qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Phil Dennis-Jordan <phil@philjordan.eu>
To: qemu-devel@nongnu.org, pbonzini@redhat.com, agraf@csgraf.de,
	graf@amazon.com, marcandre.lureau@redhat.com,
	berrange@redhat.com, thuth@redhat.com, philmd@linaro.org,
	peter.maydell@linaro.org, akihiko.odaki@daynix.com,
	phil@philjordan.eu, lists@philjordan.eu
Subject: [PATCH 23/26] hw/display/apple-gfx: Host GPU picking improvements
Date: Mon, 15 Jul 2024 23:07:02 +0200	[thread overview]
Message-ID: <20240715210705.32365-24-phil@philjordan.eu> (raw)
In-Reply-To: <20240715210705.32365-1-phil@philjordan.eu>

During startup of the PV graphics device, we need to specify the
host GPU to use for PV acceleration of the guest's graphics
operations.

On a host system, this is trivial: pick the only one. The
MTLCreateSystemDefaultDevice() function will do the right thing
in this case.

It gets a little more complicated on systems with more than one
GPU; first and foremost, this applies to x86-64 MacBook Pros
with 15/16" displays. However, with eGPUs, in theory any x86-64
Mac can gain one or more additional GPUs. In these cases, the
default is often not ideal - usually, discrete GPUs are selected.
In my tests, performance tends to be best with iGPUs, however,
and they are usually also best in terms of energy consumption.

Ideally, we will want to allow the user to manually select a GPU
if they so choose. In this patch, I am interested in picking a
sensible default. Instead of the built-in default logic, it is
now:

 1. Select a GPU with unified memory (iGPU)
 2. If (1) fails, select a GPU that is built-in, not an eGPU.
 3. If (2) fails, fall back to system default.

Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu>
---
 hw/display/apple-gfx.m | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/hw/display/apple-gfx.m b/hw/display/apple-gfx.m
index cbcbaf0006..6c92f2579b 100644
--- a/hw/display/apple-gfx.m
+++ b/hw/display/apple-gfx.m
@@ -502,6 +502,32 @@ static void apple_gfx_register_task_mapping_handlers(AppleGFXState *s,
     return mode_array;
 }
 
+static id<MTLDevice> copy_suitable_metal_device(void)
+{
+    id<MTLDevice> dev = nil;
+    NSArray<id<MTLDevice>> *devs = MTLCopyAllDevices();
+
+    /* Prefer a unified memory GPU. Failing that, pick a non-removable GPU. */
+    for (size_t i = 0; i < devs.count; ++i) {
+        if (devs[i].hasUnifiedMemory) {
+            dev = devs[i];
+            break;
+        }
+        if (!devs[i].removable) {
+            dev = devs[i];
+        }
+    }
+
+    if (dev != nil) {
+        [dev retain];
+    } else {
+        dev = MTLCreateSystemDefaultDevice();
+    }
+    [devs release];
+
+    return dev;
+}
+
 void apple_gfx_common_realize(AppleGFXState *s, PGDeviceDescriptor *desc)
 {
     PGDisplayDescriptor *disp_desc = nil;
@@ -509,7 +535,7 @@ void apple_gfx_common_realize(AppleGFXState *s, PGDeviceDescriptor *desc)
     QTAILQ_INIT(&s->tasks);
     s->render_queue = dispatch_queue_create("apple-gfx.render",
                                             DISPATCH_QUEUE_SERIAL);
-    s->mtl = MTLCreateSystemDefaultDevice();
+    s->mtl = copy_suitable_metal_device();
     s->mtl_queue = [s->mtl newCommandQueue];
 
     desc.device = s->mtl;
-- 
2.39.3 (Apple Git-146)



  parent reply	other threads:[~2024-07-17 11:09 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-15 21:06 Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 01/26] hw/vmapple/apple-gfx: Introduce ParavirtualizedGraphics.Framework support Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 02/26] hw/vmapple/apple-gfx: BQL renaming update Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 03/26] hw/display/apple-gfx: Moved from hw/vmapple/ Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 04/26] hw/display/apple-gfx: uses DEFINE_TYPES macro Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 05/26] hw/display/apple-gfx: native -> little endian memory ops Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 06/26] hw/display/apple-gfx: Removes dead/superfluous code Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 07/26] hw/display/apple-gfx: Makes set_mode thread & memory safe Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 08/26] hw/display/apple-gfx: Adds migration blocker Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 09/26] hw/display/apple-gfx: Wraps ObjC autorelease code in pool Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 10/26] hw/display/apple-gfx: Fixes ObjC new/init misuse, plugs leaks Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 11/26] hw/display/apple-gfx: Uses ObjC category extension for private property Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 12/26] hw/display/apple-gfx: Task memory mapping cleanup Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 13/26] hw/display/apple-gfx: Defines PGTask_s struct instead of casting Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 14/26] hw/display/apple-gfx: Refactoring of realize function Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 15/26] hw/display/apple-gfx: Separates generic & vmapple-specific functionality Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 16/26] hw/display/apple-gfx: Asynchronous MMIO writes on x86-64 Phil Dennis-Jordan
2024-07-15 21:26   ` Philippe Mathieu-Daudé
2024-07-16 14:29     ` Phil Dennis-Jordan
2024-07-16 14:48       ` BALATON Zoltan
2024-07-17 11:09         ` Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 17/26] hw/display/apple-gfx: Asynchronous rendering and graphics update Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 18/26] hw/display/apple-gfx: Adds PCI implementation Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 19/26] ui/cocoa: Adds non-app runloop on main thread mode Phil Dennis-Jordan
2024-07-15 21:06 ` [PATCH 20/26] hw/display/apple-gfx: Fixes cursor hotspot handling Phil Dennis-Jordan
2024-07-15 21:07 ` [PATCH 21/26] hw/display/apple-gfx: Implements texture syncing for non-UMA GPUs Phil Dennis-Jordan
2024-07-15 21:07 ` [PATCH 22/26] hw/display/apple-gfx: Replaces magic number with queried MMIO length Phil Dennis-Jordan
2024-07-15 21:07 ` Phil Dennis-Jordan [this message]
2024-07-15 21:07 ` [PATCH 24/26] hw/display/apple-gfx: Adds configurable mode list Phil Dennis-Jordan
2024-07-15 21:07 ` [PATCH 25/26] MAINTAINERS: Add myself as maintainer for apple-gfx, reviewer for HVF Phil Dennis-Jordan
2024-07-15 21:07 ` [PATCH 26/26] hw/display/apple-gfx: Removes UI pointer support check Phil Dennis-Jordan
2024-07-16  6:07 ` Akihiko Odaki
2024-07-16  6:38   ` hw/display/apple-gfx Philippe Mathieu-Daudé
2024-07-16  6:47     ` hw/display/apple-gfx Akihiko Odaki
2024-07-17 11:12       ` hw/display/apple-gfx Phil Dennis-Jordan
2024-07-17 11:16   ` Phil Dennis-Jordan

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=20240715210705.32365-24-phil@philjordan.eu \
    --to=phil@philjordan.eu \
    --cc=agraf@csgraf.de \
    --cc=akihiko.odaki@daynix.com \
    --cc=berrange@redhat.com \
    --cc=graf@amazon.com \
    --cc=lists@philjordan.eu \
    --cc=marcandre.lureau@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=philmd@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=thuth@redhat.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;
as well as URLs for NNTP newsgroup(s).