Re: [PATCH v2 08/29] drm/xe: Add dma_addr res cursor

[Matthew Brost <matthew.brost@intel.com>]

On Tue, Nov 19, 2024 at 01:15:12PM +0100, Thomas Hellström wrote:
> On Tue, 2024-10-15 at 20:24 -0700, Matthew Brost wrote:
> > From: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> > 
> > Useful for SVM ranges in SRAM and programing page tables.
> 
> We should look at providing a better commit message.
> 

Yes, this is pretty poor. Will scrub all commit messages in next and
ensure higher quality.

Matt

> > 
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> > ---
> >  drivers/gpu/drm/xe/xe_res_cursor.h | 116
> > ++++++++++++++++++++++++++++-
> >  drivers/gpu/drm/xe/xe_svm.h        |   4 +
> >  2 files changed, 118 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/xe/xe_res_cursor.h
> > b/drivers/gpu/drm/xe/xe_res_cursor.h
> > index dca374b6521c..3faa3d9adb82 100644
> > --- a/drivers/gpu/drm/xe/xe_res_cursor.h
> > +++ b/drivers/gpu/drm/xe/xe_res_cursor.h
> > @@ -30,13 +30,18 @@
> >  #include <drm/ttm/ttm_range_manager.h>
> >  #include <drm/ttm/ttm_resource.h>
> >  #include <drm/ttm/ttm_tt.h>
> > +#include "drm_pagemap.h"
> >  
> >  #include "xe_bo.h"
> >  #include "xe_device.h"
> >  #include "xe_macros.h"
> > +#include "xe_svm.h"
> >  #include "xe_ttm_vram_mgr.h"
> >  
> > -/* state back for walking over vram_mgr, stolen_mgr, and gtt_mgr
> > allocations */
> > +/**
> > + * struct xe_res_cursor - state for walking over vram_mgr,
> > stolen_mgr,
> > + * and gtt_mgr allocations
> > + */
> >  struct xe_res_cursor {
> >  	u64 start;
> >  	u64 size;
> > @@ -44,7 +49,17 @@ struct xe_res_cursor {
> >  	void *node;
> >  	u32 mem_type;
> >  	struct scatterlist *sgl;
> > +	/** @dma_addr: Current element in a struct
> > drm_pagemap_dma_addr array */
> > +	const struct drm_pagemap_dma_addr *dma_addr;
> >  	struct drm_buddy *mm;
> > +	/**
> > +	 * @dma_start: DMA start address for the current segment.
> > +	 * This may be different to @dma_addr.addr since elements in
> > +	 * the array may be coalesced to a single segment.
> > +	 */
> > +	u64 dma_start;
> > +	/** @dma_seg_size: Size of the current segment. */
> > +	u64 dma_seg_size;
> >  };
> >  
> >  static struct drm_buddy *xe_res_get_buddy(struct ttm_resource *res)
> > @@ -70,6 +85,7 @@ static inline void xe_res_first(struct ttm_resource
> > *res,
> >  				struct xe_res_cursor *cur)
> >  {
> >  	cur->sgl = NULL;
> > +	cur->dma_addr = NULL;
> >  	if (!res)
> >  		goto fallback;
> >  
> > @@ -141,6 +157,36 @@ static inline void __xe_res_sg_next(struct
> > xe_res_cursor *cur)
> >  	cur->sgl = sgl;
> >  }
> >  
> > +/**
> > + * __xe_res_dma_next() - Advance the cursor when end-of-segment is
> > reached
> > + * @cur: The cursor
> > + */
> > +static inline void __xe_res_dma_next(struct xe_res_cursor *cur)
> > +{
> > +	const struct drm_pagemap_dma_addr *addr = cur->dma_addr;
> > +	u64 start = cur->start;
> > +
> > +	while (start >= cur->dma_seg_size) {
> > +		start -= cur->dma_seg_size;
> > +		addr++;
> > +		cur->dma_seg_size = PAGE_SIZE << addr->order;
> > +	}
> > +	cur->dma_start = addr->addr;
> > +
> > +	/* Coalesce array_elements */
> > +	while (cur->dma_seg_size - start < cur->remaining) {
> > +		if (cur->dma_start + cur->dma_seg_size !=
> > addr[1].addr ||
> > +		    addr->proto != addr[1].proto)
> > +			break;
> > +		addr++;
> > +		cur->dma_seg_size += PAGE_SIZE << addr->order;
