* [PATCH] gdb: LoongArch: prepend tramp frame unwinder for signal
@ 2022-03-31 12:01 Tiezhu Yang
2022-04-04 16:43 ` Tom Tromey
0 siblings, 1 reply; 4+ messages in thread
From: Tiezhu Yang @ 2022-03-31 12:01 UTC (permalink / raw)
To: gdb-patches
Implement the "init" method of struct tramp_frame to prepend tramp
frame unwinder for signal on LoongArch.
With this patch, the following failed testcases can be fixed:
FAIL: gdb.base/annota1.exp: backtrace @ signal handler (timeout)
FAIL: gdb.base/annota3.exp: backtrace @ signal handler (pattern 2)
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
---
gdb/loongarch-linux-tdep.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 50 insertions(+)
diff --git a/gdb/loongarch-linux-tdep.c b/gdb/loongarch-linux-tdep.c
index 6c040c3..7e219c4 100644
--- a/gdb/loongarch-linux-tdep.c
+++ b/gdb/loongarch-linux-tdep.c
@@ -25,6 +25,8 @@
#include "loongarch-tdep.h"
#include "solib-svr4.h"
#include "target-descriptions.h"
+#include "trad-frame.h"
+#include "tramp-frame.h"
/* Unpack an elf_gregset_t into GDB's register cache. */
@@ -117,6 +119,50 @@ const struct regset loongarch_gregset =
loongarch_fill_gregset,
};
+/* Implement the "init" method of struct tramp_frame. */
+
+#define LOONGARCH_RT_SIGFRAME_UCONTEXT_OFFSET 128
+#define LOONGARCH_UCONTEXT_SIGCONTEXT_OFFSET 176
+
+static void
+loongarch_linux_rt_sigframe_init (const struct tramp_frame *self,
+ struct frame_info *this_frame,
+ struct trad_frame_cache *this_cache,
+ CORE_ADDR func)
+{
+ struct gdbarch *gdbarch = get_frame_arch (this_frame);
+ loongarch_gdbarch_tdep *tdep = (loongarch_gdbarch_tdep *) gdbarch_tdep (gdbarch);
+ auto regs = tdep->regs;
+
+ CORE_ADDR frame_sp = get_frame_sp (this_frame);
+ CORE_ADDR sigcontext_base = (frame_sp + LOONGARCH_RT_SIGFRAME_UCONTEXT_OFFSET
+ + LOONGARCH_UCONTEXT_SIGCONTEXT_OFFSET);
+
+ trad_frame_set_reg_addr (this_cache, regs.pc, sigcontext_base);
+ for (int i = 0; i < 32; i++)
+ trad_frame_set_reg_addr (this_cache, regs.r + i, sigcontext_base + 8 + i * 8);
+
+ trad_frame_set_id (this_cache, frame_id_build (frame_sp, func));
+}
+
+/* li.w a7, __NR_rt_sigreturn */
+#define LOONGARCH_INST_LIW_A7_RT_SIGRETURN 0x03822c0b
+/* syscall 0 */
+#define LOONGARCH_INST_SYSCALL 0x002b0000
+
+static const struct tramp_frame loongarch_linux_rt_sigframe =
+{
+ SIGTRAMP_FRAME,
+ 4,
+ {
+ { LOONGARCH_INST_LIW_A7_RT_SIGRETURN, ULONGEST_MAX },
+ { LOONGARCH_INST_SYSCALL, ULONGEST_MAX },
+ { TRAMP_SENTINEL_INSN, ULONGEST_MAX }
+ },
+ loongarch_linux_rt_sigframe_init,
+ NULL
+};
+
/* Initialize LoongArch Linux ABI info. */
static void
@@ -134,8 +180,12 @@ loongarch_linux_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch)
/* GNU/Linux uses the dynamic linker included in the GNU C Library. */
set_gdbarch_skip_solib_resolver (gdbarch, glibc_skip_solib_resolver);
+
/* Enable TLS support. */
set_gdbarch_fetch_tls_load_module_address (gdbarch, svr4_fetch_objfile_link_map);
+
+ /* Prepend tramp frame unwinder for signal. */
+ tramp_frame_prepend_unwinder (gdbarch, &loongarch_linux_rt_sigframe);
}
/* Initialize LoongArch Linux target support. */
--
2.1.0
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] gdb: LoongArch: prepend tramp frame unwinder for signal
2022-03-31 12:01 [PATCH] gdb: LoongArch: prepend tramp frame unwinder for signal Tiezhu Yang
@ 2022-04-04 16:43 ` Tom Tromey
2022-04-06 3:18 ` Tiezhu Yang
0 siblings, 1 reply; 4+ messages in thread
From: Tom Tromey @ 2022-04-04 16:43 UTC (permalink / raw)
To: Tiezhu Yang; +Cc: gdb-patches
>>>>> Tiezhu Yang <yangtiezhu@loongson.cn> writes:
> Implement the "init" method of struct tramp_frame to prepend tramp
> frame unwinder for signal on LoongArch.
