<!DOCTYPE html><html><head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> </head> <body> <div class="moz-cite-prefix">On 7/12/2024 5:55 AM, Nirmoy Das wrote: </div> <blockquote type="cite" cite="mid:29306f83-6597-4627-a296-88e9f47edaeb@linux.intel.com"> Hi Akshata, <div class="moz-cite-prefix">On 7/12/2024 8:39 AM, Akshata Jahagirdar wrote: </div> <blockquote type="cite" cite="mid:c852a9895c203552169a86bc2cdca909340961e0.1720766122.git.akshata.jahagirdar@intel.com"> <pre class="moz-quote-pre" wrap="">From: "Jahagirdar, Akshata" <a class="moz-txt-link-rfc2396E" href="mailto:akshata.jahagirdar@intel.com" moz-do-not-send="true"><akshata.jahagirdar@intel.com></a> During eviction (vram->sysmem), we use the mapping from compressed -> uncompressed. During restore (sysmem->vram), we need the mapping from uncompressed -> uncompressed. Handle logic for selecting the compressed identity map for eviction, and selecting uncompressed map for restore operations. Signed-off-by: Akshata Jahagirdar <a class="moz-txt-link-rfc2396E" href="mailto:akshata.jahagirdar@intel.com" moz-do-not-send="true"><akshata.jahagirdar@intel.com></a> Reviewed-by: Matthew Auld <a class="moz-txt-link-rfc2396E" href="mailto:matthew.auld@intel.com" moz-do-not-send="true"><matthew.auld@intel.com></a> Reviewed-by: Himal Prasad Ghimiray <a class="moz-txt-link-rfc2396E" href="mailto:himal.prasad.ghimiray@intel.com" moz-do-not-send="true"><himal.prasad.ghimiray@intel.com></a> --- drivers/gpu/drm/xe/xe_migrate.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index 3b8a334fe08f..1eee6be24423 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -715,7 +715,7 @@ static u32 xe_migrate_ccs_copy(struct xe_migrate *m, struct xe_gt *gt = m->tile->primary_gt; u32 flush_flags = 0; - if (xe_device_has_flat_ccs(gt_to_xe(gt)) && !copy_ccs && dst_is_indirect) { + if (xe_device_needs_ccs_emit(gt_to_xe(gt)) && !copy_ccs && dst_is_indirect) {</pre> </blockquote> Do we need to call xe_migrate_ccs_copy() when xe_device_needs_ccs_emit() is true ? If not then move the check for xe_device_needs_ccs_emit() to xe_migrate_copy() to ensure that xe_migrate_ccs_copy() is not even called on such platforms. <pre class="moz-quote-pre" wrap="">Regards, Nirmoy</pre> </blockquote> xe_device_needs_ccs_emit() is true for any non-BMG platform that has flat-ccs enabled. In such cases, we do need to call xe_migrate_ccs_copy() while clearing any ccs metadata in vram. BR, Akshata <blockquote type="cite" cite="mid:29306f83-6597-4627-a296-88e9f47edaeb@linux.intel.com"> <blockquote type="cite" cite="mid:c852a9895c203552169a86bc2cdca909340961e0.1720766122.git.akshata.jahagirdar@intel.com"> <pre class="moz-quote-pre" wrap=""> /* * If the src is already in vram, then it should already * have been cleared by us, or has been populated by the @@ -791,6 +791,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, bool copy_ccs = xe_device_has_flat_ccs(xe) && xe_bo_needs_ccs_pages(src_bo) && xe_bo_needs_ccs_pages(dst_bo); bool copy_system_ccs = copy_ccs && (!src_is_vram || !dst_is_vram); + bool use_comp_pat = GRAPHICS_VER(xe) >= 20 && IS_DGFX(xe) && src_is_vram && !dst_is_vram; /* Copying CCS between two different BOs is not supported yet. */ if (XE_WARN_ON(copy_ccs && src_bo != dst_bo)) @@ -833,7 +834,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, src_L0 = min(src_L0, dst_L0); - batch_size += pte_update_size(m, src_is_vram, false, src, &src_it, &src_L0, + batch_size += pte_update_size(m, src_is_vram, use_comp_pat, src, &src_it, &src_L0, &src_L0_ofs, &src_L0_pt, 0, 0, avail_pts); @@ -852,7 +853,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, /* Add copy commands size here */ batch_size += ((copy_only_ccs) ? 0 : EMIT_COPY_DW) + - ((xe_device_has_flat_ccs(xe) ? EMIT_COPY_CCS_DW : 0)); + ((xe_device_needs_ccs_emit(xe) ? EMIT_COPY_CCS_DW : 0)); bb = xe_bb_new(gt, batch_size, usm); if (IS_ERR(bb)) { </pre> </blockquote> </blockquote> </body> </html>