Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
To: igt-dev@lists.freedesktop.org
Cc: tursulin@igalia.com, zbigniew.kempczynski@intel.com,
	kamil.konieczny@linux.intel.com, lukasz.laguna@intel.com
Subject: [PATCH v3 i-g-t 2/2] benchmarks/gem_wsim: Fix mmap for discrete graphics cards
Date: Thu, 25 Jul 2024 14:26:12 +0200	[thread overview]
Message-ID: <20240725122612.1290090-3-marcin.bernatowicz@linux.intel.com> (raw)
In-Reply-To: <20240725122612.1290090-1-marcin.bernatowicz@linux.intel.com>

It appears that 'gem_mmap__wc' doesn't work for discrete graphics cards
and 'I915_MMAP_OFFSET_FIXED' is needed instead.
Adopt the mapping approach from 'lib/igt_dummyload'.

v2: Correct brace balance (Kamil)
    Remove redundant 'ptr' variable

Signed-off-by: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>
---
 benchmarks/gem_wsim.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/benchmarks/gem_wsim.c b/benchmarks/gem_wsim.c
index cd1a01a67..af30a22b3 100644
--- a/benchmarks/gem_wsim.c
+++ b/benchmarks/gem_wsim.c
@@ -1494,7 +1494,7 @@ static unsigned int create_bb(struct w_step *w, int self)
 #define TIMESTAMP (base + 0x3a8)
 	const int use_64b = gen >= 8;
 	enum { START_TS, NOW_TS };
-	uint32_t *ptr, *cs, *jmp;
+	uint32_t *cs, *jmp;
 	unsigned int r = 0;
 
 	/* Loop until CTX_TIMESTAMP - initial > target ns */
@@ -1502,7 +1502,16 @@ static unsigned int create_bb(struct w_step *w, int self)
 	gem_set_domain(fd, w->bb_handle,
 		       I915_GEM_DOMAIN_WC, I915_GEM_DOMAIN_WC);
 
-	cs = ptr = gem_mmap__wc(fd, w->bb_handle, 0, w->bb_size, PROT_WRITE);
+	if (__gem_set_caching(fd, w->bb_handle, I915_CACHING_CACHED) == 0) {
+		cs = gem_mmap__cpu(fd, w->bb_handle,
+				   0, w->bb_size,
+				   PROT_READ | PROT_WRITE);
+	} else {
+		cs = gem_mmap__device_coherent(fd,
+					       w->bb_handle,
+					       0, w->bb_size,
+					       PROT_READ | PROT_WRITE);
+	}
 
 	/* Store initial 64b timestamp: start */
 	*cs++ = MI_LOAD_REGISTER_IMM(1) | MI_CS_MMIO_DST;
-- 
2.31.1


  parent reply	other threads:[~2024-07-25 12:26 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-25 12:26 [PATCH v3 i-g-t 0/2] benchmarks/gem_wsim: support gens without relocations and mmap fix Marcin Bernatowicz
2024-07-25 12:26 ` [PATCH v3 i-g-t 1/2] benchmarks/gem_wsim: Support gens without relocations Marcin Bernatowicz
2024-07-25 12:26 ` Marcin Bernatowicz [this message]
2024-07-25 13:22 ` ✓ Fi.CI.BAT: success for benchmarks/gem_wsim: support gens without relocations and mmap fix (rev2) Patchwork
2024-07-25 13:38 ` ✓ CI.xeBAT: " Patchwork
2024-07-25 16:17 ` ✓ CI.xeFULL: " Patchwork
2024-07-26  0:44 ` ✗ Fi.CI.IGT: failure " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240725122612.1290090-3-marcin.bernatowicz@linux.intel.com \
    --to=marcin.bernatowicz@linux.intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=kamil.konieczny@linux.intel.com \
    --cc=lukasz.laguna@intel.com \
    --cc=tursulin@igalia.com \
    --cc=zbigniew.kempczynski@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox