search for: ttm_dma_populate

v2 was doing some pretty nasty things with the DMA API, so I took a different approach for this v3. As suggested, this version uses ttm_dma_populate() to populate BOs. The reason for doing this was that it would entitle us to using the DMA sync functions, but since the memory returned is already coherent anyway, we do not even need to call these functions anymore. So this series has turned into 2 small patches: - The first attempts to make it...

...-rc8 #1 [ +0.000003] Hardware name: Micro-Star International Co., Ltd. GX780/GT780/MS-1761, BIOS E1761IMS V3.01 05/02/2011 [ +0.000003] Call Trace: [ +0.000009] dump_stack+0x9f/0xe1 [ +0.000008] swiotlb_alloc_coherent+0xdf/0x150 [ +0.000010] ttm_dma_pool_get_pages+0x1ec/0x4b0 [ +0.000015] ttm_dma_populate+0x24c/0x340 [ +0.000011] ttm_tt_bind+0x23/0x50 [ +0.000006] ttm_bo_handle_move_mem+0x58c/0x5c0 [ +0.000015] ttm_bo_validate+0x152/0x190 [ +0.000004] ? ttm_bo_init_reserved+0x3d8/0x490 [ +0.000012] ? mutex_trylock+0xcd/0xe0 [ +0.000004] ? ttm_bo_handle_move_mem+0x58/0x5c0 [ +0.000007]...

On Thu, 2018-05-10 at 12:28 +0200, Mike Galbraith wrote: > On Thu, 2018-05-10 at 11:10 +0200, Mike Galbraith wrote: > > Greetings, > > > > When box is earning its keep, nouveau/swiotlb grumble.. a LOT. The > > below is from master.today. > > > > [12594.640959] nouveau 0000:01:00.0: swiotlb buffer is full (sz: 2097152 bytes) > > [12594.693000] nouveau

...ar International Co., Ltd. GX780/GT780/MS-1761, BIOS E1761IMS V3.01 05/02/2011 >> [ +0.000003] Call Trace: >> [ +0.000009] dump_stack+0x9f/0xe1 >> [ +0.000008] swiotlb_alloc_coherent+0xdf/0x150 >> [ +0.000010] ttm_dma_pool_get_pages+0x1ec/0x4b0 >> [ +0.000015] ttm_dma_populate+0x24c/0x340 >> [ +0.000011] ttm_tt_bind+0x23/0x50 >> [ +0.000006] ttm_bo_handle_move_mem+0x58c/0x5c0 >> [ +0.000015] ttm_bo_validate+0x152/0x190 >> [ +0.000004] ? ttm_bo_init_reserved+0x3d8/0x490 >> [ +0.000012] ? mutex_trylock+0xcd/0xe0 >> [ +0.000004...

...ON MS-7848/MS-7848, BIOS M7848W08.20C 09/23/2013 [ 1313.811041] Call Trace: [ 1313.811049] dump_stack+0x7c/0xb6 [ 1313.811053] swiotlb_alloc_coherent+0x13f/0x150 [ 1313.811060] ttm_dma_pool_alloc_new_pages+0x106/0x3c0 [ttm] [ 1313.811066] ttm_dma_pool_get_pages+0x10a/0x1e0 [ttm] [ 1313.811070] ttm_dma_populate+0x21f/0x2f0 [ttm] [ 1313.811075] ttm_tt_bind+0x2f/0x60 [ttm] [ 1313.811079] ttm_bo_handle_move_mem+0x51f/0x580 [ttm] [ 1313.811084] ? ttm_bo_handle_move_mem+0x5/0x580 [ttm] [ 1313.811088] ttm_bo_validate+0x10c/0x120 [ttm] [ 1313.811092] ? ttm_bo_validate+0x5/0x120 [ttm] [ 1313.811106] ? drm_m...

