Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Zbigniew Kempczyński" <zbigniew.kempczynski@intel.com>
To: igt-dev@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Subject: [igt-dev] [PATCH i-g-t v10 08/31] lib/intel_allocator_msgchannel: Scale to 4k of parallel clients
Date: Fri, 20 Nov 2020 12:31:08 +0100	[thread overview]
Message-ID: <20201120113131.67971-9-zbigniew.kempczynski@intel.com> (raw)
In-Reply-To: <20201120113131.67971-1-zbigniew.kempczynski@intel.com>

When playing with multiprocess mode in allocator we're currently
using sysvipc message queues in blocking mode (request/response).
We can calculate then what is maximum depth for the queue for
requested number of children. Change alters kernel queue depth
to cover 4k users (1 is main thread and 4095 are children).

We're still prone to unlimited wait in allocator thread
(more than 4095 children successfully send the messages)
but we're going to address this later.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Cc: Andrzej Turko <andrzej.turko@linux.intel.com>
Cc: Dominik Grzegorzek <dominik.grzegorzek@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
---
 lib/intel_allocator_msgchannel.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/lib/intel_allocator_msgchannel.c b/lib/intel_allocator_msgchannel.c
index 084a977d..9927e44c 100644
--- a/lib/intel_allocator_msgchannel.c
+++ b/lib/intel_allocator_msgchannel.c
@@ -12,6 +12,7 @@ extern __thread pid_t child_tid;
 
 #define FTOK_IGT_ALLOCATOR_KEY "/tmp/igt.allocator.key"
 #define FTOK_IGT_ALLOCATOR_PROJID 2020
+#define MAXQLEN 4096
 
 #define ALLOCATOR_REQUEST 1
 
@@ -57,6 +58,13 @@ static void msgqueue_init(struct msg_channel *channel)
 	queue = msgget(key, IPC_CREAT);
 	igt_debug("msg queue: %d\n", queue);
 
+	igt_assert(msgctl(queue, IPC_STAT, &qstat) == 0);
+	igt_debug("msg size in bytes: %lu\n", qstat.msg_qbytes);
+	qstat.msg_qbytes = MAXQLEN * sizeof(struct msgqueue_buf);
+	igt_debug("resizing queue to support %d requests\n", MAXQLEN);
+	igt_assert_f(msgctl(queue, IPC_SET, &qstat) == 0,
+		     "Couldn't change queue size to %lu\n", qstat.msg_qbytes);
+
 	msgdata = calloc(1, sizeof(*msgdata));
 	igt_assert(msgdata);
 	msgdata->key = key;
-- 
2.26.0

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

  parent reply	other threads:[~2020-11-20 11:31 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-20 11:31 [igt-dev] [PATCH i-g-t v10 00/31] Introduce IGT allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 01/31] lib/igt_list: igt_hlist implementation Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 02/31] lib/igt_map: Introduce igt_map Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 03/31] lib/igt_core: Track child process pid and tid Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 04/31] lib/intel_allocator_simple: Add simple allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 05/31] lib/intel_allocator_random: Add random allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 06/31] lib/intel_allocator: Add intel_allocator core Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 07/31] lib/intel_allocator: Try to stop smoothly instead of deinit Zbigniew Kempczyński
2020-11-20 11:31 ` Zbigniew Kempczyński [this message]
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 09/31] lib/intel_bufops: Removes handle from allocator, change size Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 10/31] lib/intel_bufops: Add init with handle and size function Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 11/31] lib/intel_batchbuffer: Integrate intel_bb with allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 12/31] lib/intel_batchbuffer: Add tracking intel_buf to intel_bb Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 13/31] lib/intel_aux_pgtable: Get addresses for aux table from an allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 14/31] lib/igt_fb: Initialize intel_buf with same size as fb Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 15/31] tests/api_intel_bb: Modify test to verify intel_bb with allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 16/31] tests/api_intel_bb: Add compressed->compressed copy Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 17/31] tests/api_intel_allocator: Simple allocator test suite Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 18/31] tests/gem|kms: Remove intel_bb from fixture Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 19/31] tests/gem_mmap_offset: Use intel_buf wrapper code instead direct Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 20/31] tests/gem_ppgtt: Adopt test to use intel_bb with allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 21/31] tests/gem_render_copy_redux: Adopt to use with intel_bb and allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 22/31] tests/perf.c: Remove buffer from batch Zbigniew Kempczyński
2020-11-20 12:06   ` Lionel Landwerlin
2020-11-20 12:17     ` Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 23/31] tests/gem_linear_blits: Use intel allocator Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 24/31] lib/ioctl_wrappers: Add gem_has_relocations() check Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 25/31] lib/intel_batchbuffer: Use relocations in intel-bb up to gen12 Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 26/31] tests/api_intel_*: Adopt to use relocations as default " Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 27/31] tests/gem_ppgtt: Migrate memory check out of render blits Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 28/31] tests/api_intel_bb: Remove check-canonical test Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 29/31] tests/api_intel_bb: Use allocator in delta-check test Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 30/31] lib/intel_allocator: Separate allocator multiprocess start Zbigniew Kempczyński
2020-11-20 11:31 ` [igt-dev] [PATCH i-g-t v10 31/31] tests/api_intel_allocator: Prepare to run with sanitizer Zbigniew Kempczyński
2020-11-20 13:28 ` [igt-dev] ✗ Fi.CI.BAT: failure for Introduce IGT allocator (rev11) Patchwork
2020-11-20 17:34   ` Zbigniew Kempczyński

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=20201120113131.67971-9-zbigniew.kempczynski@intel.com \
    --to=zbigniew.kempczynski@intel.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=igt-dev@lists.freedesktop.org \
    /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