ecba102,72dd41e7a7d6..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@@ -1566,7 -1527,10 +1527,6 @@@ static enum dc_status apply_single_cont
  				&pipe_ctx->stream->audio_info);
  	}
  
- #if defined(CONFIG_DRM_AMD_DC_DCN)
 -	/* make sure no pipes syncd to the pipe being enabled */
 -	if (!pipe_ctx->stream->apply_seamless_boot_optimization && dc->config.use_pipe_ctx_sync_logic)
 -		check_syncd_pipes_for_disabled_master_pipe(dc, context, pipe_ctx->pipe_idx);
 -
  	/* DCN3.1 FPGA Workaround
  	 * Need to enable HPO DP Stream Encoder before setting OTG master enable.
  	 * To do so, move calling function enable_stream_timing to only be done AFTER calling

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2022-01-26 23:29 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2022-01-26 23:29 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linux Kernel Mailing List, Linux Next Mailing List

[-- Attachment #1: Type: text/plain, Size: 2113 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c

between commit:

  b95dc06af3e6 ("drm/amdgpu: disable runpm if we are the primary adapter")

from Linus' tree and commit:

  243c719e872a ("drm/amdgpu: handle BACO synchronization with secondary funcs")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b21bcdc97460,f3cada3f743b..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@@ -1893,26 -1947,28 +1948,48 @@@ MODULE_DEVICE_TABLE(pci, pciidlist)
  
  static const struct drm_driver amdgpu_kms_driver;
  
 +static bool amdgpu_is_fw_framebuffer(resource_size_t base,
 +				     resource_size_t size)
 +{
 +	bool found = false;
 +#if IS_REACHABLE(CONFIG_FB)
 +	struct apertures_struct *a;
 +
 +	a = alloc_apertures(1);
 +	if (!a)
 +		return false;
 +
 +	a->ranges[0].base = base;
 +	a->ranges[0].size = size;
 +
 +	found = is_firmware_framebuffer(a);
 +	kfree(a);
 +#endif
 +	return found;
 +}
 +
+ static void amdgpu_get_secondary_funcs(struct amdgpu_device *adev)
+ {
+ 	struct pci_dev *p = NULL;
+ 	int i;
+ 
+ 	/* 0 - GPU
+ 	 * 1 - audio
+ 	 * 2 - USB
+ 	 * 3 - UCSI
+ 	 */
+ 	for (i = 1; i < 4; i++) {
+ 		p = pci_get_domain_bus_and_slot(pci_domain_nr(adev->pdev->bus),
+ 						adev->pdev->bus->number, i);
+ 		if (p) {
+ 			pm_runtime_get_sync(&p->dev);
+ 			pm_runtime_mark_last_busy(&p->dev);
+ 			pm_runtime_put_autosuspend(&p->dev);
+ 			pci_dev_put(p);
+ 		}
+ 	}
+ }
+ 
  static int amdgpu_pci_probe(struct pci_dev *pdev,
  			    const struct pci_device_id *ent)
  {

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2022-01-11 23:09 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2022-01-11 23:09 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linus Torvalds, Linux Kernel Mailing List,
	Linux Next Mailing List, Wenjing Liu

[-- Attachment #1: Type: text/plain, Size: 4678 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got conflicts in:

  drivers/gpu/drm/amd/display/dc/core/dc_resource.c
  drivers/gpu/drm/amd/display/dc/inc/resource.h

between commit:

  75b950ef6166 ("Revert "drm/amd/display: Fix for otg synchronization logic"")

from Linus' tree and commit:

  580013b2cef8 ("drm/amd/display: unhard code link to phy idx mapping in dc link and clean up")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index de5c7d1e0267,b3912ff9dc91..000000000000
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@@ -3216,3 -3216,90 +3216,36 @@@ struct hpo_dp_link_encoder *resource_ge
  	return hpo_dp_link_enc;
  }
  #endif
+ 
 -void reset_syncd_pipes_from_disabled_pipes(struct dc *dc,
 -		struct dc_state *context)
 -{
 -	int i, j;
 -	struct pipe_ctx *pipe_ctx_old, *pipe_ctx, *pipe_ctx_syncd;
 -
 -	/* If pipe backend is reset, need to reset pipe syncd status */
 -	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 -		pipe_ctx_old =	&dc->current_state->res_ctx.pipe_ctx[i];
 -		pipe_ctx = &context->res_ctx.pipe_ctx[i];
 -
 -		if (!pipe_ctx_old->stream)
 -			continue;
 -
 -		if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe)
 -			continue;
 -
 -		if (!pipe_ctx->stream ||
 -				pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) {
 -
 -			/* Reset all the syncd pipes from the disabled pipe */
 -			for (j = 0; j < dc->res_pool->pipe_count; j++) {
 -				pipe_ctx_syncd = &context->res_ctx.pipe_ctx[j];
 -				if ((GET_PIPE_SYNCD_FROM_PIPE(pipe_ctx_syncd) == pipe_ctx_old->pipe_idx) ||
 -					!IS_PIPE_SYNCD_VALID(pipe_ctx_syncd))
 -					SET_PIPE_SYNCD_TO_PIPE(pipe_ctx_syncd, j);
 -			}
 -		}
 -	}
 -}
 -
 -void check_syncd_pipes_for_disabled_master_pipe(struct dc *dc,
 -	struct dc_state *context,
 -	uint8_t disabled_master_pipe_idx)
 -{
 -	int i;
 -	struct pipe_ctx *pipe_ctx, *pipe_ctx_check;
 -
 -	pipe_ctx = &context->res_ctx.pipe_ctx[disabled_master_pipe_idx];
 -	if ((GET_PIPE_SYNCD_FROM_PIPE(pipe_ctx) != disabled_master_pipe_idx) ||
 -		!IS_PIPE_SYNCD_VALID(pipe_ctx))
 -		SET_PIPE_SYNCD_TO_PIPE(pipe_ctx, disabled_master_pipe_idx);
 -
 -	/* for the pipe disabled, check if any slave pipe exists and assert */
 -	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 -		pipe_ctx_check = &context->res_ctx.pipe_ctx[i];
 -
 -		if ((GET_PIPE_SYNCD_FROM_PIPE(pipe_ctx_check) == disabled_master_pipe_idx) &&
 -			IS_PIPE_SYNCD_VALID(pipe_ctx_check) && (i != disabled_master_pipe_idx))
 -			DC_ERR("DC: Failure: pipe_idx[%d] syncd with disabled master pipe_idx[%d]\n",
 -				i, disabled_master_pipe_idx);
 -	}
 -}
 -
+ uint8_t resource_transmitter_to_phy_idx(const struct dc *dc, enum transmitter transmitter)
+ {
+ 	/* TODO - get transmitter to phy idx mapping from DMUB */
+ 	uint8_t phy_idx = transmitter - TRANSMITTER_UNIPHY_A;
+ 
+ #if defined(CONFIG_DRM_AMD_DC_DCN)
+ 	if (dc->ctx->dce_version == DCN_VERSION_3_1 &&
+ 			dc->ctx->asic_id.hw_internal_rev == YELLOW_CARP_B0) {
+ 		switch (transmitter) {
+ 		case TRANSMITTER_UNIPHY_A:
+ 			phy_idx = 0;
+ 			break;
+ 		case TRANSMITTER_UNIPHY_B:
+ 			phy_idx = 1;
+ 			break;
+ 		case TRANSMITTER_UNIPHY_C:
+ 			phy_idx = 5;
+ 			break;
+ 		case TRANSMITTER_UNIPHY_D:
+ 			phy_idx = 6;
+ 			break;
+ 		case TRANSMITTER_UNIPHY_E:
+ 			phy_idx = 4;
+ 			break;
+ 		default:
+ 			phy_idx = 0;
+ 			break;
+ 		}
+ 	}
+ #endif
+ 	return phy_idx;
+ }
diff --cc drivers/gpu/drm/amd/display/dc/inc/resource.h
index e589cbe67307,028180f58f71..000000000000
--- a/drivers/gpu/drm/amd/display/dc/inc/resource.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h
@@@ -208,4 -212,12 +208,6 @@@ struct hpo_dp_link_encoder *resource_ge
  		const struct dc_link *link);
  #endif
  