...3acc35e3805..2cdaea58678d 100644 --- a/drm/nouveau/nouveau_bo.c +++ b/drm/nouveau/nouveau_bo.c @@ -1502,7 +1502,7 @@ nouveau_ttm_tt_populate(struct ttm_tt *ttm) } #endif -#ifdef CONFIG_SWIOTLB +#if IS_ENABLED(CONFIG_SWIOTLB) && IS_ENABLED(CONFIG_X86) if (swiotlb_nr_tbl()) { return ttm_dma_populate((void *)ttm, dev->dev); } @@ -1570,7 +1570,7 @@ nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm) } #endif -#ifdef CONFIG_SWIOTLB +#if IS_ENABLED(CONFIG_SWIOTLB) && IS_ENABLED(CONFIG_X86) if (swiotlb_nr_tbl()) { ttm_dma_unpopulate((void *)ttm, dev->dev); return; -- 2.7.2

...W08.20C 09/23/2013 > [ 1313.811041] Call Trace: > [ 1313.811049] dump_stack+0x7c/0xb6 > [ 1313.811053] swiotlb_alloc_coherent+0x13f/0x150 > [ 1313.811060] ttm_dma_pool_alloc_new_pages+0x106/0x3c0 [ttm] > [ 1313.811066] ttm_dma_pool_get_pages+0x10a/0x1e0 [ttm] > [ 1313.811070] ttm_dma_populate+0x21f/0x2f0 [ttm] > [ 1313.811075] ttm_tt_bind+0x2f/0x60 [ttm] > [ 1313.811079] ttm_bo_handle_move_mem+0x51f/0x580 [ttm] > [ 1313.811084] ? ttm_bo_handle_move_mem+0x5/0x580 [ttm] > [ 1313.811088] ttm_bo_validate+0x10c/0x120 [ttm] > [ 1313.811092] ? ttm_bo_validate+0x5/0x120 [ttm...

...dware name: Micro-Star International Co., Ltd. GX780/GT780/MS-1761, BIOS E1761IMS V3.01 05/02/2011 > [ +0.000003] Call Trace: > [ +0.000009] dump_stack+0x9f/0xe1 > [ +0.000008] swiotlb_alloc_coherent+0xdf/0x150 > [ +0.000010] ttm_dma_pool_get_pages+0x1ec/0x4b0 > [ +0.000015] ttm_dma_populate+0x24c/0x340 > [ +0.000011] ttm_tt_bind+0x23/0x50 > [ +0.000006] ttm_bo_handle_move_mem+0x58c/0x5c0 > [ +0.000015] ttm_bo_validate+0x152/0x190 > [ +0.000004] ? ttm_bo_init_reserved+0x3d8/0x490 > [ +0.000012] ? mutex_trylock+0xcd/0xe0 > [ +0.000004] ? ttm_bo_handle_move_...

...ed 4.16.0-0.rc2.git0.1.fc28.x86_64 #1 [ 1035.437024] Hardware name: NVIDIA MCP7A/MCP7A, BIOS 6.00 PG 04/22/2009 [ 1035.437025] Call Trace: [ 1035.437036] dump_stack+0x5c/0x85 [ 1035.437040] swiotlb_alloc_coherent+0x1c3/0x1e0 [ 1035.437052] ttm_dma_pool_get_pages+0x21a/0x620 [ttm] [ 1035.437057] ttm_dma_populate+0xdd/0x390 [ttm] [ 1035.437062] ttm_tt_bind+0x2e/0x60 [ttm] [ 1035.437067] ttm_bo_handle_move_mem+0x4cf/0x550 [ttm] [ 1035.437073] ttm_bo_validate+0x119/0x130 [ttm] [ 1035.437104] ? drm_get_edid_switcheroo+0x16/0x40 [drm] [ 1035.437109] ttm_bo_init_reserved+0x334/0x380 [ttm] [ 1035.437114] ?...

...to an ever repeating warning that the caller of swiotlb_alloc() explicitly asked to be squelched. Pass the caller's request for silence onward. Xorg-3170 [006] .... 963.866098: swiotlb_alloc+0x1d/0x1a0: gfp & __GFP_NOWARN Xorg-3170 [006] .... 963.866101: <stack trace> => ttm_dma_populate+0x250/0x310 [ttm] => ttm_tt_populate+0x28/0x70 [ttm] => ttm_tt_bind+0x26/0x60 [ttm] => ttm_bo_handle_move_mem+0x51a/0x580 [ttm] => ttm_bo_validate+0xfa/0x110 [ttm] => ttm_bo_init_reserved+0x296/0x450 [ttm] => ttm_bo_init+0x73/0xd0 [ttm] => nouveau_bo_new+0x3eb/0x5c0 [nou...