> > +	}
> > +
> > +	cur->dma_addr = addr;
> > +	cur->start = start;
> > +	cur->size = cur->dma_seg_size - start;
> > +}
> > +
> >  /**
> >   * xe_res_first_sg - initialize a xe_res_cursor with a scatter
> > gather table
> >   *
> > @@ -160,11 +206,42 @@ static inline void xe_res_first_sg(const struct
> > sg_table *sg,
> >  	cur->start = start;
> >  	cur->remaining = size;
> >  	cur->size = 0;
> > +	cur->dma_addr = NULL;
> >  	cur->sgl = sg->sgl;
> >  	cur->mem_type = XE_PL_TT;
> >  	__xe_res_sg_next(cur);
> >  }
> >  
> > +/**
> > + * xe_res_first_dma - initialize a xe_res_cursor with dma_addr array
> > + *
> > + * @dma_addr: struct drm_pagemap_dma_addr array to walk
> > + * @start: Start of the range
> > + * @size: Size of the range
> > + * @cur: cursor object to initialize
> > + *
> > + * Start walking over the range of allocations between @start and
> > @size.
> > + */
> > +static inline void xe_res_first_dma(const struct
> > drm_pagemap_dma_addr *dma_addr,
> > +				    u64 start, u64 size,
> > +				    struct xe_res_cursor *cur)
> > +{
> > +	XE_WARN_ON(!dma_addr);
> > +	XE_WARN_ON(!IS_ALIGNED(start, PAGE_SIZE) ||
> > +		   !IS_ALIGNED(size, PAGE_SIZE));
> > +
> > +	cur->node = NULL;
> > +	cur->start = start;
> > +	cur->remaining = size;
> > +	cur->dma_seg_size = PAGE_SIZE << dma_addr->order;
> > +	cur->dma_start = 0;
> > +	cur->size = 0;
> > +	cur->dma_addr = dma_addr;
> > +	__xe_res_dma_next(cur);
> > +	cur->sgl = NULL;
> > +	cur->mem_type = XE_PL_TT;
> > +}
> > +
> >  /**
> >   * xe_res_next - advance the cursor
> >   *
> > @@ -191,6 +268,12 @@ static inline void xe_res_next(struct
> > xe_res_cursor *cur, u64 size)
> >  		return;
> >  	}
> >  
> > +	if (cur->dma_addr) {
> > +		cur->start += size;
> > +		__xe_res_dma_next(cur);
> > +		return;
> > +	}
> > +
> >  	if (cur->sgl) {
> >  		cur->start += size;
> >  		__xe_res_sg_next(cur);
> > @@ -232,6 +315,35 @@ static inline void xe_res_next(struct
> > xe_res_cursor *cur, u64 size)
> >   */
> >  static inline u64 xe_res_dma(const struct xe_res_cursor *cur)
> >  {
> > -	return cur->sgl ? sg_dma_address(cur->sgl) + cur->start :
> > cur->start;
> > +	if (cur->dma_addr)
> > +		return cur->dma_start + cur->start;
> > +	else if (cur->sgl)
> > +		return sg_dma_address(cur->sgl) + cur->start;
> > +	else
> > +		return cur->start;
> > +}
> > +
> > +/**
> > + * xe_res_is_vram() - Whether the cursor current dma address points
> > to
> > + * same-device VRAM
> > + * @cur: The cursor.
> > + *
> > + * Return: true iff the address returned by xe_res_dma() points to
> > internal vram.
> > + */
> > +static inline bool xe_res_is_vram(const struct xe_res_cursor *cur)
> > +{
> > +	if (cur->dma_addr)
> > +		return cur->dma_addr->proto == XE_INTERCONNECT_VRAM;
> > +
> > +	switch (cur->mem_type) {
> > +	case XE_PL_STOLEN:
> > +	case XE_PL_VRAM0:
> > +	case XE_PL_VRAM1:
> > +		return true;
> > +	default:
> > +		break;
> > +	}
> > +
> > +	return false;
> >  }
> >  #endif
> > diff --git a/drivers/gpu/drm/xe/xe_svm.h
> > b/drivers/gpu/drm/xe/xe_svm.h
> > index 979f2322eeba..376e86876a11 100644
> > --- a/drivers/gpu/drm/xe/xe_svm.h
> > +++ b/drivers/gpu/drm/xe/xe_svm.h
> > @@ -6,6 +6,10 @@
> >  #ifndef _XE_SVM_H_
> >  #define _XE_SVM_H_
> >  
> > +#include "drm_pagemap.h"
> > +
> > +#define XE_INTERCONNECT_VRAM DRM_INTERCONNECT_DRIVER
> > +
> >  struct xe_vm;
> >  
> >  int xe_svm_init(struct xe_vm *vm);
>