 -void reset_syncd_pipes_from_disabled_pipes(struct dc *dc,
 -	struct dc_state *context);
 -
 -void check_syncd_pipes_for_disabled_master_pipe(struct dc *dc,
 -	struct dc_state *context,
 -	uint8_t disabled_master_pipe_idx);
+ uint8_t resource_transmitter_to_phy_idx(const struct dc *dc, enum transmitter transmitter);
+ 
  #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2021-09-02  1:22 Stephen Rothwell
@ 2021-09-10  1:16 ` Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2021-09-10  1:16 UTC (permalink / raw)
  To: Dave Airlie
  Cc: Alex Deucher, Alex Deucher, Linux Kernel Mailing List,
	Linux Next Mailing List, Luben Tuikov, DRI

[-- Attachment #1: Type: text/plain, Size: 987 bytes --]

Hi all,

On Thu, 2 Sep 2021 11:22:08 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Today's linux-next merge of the amdgpu tree got a conflict in:
> 
>   drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
> 
> between commit:
> 
>   477f70cd2a67 ("Merge tag 'drm-next-2021-08-31-1' of git://anongit.freedesktop.org/drm/drm")
> 
> from Linus' tree and commit:
> 
>   cc947bf91bad ("drm/amdgpu: Process any VBIOS RAS EEPROM address")
> 
> from the amdgpu tree.
> 
> I fixed it up (I used the latter version) and can carry the fix as
> necessary. This is now fixed as far as linux-next is concerned, but any
> non trivial conflicts should be mentioned to your upstream maintainer
> when your tree is submitted for merging.  You may also want to consider
> cooperating with the maintainer of the conflicting tree to minimise any
> particularly complex conflicts.

This is now a conflict between the drm tree and Linus' tree.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2021-09-02  1:22 Stephen Rothwell
  2021-09-10  1:16 ` Stephen Rothwell
  0 siblings, 1 reply; 29+ messages in thread
From: Stephen Rothwell @ 2021-09-02  1:22 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linux Kernel Mailing List, Linux Next Mailing List,
	Luben Tuikov

[-- Attachment #1: Type: text/plain, Size: 798 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c

between commit:

  477f70cd2a67 ("Merge tag 'drm-next-2021-08-31-1' of git://anongit.freedesktop.org/drm/drm")

from Linus' tree and commit:

  cc947bf91bad ("drm/amdgpu: Process any VBIOS RAS EEPROM address")

from the amdgpu tree.

I fixed it up (I used the latter version) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2021-07-06  0:52 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2021-07-06  0:52 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Alex Sierra, Linux Kernel Mailing List,
	Linux Next Mailing List

[-- Attachment #1: Type: text/plain, Size: 1372 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c

between commit:

  e058a84bfddc ("Merge tag 'drm-next-2021-07-01' of git://anongit.freedesktop.org/drm/drm")

from the origin tree and commit:

  8c21fc49a8e6 ("drm/amdkfd: add owner ref param to get hmm pages")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 307d22abd814,2e9ad6e0dfbb..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@@ -695,9 -692,8 +691,9 @@@ int amdgpu_ttm_tt_get_user_pages(struc
  	readonly = amdgpu_ttm_tt_is_readonly(ttm);
  	r = amdgpu_hmm_range_get_pages(&bo->notifier, mm, pages, start,
  				       ttm->num_pages, &gtt->range, readonly,
- 				       true);
 -				       false, NULL);
 -out_putmm:
++				       true, NULL);
 +out_unlock:
 +	mmap_read_unlock(mm);
  	mmput(mm);
  
  	return r;

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2021-01-15  0:53 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2021-01-15  0:53 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linus Torvalds, Linux Kernel Mailing List,
	Linux Next Mailing List, Wesley Chalmers

[-- Attachment #1: Type: text/plain, Size: 773 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c

between commit:

  61d791365b72 ("drm/amd/display: avoid uninitialized variable warning")

from Linus' tree and commit:

  f03e80d2e82c ("drm/amd/display: Initialize stack variable")

from the amdgpu tree.

I fixed it up (I used the latter version) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2021-01-08  1:04 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2021-01-08  1:04 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Linux Kernel Mailing List, Linux Next Mailing List,
	Maxime Ripard, Tian Tao

[-- Attachment #1: Type: text/plain, Size: 1255 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c

between commit:

  eca22edb37d2 ("drm: Pass the full state to connectors atomic functions")

from Linus' tree and commit:

  540b60f9fc7f ("drm/amd/display: remove unused including <linux/version.h>")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 8ab0b9060d2b,5340c41b85f0..000000000000
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@@ -23,8 -23,6 +23,7 @@@
   *
   */
  
- #include <linux/version.h>
 +#include <drm/drm_atomic.h>
  #include <drm/drm_atomic_helper.h>
  #include <drm/drm_dp_mst_helper.h>
  #include <drm/drm_dp_helper.h>

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2020-11-03  3:10 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2020-11-03  3:10 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Dave Airlie, Deepak R Varma,
	Linux Kernel Mailing List, Linux Next Mailing List

[-- Attachment #1: Type: text/plain, Size: 2145 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c

between commit:

  0a667b500703 ("drm/ttm: remove bdev from ttm_tt")

from Linus' tree and commit:

  c4c5ae67d179 ("drm/amdgpu/amdgpu: use "*" adjacent to data name")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 8039d2399584,69ebd07f3eee..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@@ -63,16 -63,13 +63,16 @@@
  
  #define AMDGPU_TTM_VRAM_MAX_DW_READ	(size_t)128
  
 +static int amdgpu_ttm_backend_bind(struct ttm_bo_device *bdev,
 +				   struct ttm_tt *ttm,
 +				   struct ttm_resource *bo_mem);
 +
  static int amdgpu_ttm_init_on_chip(struct amdgpu_device *adev,
  				    unsigned int type,
- 				    uint64_t size)
+ 				    uint64_t size_in_page)
  {
  	return ttm_range_man_init(&adev->mman.bdev, type,
- 				  false, size >> PAGE_SHIFT);
 -				  TTM_PL_FLAG_UNCACHED, TTM_PL_FLAG_UNCACHED,
+ 				  false, size_in_page);
  }
  
  /**
@@@ -1107,12 -1111,11 +1107,12 @@@ gart_bind_fail
   * Called by ttm_tt_bind() on behalf of ttm_bo_handle_move_mem().
   * This handles binding GTT memory to the device address space.
   */
 -static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm,
 +static int amdgpu_ttm_backend_bind(struct ttm_bo_device *bdev,
 +				   struct ttm_tt *ttm,
  				   struct ttm_resource *bo_mem)
  {
 -	struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
 +	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
- 	struct amdgpu_ttm_tt *gtt = (void*)ttm;
+ 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
  	uint64_t flags;
  	int r = 0;
  

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2020-10-27  0:48 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2020-10-27  0:48 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Christian König, Linux Kernel Mailing List,
	Linux Next Mailing List, xinhui pan, Dave Airlie, DRI

[-- Attachment #1: Type: text/plain, Size: 1570 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c

between commits:

  0fe438cec9e1 ("drm/ttm: remove default caching")
  9c3006a4cc1b ("drm/ttm: remove available_caching")

from Linus' tree and commit:

  d836917da7e5 ("drm/amdgpu: Fix size calculation when init onchip memory")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 8039d2399584,11dd3d9eac15..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@@ -63,16 -63,13 +63,16 @@@
  
  #define AMDGPU_TTM_VRAM_MAX_DW_READ	(size_t)128
  
 +static int amdgpu_ttm_backend_bind(struct ttm_bo_device *bdev,
 +				   struct ttm_tt *ttm,
 +				   struct ttm_resource *bo_mem);
 +
  static int amdgpu_ttm_init_on_chip(struct amdgpu_device *adev,
  				    unsigned int type,
- 				    uint64_t size)
+ 				    uint64_t size_in_page)
  {
  	return ttm_range_man_init(&adev->mman.bdev, type,
- 				  false, size >> PAGE_SHIFT);
 -				  TTM_PL_FLAG_UNCACHED, TTM_PL_FLAG_UNCACHED,
+ 				  false, size_in_page);
  }
  
  /**

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2020-10-16  1:48 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2020-10-16  1:48 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Alex Deucher, Evan Quan, Linux Kernel Mailing List,
	Linux Next Mailing List

[-- Attachment #1: Type: text/plain, Size: 1103 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c

between commit:

  3cb9d2416ccd ("drm/amd/pm: setup APU dpm clock table in SMU HW initialization")

from Linus' tree and commit:

  44d6e160f60a ("drm/amd/pm: setup APU dpm clock table in SMU HW initialization")

from the amdgpu tree.

