* [PATCH] Fix rtx_code_size static initialization order fiasco
@ 2018-10-25 14:13 Ilya Leoshkevich
2018-10-25 14:14 ` Richard Biener
2018-10-25 14:27 ` Richard Earnshaw (lists)
0 siblings, 2 replies; 3+ messages in thread
From: Ilya Leoshkevich @ 2018-10-25 14:13 UTC (permalink / raw)
To: gcc-patches
Cc: krebbel, rdapp, rsandifo, Richard.Earnshaw, rguenther, Ilya Leoshkevich
Bootstrapped and regtested on s390x-redhat-linux.
r264556 and r264537 changed the format of EQ_ATTR_ALT RTXs to "ww",
which also required adjusting rtx_code_size initializer. In order to
simplify things, the list of rtx_codes known to use HOST_WIDE_INTs was
replaced by the format string check. However, unlike the old one, this
new check cannot be always performed at compile time, in which case a
static constructor is generated. This may lead to a static
initialization order fiasco with respect to other static constructors
in the compiler, in case of PR87747, cselib's pool_allocator.
gcc/ChangeLog:
2018-10-25 Ilya Leoshkevich <iii@linux.ibm.com>
PR bootstrap/87747
* rtl.c (RTX_CODE_HWINT_P_1): New helper macro.
(RTX_CODE_HWINT_P): New macro.
(rtx_code_size): Use RTX_CODE_HWINT_P ().
---
gcc/rtl.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/gcc/rtl.c b/gcc/rtl.c
index ca5c25c422f..86a40b11cf4 100644
--- a/gcc/rtl.c
+++ b/gcc/rtl.c
@@ -106,11 +106,23 @@ const enum rtx_class rtx_class[NUM_RTX_CODE] = {
#undef DEF_RTL_EXPR
};
+/* Whether rtxs with the given code code store data in the hwint field. */
+
+#define RTX_CODE_HWINT_P_1(ENUM) \
+ ((ENUM) == CONST_INT || (ENUM) == CONST_DOUBLE \
+ || (ENUM) == CONST_FIXED || (ENUM) == CONST_WIDE_INT)
+#ifdef GENERATOR_FILE
+#define RTX_CODE_HWINT_P(ENUM) \
+ (RTX_CODE_HWINT_P_1 (ENUM) || (ENUM) == EQ_ATTR_ALT)
+#else
+#define RTX_CODE_HWINT_P RTX_CODE_HWINT_P_1
+#endif
+
/* Indexed by rtx code, gives the size of the rtx in bytes. */
const unsigned char rtx_code_size[NUM_RTX_CODE] = {
#define DEF_RTL_EXPR(ENUM, NAME, FORMAT, CLASS) \
- ((FORMAT)[0] == 'w' \
+ (RTX_CODE_HWINT_P (ENUM) \
? RTX_HDR_SIZE + (sizeof FORMAT - 1) * sizeof (HOST_WIDE_INT) \
: (ENUM) == REG \
? RTX_HDR_SIZE + sizeof (reg_info) \
--
2.19.0
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] Fix rtx_code_size static initialization order fiasco
2018-10-25 14:13 [PATCH] Fix rtx_code_size static initialization order fiasco Ilya Leoshkevich
@ 2018-10-25 14:14 ` Richard Biener
2018-10-25 14:27 ` Richard Earnshaw (lists)
1 sibling, 0 replies; 3+ messages in thread
From: Richard Biener @ 2018-10-25 14:14 UTC (permalink / raw)
To: Ilya Leoshkevich; +Cc: gcc-patches, krebbel, rdapp, rsandifo, Richard.Earnshaw
On Thu, 25 Oct 2018, Ilya Leoshkevich wrote:
> Bootstrapped and regtested on s390x-redhat-linux.
>
> r264556 and r264537 changed the format of EQ_ATTR_ALT RTXs to "ww",
> which also required adjusting rtx_code_size initializer. In order to
> simplify things, the list of rtx_codes known to use HOST_WIDE_INTs was
> replaced by the format string check. However, unlike the old one, this
> new check cannot be always performed at compile time, in which case a
> static constructor is generated. This may lead to a static
> initialization order fiasco with respect to other static constructors
> in the compiler, in case of PR87747, cselib's pool_allocator.
OK.
Thanks,
Richard.
> gcc/ChangeLog:
>
> 2018-10-25 Ilya Leoshkevich <iii@linux.ibm.com>
>
> PR bootstrap/87747
> * rtl.c (RTX_CODE_HWINT_P_1): New helper macro.
> (RTX_CODE_HWINT_P): New macro.
> (rtx_code_size): Use RTX_CODE_HWINT_P ().
