From: Ahmad Fatoum <a.fatoum@barebox.org>
To: barebox@lists.infradead.org
Cc: Ahmad Fatoum <a.fatoum@barebox.org>
Subject: [PATCH 5/9] lib: term: fix term_getsize cursor restore
Date: Fri, 1 May 2026 08:53:59 +0200 [thread overview]
Message-ID: <20260501070625.952091-6-a.fatoum@barebox.org> (raw)
In-Reply-To: <20260501070625.952091-1-a.fatoum@barebox.org>
term_getsize() sends \e7 (DECSC) per-cdev to Save Cursor position, but
fails to use \e8 (DECRC) to Restore Cursor on each console afterwards.
Instead, it attempts to restore with term_setpos(0, 0) which uses printf()
internally and thus hits all stdout consoles, including ones that were
never queried like fbconsole.
Fix this by making cursor saving/restoring per-console by means of
DECSC/DECRC and thereby removing the term_setpos(0, 0) call entirely.
Both existent callers are safe without the term_setpos(0, 0):
- edit.c immediately clears the screen with \e[2J and positions its own
cursor
- watch.c sends \e[H\e[J (home + clear) before calling term_getsize
Also send DECRC on parse failure to restore cursor even when the terminal
didn't respond to the position query.
Signed-off-by: Ahmad Fatoum <a.fatoum@barebox.org>
---
lib/term.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/term.c b/lib/term.c
index 3949aeb45136..0a656d79e725 100644
--- a/lib/term.c
+++ b/lib/term.c
@@ -76,6 +76,7 @@ int term_cdev_get_size(struct console_device *cdev,
int *screenwidth, int *screenheight)
{
const char esc[] = "\e7" "\e[r" "\e[999;999H" "\e[6n";
+ const char restore[] = "\e8";
int ret, params[2];
if (!(cdev->f_active & CONSOLE_STDIN))
@@ -88,6 +89,9 @@ int term_cdev_get_size(struct console_device *cdev,
console_puts(cdev, esc);
ret = term_cdev_read_reply(cdev, params, 2, 'R');
+
+ console_puts(cdev, restore);
+
if (ret)
return ret;
@@ -113,8 +117,6 @@ int term_getsize(int *screenwidth, int *screenheight)
found = true;
}
- term_setpos(0, 0);
-
if (!found)
return -ENOENT;
--
2.47.3
next prev parent reply other threads:[~2026-05-01 7:07 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-01 6:53 [PATCH 0/9] lib: term: fix fbconsole cursor desynchronization Ahmad Fatoum
2026-05-01 6:53 ` [PATCH 1/9] lib: term: avoid printing NUL with new new console_puts API Ahmad Fatoum
2026-05-01 6:53 ` [PATCH 2/9] lib: term: return error code from term_getsize() Ahmad Fatoum
2026-05-01 6:53 ` [PATCH 3/9] lib: term: add per-console terminal response parser Ahmad Fatoum
2026-05-01 6:53 ` [PATCH 4/9] lib: term: factor out single cdev handling from term_getsize Ahmad Fatoum
2026-05-01 6:53 ` Ahmad Fatoum [this message]
2026-05-01 6:54 ` [PATCH 6/9] console: add get_size callback for direct size reporting Ahmad Fatoum
2026-05-01 6:54 ` [PATCH 7/9] video: fbconsole: implement get_size Ahmad Fatoum
2026-05-01 6:54 ` [PATCH 8/9] console: add per-console terminal.size parameter Ahmad Fatoum
2026-05-01 7:19 ` [PATCH 9/9] efi: reimplement query_console_serial using term_getsize Ahmad Fatoum
2026-05-07 10:38 ` [PATCH 0/9] lib: term: fix fbconsole cursor desynchronization Sascha Hauer
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=20260501070625.952091-6-a.fatoum@barebox.org \
--to=a.fatoum@barebox.org \
--cc=barebox@lists.infradead.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