From: Jakub Jelinek <jakub@redhat.com>
To: Jonathan Yong <10walls@gmail.com>, Xi Ruoyao <xry111@xry111.site>,
Gcc Patch List <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] Change gcc/ira-conflicts.cc build_conflict_bit_table to use size_t/%zu
Date: Thu, 1 Feb 2024 16:25:20 +0100 [thread overview]
Message-ID: <Zbu34FTR73081bMb@tucnak> (raw)
In-Reply-To: <Zbuw9xXm85HZrgAP@tucnak>
On Thu, Feb 01, 2024 at 03:55:51PM +0100, Jakub Jelinek wrote:
> No, besides the formatting being incorrect both in ChangeLog and in the
> patch, this pessimizes ILP32 hosts unnecessarily.
So like this instead?
2024-02-01 Jakub Jelinek <jakub@redhat.com>
* hwint.h (GCC_PRISZ, fmt_size_t, HOST_SIZE_T_PRINT_DEC,
HOST_SIZE_T_PRINT_UNSIGNED, HOST_SIZE_T_PRINT_HEX,
HOST_SIZE_T_PRINT_HEX_PURE): Define.
* ira-conflicts.cc (build_conflict_bit_table): Use it. Formatting
fixes.
--- gcc/hwint.h.jj 2024-01-03 11:51:32.676715299 +0100
+++ gcc/hwint.h 2024-02-01 16:22:53.037013522 +0100
@@ -115,6 +115,27 @@ typedef HOST_WIDE_INT __gcc_host_wide_in
#define HOST_WIDE_INT_PRINT_DOUBLE_HEX "0x%" PRIx64 "%016" PRIx64
#define HOST_WIDE_INT_PRINT_PADDED_HEX "%016" PRIx64
+/* Similarly format modifier for printing size_t. As not all hosts support
+ z modifier in printf, use GCC_PRISZ and cast argument to fmt_size_t.
+ So, instead of doing fprintf ("%zu\n", sizeof (x) * y); use
+ fprintf (HOST_SIZE_T_PRINT_UNSIGNED "\n",
+ (fmt_size_t) (sizeof (x) * y)); */
+#if SIZE_MAX <= INT_MAX
+# define GCC_PRISZ ""
+# define fmt_size_t unsigned int
+#elif SIZE_MAX <= LONG_MAX
+# define GCC_PRISZ HOST_LONG_FORMAT
+# define fmt_size_t unsigned long int
+#else
+# define GCC_PRISZ HOST_LONG_LONG_FORMAT
+# define fmt_size_t unsigned long long int
+#endif
+
+#define HOST_SIZE_T_PRINT_DEC "%" GCC_PRISZ "d"
+#define HOST_SIZE_T_PRINT_UNSIGNED "%" GCC_PRISZ "u"
+#define HOST_SIZE_T_PRINT_HEX "%#" GCC_PRISZ "x"
+#define HOST_SIZE_T_PRINT_HEX_PURE "%" GCC_PRISZ "x"
+
/* Define HOST_WIDEST_FAST_INT to the widest integer type supported
efficiently in hardware. (That is, the widest integer type that fits
in a hardware register.) Normally this is "long" but on some hosts it
--- gcc/ira-conflicts.cc.jj 2024-01-03 11:51:31.748728178 +0100
+++ gcc/ira-conflicts.cc 2024-02-01 16:12:02.156137005 +0100
@@ -115,10 +115,10 @@ build_conflict_bit_table (void)
> (uint64_t) param_ira_max_conflict_table_size * 1024 * 1024)
{
if (internal_flag_ira_verbose > 0 && ira_dump_file != NULL)
- fprintf
- (ira_dump_file,
- "+++Conflict table will be too big(>%dMB) -- don't use it\n",
- param_ira_max_conflict_table_size);
+ fprintf (ira_dump_file,
+ "+++Conflict table will be too big(>%dMB) "
+ "-- don't use it\n",
+ param_ira_max_conflict_table_size);
return false;
}
}
@@ -148,11 +148,13 @@ build_conflict_bit_table (void)
object_set_words = (ira_objects_num + IRA_INT_BITS - 1) / IRA_INT_BITS;
if (internal_flag_ira_verbose > 0 && ira_dump_file != NULL)
- fprintf
- (ira_dump_file,
- "+++Allocating %ld bytes for conflict table (uncompressed size %ld)\n",
- (long) allocated_words_num * sizeof (IRA_INT_TYPE),
- (long) object_set_words * ira_objects_num * sizeof (IRA_INT_TYPE));
+ fprintf (ira_dump_file,
+ "+++Allocating " HOST_SIZE_T_PRINT_UNSIGNED
+ " bytes for conflict table (uncompressed size "
+ HOST_SIZE_T_PRINT_UNSIGNED ")\n",
+ (fmt_size_t) (sizeof (IRA_INT_TYPE) * allocated_words_num),
+ (fmt_size_t) (sizeof (IRA_INT_TYPE) * object_set_words
+ * ira_objects_num));
objects_live = sparseset_alloc (ira_objects_num);
for (i = 0; i < ira_max_point; i++)
Jakub
next prev parent reply other threads:[~2024-02-01 15:25 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-01 12:45 Jonathan Yong
2024-02-01 13:01 ` Jakub Jelinek
2024-02-01 13:06 ` Xi Ruoyao
2024-02-01 13:42 ` Jonathan Yong
2024-02-01 13:55 ` Jakub Jelinek
2024-02-01 14:33 ` Xi Ruoyao
2024-02-01 14:53 ` Jonathan Yong
2024-02-01 14:55 ` Jakub Jelinek
2024-02-01 15:25 ` Jakub Jelinek [this message]
2024-02-01 15:33 ` Jonathan Yong
2024-02-02 23:43 ` Jonathan Yong
2024-02-03 0:03 ` Jakub Jelinek
2024-02-09 10:33 ` Richard Biener
2024-02-01 13:13 ` Jakub Jelinek
2024-02-01 14:53 ` Jakub Jelinek
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=Zbu34FTR73081bMb@tucnak \
--to=jakub@redhat.com \
--cc=10walls@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=xry111@xry111.site \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).