All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Alex Bennée" <alex.bennee@linaro.org>
To: Nicolas Cavallari <nicolas.cavallari@green-communications.fr>
Cc: Heiko Thiery <heiko.thiery@gmail.com>,
	 buildroot@buildroot.org,  Julien Olivain <ju.o@free.fr>,
	 Eric Le Bihan <eric.le.bihan.dev@free.fr>
Subject: Re: [Buildroot] [PATCH v6 4/5] package/vkmark: add vkmark benchmarking tool
Date: Wed, 11 Jun 2025 17:39:41 +0100	[thread overview]
Message-ID: <87tt4mi65u.fsf@draig.linaro.org> (raw)
In-Reply-To: <274904fe-4703-4571-88db-d2f3e1539d4e@green-communications.fr> (Nicolas Cavallari's message of "Wed, 11 Jun 2025 17:39:06 +0200")

Nicolas Cavallari <nicolas.cavallari@green-communications.fr> writes:

> On 11/06/2025 15:55, Alex Bennée wrote:
>> Heiko Thiery <heiko.thiery@gmail.com> writes:
>> 
>>> Hi Alex,
>>>
>>> Am Mi., 11. Juni 2025 um 13:16 Uhr schrieb Alex Bennée <alex.bennee@linaro.org>:
>>>
>>>   We build from a recent HEAD to ensure we have the fixes to better
>>>   handle the missing KHR_display extension.
>>>
>>>   Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>>>
>>> I tried to build vkmark against, now with wayland enabled and I see this:
>>>
>>> Program /usr/bin/wayland-scanner found: NO
>>>
>>> ../out-weston/build/vkmark-36e7d9b2ecf723e876add65534e95f55ec1bc79d/src/meson.build:116:22: ERROR: Program
>>> '/usr/bin/wayland-scanner' not found or not executable
>> Hmm this seems odd. At least wayland-scanner has a pkgconf variable:
>>    🕙14:51:21 alex@draig:buildroot.git/builds/x86_64  on 
>> add-virtio-vulkan-vkmark:master [$!?⇡]
>>    ➜  pkg-config --variable=wayland_scanner  wayland-scanner
>>    /usr/bin/wayland-scanner
>>    🕙14:52:16 alex@draig:buildroot.git/builds/x86_64  on  add-virtio-vulkan-vkmark:master [$!?⇡]
>>    ➜  ./host/bin/pkgconf --variable=wayland_scanner  wayland-scanner
>>    /home/alex/lsrc/tests/buildroot.git/builds/x86_64/host/bin/wayland-scanner
>>    🕙14:52:29 alex@draig:buildroot.git/builds/x86_64  on  add-virtio-vulkan-vkmark:master [$!?⇡]
>>    ➜
>> And the meson should find it with the normal dependency logic:
>>      wayland_scanner =
>> find_program(wayland_scanner_dep.get_variable('wayland_scanner'))
>
> wayland_scanner_dep is defined as:
>
> wayland_scanner_dep = dependency('wayland-scanner', required :
> get_option('wayland') == 'true')
>
> So it is searching for a wayland-scanner compiled for the target, not
> for one on the host. It should probably changed to
>
> wayland_scanner_dep = dependency('wayland-scanner', required :
> get_option('wayland') == 'true', native: true)

That works:

Build-time dependency wayland-scanner found: YES 1.23.1                                                                                                                       
Run-time dependency libdrm found: YES 2.4.124                                                                                                                                 
Run-time dependency gbm found: YES 25.1.3                                                                                                                                     
Program python3 found: YES (/home/alex/lsrc/tests/buildroot.git/builds/arm64/host/bin/python3)                                                                                
Program /home/alex/lsrc/tests/buildroot.git/builds/arm64/host/bin/wayland-scanner found: YES (/home/alex/lsrc/tests/buildroot.git/builds/arm64/host/bin/wayland-scanner)    


--8<---------------cut here---------------start------------->8---
modified   package/vkmark/0002-meson-fix-cross-compilation-50-75.patch
@@ -1,4 +1,4 @@
-From dc9a0c2d8687501ddf97c97599fa7a69d07d3d3d Mon Sep 17 00:00:00 2001
+From 0cb1a627ccb2da3c0cfd287e407c3e85fb750c94 Mon Sep 17 00:00:00 2001
 From: =?UTF-8?q?Alex=20Benn=C3=A9e?= <alex.bennee@linaro.org>
 Date: Fri, 6 Jun 2025 17:48:40 +0100
 Subject: [PATCH 2/2] meson: fix cross-compilation (#50, #75)
@@ -6,15 +6,33 @@ MIME-Version: 1.0
 Content-Type: text/plain; charset=UTF-8
 Content-Transfer-Encoding: 8bit
 
