<!DOCTYPE html><html><head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> </head> <body> Hi Michal, <div class="moz-cite-prefix">On 6/25/2024 9:55 PM, Michal Wajdeczko wrote: </div> <blockquote type="cite" cite="mid:7c2a77ab-64ab-4244-b533-cdbe0a51bcc7@intel.com"> <pre class="moz-quote-pre" wrap="">On 25.06.2024 10:49, Nirmoy Das wrote: </pre> <blockquote type="cite"> <pre class="moz-quote-pre" wrap="">GuC TLB invalidation depends on GuC to process the request from the CT queue and then the real time to invalidate TLB. Add a function to return overestimated possible time a TLB inval H2G might take which can be used as timeout value for TLB invalidation wait time. Closes: <a class="moz-txt-link-freetext" href="https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1622">https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1622</a> Cc: Matthew Brost <a class="moz-txt-link-rfc2396E" href="mailto:matthew.brost@intel.com"><matthew.brost@intel.com></a> Suggested-by: Daniele Ceraolo Spurio <a class="moz-txt-link-rfc2396E" href="mailto:daniele.ceraolospurio@intel.com"><daniele.ceraolospurio@intel.com></a> Signed-off-by: Nirmoy Das <a class="moz-txt-link-rfc2396E" href="mailto:nirmoy.das@intel.com"><nirmoy.das@intel.com></a> --- drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 2 +- drivers/gpu/drm/xe/xe_guc_ct.c | 12 ++++++++++++ drivers/gpu/drm/xe/xe_guc_ct.h | 2 ++ 3 files changed, 15 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c index e1f1ccb01143..fa61070d6201 100644 --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c +++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c @@ -17,7 +17,7 @@ #include "xe_trace.h" #include "regs/xe_guc_regs.h" -#define TLB_TIMEOUT (HZ / 4) +#define TLB_TIMEOUT xe_guc_tlb_timeout_jiffies() </pre> </blockquote> <pre class="moz-quote-pre" wrap="">if this is no longer a macro then maybe better get rid of it and use function call directly ?</pre> </blockquote> Wanted to keep the diff to minimal. <blockquote type="cite" cite="mid:7c2a77ab-64ab-4244-b533-cdbe0a51bcc7@intel.com"> <blockquote type="cite"> <pre class="moz-quote-pre" wrap=""> static void xe_gt_tlb_fence_timeout(struct work_struct *work) { diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index b4137fe195a4..e30c0da86acc 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -112,6 +112,18 @@ ct_to_xe(struct xe_guc_ct *ct) #define CTB_G2H_BUFFER_SIZE (4 * CTB_H2G_BUFFER_SIZE) #define G2H_ROOM_BUFFER_SIZE (CTB_G2H_BUFFER_SIZE / 4) +/** + * xe_guc_tlb_timeout_jiffies - Calculate the maximum time to process a tlb inval command + * + * This function computes the maximum time to process a tlb inval H2G commands + * in jiffies. A 4KB buffer full of commands takes a little over a second to process, </pre> </blockquote> <pre class="moz-quote-pre" wrap="">these numbers looks like from experiment - what commands did you use? do we have SLA from the GuC for max processing time of any command?</pre> </blockquote> This is from i915 which was added with <a class="moz-txt-link-freetext" href="https://patchwork.freedesktop.org/patch/563102/">https://patchwork.freedesktop.org/patch/563102/</a> <blockquote type="cite" cite="mid:7c2a77ab-64ab-4244-b533-cdbe0a51bcc7@intel.com"> <blockquote type="cite"> <pre class="moz-quote-pre" wrap="">+ * so this time is set to 2 seconds to be safe. </pre> </blockquote> <pre class="moz-quote-pre" wrap="">nit: missing "Return:" tag </pre> <blockquote type="cite"> <pre class="moz-quote-pre" wrap="">+ */ +long xe_guc_tlb_timeout_jiffies(void) </pre> </blockquote> <pre class="moz-quote-pre" wrap="">this function doesn't fit into xe_guc_ct component at all if you need to reflect CTB size in TLB timeout calculation then: a) from xe_guc_ct export function that will tell you max number of commands it can queue (or are already in the CTB queue) b) in xe_guc_tlb import above function and apply additional timeout per each H2G action in flight</pre> </blockquote> Currently only information we have which is also validated /r-b-ed by GuC team is the above calculation. I can export xe_guc_ct_max_queue_time_jiffies and use that in xe_guc_tlb so the function fits their component. Regards, Nirmoy <blockquote type="cite" cite="mid:7c2a77ab-64ab-4244-b533-cdbe0a51bcc7@intel.com"> <blockquote type="cite"> <pre class="moz-quote-pre" wrap="">+{ + return (CTB_H2G_BUFFER_SIZE * HZ) / SZ_2K; +} + static size_t guc_ct_size(void) { return 2 * CTB_DESC_SIZE + CTB_H2G_BUFFER_SIZE + diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h index 105bb8e99a8d..a9755574d6c9 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.h +++ b/drivers/gpu/drm/xe/xe_guc_ct.h @@ -64,4 +64,6 @@ xe_guc_ct_send_block_no_fail(struct xe_guc_ct *ct, const u32 *action, u32 len) return xe_guc_ct_send_recv_no_fail(ct, action, len, NULL); } +long xe_guc_tlb_timeout_jiffies(void); + #endif </pre> </blockquote> </blockquote> </body> </html>