> With this patch, the following failed testcases can be fixed:
> FAIL: gdb.base/annota1.exp: backtrace @ signal handler (timeout)
> FAIL: gdb.base/annota3.exp: backtrace @ signal handler (pattern 2)
Thank you for the patch. This is ok, though as loongarch maintainer you
can self-approve patches that only touch this port's files.
Because LoongArch is a new port, I think it's fine if you want to apply
this to the gdb 12 branch as well.
Tom
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] gdb: LoongArch: prepend tramp frame unwinder for signal
2022-04-04 16:43 ` Tom Tromey
@ 2022-04-06 3:18 ` Tiezhu Yang
2022-04-06 14:56 ` Tom Tromey
0 siblings, 1 reply; 4+ messages in thread
From: Tiezhu Yang @ 2022-04-06 3:18 UTC (permalink / raw)
To: Tom Tromey; +Cc: gdb-patches
On 04/05/2022 12:43 AM, Tom Tromey wrote:
>>>>>> Tiezhu Yang <yangtiezhu@loongson.cn> writes:
>
>> Implement the "init" method of struct tramp_frame to prepend tramp
>> frame unwinder for signal on LoongArch.
>
>> With this patch, the following failed testcases can be fixed:
>
>> FAIL: gdb.base/annota1.exp: backtrace @ signal handler (timeout)
>> FAIL: gdb.base/annota3.exp: backtrace @ signal handler (pattern 2)
>
> Thank you for the patch. This is ok, though as loongarch maintainer you
> can self-approve patches that only touch this port's files.
Hi Tom,
Thank you very much, I will push this patch later.
If I am understanding you correctly, what you said means that
if the patches only touch files related with LoongArch, I can
push them to the master first, and then send the patches with
"COMMITTED" in the subject line to gdb-patches@sourceware.org,
right?
>
> Because LoongArch is a new port, I think it's fine if you want to apply
> this to the gdb 12 branch as well.
>
> Tom
>
I think it is enough to apply the patches to the master, thank
you.
Thanks,
Tiezhu
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] gdb: LoongArch: prepend tramp frame unwinder for signal
2022-04-06 3:18 ` Tiezhu Yang
@ 2022-04-06 14:56 ` Tom Tromey
0 siblings, 0 replies; 4+ messages in thread
From: Tom Tromey @ 2022-04-06 14:56 UTC (permalink / raw)
To: Tiezhu Yang; +Cc: Tom Tromey, gdb-patches
> If I am understanding you correctly, what you said means that
> if the patches only touch files related with LoongArch, I can
> push them to the master first, and then send the patches with
> "COMMITTED" in the subject line to gdb-patches@sourceware.org,
> right?
Yes. Normally I send the email first, then push. But it really doesn't
matter hugely since email is asynchronous.
thanks,
Tom
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2022-04-06 14:58 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-03-31 12:01 [PATCH] gdb: LoongArch: prepend tramp frame unwinder for signal Tiezhu Yang
2022-04-04 16:43 ` Tom Tromey
2022-04-06 3:18 ` Tiezhu Yang
2022-04-06 14:56 ` Tom Tromey
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 read-only IMAP folder(s) and NNTP newsgroup(s).