So this is one of the things that can go wrong when you have the same
patch obn both sides of a merge ... due to the context not verying
very much, the chunk of code that is moved into smu_smc_hw_setup()
by these identical patches was inserted twice by the automatic git
mrege resolution.

I fixed it up (I removed the second copy of the code) and can carry the
fix as necessary. This is now fixed as far as linux-next is concerned,
but any non trivial conflicts should be mentioned to your upstream
maintainer when your tree is submitted for merging.  You may also want
to consider cooperating with the maintainer of the conflicting tree to
minimise any particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2020-08-25  1:35 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2020-08-25  1:35 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List, Luben Tuikov,
	Aaron Ma

[-- Attachment #1: Type: text/plain, Size: 1349 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

between commit:

  3b2e973dff59 ("drm/amd/display: add dmcub check on RENOIR")

from Linus' tree and commit:

  4a580877bdcb ("drm/amdgpu: Get DRM dev from adev by inline-f")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index df9338257ae0,785f21ea35df..000000000000
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@@ -1383,8 -1389,8 +1389,8 @@@ static int dm_late_init(void *handle
  	struct dmcu *dmcu = NULL;
  	bool ret = true;
  
 -	if (!adev->dm.fw_dmcu)
 +	if (!adev->dm.fw_dmcu && !adev->dm.dmub_fw)
- 		return detect_mst_link_for_all_connectors(adev->ddev);
+ 		return detect_mst_link_for_all_connectors(adev_to_drm(adev));
  
  	dmcu = adev->dm.dc->res_pool->dmcu;
  

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2020-07-23  8:01 ` Daniel Vetter
@ 2020-07-23 13:34   ` Alex Deucher
  0 siblings, 0 replies; 29+ messages in thread
From: Alex Deucher @ 2020-07-23 13:34 UTC (permalink / raw)
  To: Daniel Vetter
  Cc: Stephen Rothwell, Dave Airlie, DRI, Linux Next Mailing List,
	Linux Kernel Mailing List

On Thu, Jul 23, 2020 at 4:01 AM Daniel Vetter <daniel@ffwll.ch> wrote:
>
> On Wed, Jul 22, 2020 at 6:34 AM Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > Hi all,
> >
> > [I can't find a previous email about this, sorry ...]
> >
> > There is a semantic conflict between Linus' tree and the amdgpu tree
> > between commit
> >
> >   d7a6634a4cfb ("drm/amdgpu/atomfirmware: fix vram_info fetching for renoir")
> >
> > from Linus' tree and commts
> >
> >   fe098a5d6443 ("drm/amdgpu/atomfirmware: fix vram_info fetching for renoir")
> >   836dab851903 ("drm/amdgpu/atomfirmware: update vram info handling for renoir")
> >
> > The automted git merge leaves two "case 12" labels.  I have been
> > reverting commit d7a6634a4cfb since July 3 ... This will need to be
> > fixed up when the amdgpu tree is next merged into the drm tree, or a back
> > merge of d7a6634a4cfb could be done into the amdgpu tree and the older
> > "case 12" label removed in that merge.
>
> Yeah this is a bit nasty, I just crashed over it too (btw moved to
> drm-next now). Alex is this the right conflict resolution for -next? I
> think we should bake this in with a backmerge ...

Yes, you can drop either d7a6634a4cfb or fe098a5d6443.  I just
cherry-picked the commit from -next back to 5.8 and stable since it
was a bug fix.

Alex

> -Daniel
>
> >
> > --
> > Cheers,
> > Stephen Rothwell
> > _______________________________________________
> > dri-devel mailing list
> > dri-devel@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
>
>
>
> --
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2020-07-22  4:34 Stephen Rothwell
@ 2020-07-23  8:01 ` Daniel Vetter
  2020-07-23 13:34   ` Alex Deucher
  0 siblings, 1 reply; 29+ messages in thread
From: Daniel Vetter @ 2020-07-23  8:01 UTC (permalink / raw)
  To: Stephen Rothwell
  Cc: Alex Deucher, Dave Airlie, DRI, Linux Next Mailing List,
	Linux Kernel Mailing List

On Wed, Jul 22, 2020 at 6:34 AM Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Hi all,
>
> [I can't find a previous email about this, sorry ...]
>
> There is a semantic conflict between Linus' tree and the amdgpu tree
> between commit
>
>   d7a6634a4cfb ("drm/amdgpu/atomfirmware: fix vram_info fetching for renoir")
>
> from Linus' tree and commts
>
>   fe098a5d6443 ("drm/amdgpu/atomfirmware: fix vram_info fetching for renoir")
>   836dab851903 ("drm/amdgpu/atomfirmware: update vram info handling for renoir")
>
> The automted git merge leaves two "case 12" labels.  I have been
> reverting commit d7a6634a4cfb since July 3 ... This will need to be
> fixed up when the amdgpu tree is next merged into the drm tree, or a back
> merge of d7a6634a4cfb could be done into the amdgpu tree and the older
> "case 12" label removed in that merge.

Yeah this is a bit nasty, I just crashed over it too (btw moved to
drm-next now). Alex is this the right conflict resolution for -next? I
think we should bake this in with a backmerge ...
-Daniel

>
> --
> Cheers,
> Stephen Rothwell
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/dri-devel



-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2020-07-22  4:34 Stephen Rothwell
  2020-07-23  8:01 ` Daniel Vetter
  0 siblings, 1 reply; 29+ messages in thread
From: Stephen Rothwell @ 2020-07-22  4:34 UTC (permalink / raw)
  To: Alex Deucher, Dave Airlie, DRI
  Cc: Linux Next Mailing List, Linux Kernel Mailing List

[-- Attachment #1: Type: text/plain, Size: 787 bytes --]

Hi all,

[I can't find a previous email about this, sorry ...]

There is a semantic conflict between Linus' tree and the amdgpu tree
between commit

  d7a6634a4cfb ("drm/amdgpu/atomfirmware: fix vram_info fetching for renoir")

from Linus' tree and commts

  fe098a5d6443 ("drm/amdgpu/atomfirmware: fix vram_info fetching for renoir")
  836dab851903 ("drm/amdgpu/atomfirmware: update vram info handling for renoir")

The automted git merge leaves two "case 12" labels.  I have been
reverting commit d7a6634a4cfb since July 3 ... This will need to be
fixed up when the amdgpu tree is next merged into the drm tree, or a back
merge of d7a6634a4cfb could be done into the amdgpu tree and the older
"case 12" label removed in that merge.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2019-12-12  0:03 Stephen Rothwell
  2019-12-12  0:11 ` Stephen Rothwell
@ 2019-12-12  0:16 ` Timothy Pearson
  1 sibling, 0 replies; 29+ messages in thread
From: Timothy Pearson @ 2019-12-12  0:16 UTC (permalink / raw)
  To: Stephen Rothwell
  Cc: Alex Deucher, Linux Next Mailing List, linux-kernel, Nick Desaulniers

The new version looks OK from my side.