...6_64 #1 > [ 1035.437024] Hardware name: NVIDIA MCP7A/MCP7A, BIOS 6.00 PG 04/22/2009 > [ 1035.437025] Call Trace: > [ 1035.437036] dump_stack+0x5c/0x85 > [ 1035.437040] swiotlb_alloc_coherent+0x1c3/0x1e0 > [ 1035.437052] ttm_dma_pool_get_pages+0x21a/0x620 [ttm] > [ 1035.437057] ttm_dma_populate+0xdd/0x390 [ttm] > [ 1035.437062] ttm_tt_bind+0x2e/0x60 [ttm] > [ 1035.437067] ttm_bo_handle_move_mem+0x4cf/0x550 [ttm] > [ 1035.437073] ttm_bo_validate+0x119/0x130 [ttm] > [ 1035.437104] ? drm_get_edid_switcheroo+0x16/0x40 [drm] > [ 1035.437109] ttm_bo_init_reserved+0x334/0x380...

...name: MEDION MS-7848/MS-7848, BIOS M7848W08.20C 09/23/2013 [ 6253.341543] Call Trace: [ 6253.341553] dump_stack+0x78/0xb3 [ 6253.341559] swiotlb_alloc+0x134/0x170 [ 6253.341567] ttm_dma_pool_alloc_new_pages+0x161/0x3c0 [ttm] [ 6253.341574] ttm_dma_pool_get_pages+0xe0/0x1c0 [ttm] [ 6253.341580] ttm_dma_populate+0x250/0x310 [ttm] [ 6253.341586] ttm_tt_populate+0x28/0x70 [ttm] [ 6253.341591] ttm_tt_bind+0x26/0x60 [ttm] [ 6253.341596] ttm_bo_handle_move_mem+0x51a/0x580 [ttm] [ 6253.341612] ? drm_mm_insert_node_in_range+0x42b/0x480 [drm] [ 6253.341617] ttm_bo_validate+0xfa/0x110 [ttm] [ 6253.341622] ? _...

...m) dev = drm->dev; pdev = device->dev; - /* - * Objects matching this condition have been marked as force_coherent, - * so use the DMA API for them. - */ - if (!nvxx_device(&drm->device)->func->cpu_coherent && - ttm->caching_state == tt_uncached) - return ttm_dma_populate(ttm_dma, dev->dev); - #if IS_ENABLED(CONFIG_AGP) if (drm->agp.bridge) { return ttm_agp_tt_populate(ttm); @@ -1557,16 +1516,6 @@ nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm) dev = drm->dev; pdev = device->dev; - /* - * Objects matching this condition have been marked as for...

It has been a couple of months since v4 - apologies for this. v4 has not received many comments, but this version addresses them and makes a new attempt at pushing the critical bit for GK20A and Nouveau on ARM in general. As a reminder, this series addresses the memory coherency issue that we are seeing on ARM platforms. Contrary to x86 which invalidates the PCI caches whenever a write is made by

https://bugs.freedesktop.org/show_bug.cgi?id=81485 Priority: medium Bug ID: 81485 Assignee: nouveau at lists.freedesktop.org Summary: dmesg spam on startup QA Contact: xorg-team at lists.x.org Severity: normal Classification: Unclassified OS: All Reporter: fry.kun at gmail.com Hardware: Other

Another revision of this patchset critical for GK20A to operate. Previous attempts were exclusively using either TTM's regular page allocator or the DMA API one. Both have their advantages and drawbacks: the page allocator is fast but requires explicit synchronization on non-coherent architectures, whereas the DMA allocator always returns coherent memory, but is also slower, creates a

Hi guys, Well let's face it AGP is a total headache to maintain and dead for at least 10+ years. We have a lot of x86 specific stuff in the architecture independent graphics memory management to get the caching right, abusing the DMA API on multiple occasions, need to distinct between AGP and driver specific page tables etc etc... So the idea here is to just go ahead and remove the support