> ---
> gcc/rtl.c | 14 +++++++++++++-
> 1 file changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/gcc/rtl.c b/gcc/rtl.c
> index ca5c25c422f..86a40b11cf4 100644
> --- a/gcc/rtl.c
> +++ b/gcc/rtl.c
> @@ -106,11 +106,23 @@ const enum rtx_class rtx_class[NUM_RTX_CODE] = {
> #undef DEF_RTL_EXPR
> };
>
> +/* Whether rtxs with the given code code store data in the hwint field. */
> +
> +#define RTX_CODE_HWINT_P_1(ENUM) \
> + ((ENUM) == CONST_INT || (ENUM) == CONST_DOUBLE \
> + || (ENUM) == CONST_FIXED || (ENUM) == CONST_WIDE_INT)
> +#ifdef GENERATOR_FILE
> +#define RTX_CODE_HWINT_P(ENUM) \
> + (RTX_CODE_HWINT_P_1 (ENUM) || (ENUM) == EQ_ATTR_ALT)
> +#else
> +#define RTX_CODE_HWINT_P RTX_CODE_HWINT_P_1
> +#endif
> +
> /* Indexed by rtx code, gives the size of the rtx in bytes. */
>
> const unsigned char rtx_code_size[NUM_RTX_CODE] = {
> #define DEF_RTL_EXPR(ENUM, NAME, FORMAT, CLASS) \
> - ((FORMAT)[0] == 'w' \
> + (RTX_CODE_HWINT_P (ENUM) \
> ? RTX_HDR_SIZE + (sizeof FORMAT - 1) * sizeof (HOST_WIDE_INT) \
> : (ENUM) == REG \
> ? RTX_HDR_SIZE + sizeof (reg_info) \
>
--
Richard Biener <rguenther@suse.de>
SUSE LINUX GmbH, GF: Felix Imendoerffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nuernberg)
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] Fix rtx_code_size static initialization order fiasco
2018-10-25 14:13 [PATCH] Fix rtx_code_size static initialization order fiasco Ilya Leoshkevich
2018-10-25 14:14 ` Richard Biener
@ 2018-10-25 14:27 ` Richard Earnshaw (lists)
1 sibling, 0 replies; 3+ messages in thread
From: Richard Earnshaw (lists) @ 2018-10-25 14:27 UTC (permalink / raw)
To: Ilya Leoshkevich, gcc-patches; +Cc: krebbel, rdapp, rsandifo, rguenther
On 25/10/2018 14:29, Ilya Leoshkevich wrote:
> Bootstrapped and regtested on s390x-redhat-linux.
>
> r264556 and r264537 changed the format of EQ_ATTR_ALT RTXs to "ww",
> which also required adjusting rtx_code_size initializer. In order to
> simplify things, the list of rtx_codes known to use HOST_WIDE_INTs was
> replaced by the format string check. However, unlike the old one, this
> new check cannot be always performed at compile time, in which case a
> static constructor is generated. This may lead to a static
> initialization order fiasco with respect to other static constructors
> in the compiler, in case of PR87747, cselib's pool_allocator.
>
> gcc/ChangeLog:
>
> 2018-10-25 Ilya Leoshkevich <iii@linux.ibm.com>
>
> PR bootstrap/87747
> * rtl.c (RTX_CODE_HWINT_P_1): New helper macro.
> (RTX_CODE_HWINT_P): New macro.
> (rtx_code_size): Use RTX_CODE_HWINT_P ().
> ---
> gcc/rtl.c | 14 +++++++++++++-
> 1 file changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/gcc/rtl.c b/gcc/rtl.c
> index ca5c25c422f..86a40b11cf4 100644
> --- a/gcc/rtl.c
> +++ b/gcc/rtl.c
> @@ -106,11 +106,23 @@ const enum rtx_class rtx_class[NUM_RTX_CODE] = {
> #undef DEF_RTL_EXPR
> };
>
> +/* Whether rtxs with the given code code store data in the hwint field. */
> +
> +#define RTX_CODE_HWINT_P_1(ENUM) \
> + ((ENUM) == CONST_INT || (ENUM) == CONST_DOUBLE \
> + || (ENUM) == CONST_FIXED || (ENUM) == CONST_WIDE_INT)
> +#ifdef GENERATOR_FILE
> +#define RTX_CODE_HWINT_P(ENUM) \
> + (RTX_CODE_HWINT_P_1 (ENUM) || (ENUM) == EQ_ATTR_ALT)
> +#else
> +#define RTX_CODE_HWINT_P RTX_CODE_HWINT_P_1
> +#endif
> +
> /* Indexed by rtx code, gives the size of the rtx in bytes. */
>
> const unsigned char rtx_code_size[NUM_RTX_CODE] = {
> #define DEF_RTL_EXPR(ENUM, NAME, FORMAT, CLASS) \
> - ((FORMAT)[0] == 'w' \
> + (RTX_CODE_HWINT_P (ENUM) \
> ? RTX_HDR_SIZE + (sizeof FORMAT - 1) * sizeof (HOST_WIDE_INT) \
> : (ENUM) == REG \
> ? RTX_HDR_SIZE + sizeof (reg_info) \
>
I can confirm that with this patch we now get past the previous point of
failure on the affected machine. Thanks for turning this around so quickly.
R.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-10-25 13:58 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-10-25 14:13 [PATCH] Fix rtx_code_size static initialization order fiasco Ilya Leoshkevich
2018-10-25 14:14 ` Richard Biener
2018-10-25 14:27 ` Richard Earnshaw (lists)
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).