----- Original Message -----
> From: "Stephen Rothwell" <sfr@canb.auug.org.au>
> To: "Alex Deucher" <alexdeucher@gmail.com>
> Cc: "Linux Next Mailing List" <linux-next@vger.kernel.org>, "linux-kernel" <linux-kernel@vger.kernel.org>, "Nick
> Desaulniers" <ndesaulniers@google.com>, "Timothy Pearson" <tpearson@raptorengineering.com>
> Sent: Wednesday, December 11, 2019 6:03:37 PM
> Subject: linux-next: manual merge of the amdgpu tree with Linus' tree

> Hi all,
> 
> Today's linux-next merge of the amdgpu tree got conflicts in:
> 
>  drivers/gpu/drm/amd/display/dc/calcs/Makefile
>  drivers/gpu/drm/amd/display/dc/dcn20/Makefile
>  drivers/gpu/drm/amd/display/dc/dcn21/Makefile
>  drivers/gpu/drm/amd/display/dc/dml/Makefile
>  drivers/gpu/drm/amd/display/dc/dsc/Makefile
> 
> between commits:
> 
>  c868868f6b6a ("drm/amdgpu: fix stack alignment ABI mismatch for Clang")
>  00db297106e8 ("drm/amdgpu: fix stack alignment ABI mismatch for GCC 7.1+")
> 
> from Linus' tree and commit:
> 
>  86462415d58d ("amdgpu: Enable initial DCN support on POWER")
> 
> from the amdgpu tree.
> 
> I fixed it up (I think .. see below) and can carry the fix as
> necessary. This is now fixed as far as linux-next is concerned, but any
> non trivial conflicts should be mentioned to your upstream maintainer
> when your tree is submitted for merging.  You may also want to consider
> cooperating with the maintainer of the conflicting tree to minimise any
> particularly complex conflicts.
> 
> --
> Cheers,
> Stephen Rothwell
> 
> diff --cc drivers/gpu/drm/amd/display/dc/calcs/Makefile
> index 26c6d735cdc7,4d3006bd4337..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/calcs/Makefile
> @@@ -24,22 -25,23 +25,30 @@@
>  # It calculates Bandwidth and Watermarks values for HW programming
>  #
>  
> -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
> -	cc_stack_align := -mpreferred-stack-boundary=4
> -else ifneq ($(call cc-option, -mstack-alignment=16),)
> -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
> +calcs_ccflags := -mhard-float -msse
> + endif
> +
> -ifdef CONFIG_X86
> -calcs_ccflags := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++calcs_ccflags := -mhard-float -maltivec
> ++endif
>  
> -ifdef CONFIG_CC_IS_CLANG
> -calcs_ccflags += -msse2
> +ifdef CONFIG_CC_IS_GCC
> +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
> +IS_OLD_GCC = 1
>  endif
>  endif
>  
> -ifdef CONFIG_PPC64
> -calcs_ccflags := -mhard-float -maltivec $(cc_stack_align)
> +ifdef IS_OLD_GCC
> +# Stack alignment mismatch, proceed with caution.
> +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
> +# (8B stack alignment).
> +calcs_ccflags += -mpreferred-stack-boundary=4
> +else
> ++ifdef CONFIG_X86
> +calcs_ccflags += -msse2
> +endif
> + endif
>  
>  CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_ccflags)
>  CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_ccflags)
> diff --cc drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> index 63f3bddba7da,07f652d40f86..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> @@@ -6,26 -7,25 +7,32 @@@ DCN20 = dcn20_resource.o dcn20_init.o d
>  		dcn20_stream_encoder.o dcn20_link_encoder.o dcn20_dccg.o \
>  		dcn20_vmid.o dcn20_dwb.o dcn20_dwb_scl.o
>  
> - ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT
>  DCN20 += dcn20_dsc.o
> - endif
>  
> -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
> -	cc_stack_align := -mpreferred-stack-boundary=4
> -else ifneq ($(call cc-option, -mstack-alignment=16),)
> -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
> +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse
> + endif
> +
> -ifdef CONFIG_X86
> -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse
> $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec
> ++endif
>  
> -ifdef CONFIG_CC_IS_CLANG
> -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
> +ifdef CONFIG_CC_IS_GCC
> +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
> +IS_OLD_GCC = 1
>  endif
>  endif
>  
> -ifdef CONFIG_PPC64
> -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec
> $(cc_stack_align)
> +ifdef IS_OLD_GCC
> +# Stack alignment mismatch, proceed with caution.
> +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
> +# (8B stack alignment).
> +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -mpreferred-stack-boundary=4
> +else
> ++ifdef CONFIG_X86
> +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
> +endif
> + endif
>  
>  AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20))
>  
> diff --cc drivers/gpu/drm/amd/display/dc/dcn21/Makefile
> index 14113ccf498d,041464d001bd..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
> @@@ -1,24 -2,26 +2,33 @@@
>  #
>  # Makefile for DCN21.
>  
> - DCN21 = dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o dcn21_hwseq.o
> dcn21_link_encoder.o
> + DCN21 = dcn21_init.o dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o \
> + 	 dcn21_hwseq.o dcn21_link_encoder.o
>  
> -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
> -	cc_stack_align := -mpreferred-stack-boundary=4
> -else ifneq ($(call cc-option, -mstack-alignment=16),)
> -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
> +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse
> + endif
> +
> -ifdef CONFIG_X86
> -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse
> $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec
> ++endif
>  
> -ifdef CONFIG_CC_IS_CLANG
> -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2
> +ifdef CONFIG_CC_IS_GCC
> +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
> +IS_OLD_GCC = 1
>  endif
>  endif
>  
> -ifdef CONFIG_PPC64
> -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec
> $(cc_stack_align)
> +ifdef IS_OLD_GCC
> +# Stack alignment mismatch, proceed with caution.
> +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
> +# (8B stack alignment).
> +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -mpreferred-stack-boundary=4
> +else
> ++ifdef CONFIG_X86
> +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2
> +endif
> + endif
>  
>  AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21))
>  
> diff --cc drivers/gpu/drm/amd/display/dc/dml/Makefile
> index 8df251626e22,82c8978c81ab..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile
> @@@ -24,22 -25,23 +25,30 @@@
>  # It provides the general basic services required by other DAL
>  # subcomponents.
>  
> -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
> -	cc_stack_align := -mpreferred-stack-boundary=4
> -else ifneq ($(call cc-option, -mstack-alignment=16),)
> -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
> +dml_ccflags := -mhard-float -msse
> + endif
> +
> -ifdef CONFIG_X86
> -dml_ccflags := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++dml_ccflags := -mhard-float -maltivec
> ++endif
>  
> -ifdef CONFIG_CC_IS_CLANG
> -dml_ccflags += -msse2
> +ifdef CONFIG_CC_IS_GCC
> +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
> +IS_OLD_GCC = 1
>  endif
>  endif
>  
> -ifdef CONFIG_PPC64
> -dml_ccflags := -mhard-float -maltivec $(cc_stack_align)
> +ifdef IS_OLD_GCC
> +# Stack alignment mismatch, proceed with caution.
> +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
> +# (8B stack alignment).
> +dml_ccflags += -mpreferred-stack-boundary=4
> +else
> ++ifdef CONFIG_X86
> +dml_ccflags += -msse2
> +endif
> + endif
>  
>  CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags)
>  
> diff --cc drivers/gpu/drm/amd/display/dc/dsc/Makefile
> index 970737217e53,08edd919ec82..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile
> @@@ -1,22 -2,23 +2,30 @@@
>  #
>  # Makefile for the 'dsc' sub-component of DAL.
>  
> -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
> -	cc_stack_align := -mpreferred-stack-boundary=4
> -else ifneq ($(call cc-option, -mstack-alignment=16),)
> -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
> +dsc_ccflags := -mhard-float -msse
> + endif
> +
> -ifdef CONFIG_X86
> -dsc_ccflags := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++dsc_ccflags := -mhard-float -maltivec
> ++endif
>  
> -ifdef CONFIG_CC_IS_CLANG
> -dsc_ccflags += -msse2
> +ifdef CONFIG_CC_IS_GCC
> +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
> +IS_OLD_GCC = 1
>  endif
>  endif
>  
> -ifdef CONFIG_PPC64
> -dsc_ccflags := -mhard-float -maltivec $(cc_stack_align)
> +ifdef IS_OLD_GCC
> +# Stack alignment mismatch, proceed with caution.
> +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
> +# (8B stack alignment).
> +dsc_ccflags += -mpreferred-stack-boundary=4
> +else
> ++ifdef CONFIG_X86
> +dsc_ccflags += -msse2
> +endif
> + endif
>  
>  CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_ccflags)
>   CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc_dpi.o := $(dsc_ccflags)

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2019-12-12  0:03 Stephen Rothwell
@ 2019-12-12  0:11 ` Stephen Rothwell
  2019-12-12  0:16 ` Timothy Pearson
  1 sibling, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-12-12  0:11 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List,
	Nick Desaulniers, Timothy Pearson, Dave Airlie

[-- Attachment #1: Type: text/plain, Size: 9478 bytes --]

Hi all,

Just cc'ing Dave.

