From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: alex.bennee@linaro.org, laurent@vivier.eu
Subject: [PATCH v2 11/12] linux-user/sparc: Fix cpu_clone_regs_*
Date: Fri, 25 Oct 2019 07:39:20 -0400 [thread overview]
Message-ID: <20191025113921.9412-12-richard.henderson@linaro.org> (raw)
In-Reply-To: <20191025113921.9412-1-richard.henderson@linaro.org>
We failed to set the secondary return value in %o1
we failed to advance the PC past the syscall,
we failed to adjust regwptr into the new structure,
we stored the stack pointer into the wrong register.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
v2: Take CLONE_VM into account when performing syscall return.
---
linux-user/sparc/target_cpu.h | 47 ++++++++++++++++++++++++++++-------
1 file changed, 38 insertions(+), 9 deletions(-)
diff --git a/linux-user/sparc/target_cpu.h b/linux-user/sparc/target_cpu.h
index 8ff706adce..14b2158969 100644
--- a/linux-user/sparc/target_cpu.h
+++ b/linux-user/sparc/target_cpu.h
@@ -23,22 +23,51 @@
static inline void cpu_clone_regs_child(CPUSPARCState *env, target_ulong newsp,
unsigned flags)
{
- if (newsp) {
- env->regwptr[22] = newsp;
- }
- /* syscall return for clone child: 0, and clear CF since
- * this counts as a success return value.
+ /*
+ * After cpu_copy, env->regwptr is pointing into the old env.
+ * Update the new cpu to use its own register window.
*/
- env->regwptr[0] = 0;
-#if defined(TARGET_SPARC64) && !defined(TARGET_ABI32)
- env->xcc &= ~PSR_CARRY;
+ env->regwptr = env->regbase + (env->cwp * 16);
+
+ if (newsp) {
+ /* When changing stacks, do it with clean register windows. */
+#ifdef TARGET_SPARC64
+ env->cansave = env->nwindows - 2;
+ env->cleanwin = env->nwindows - 2;
+ env->canrestore = 0;
#else
- env->psr &= ~PSR_CARRY;
+ env->wim = 1 << env->cwp;
#endif
+ /* ??? The kernel appears to copy one stack frame to the new stack. */
+ /* ??? The kernel force aligns the new stack. */
+ env->regwptr[WREG_SP] = newsp;
+ }
+
+ if (flags & CLONE_VM) {
+ /*
+ * Syscall return for clone child: %o0 = 0 and clear CF since this
+ * counts as a success return value. Advance the PC past the syscall.
+ * For fork child, all of this happens in cpu_loop, and we must not
+ * do the pc advance twice.
+ */
+ env->regwptr[WREG_O0] = 0;
+#if defined(TARGET_SPARC64) && !defined(TARGET_ABI32)
+ env->xcc &= ~PSR_CARRY;
+#else
+ env->psr &= ~PSR_CARRY;
+#endif
+ env->pc = env->npc;
+ env->npc = env->npc + 4;
+ }
+
+ /* Set the second return value for the child: %o1 = 1. */
+ env->regwptr[WREG_O1] = 1;
}
static inline void cpu_clone_regs_parent(CPUSPARCState *env, unsigned flags)
{
+ /* Set the second return value for the parent: %o1 = 0. */
+ env->regwptr[WREG_O1] = 0;
}
static inline void cpu_set_tls(CPUSPARCState *env, target_ulong newtls)
--
2.17.1
next prev parent reply other threads:[~2019-10-25 11:48 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-25 11:39 [PATCH v2 00/12] linux-user sparc fixes Richard Henderson
2019-10-25 11:39 ` [PATCH v2 01/12] scripts/qemu-binfmt-conf: Update for sparc64 Richard Henderson
2019-11-05 9:43 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 02/12] tests/tcg/multiarch/linux-test: Fix error check for shmat Richard Henderson
2019-10-25 12:59 ` Philippe Mathieu-Daudé
2019-10-25 13:00 ` Philippe Mathieu-Daudé
2019-11-05 9:45 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 03/12] target/sparc: Define an enumeration for accessing env->regwptr Richard Henderson
2019-11-05 9:45 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 04/12] linux-user/sparc: Use WREG constants in sparc/target_cpu.h Richard Henderson
2019-10-25 12:09 ` Laurent Vivier
2019-11-05 9:49 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 05/12] linux-user/sparc: Begin using WREG constants in sparc/signal.c Richard Henderson
2019-10-25 12:16 ` Laurent Vivier
2019-11-05 9:49 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 06/12] linux-user/sparc: Use WREG_SP constant " Richard Henderson
2019-10-25 12:19 ` Laurent Vivier
2019-11-05 9:50 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 07/12] linux-user/sparc: Fix WREG usage in setup_frame Richard Henderson
2019-11-05 9:50 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 08/12] linux-user/sparc64: Fix target_signal_frame Richard Henderson
2019-10-25 12:47 ` Laurent Vivier
2019-10-25 13:38 ` Richard Henderson
2019-10-25 13:43 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 09/12] linux-user: Rename cpu_clone_regs to cpu_clone_regs_child Richard Henderson
2019-10-25 13:03 ` Philippe Mathieu-Daudé
2019-11-05 9:52 ` Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 10/12] linux-user: Introduce cpu_clone_regs_parent Richard Henderson
2019-10-25 13:04 ` Philippe Mathieu-Daudé
2019-11-05 9:52 ` Laurent Vivier
2019-10-25 11:39 ` Richard Henderson [this message]
2019-11-05 9:54 ` [PATCH v2 11/12] linux-user/sparc: Fix cpu_clone_regs_* Laurent Vivier
2019-10-25 11:39 ` [PATCH v2 12/12] linux-user/alpha: Set r20 secondary return value Richard Henderson
2019-10-25 12:56 ` Laurent Vivier
2019-11-05 9:53 ` Laurent Vivier
2019-10-25 12:50 ` [PATCH v2 00/12] linux-user sparc fixes Laurent Vivier
2019-10-26 7:02 ` no-reply
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=20191025113921.9412-12-richard.henderson@linaro.org \
--to=richard.henderson@linaro.org \
--cc=alex.bennee@linaro.org \
--cc=laurent@vivier.eu \
--cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).