-Now we have updated the minimum meson version we can apply sys_root to
-the prefix variable to ensure we can find the right vk.xml when
-cross-compiling.
+Currently there are a few problems getting in the way of cross
+compilation:
+
+  - we need to apply sys_root to the prefix variable to find the right vk.xml
+  - we need to force the native wayland-scanner to generate headers
 
 Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
 ---
+ meson.build     | 4 +++-
  src/meson.build | 4 +++-
- 1 file changed, 3 insertions(+), 1 deletion(-)
+ 2 files changed, 6 insertions(+), 2 deletions(-)
 
+diff --git a/meson.build b/meson.build
+index 2d08d53..6cb56f5 100644
+--- a/meson.build
++++ b/meson.build
+@@ -29,7 +29,9 @@ xcb_icccm_dep = dependency('xcb-icccm', required : get_option('xcb') == 'true')
+ wayland_client_dep = dependency('wayland-client', required : get_option('wayland') == 'true')
+ wayland_protocols_dep = dependency('wayland-protocols', version : '>= 1.12',
+                                    required : get_option('wayland') == 'true')
+-wayland_scanner_dep = dependency('wayland-scanner', required : get_option('wayland') == 'true')
++wayland_scanner_dep = dependency('wayland-scanner',
++                                 required : get_option('wayland') == 'true',
++                                 native: true)
+ libdrm_dep = dependency('libdrm', required : get_option('kms') == 'true')
+ gbm_dep = dependency('gbm', required : get_option('kms') == 'true')
+ 
 diff --git a/src/meson.build b/src/meson.build
 index a664ddc..b0e8c4a 100644
 --- a/src/meson.build
--8<---------------cut here---------------end--------------->8---



-- 
Alex Bennée
Virtualisation Tech Lead @ Linaro
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

  reply	other threads:[~2025-06-11 16:39 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-11 11:16 [Buildroot] [PATCH v6 0/5] Add vkmark graphics benchmarking tool Alex Bennée
2025-06-11 11:16 ` [Buildroot] [PATCH v6 1/5] package/{mesa3d, mesa3d-headers}: bump version to 25.1.3 Alex Bennée
2025-06-11 11:16 ` [Buildroot] [PATCH v6 2/5] package/vulkan-sdk: new package Alex Bennée
2025-06-14 10:51   ` Julien Olivain
2025-06-11 11:16 ` [Buildroot] [PATCH v6 3/5] package/vulkan-[headers|loader|tools]: bump to 1.4.317 release Alex Bennée
2025-06-14 11:11   ` Julien Olivain
2025-06-11 11:16 ` [Buildroot] [PATCH v6 4/5] package/vkmark: add vkmark benchmarking tool Alex Bennée
2025-06-11 13:11   ` Heiko Thiery
2025-06-11 13:55     ` Alex Bennée
2025-06-11 13:57       ` Heiko Thiery
2025-06-11 15:39       ` Nicolas Cavallari
2025-06-11 16:39         ` Alex Bennée [this message]
2025-06-14 10:59   ` Julien Olivain
2025-06-11 11:16 ` [Buildroot] [PATCH v6 5/5] support/testing: add test for vkmark package Alex Bennée
2025-06-14 11:05   ` Julien Olivain
     [not found] ` <20250611111640.702572-2-alex.bennee__11620.3121536941$1749640635$gmane$org@linaro.org>
2025-06-11 21:38   ` [Buildroot] [PATCH v6 1/5] package/{mesa3d, mesa3d-headers}: bump version to 25.1.3 Bernd Kuhls
2025-06-12  9:42     ` Alex Bennée

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=87tt4mi65u.fsf@draig.linaro.org \
    --to=alex.bennee@linaro.org \
    --cc=buildroot@buildroot.org \
    --cc=eric.le.bihan.dev@free.fr \
    --cc=heiko.thiery@gmail.com \
    --cc=ju.o@free.fr \
    --cc=nicolas.cavallari@green-communications.fr \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.