On Thu, 12 Dec 2019 11:03:37 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Today's linux-next merge of the amdgpu tree got conflicts in:
> 
>   drivers/gpu/drm/amd/display/dc/calcs/Makefile
>   drivers/gpu/drm/amd/display/dc/dcn20/Makefile
>   drivers/gpu/drm/amd/display/dc/dcn21/Makefile
>   drivers/gpu/drm/amd/display/dc/dml/Makefile
>   drivers/gpu/drm/amd/display/dc/dsc/Makefile
> 
> between commits:
> 
>   c868868f6b6a ("drm/amdgpu: fix stack alignment ABI mismatch for Clang")
>   00db297106e8 ("drm/amdgpu: fix stack alignment ABI mismatch for GCC 7.1+")
> 
> from Linus' tree and commit:
> 
>   86462415d58d ("amdgpu: Enable initial DCN support on POWER")
> 
> from the amdgpu tree.
> 
> I fixed it up (I think .. see below) and can carry the fix as
> necessary. This is now fixed as far as linux-next is concerned, but any
> non trivial conflicts should be mentioned to your upstream maintainer
> when your tree is submitted for merging.  You may also want to consider
> cooperating with the maintainer of the conflicting tree to minimise any
> particularly complex conflicts.
> 
> -- 
> Cheers,
> Stephen Rothwell
> 
> diff --cc drivers/gpu/drm/amd/display/dc/calcs/Makefile
> index 26c6d735cdc7,4d3006bd4337..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/calcs/Makefile
> @@@ -24,22 -25,23 +25,30 @@@
>   # It calculates Bandwidth and Watermarks values for HW programming
>   #
>   
>  -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
>  -	cc_stack_align := -mpreferred-stack-boundary=4
>  -else ifneq ($(call cc-option, -mstack-alignment=16),)
>  -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
>  +calcs_ccflags := -mhard-float -msse
> + endif
> + 
>  -ifdef CONFIG_X86
>  -calcs_ccflags := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++calcs_ccflags := -mhard-float -maltivec
> ++endif
>   
>  -ifdef CONFIG_CC_IS_CLANG
>  -calcs_ccflags += -msse2
>  +ifdef CONFIG_CC_IS_GCC
>  +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
>  +IS_OLD_GCC = 1
>   endif
>   endif
>   
>  -ifdef CONFIG_PPC64
>  -calcs_ccflags := -mhard-float -maltivec $(cc_stack_align)
>  +ifdef IS_OLD_GCC
>  +# Stack alignment mismatch, proceed with caution.
>  +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
>  +# (8B stack alignment).
>  +calcs_ccflags += -mpreferred-stack-boundary=4
>  +else
> ++ifdef CONFIG_X86
>  +calcs_ccflags += -msse2
>  +endif
> + endif
>   
>   CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_ccflags)
>   CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_ccflags)
> diff --cc drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> index 63f3bddba7da,07f652d40f86..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> @@@ -6,26 -7,25 +7,32 @@@ DCN20 = dcn20_resource.o dcn20_init.o d
>   		dcn20_stream_encoder.o dcn20_link_encoder.o dcn20_dccg.o \
>   		dcn20_vmid.o dcn20_dwb.o dcn20_dwb_scl.o
>   
> - ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT
>   DCN20 += dcn20_dsc.o
> - endif
>   
>  -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
>  -	cc_stack_align := -mpreferred-stack-boundary=4
>  -else ifneq ($(call cc-option, -mstack-alignment=16),)
>  -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
>  +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse
> + endif
> + 
>  -ifdef CONFIG_X86
>  -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec
> ++endif
>   
>  -ifdef CONFIG_CC_IS_CLANG
>  -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
>  +ifdef CONFIG_CC_IS_GCC
>  +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
>  +IS_OLD_GCC = 1
>   endif
>   endif
>   
>  -ifdef CONFIG_PPC64
>  -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec $(cc_stack_align)
>  +ifdef IS_OLD_GCC
>  +# Stack alignment mismatch, proceed with caution.
>  +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
>  +# (8B stack alignment).
>  +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -mpreferred-stack-boundary=4
>  +else
> ++ifdef CONFIG_X86
>  +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
>  +endif
> + endif
>   
>   AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20))
>   
> diff --cc drivers/gpu/drm/amd/display/dc/dcn21/Makefile
> index 14113ccf498d,041464d001bd..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
> @@@ -1,24 -2,26 +2,33 @@@
>   #
>   # Makefile for DCN21.
>   
> - DCN21 = dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o dcn21_hwseq.o dcn21_link_encoder.o
> + DCN21 = dcn21_init.o dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o \
> + 	 dcn21_hwseq.o dcn21_link_encoder.o
>   
>  -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
>  -	cc_stack_align := -mpreferred-stack-boundary=4
>  -else ifneq ($(call cc-option, -mstack-alignment=16),)
>  -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
>  +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse
> + endif
> + 
>  -ifdef CONFIG_X86
>  -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec
> ++endif
>   
>  -ifdef CONFIG_CC_IS_CLANG
>  -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2
>  +ifdef CONFIG_CC_IS_GCC
>  +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
>  +IS_OLD_GCC = 1
>   endif
>   endif
>   
>  -ifdef CONFIG_PPC64
>  -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec $(cc_stack_align)
>  +ifdef IS_OLD_GCC
>  +# Stack alignment mismatch, proceed with caution.
>  +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
>  +# (8B stack alignment).
>  +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -mpreferred-stack-boundary=4
>  +else
> ++ifdef CONFIG_X86
>  +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2
>  +endif
> + endif
>   
>   AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21))
>   
> diff --cc drivers/gpu/drm/amd/display/dc/dml/Makefile
> index 8df251626e22,82c8978c81ab..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile
> @@@ -24,22 -25,23 +25,30 @@@
>   # It provides the general basic services required by other DAL
>   # subcomponents.
>   
>  -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
>  -	cc_stack_align := -mpreferred-stack-boundary=4
>  -else ifneq ($(call cc-option, -mstack-alignment=16),)
>  -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
>  +dml_ccflags := -mhard-float -msse
> + endif
> + 
>  -ifdef CONFIG_X86
>  -dml_ccflags := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++dml_ccflags := -mhard-float -maltivec
> ++endif
>   
>  -ifdef CONFIG_CC_IS_CLANG
>  -dml_ccflags += -msse2
>  +ifdef CONFIG_CC_IS_GCC
>  +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
>  +IS_OLD_GCC = 1
>   endif
>   endif
>   
>  -ifdef CONFIG_PPC64
>  -dml_ccflags := -mhard-float -maltivec $(cc_stack_align)
>  +ifdef IS_OLD_GCC
>  +# Stack alignment mismatch, proceed with caution.
>  +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
>  +# (8B stack alignment).
>  +dml_ccflags += -mpreferred-stack-boundary=4
>  +else
> ++ifdef CONFIG_X86
>  +dml_ccflags += -msse2
>  +endif
> + endif
>   
>   CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags)
>   
> diff --cc drivers/gpu/drm/amd/display/dc/dsc/Makefile
> index 970737217e53,08edd919ec82..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile
> @@@ -1,22 -2,23 +2,30 @@@
>   #
>   # Makefile for the 'dsc' sub-component of DAL.
>   
>  -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
>  -	cc_stack_align := -mpreferred-stack-boundary=4
>  -else ifneq ($(call cc-option, -mstack-alignment=16),)
>  -	cc_stack_align := -mstack-alignment=16
> ++ifdef CONFIG_X86
>  +dsc_ccflags := -mhard-float -msse
> + endif
> + 
>  -ifdef CONFIG_X86
>  -dsc_ccflags := -mhard-float -msse $(cc_stack_align)
> ++ifdef CONFIG_PPC64
> ++dsc_ccflags := -mhard-float -maltivec
> ++endif
>   
>  -ifdef CONFIG_CC_IS_CLANG
>  -dsc_ccflags += -msse2
>  +ifdef CONFIG_CC_IS_GCC
>  +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
>  +IS_OLD_GCC = 1
>   endif
>   endif
>   
>  -ifdef CONFIG_PPC64
>  -dsc_ccflags := -mhard-float -maltivec $(cc_stack_align)
>  +ifdef IS_OLD_GCC
>  +# Stack alignment mismatch, proceed with caution.
>  +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
>  +# (8B stack alignment).
>  +dsc_ccflags += -mpreferred-stack-boundary=4
>  +else
> ++ifdef CONFIG_X86
>  +dsc_ccflags += -msse2
>  +endif
> + endif
>   
>   CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_ccflags)
>   CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc_dpi.o := $(dsc_ccflags)

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2019-12-11 23:37 Stephen Rothwell
@ 2019-12-12  0:10 ` Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-12-12  0:10 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List, Yintian Tao,
	Jason Gunthorpe, Dave Airlie

[-- Attachment #1: Type: text/plain, Size: 1526 bytes --]

Hi all,

Just cc'ing Dave.

On Thu, 12 Dec 2019 10:37:38 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Today's linux-next merge of the amdgpu tree got a conflict in:
> 
>   drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> 
> between commit:
> 
>   62914a99dee5 ("drm/amdgpu: Use mmu_interval_insert instead of hmm_mirror")
> 
> from Linus' tree and commit:
> 
>   8497d2bcdee1 ("drm/amd/powerplay: enable pp one vf mode for vega10")
> 
> from the amdgpu tree.
> 
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
> 
> -- 
> Cheers,
> Stephen Rothwell
> 
> diff --cc drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index c17505fba988,8992506541d8..000000000000
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@@ -2794,8 -2828,6 +2828,7 @@@ int amdgpu_device_init(struct amdgpu_de
>   	mutex_init(&adev->virt.vf_errors.lock);
>   	hash_init(adev->mn_hash);
>   	mutex_init(&adev->lock_reset);
>  +	mutex_init(&adev->notifier_lock);
> - 	mutex_init(&adev->virt.dpm_mutex);
>   	mutex_init(&adev->psp.mutex);
>   
>   	r = amdgpu_device_check_arguments(adev);

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-12-12  0:03 Stephen Rothwell
  2019-12-12  0:11 ` Stephen Rothwell
  2019-12-12  0:16 ` Timothy Pearson
  0 siblings, 2 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-12-12  0:03 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List,
	Nick Desaulniers, Timothy Pearson

[-- Attachment #1: Type: text/plain, Size: 8811 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got conflicts in:

  drivers/gpu/drm/amd/display/dc/calcs/Makefile
  drivers/gpu/drm/amd/display/dc/dcn20/Makefile
  drivers/gpu/drm/amd/display/dc/dcn21/Makefile
  drivers/gpu/drm/amd/display/dc/dml/Makefile
  drivers/gpu/drm/amd/display/dc/dsc/Makefile

between commits:

  c868868f6b6a ("drm/amdgpu: fix stack alignment ABI mismatch for Clang")
  00db297106e8 ("drm/amdgpu: fix stack alignment ABI mismatch for GCC 7.1+")

from Linus' tree and commit:

  86462415d58d ("amdgpu: Enable initial DCN support on POWER")

from the amdgpu tree.

I fixed it up (I think .. see below) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/display/dc/calcs/Makefile
index 26c6d735cdc7,4d3006bd4337..000000000000
--- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/calcs/Makefile
@@@ -24,22 -25,23 +25,30 @@@
  # It calculates Bandwidth and Watermarks values for HW programming
  #
  
 -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
 -	cc_stack_align := -mpreferred-stack-boundary=4
 -else ifneq ($(call cc-option, -mstack-alignment=16),)
 -	cc_stack_align := -mstack-alignment=16
++ifdef CONFIG_X86
 +calcs_ccflags := -mhard-float -msse
+ endif
+ 
 -ifdef CONFIG_X86
 -calcs_ccflags := -mhard-float -msse $(cc_stack_align)
++ifdef CONFIG_PPC64
++calcs_ccflags := -mhard-float -maltivec
++endif
  
 -ifdef CONFIG_CC_IS_CLANG
 -calcs_ccflags += -msse2
 +ifdef CONFIG_CC_IS_GCC
 +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 +IS_OLD_GCC = 1
  endif
  endif
  
 -ifdef CONFIG_PPC64
 -calcs_ccflags := -mhard-float -maltivec $(cc_stack_align)
 +ifdef IS_OLD_GCC
 +# Stack alignment mismatch, proceed with caution.
 +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
 +# (8B stack alignment).
 +calcs_ccflags += -mpreferred-stack-boundary=4
 +else
++ifdef CONFIG_X86
 +calcs_ccflags += -msse2
 +endif
+ endif
  
  CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_ccflags)
  CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_ccflags)
diff --cc drivers/gpu/drm/amd/display/dc/dcn20/Makefile
index 63f3bddba7da,07f652d40f86..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
@@@ -6,26 -7,25 +7,32 @@@ DCN20 = dcn20_resource.o dcn20_init.o d
  		dcn20_stream_encoder.o dcn20_link_encoder.o dcn20_dccg.o \
  		dcn20_vmid.o dcn20_dwb.o dcn20_dwb_scl.o
  
- ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT
  DCN20 += dcn20_dsc.o
- endif
  
 -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
 -	cc_stack_align := -mpreferred-stack-boundary=4
 -else ifneq ($(call cc-option, -mstack-alignment=16),)
 -	cc_stack_align := -mstack-alignment=16
++ifdef CONFIG_X86
 +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse
+ endif
+ 
 -ifdef CONFIG_X86
 -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse $(cc_stack_align)
++ifdef CONFIG_PPC64
++CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec
++endif
  
 -ifdef CONFIG_CC_IS_CLANG
 -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
 +ifdef CONFIG_CC_IS_GCC
 +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 +IS_OLD_GCC = 1
  endif
  endif
  
 -ifdef CONFIG_PPC64
 -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec $(cc_stack_align)
 +ifdef IS_OLD_GCC
 +# Stack alignment mismatch, proceed with caution.
 +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
 +# (8B stack alignment).
 +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -mpreferred-stack-boundary=4
 +else
++ifdef CONFIG_X86
 +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
 +endif
+ endif
  
  AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20))
  
diff --cc drivers/gpu/drm/amd/display/dc/dcn21/Makefile
index 14113ccf498d,041464d001bd..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
@@@ -1,24 -2,26 +2,33 @@@
  #
  # Makefile for DCN21.
  
- DCN21 = dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o dcn21_hwseq.o dcn21_link_encoder.o
+ DCN21 = dcn21_init.o dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o \
+ 	 dcn21_hwseq.o dcn21_link_encoder.o
  
 -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
 -	cc_stack_align := -mpreferred-stack-boundary=4
 -else ifneq ($(call cc-option, -mstack-alignment=16),)
 -	cc_stack_align := -mstack-alignment=16
++ifdef CONFIG_X86
 +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse
+ endif
+ 
 -ifdef CONFIG_X86
 -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse $(cc_stack_align)
++ifdef CONFIG_PPC64
++CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec
++endif
  
 -ifdef CONFIG_CC_IS_CLANG
 -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2
 +ifdef CONFIG_CC_IS_GCC
 +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 +IS_OLD_GCC = 1
  endif
  endif
  
 -ifdef CONFIG_PPC64
 -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec $(cc_stack_align)
 +ifdef IS_OLD_GCC
 +# Stack alignment mismatch, proceed with caution.
 +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
 +# (8B stack alignment).
 +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -mpreferred-stack-boundary=4
 +else
++ifdef CONFIG_X86
 +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2
 +endif
+ endif
  
  AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21))
  
diff --cc drivers/gpu/drm/amd/display/dc/dml/Makefile
index 8df251626e22,82c8978c81ab..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dml/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile
@@@ -24,22 -25,23 +25,30 @@@
  # It provides the general basic services required by other DAL
  # subcomponents.
  
 -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
 -	cc_stack_align := -mpreferred-stack-boundary=4
 -else ifneq ($(call cc-option, -mstack-alignment=16),)
 -	cc_stack_align := -mstack-alignment=16
++ifdef CONFIG_X86
 +dml_ccflags := -mhard-float -msse
+ endif
+ 
 -ifdef CONFIG_X86
 -dml_ccflags := -mhard-float -msse $(cc_stack_align)
++ifdef CONFIG_PPC64
++dml_ccflags := -mhard-float -maltivec
++endif
  
 -ifdef CONFIG_CC_IS_CLANG
 -dml_ccflags += -msse2
 +ifdef CONFIG_CC_IS_GCC
 +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 +IS_OLD_GCC = 1
  endif
  endif
  
 -ifdef CONFIG_PPC64
 -dml_ccflags := -mhard-float -maltivec $(cc_stack_align)
 +ifdef IS_OLD_GCC
 +# Stack alignment mismatch, proceed with caution.
 +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
 +# (8B stack alignment).
 +dml_ccflags += -mpreferred-stack-boundary=4
 +else
++ifdef CONFIG_X86
 +dml_ccflags += -msse2
 +endif
+ endif
  
  CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags)
  
diff --cc drivers/gpu/drm/amd/display/dc/dsc/Makefile
index 970737217e53,08edd919ec82..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile
@@@ -1,22 -2,23 +2,30 @@@
  #
  # Makefile for the 'dsc' sub-component of DAL.
  
 -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
 -	cc_stack_align := -mpreferred-stack-boundary=4
 -else ifneq ($(call cc-option, -mstack-alignment=16),)
 -	cc_stack_align := -mstack-alignment=16
++ifdef CONFIG_X86
 +dsc_ccflags := -mhard-float -msse
+ endif
+ 
 -ifdef CONFIG_X86
 -dsc_ccflags := -mhard-float -msse $(cc_stack_align)
++ifdef CONFIG_PPC64
++dsc_ccflags := -mhard-float -maltivec
++endif
  
 -ifdef CONFIG_CC_IS_CLANG
 -dsc_ccflags += -msse2
 +ifdef CONFIG_CC_IS_GCC
 +ifeq ($(call cc-ifversion, -lt, 0701, y), y)
 +IS_OLD_GCC = 1
  endif
  endif
  
 -ifdef CONFIG_PPC64
 -dsc_ccflags := -mhard-float -maltivec $(cc_stack_align)
 +ifdef IS_OLD_GCC
 +# Stack alignment mismatch, proceed with caution.
 +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
 +# (8B stack alignment).
 +dsc_ccflags += -mpreferred-stack-boundary=4
 +else
++ifdef CONFIG_X86
 +dsc_ccflags += -msse2
 +endif
+ endif
  
  CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_ccflags)
  CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc_dpi.o := $(dsc_ccflags)

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-12-11 23:37 Stephen Rothwell
  2019-12-12  0:10 ` Stephen Rothwell
  0 siblings, 1 reply; 29+ messages in thread
From: Stephen Rothwell @ 2019-12-11 23:37 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List, Yintian Tao,
	Jason Gunthorpe

[-- Attachment #1: Type: text/plain, Size: 1309 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

between commit:

  62914a99dee5 ("drm/amdgpu: Use mmu_interval_insert instead of hmm_mirror")

from Linus' tree and commit:

  8497d2bcdee1 ("drm/amd/powerplay: enable pp one vf mode for vega10")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index c17505fba988,8992506541d8..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@@ -2794,8 -2828,6 +2828,7 @@@ int amdgpu_device_init(struct amdgpu_de
  	mutex_init(&adev->virt.vf_errors.lock);
  	hash_init(adev->mn_hash);
  	mutex_init(&adev->lock_reset);
 +	mutex_init(&adev->notifier_lock);
- 	mutex_init(&adev->virt.dpm_mutex);
  	mutex_init(&adev->psp.mutex);
  
  	r = amdgpu_device_check_arguments(adev);

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: linux-next: manual merge of the amdgpu tree with Linus' tree
  2019-10-28  0:57 Stephen Rothwell
@ 2019-10-28  1:13 ` Masahiro Yamada
  0 siblings, 0 replies; 29+ messages in thread
From: Masahiro Yamada @ 2019-10-28  1:13 UTC (permalink / raw)
  To: Stephen Rothwell
  Cc: Alex Deucher, Linux Next Mailing List, Linux Kernel Mailing List,
	Nick Desaulniers

On Mon, Oct 28, 2019 at 9:57 AM Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Hi all,
>
> Today's linux-next merge of the amdgpu tree got a conflict in:
>
>   drivers/gpu/drm/amd/display/dc/dcn20/Makefile
>
> between commit:
>
>   54b8ae66ae1a ("kbuild: change *FLAGS_<basetarget>.o to take the path relative to $(obj)")
>
> from Linus' tree and commits:
>
>   4f952528add3 ("drm/amdgpu: fix stack alignment ABI mismatch for Clang")
>   971463bea55c ("drm/amdgpu: fix stack alignment ABI mismatch for GCC 7.1+")
>   101d09f07966 ("drm/amdgpu: enable -msse2 for GCC 7.1+ users")
>
> from the amdgpu tree.
>
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.


I believe subsystems should queue up patches
against -rc1 or a later tag.

Developing based on an older version
causes unneeded conflicts like this.




> --
> Cheers,
> Stephen Rothwell
>
> diff --cc drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> index ddb8d5649e79,be3a614963c6..000000000000
> --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
> @@@ -10,16 -10,21 +10,21 @@@ ifdef CONFIG_DRM_AMD_DC_DSC_SUPPOR
>   DCN20 += dcn20_dsc.o
>   endif
>
> - ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
> -       cc_stack_align := -mpreferred-stack-boundary=4
> - else ifneq ($(call cc-option, -mstack-alignment=16),)
> -       cc_stack_align := -mstack-alignment=16
> - endif
>  -CFLAGS_dcn20_resource.o := -mhard-float -msse
> ++CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse
>
> - CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse $(cc_stack_align)
> + ifdef CONFIG_CC_IS_GCC
> + ifeq ($(call cc-ifversion, -lt, 0701, y), y)
> + IS_OLD_GCC = 1
> + endif
> + endif
>
> - ifdef CONFIG_CC_IS_CLANG
> + ifdef IS_OLD_GCC
> + # Stack alignment mismatch, proceed with caution.
> + # GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
> + # (8B stack alignment).
>  -CFLAGS_dcn20_resource.o += -mpreferred-stack-boundary=4
> ++CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -mpreferred-stack-boundary=4
> + else
>  -CFLAGS_dcn20_resource.o += -msse2
>  +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
>   endif
>
>   AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20))



-- 
Best Regards
Masahiro Yamada

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-10-28  1:11 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-10-28  1:11 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List, Kevin Wang,
	Evan Quan

[-- Attachment #1: Type: text/plain, Size: 829 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h

between commit:

  155f85c0d568 ("drm/amd/powerplay: remove duplicate macro smu_get_uclk_dpm_states in amdgpu_smu.h")

from Linus' tree and commit:

  18c1d3cee4c5 ("drm/amd/powerplay: split out those internal used swSMU APIs V2")

from the amdgpu tree.

I fixed it up (I just used the amdgpu tree version) and can carry the
fix as necessary. This is now fixed as far as linux-next is concerned,
but any non trivial conflicts should be mentioned to your upstream
maintainer when your tree is submitted for merging.  You may also want
to consider cooperating with the maintainer of the conflicting tree to
minimise any particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-10-28  1:03 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-10-28  1:03 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List,
	Arnd Bergmann, Nick Desaulniers

[-- Attachment #1: Type: text/plain, Size: 2593 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/display/dc/dcn21/Makefile

between commits:

  d7b0827f28ab ("Merge tag 'kbuild-v5.4' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild")
  c0e70e10b11b ("drm/amd/display: fix dcn21 Makefile for clang")

from Linus' tree and commits:

  618f55831b84 ("drm/amd/display: fix dcn21 Makefile for clang")
  4f952528add3 ("drm/amdgpu: fix stack alignment ABI mismatch for Clang")
  971463bea55c ("drm/amdgpu: fix stack alignment ABI mismatch for GCC 7.1+")
  101d09f07966 ("drm/amdgpu: enable -msse2 for GCC 7.1+ users")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/display/dc/dcn21/Makefile
index ef673bffc241,feb7e705e792..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile
@@@ -1,18 -1,23 +1,23 @@@
  #
  # Makefile for DCN21.
  
- DCN21 = dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o
+ DCN21 = dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o dcn21_hwseq.o dcn21_link_encoder.o
  
- ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
- 	cc_stack_align := -mpreferred-stack-boundary=4
- else ifneq ($(call cc-option, -mstack-alignment=16),)
- 	cc_stack_align := -mstack-alignment=16
- endif
 -CFLAGS_dcn21_resource.o := -mhard-float -msse
++CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse
  
- CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse $(cc_stack_align)
+ ifdef CONFIG_CC_IS_GCC
+ ifeq ($(call cc-ifversion, -lt, 0701, y), y)
+ IS_OLD_GCC = 1
+ endif
+ endif
  
- ifdef CONFIG_CC_IS_CLANG
+ ifdef IS_OLD_GCC
+ # Stack alignment mismatch, proceed with caution.
+ # GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
+ # (8B stack alignment).
 -CFLAGS_dcn21_resource.o += -mpreferred-stack-boundary=4
++CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -mpreferred-stack-boundary=4
+ else
 -CFLAGS_dcn21_resource.o += -msse2
 +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2
  endif
  
  AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21))

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-10-28  0:57 Stephen Rothwell
  2019-10-28  1:13 ` Masahiro Yamada
  0 siblings, 1 reply; 29+ messages in thread
From: Stephen Rothwell @ 2019-10-28  0:57 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List,
	Masahiro Yamada, Nick Desaulniers

[-- Attachment #1: Type: text/plain, Size: 2327 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/display/dc/dcn20/Makefile

between commit:

  54b8ae66ae1a ("kbuild: change *FLAGS_<basetarget>.o to take the path relative to $(obj)")

from Linus' tree and commits:

  4f952528add3 ("drm/amdgpu: fix stack alignment ABI mismatch for Clang")
  971463bea55c ("drm/amdgpu: fix stack alignment ABI mismatch for GCC 7.1+")
  101d09f07966 ("drm/amdgpu: enable -msse2 for GCC 7.1+ users")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/display/dc/dcn20/Makefile
index ddb8d5649e79,be3a614963c6..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile
@@@ -10,16 -10,21 +10,21 @@@ ifdef CONFIG_DRM_AMD_DC_DSC_SUPPOR
  DCN20 += dcn20_dsc.o
  endif
  
- ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
- 	cc_stack_align := -mpreferred-stack-boundary=4
- else ifneq ($(call cc-option, -mstack-alignment=16),)
- 	cc_stack_align := -mstack-alignment=16
- endif
 -CFLAGS_dcn20_resource.o := -mhard-float -msse
++CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse
  
- CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse $(cc_stack_align)
+ ifdef CONFIG_CC_IS_GCC
+ ifeq ($(call cc-ifversion, -lt, 0701, y), y)
+ IS_OLD_GCC = 1
+ endif
+ endif
  
- ifdef CONFIG_CC_IS_CLANG
+ ifdef IS_OLD_GCC
+ # Stack alignment mismatch, proceed with caution.
+ # GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
+ # (8B stack alignment).
 -CFLAGS_dcn20_resource.o += -mpreferred-stack-boundary=4
++CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -mpreferred-stack-boundary=4
+ else
 -CFLAGS_dcn20_resource.o += -msse2
 +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2
  endif
  
  AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20))

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-10-03  0:19 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-10-03  0:19 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List,
	Jason Gunthorpe, Yong Zhao

[-- Attachment #1: Type: text/plain, Size: 1258 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdkfd/kfd_priv.h

between commit:

  471f39020569 ("drm/amdkfd: use mmu_notifier_put")

from Linus' tree and commit:

  4b3cece1863c ("drm/amdkfd: Use hex print format for pasid")

from the amdgpu tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index c89326125d71,fc8b81208368..000000000000
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@@ -687,7 -688,10 +688,7 @@@ struct kfd_process 
  	/* We want to receive a notification when the mm_struct is destroyed */
  	struct mmu_notifier mmu_notifier;
  
- 	unsigned int pasid;
 -	/* Use for delayed freeing of kfd_process structure */
 -	struct rcu_head	rcu;
 -
+ 	uint16_t pasid;
  	unsigned int doorbell_index;
  
  	/*

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-06-17  3:12 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-06-17  3:12 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List, Yintian Tao,
	Prike Liang, Trigger Huang

[-- Attachment #1: Type: text/plain, Size: 1090 bytes --]

Hi Alex,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c

between commits:

  192905989302 ("drm/amd/amdgpu: add RLC firmware to support raven1 refresh")
  f3a5231c8f14 ("drm/amdgpu: return 0 by default in amdgpu_pm_load_smu_firmware")

from Linus' tree and commits:

  80f41f84ae2c ("drm/amd/amdgpu: add RLC firmware to support raven1 refresh")
  4a39ec6ac5f3 ("drm/amdgpu: fix pm_load_smu_firmware for SR-IOV")
  e9bc1bf7916e ("drm/amdgpu: register pm sysfs for sriov (v2)")
  0079f82e710c ("drm/amdgpu: return 0 by default in amdgpu_pm_load_smu_firmware")

from the amdgpu tree.

I fixed it up (I used the latter version) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.



-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* linux-next: manual merge of the amdgpu tree with Linus' tree
@ 2019-03-18 23:55 Stephen Rothwell
  0 siblings, 0 replies; 29+ messages in thread
From: Stephen Rothwell @ 2019-03-18 23:55 UTC (permalink / raw)
  To: Alex Deucher
  Cc: Linux Next Mailing List, Linux Kernel Mailing List,
	Nicholas Kazlauskas, Dave Airlie, Mario Kleiner, David Francis

[-- Attachment #1: Type: text/plain, Size: 917 bytes --]

Hi all,

Today's linux-next merge of the amdgpu tree got a conflict in:

  drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

between commits:

  d63716658ac1 ("drm/amd/display: Use vrr friendly pageflip throttling in DC.")
  634092b1b9f6 ("drm/amd/display: Use vrr friendly pageflip throttling in DC.")

from Linus' tree and commit:

  7b19bba58f77 ("drm/amd/display: Use vrr friendly pageflip throttling in DC.")

and following commits from the amdgpu tree.

I fixed it up (I used the version from the amdgpu tree) and can carry the
fix as necessary. This is now fixed as far as linux-next is concerned,
but any non trivial conflicts should be mentioned to your upstream
maintainer when your tree is submitted for merging.  You may also want
to consider cooperating with the maintainer of the conflicting tree to
minimise any particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

end of thread, other threads:[~2022-01-26 23:32 UTC | newest]

Thread overview: 29+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-02 15:10 linux-next: manual merge of the amdgpu tree with Linus' tree Mark Brown
2021-08-02 15:48 ` Mark Brown
  -- strict thread matches above, loose matches on Subject: below --
2022-01-26 23:32 Stephen Rothwell
2022-01-26 23:29 Stephen Rothwell
2022-01-11 23:09 Stephen Rothwell
2021-09-02  1:22 Stephen Rothwell
2021-09-10  1:16 ` Stephen Rothwell
2021-07-06  0:52 Stephen Rothwell
2021-01-15  0:53 Stephen Rothwell
2021-01-08  1:04 Stephen Rothwell
2020-11-03  3:10 Stephen Rothwell
2020-10-27  0:48 Stephen Rothwell
2020-10-16  1:48 Stephen Rothwell
2020-08-25  1:35 Stephen Rothwell
2020-07-22  4:34 Stephen Rothwell
2020-07-23  8:01 ` Daniel Vetter
2020-07-23 13:34   ` Alex Deucher
2019-12-12  0:03 Stephen Rothwell
2019-12-12  0:11 ` Stephen Rothwell
2019-12-12  0:16 ` Timothy Pearson
2019-12-11 23:37 Stephen Rothwell
2019-12-12  0:10 ` Stephen Rothwell
2019-10-28  1:11 Stephen Rothwell
2019-10-28  1:03 Stephen Rothwell
2019-10-28  0:57 Stephen Rothwell
2019-10-28  1:13 ` Masahiro Yamada
2019-10-03  0:19 Stephen Rothwell
2019-06-17  3:12 Stephen Rothwell
2019-03-18 23:55 Stephen Rothwell

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).